uva-dare (digital academic repository) epidemiology and ... · revisited - plasma renin activity,...

UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Epidemiology and pathophysiology of malignant hypertension

van den Born, B.J.H.

Link to publication

Citation for published version (APA):van den Born, B. J. H. (2007). Epidemiology and pathophysiology of malignant hypertension.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 11 Sep 2020

https://dare.uva.nl/personal/pure/en/publications/epidemiology-and-pathophysiology-of-malignant-hypertension(d058a5a9-9ad0-40cd-9878-cbdb4426f607).html

Epidemiology and Pathophysiology of

Malignant Hypertension

Bert-Jan H. van den Born

Epidemiology and pathophysiology of malignant hypertension Author: van den Born, B.J.H. Cover: Dijkdoorbraak bij Papendrecht, Zuid-Holland, 1953 Printed by: Print Partners Ipskamp BV, Enschede ISBN: 978-90-9022402-2 The publication of this thesis was financially supported by Amstol Stichting, Astellas Pharma, Astra Zeneca, Bayer, Boehringer Ingelheim, Bristol-Meyers Squibb, Menarini Pharma, Merck Sharp & Dohme, Novartis Pharma, Pfizer, Sankyo Pharma, Sanofi Aventis, Solvay, and Zambon Financial support from the Amsterdam Pathophysiology of the Circulation Research Foundation is gratefully acknowledged.

Epidemiology and Pathophysiology of

Malignant Hypertension

Academisch Proefschrift

Ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam op gezag van de Rector Magnificus prof. dr. D.C. van den Boom

ten overstaan van een door het college voor promoties ingestelde commissie,

in het openbaar te verdedigen in de Agnietenkapel.

op

vrijdag 9 november 2007, te 14.00 uur

door

Bert-Jan van den Born

geboren te Ochten

Promotiecommissie Promotor Prof. dr. M.M. Levi Co-promotores Dr. G.A. van Montfrans Prof. dr. R.P. Koopmans Overige leden Prof. dr. H.R. Büller Dr. J.C.M. Meijers Dr. A.C. van den Meiracker Prof. dr. G.J. Navis Prof. dr. J. Stam Prof. dr. A.H. Zwinderman Faculteit der Geneeskunde

Contents Chapter 1. General introduction: malignant hypertension Part I. Epidemiology of malignant hypertension Chapter 2. Value of routine funduscopy in patients with hypertension - systematic review - Chapter 3. Ethnic disparities in the incidence, presentation and compli-

cations of malignant hypertension Chapter 4. Thrombotic microangiopathy and renal failure in malignant

hypertension Chapter 5. The M235T polymorphism in the angiotensinogen gene is

associated with the risk of malignant hypertension Part II. Pathophysiology of malignant hypertension Chapter 6. The renin-angiotensin system in malignant hypertension

revisited - plasma renin activity, thrombotic microangiopathy and renal failure in malignant hypertension

Chapter 7. The severity of thrombotic microangiopathy in patients with

malignant hypertension is associated with reduced activity of the von Willebrand factor cleaving protease (ADAMTS13)

Chapter 8. Impaired cerebral autoregulation in patients with malignant

hypertension

Chapter 9. Cerebral and systemic hemodynamics during acute blood pressure reduction with sodium nitroprusside or labetalol in patients with malignant hypertension

Chapter 10. Summary and discussion Chapter 11. Samenvatting en discussie Dankwoord

Voor mijn ouders

Chapter 1

Malignant Hypertension Bert-Jan H van den Born, Richard P Koopmans and Gert A van Montfrans

Chapter 1

10

Summary Malignant hypertension is a hypertensive emergency characterized by severe hypertension and ischemic retinal lesions. The complications of malignant hypertension differ from chronic uncontrolled hypertension and include hyper-tensive encephalopathy, thrombotic microangiopathy and acute renal failure. Although the incidence of malignant hypertension has declined after the introduction of anti-hypertensive agents, it can still be relatively common in multi-ethnic populations. A history of hypertension and several underlying conditions such as renal parenchymal and renovascular disease increase the risk of malignant hypertension. However, the reason why only few patients progress to the malignant phase is incompletely understood. Pressure natriuresis and activation of the renin-angiotensin system precede the development of malignant hypertension. Activation of the renin-angiotensin system may result from ischemia of the renovascular bed and volume depletion. The treatment of malignant hypertension requires an immediate but controlled reduction of blood pressure to safe, but not normal blood pressure levels. Initially a ~25% reduction in mean arterial pressure is recommended within minutes to hours. This strategy is based on the observation that the curve of cerebral autoregulation may be shifted towards the right in these patients. In the controlled reduction of blood pressure, sodium nitroprusside (SNP) is superior due to its short half-life, but its use may be hampered by the risk of cyanide toxicity and increase in intracranial pressure. Alternatively, labetalol and nicardipine can be used in patients who are at increased risk for the complications of SNP such as patients with hypertensive encephalopathy and those with renal and liver insufficiency.

Malignant hypertension

11

Towards a definition Malignant hypertension was first described in 1914 by Volhard who noted the association of severe hypertension and renal ischemia with retinopathy.1 In 1939 this finding was supported by Keith, Wagener and Barker in an observational study of 219 hypertensive patients.2 They found, in an era where treatment of hypertension was not yet possible, that the survival of hypertensive patients with bilateral optic disc edema (grade IV retinopathy) was greatly impaired compared to those having no or minimal retinal lesions. More than 90% of patients with grade IV retinopathy had died of renal, cardiac and cerebral complications after 1 year. Patients with cotton-wool patches and retinal hemorrhages (grade III) had an intermediate survival with 65% deceased after 1 year. Since, hyper-tension with grade IV retinopathy has been considered malignant hypertension. The term accelerated or premalignant hypertension was used for patients with severe hypertension and grade III retinopathy. Later, after effective anti-hypertensive therapy became available, the term accelerated hypertension was discarded because it was recognized that the clinical features and prognosis of hypertensive patients with grade III and IV retinopathy was similar and quite different from patients with no or unilateral retinopathy.3, 4 The evidence that the clinical characteristics and prognosis of patients with grade III and IV retinopathy are comparable is supported by the fact that both grades share a common etiological background. The poor renal outcome and death from cerebrovascular accidents in severely hypertensive patients with grade III and IV retinopathy suggests that these retinal changes are an indicator of imminent ischemic complications elsewhere. However, not all patients with a combination of cotton wool patches, flame shaped hemorrhages or papilledema have malignant hypertension; vice versa not all patients with hypertensive encephalopathy, a hypertensive emergency commonly associated with malignant hypertension, have retinal abnormalities.5

Currently, there is no gold standard to differentiate a true hypertensive emergency from a hypertensive urgency. A decision model, incorporating selected clinical, hemodynamic and biochemical markers may be the best way in differentiating a true hypertensive emergency from a hypertensive urgency. However, to ensure adequate comparison of patients the definition of malignant hypertension, namely severe hypertension with bilateral presence of grade III and IV hypertensive retinopathy, as still endorsed in international hypertension guidelines and the WHO is used in this thesis.

Chapter 1

12

Histopathological findings The pathological hallmark of malignant hypertension is fibrinoid necrosis and myointimal proliferation of small arteries and arterioles (up to a diameter of 0.2 mm). The fibrinoid depositions are caused by continued seepage of fibrin and serum proteins through a necrotic vessel wall into surrounding viable tissue and are sometimes accompanied by thrombosis of the arterial lumen.6, 7 The myointimal proliferation consists of smooth muscle cell hyperplasia of the subintimal layer often accompanied by narrowing of the arterial lumen.7 Fibrinoid necrosis is not invariably present in malignant hypertension,7 and can also be found in other conditions such as scleroderma, TTP-HUS and anti-phospholipid antibody syndrome.8-10 Myointimal proliferation is, to a lesser degree, also observed in severely hypertensive patients without retinal abnormalities.6, 7 In an autopsy series of 20 patients who died from hypertensive encephalopathy or malignant hypertension, the pathological findings in the retina paralleled the histopathological findings in the kidney and brain with regard to the presence of fibrinoid necrosis and subintimal hyperplasia.11 This suggests that the vascular lesions observed in malignant hypertension affects organs to a similar extent. Another autopsy study has also described fibrinoid necrosis and myointimal proliferation in the vasculature of other organs such as the pancreas, adrenal glands, intestine and liver.12

The complications of malignant hypertension differ from chronic (uncon-trolled) hypertension in the sense that retinopathy, encephalopathy and acute renal failure are not observed in the latter. Therefore the histological findings of these target organs in malignant hypertension will be discussed in more detail. Retina Most histopathological evidence on the pathogenesis of the retinal arteriolar lesions and retinopathy of malignant hypertension comes from autopsy studies in monkeys with induced hypertension by clipping the renal arteries. The retinopathy observed in these animals includes cotton wool spots, flame shaped hemorrhages, optic nerve edema and choroidopathy (figure 1). Although information on hypertensive retinopathy from autopsy studies in humans is scarce, the retinal pathology of patients dying from malignant hypertension or hypertensive encephalopathy and animals with induced hypertension seems to be identical.11, 13, 14


13

Fig.1

Retina of a patient showing flame-shaped hemorrhages, soft exudates and papilledema with macular hard exudates as a result of lipid depositions. Histologically, the precapillary retinal arteries and arterioles are unevenly affected with some arteries appearing normal and others showing extensive fibrinoid necrosis.13 Cotton wool spots are the earliest and most commonly observed retinal lesions, whereas optic nerve edema and choroidopathy appear late.14 Cotton wool spots consist of accumulations of cytoplasmic debris in the retinal nerve-fibre layer.15 These accumulations are caused by obstruction of axoplasmic transport in ganglion cell axons resulting from occluded retinal arterioles.15, 16 Retinal hemorrhages develop as a result of ischemia of small retinal vessels with subsequent loss of integrity of the vessel wall and extravasation of erythrocytes.13 These bleedings follow the direction of the retinal nerve fiber layer and therefore become visible as flame shaped hemorrhages.17 Hypertensive choroidopathy results from ischemia of the choroidal vascular bed.18 The optic disc edema observed in autopsies of monkeys with malignant hypertension appears to result from axonal hydropic swelling secondary to ischemic infarction and is, in contrast to previous beliefs,

Chapter 1

14

not caused by increased intracranial pressure.19 Thus the retinal lesions observed in malignant hypertension (cotton wool spots, flame shaped hemorrhages and optic nerve edema) all result from ischemia and therefore appear to share a common etiological background. Brain Brain autopsies have shown that fibrinoid necrosis is present in more than half of the patients who died from complications related to hypertensive encephalo-pathy and malignant hypertension.11 Focal hyalinization and fibrous thickening of the vessel wall were invariably present, sometimes accompanied by narrowing and occlusion of the vessel lumen. These abnormalities were not observed in hypertensive patients without retinal abnormalities.

In all patients with malignant hypertension, multiple zones of micro-infarctions were observed in all parts of the brain, especially in the brainstem and basal ganglia, and to a lesser extent in the cerebral cortex and spinal cord. These zones were associated with either fibrinoid necrosis, fibrin occlusion or focal irregularities of the vessel wall. Brain tissue showed pallor or necrosis of neurons. In more advanced stages these zones consisted of microglial cells and astrocytes in the presence of fibrous material while both axis cylinders and myelin had disappeared. Although increased cerebral spinal fluid pressure and cerebral edema are observed in patients who died from hypertensive encephalo-pathy related to malignant hypertension, the degree of cerebral edema (in the absence of intracerebral hemorrhage) is usually limited.11, 20 Kidney Light and electron microscopy of renal biopsy specimens of patients with malignant hypertension and renal insufficiency show marked intimal hyperplasia with hyalinization and medial hypertrophy with narrowing of the arterial lumen. Fibrinoid necrosis may be present in the afferent arterioles and distal interlobular arteries.6 Renal parenchymal lesions consist of ischemic glomerular changes as evidenced by collapsed glomerular tufts and collagen depositions in Bowman’s space and patchy necrosis of tubular cells with blood and fibrin depositions in the lumina of convoluted tubules.7, 21 Juxtaglomerular cells, responsible for renin production and secretion, show hypergranularity and have a hyperplastic appearance consistent with activation of the renin-angiotensin system in malignant hypertension.7, 22


15

Pathogenesis Malignant hypertension is considered a renin-mediated type of hypertension because of the, often marked, activation of the renin-angiotensin aldosterone system (RAAS).23-25 However, RAAS activation is not an exclusive feature of malignant hypertension as patients fulfilling the WHO criteria for malignant hypertension may have normal plasma renin activity (PRA).25-27 These variations in RAAS activation are incompletely understood. In experimental models it has been demonstrated that, apart from angiotensin II, other blood pressure stimulating factors, such as deoxycorticosterone, noradrenalin or decreased NO, may result in malignant hypertension.28-30 In humans, this is best demonstrated by patients with an aldosterone producing adenoma (Conn’s syndrome) or pheochromocytoma. In Conn’s syndrome, renin secretion is suppressed because of an increased autonomous production of aldosterone. Despite an intially suppressed renin secretion, these patients may develop malignant hypertension and profound activation of the renin-angiotensin system.31, 32 In patients with a pheochromocytoma, surgical removal of the tumor may result in normalization of blood pressure despite an initial presentation with malignant hypertension. These data suggest that a single blood pressure increasing stimulus, not directly related to the renin-angiotensin system, may lead to malignant hypertension and RAAS activation.33 Therefore, it is likely that the activation of the renin-angiotensin system in malignant hypertension is a secondary event. Possible mechanisms responsible for the (paradoxical) activation of the renin-angiotensin system may include sodium depletion and renovascular ischemia.

Experiments in animals and observations in humans have shown that malignant hypertension is preceded by sodium depletion and activation of the renin-angiotensin system. In uninephrectomized dogs, a modest blood pressure increase elicited by chronic intra-renal infusion of either norepinephrine or angiotensin II leads to sodium retention.34, 35 However, intra-renal infusion of increasing doses of norepinephrine leads to natriuresis and weight loss, followed by malignant hypertension, intravascular hemolysis and an activated RAAS.29 The natriuresis which follows a blood pressure increase has also been demonstrated in rats with a unilateral clip on the renal artery and an untouched contralateral kidney. In these animals, sodium excretion and weight loss precede the development of an activated RAAS and malignant hypertension.36 Administration of isotonic saline ameliorates the malignant course compared to rats receiving only water, likely by lowering angiotensin II.36, 37 Uninephrecto-mized rats with a clip on the renal artery can exhibit profound activation of the

Chapter 1

16

renin-angiotensin system without concomitant natriuresis.38, 39 However, the RAAS activation in these animals is not as great as in unilaterally clipped rats with an intact contralateral kidney. This is likely due to restoration of renal perfusion as a result of sodium retention, thereby aborting further stimulation of the renin-angiotensin system.

In humans, the association between RAAS activation and volume depletion is best demonstrated by the hyponatremic-hypertensive syndrome. This syndrome is mainly observed in patients with severe or malignant hypertension and renal artery stenosis. Apart from severe hypertension these patients have an absolute sodium deficiency and an activated RAAS.40 The degree of RAAS activation in these patients is correlated with the severity of hyponatremia. This is thought to be the result of an angiotensin II mediated stimulation of anti-diuretic hormone (ADH).41, 42 Although hyponatremia is not a common feature in malignant hypertension, volume depletion and increased ADH levels are frequently observed in patients with malignant hypertension.43 The mechanisms responsible for the blood pressure induced sodium excretion are incompletely understood as autoregulation of renal blood flow is well preserved by increasing renal afferent tone despite large differences in blood pressure or angiotensin II.44, 45 More recent studies however, have shown that renal medullary blood flow is less well autoregulated than renal cortical blood flow suggesting that differences in renal medullary flow may be responsible for the blood pressure induced changes in sodium excretion.45, 46

Activation of the renin-angiotensin system, either by volume depletion or renovascular ischemia or both, may lead to a further blood pressure increase by promoting renovascular ischemia, thereby stimulating further renin release from juxtaglomerular cells, inducing a vicious circle as proposed by Wilson and others.47, 48 The association between vascular damage and an angiotensin II mediated blood pressure increase has been well established in animals and humans. Transgenic rats harbouring the human renin and angiotensinogen genes develop severe hypertension and profound endothelial damage characterized by the activation of proinflammatory and procoagulant pathways.49, 50 In humans, both histopathological and angiographic evidence suggests that the degree of vascular damage is associated with the degree of renin release.7, 22, 51 The thrombotic microangiopathy (TMA) of malignant hypertension, as evidence of profound vascular damage seems associated with the degree of RAAS activation and the presence of fibrinoid necrosis in kidney biopsies.26, 52 Treatment with either an ACE-inhibitor or direct vasodilator such as sodium nitroprusside (SNP) are successful in the treatment of malignant hypertension and lead to resolution


17

of the vascular damage associated with malignant hypertension.26, 53 This indicates that ongoing activation of the renin-angiotensin system can be aborted by both RAAS dependent and RAAS-independent anti-hypertensive treatment.27,

54 Vasodilation and amelioration of vascular damage resulting in subsequent improvement of blood flow to ischemic areas in the kidney and termination of pressure natriuresis may all be responsible for cessation of the ongoing renin release from juxtaglomerular cells. The exact mechanisms however leading to both the initiation and cessation of the proposed vicious circle of malignant hypertension need further elucidation. Risk factors and causes Risk factors A history of poorly controlled or untreated hypertension is the most important risk factor of malignant hypertension. In the Malignant Hypertension Register from West Birmingham, only 29% of admitted patients had received one or more antihypertensive drugs prior to admission, whilst 45% were previously diagnosed with hypertension.55 Left ventricular hypertrophy was present on the electrocardiogram in 77% at admission suggesting that hypertension was present in more patients before they presented with malignant hypertension. Malignant hypertension is more frequently observed among patients from sub-Saharan African descent.55 This may be explained by differences in hypertension awareness and control in this population. A previous study in the US showed that blacks presenting with severe uncontrolled hypertension at the emergency department were less likely to have a primary care physician or health insurance than ethnicity matched hypertensive controls.56 However, data on ethnic differences regarding hypertension awareness and treatment in relation to severe uncontrolled or malignant hypertension are lacking. Previous studies have suggested an association with smoking and malignant hypertension, patients with malignant hypertension being more likely to smoke than hypertensive or normotensive controls irrespective of their age, sex or ethnic background.57, 58 However, differences in socioeconomic background and adherence to anti-hypertensive therapy may have confounded these associations. The association between dyslipidemia and malignant hypertension has been examined in one case control study.59 Methodological shortcomings however limit any conclusion drawn from these data. So, apart from a previous history of hypertension, definitive associations of cardiovascular risk factors with malignant hypertension are lacking. As malignant hypertension can be

Chapter 1

18

considered an extreme phenotype of renin-mediated hypertension, two studies have examined the association between malignant hypertension and poly-morphisms in the renin-angiotensin system. One study found a possible association between the ACE I/D polymorphism and the risk of malignant hypertension in white subjects compared to hypertensive controls.60 This finding however could not be reproduced in a later study, although the ACE DD genotype was more frequent in malignant hypertensive patients compared to normotensive controls.61 Both studies were hampered however by the lack of comprehensive information on the clinical characteristics and selection of patients. Therefore, definitive genetic associations with malignant hypertension risk remain to be determined. Secondary causes Basically any drug or condition that increases blood pressure can contribute to the development of malignant hypertension. Yet, there are some drugs and secondary causes that may accelerate blood pressure to such an extent that malignant hypertension follows. These underlying causes are summarized in table 1. Several retrospective studies have shown that primary renal and renovascular disease account for most of the secondary causes of malignant hypertension. However, the prevalence of these conditions varies between 11% to 67% for renal parenchymal disease and 2% to 33% for renovascular disease.55, 62-67 These differences may be explained by differences in definition, efforts to demonstrate an underlying cause, selection and referral bias, and the control of hypertension in the population at large. In populations with more vigorous blood pressure control patients with essential hypertension are more likely to be detected, treated and controlled thus preventing them from developing malignant hypertension. This may also explain the seemingly low prevalence of secondary causes in black patients with malignant hypertension.68

The prevalence of endocrine causes of malignant hypertension such as pheochromocytoma, Conn’s and Cushing’s syndrome is invariably low, accounting for less then 5% of the total number of cases.55, 63-66 Although rare, oral contraceptives seem to be involved in the development of malignant hyper-tension in some women. Cessation of oral contraceptives has led to a remarkable recovery of malignant hypertension with tapering of previously instituted antihypertensive medication.69-72 In some cases normal blood pressure values have been recorded just prior to the prescription of anti-contraceptives.69, 70, 72


19

Table 1 Conditions associated with malignant hypertension Renal parenchymal disease Primary renal disease (glomerulonephritis, glomerulopathy, tubulointerstitial) Systemic disorders with renal involvement (scleroderma, SLE, HUS-TTP syndrome, vasculitis) Inherited kidney diseases (FPKD, Alport’s syndrome) Renovascular disease Atherosclerotic Fibromuscular dysplasia Vasculitis (Takayasu, PAN) Endocrine disorders Pheochromocytoma Cushing’s syndrome Renin secreting tumors Mineralocorticoid hypertension Drugs and intoxications Cocaine, amphetamines, cyclosporine, erythropoietin, MAO-interactions, clonidine withdrawal Autonomic hyper-reactivity Guillain-Barre syndrome Spinal cord injury Baroreflex failure Eclampsia

SLE: systemic lupus erythematosus, HUS: hemolytic uremic syndrome, TTP: throm-botic thrombocytopenic purpura, FPKD: familial polycystic kidney disease), PAN: polyarteritis nodosa, MAO: monoamine-oxidase. Clinical perspectives The clinical presentation of malignant hypertension varies from an asymptomatic presentation to a combination of hypertensive encephalopathy, renal insufficiency and TMA. The most recent and comprehensive data on the clinical features and complications of malignant hypertension are derived from two cohorts in the UK, the West Birmingham Malignant Hypertension Register and the malignant hypertension cohort of the Glasgow Blood Pressure Clinic.73,

Chapter 1

20

74 Both cohorts are derived from multi-ethnic populations. The incidence of malignant hypertension in the West Birmingham cohort remained practically unchanged between 1970 and 1993 and was estimated at 2 per 100.000 inhabitants per year.55 The most common presenting complaints included headache, believed to result from increased intracranial pressure, and visual disturbances. Other frequent presenting complaints include gastrointestinal symptoms (nausea, vomiting and abdominal pain) and pulmonary or peripheral edema as evidence of congestive heart failure.74 The complications of malignant hypertension differ from those of chronic uncontrolled hypertension and include hypertensive encephalopathy, TMA and acute renal failure. These specific complications will now be discussed in more detail. Hypertensive encephalopathy In the Glasgow Blood Pressure Clinic, 17% of malignant hypertensive patients presented with the neurological symptoms of a hypertensive encephalopathy.74 These symptoms included altered consciousness (delirium, agitation, stupor), seizures, cortical blindness and coma; focal neurological signs were uncommon. Untreated, hypertensive encephalopathy leads to irreversible neurological damage and death.11 Hypertensive encephalopathy is believed to result from a breakthrough of cerebral autoregulation (CA).75, 76 Normally, cerebral blood flow is kept constant despite large variations in blood pressure (figure 2, solid line).75, 76 In severe uncontrolled hypertension blood flow cannot be maintained resulting in a rising intracranial pressure and, eventually, cerebral edema.77 The rate of change in blood pressure rather than the absolute height seems more important in the development of hypertensive encephalopathy: in situations of a rapid blood pressure increase, such as in patients with eclampsia or acute glomerulonephritis, hypertensive encephalopathy may develop at much lower blood pressure levels than in patients with pre-existent (uncontrolled) hypertension.78 These differences may be explained by the capacity of CA to adapt to the higher blood pressure values. In uncontrolled chronic hypertension, such an adaptation has been demonstrated by a rightward shift of the CA curve (figure 2, dotted line).76, 79 The etiology of cerebral edema in hypertensive encephalopathy is likely passive distension of cerebral arterioles. This is supported by imaging studies using SPECT, CT and MRI imaging which have shown co-localization of areas with cerebral edema and hyperperfusion.80 The posterior regions of the brain are predominantly affected and the cerebral edema in this region can best be made visible on MRI with FLAIR imaging showing edema as areas with increased signal intensity (figure 3).78, 80 In some patients


21

the brainstem may also be involved leading to compression of the cerebral aqueduct and an obstructive hydrocephalus.81-83 The reason why the posterior regions of the brain are predominantly affected in hypertensive encephalopathy may be the scarce innervation of this part of the brain by the sympathetic nervous system. Autopsy studies have shown that sympathetic innervation is less pronounced in the outflow tract of the vertebral arteries than that of the middle cerebral artery.84 In healthy volunteers, posterior cerebral artery flow is subject to a lower damping of blood pressure oscillations compared to flow in the middle cerebral artery.85 Although hypertensive encephalopathy may accompany malignant hypertension, the retinal findings which corroborate malignant hypertension do not need to be present.5 A possible explanation for this finding is that the retina, innervated by a branch from the middle cerebral artery, is more protected from the rising blood pressure due to the afore mentioned differences in sympathetic innervation. Fig. 2

Schematic representation of cerebral autoregulation (CA). Normally, cerebral blood flow is held constant despite large variations in blood pressure (solid line). In case of an acute and severe blood pressure elevation, cerebral blood flow cannot be maintained resulting in a breakthrough of CA and cerebral hyperperfusion. In patients with (uncontrolled) chronic hypertension the CA curve is shifted towards the right (dotted line) as a result of adaptation to the higher blood pressure levels protecting the brain from autoregulation breakthrough.

Chapter 1

22

Fig. 3

MRI of a patient with hypertensive encephalopathy showing an axial fluid attenuated inversion recovery (FLAIR)-image at the level of the occipital and temporal lobes. Areas with increased signal intensity are visible in the cortical and subcortical occipital region. Hypertensive encephalopathy and the cerebral white matter lesions which may accompany hypertensive encephalopathy are fully reversible with timely and appropriate blood pressure treatment.78 Therefore, this syndrome has been described as reversible posterior leucoencephalopathy in the literature. Apart from hypertension, other causes have been implicated in this syndrome, including cytotoxic therapy (cyclosporine, tacrolimus, cisplatin), erythropoietin, bevacizumab (a monoclonal antibody that binds and inhibits VEGF),86-89 thrombotic thrombocytopenic purpura (TTP),90 and carotid endarteriectomy hyperperfusion syndrome.91 Therefore, the causes of this syndrome are probably multifactorial including, apart from hypertension, endothelial activation or damage.


23

Thrombotic microangiopathy Malignant hypertension is frequently associated with a Coombs negative hemo-lysis, which is characterized by low platelets, an increased lactic dehydrogenase (LDH) and schistocytes in a peripheral blood smear (figure 4).26, 52 In the Glasgow cohort, 27% of malignant hypertensive patients had evidence of intra-vascular hemolysis defined by the presence of schistocytes.74 Previous studies have suggested that the presence of TMA is associated with more severe renal insufficiency at presentation, but that recovery of renal failure may occur with blood pressure control.21, 92-94 More recent studies have shown that both pro-coagulant and proinflammatory pathways are important in the development of the vascular abnormalities that are observed in malignant hypertension. Rats double transgenic for the human renin and angiotensinogen genes develop severe hypertension and vascular lesions indistinguishable from that observed in humans with malignant hypertension.95 Marked expression of adhesion molecules (ICAM, VCAM) are present in the interstitium, intima and adventitia of small renal arteries in these animals.49 This is followed by monocyte and macrophage infiltration and increased expression of plasminogen activator inhibitor-1 (PAI-1) and fibronectin.49 Either blood pressure treatment or direct inhibition of nuclear factor (NF) κβ, a key regulator of adhesion molecules and proinflammatory cytokines, leads to attenuation of these vascular changes.50, 96

In humans with malignant hypertension marked increases are observed in circulating levels of adhesion molecules, soluble P-selectin and von Willebrand factor (VWF) relative to normotensive or hypertensive controls.97 How the marked endothelial activation in malignant hypertension leads to fragmentation of red blood cells is not clear. Apart from malignant hypertension, there are only a few other conditions associated with TMA, notably thrombotic thrombo-cytopenic purpura (TTP), hemolytic uremic syndrome (HUS), HELLP syndrome, antiphospholipid syndrome and severe sepsis. The TMA in patients with TTP is due to a congenital or acquired deficiency of ADAMTS13.98, 99 ADAMTS13 is a metalloproteinase which cleaves VWF multimers released from the Weibel-Palade bodies of endothelial cells and platelet alpha granules.100, 101 Deficiency of ADAMTS13 leads to the accumulation of un-usually large VWF multimers and can induce platelet aggregation, thrombosis, and thrombocytopenia under high fluid shear rates.101 Deficiency of ADAMTS13 has been reported in several other conditions such as sepsis with diffuse intravascular coagulation, multi-organ failure and malignancy.102, 103 Whether deficiency of ADAMTS13 may be involved in the pathogenesis of the TMA of malignant hypertension is unclear. It is conceivable that high fluid shear

Chapter 1

24

rates and endothelial activation as evidenced by high levels of VWF in malignant hypertension may affect ADAMTS13 activity and lead to an increased presence of large prothrombogenic VWF multimers. Fig. 4

Peripheral blood smear showing fragmented red blood cells (arrows). Original magnification 1250x. Renal failure Renal dysfunction frequently accompanies malignant hypertension and patients may present with acute renal failure requiring immediate dialysis. The degree of renal insufficiency at presentation is an important determinant of future renal failure, patients with higher serum creatinine levels having a greater risk of renal failure.62, 104 However, in a subset of patients presenting with malignant hyper-tension and renal dysfunction, recovery of renal function may occur after adequate blood pressure control.92, 104, 105 21, 94, 106 These patients more often present with oliguric renal failure, normal sized kidneys and evidence of TMA at presentation.21, 92-94 Renal biopsies of patients recovering from renal failure show that renal arterioles are primarily affected with secondary ischemic changes of glomeruli due to subintimal proliferation and fibrinoid necrosis of small renal arteries and arterioles.21, 93 Loss of tubular cells suggestive of acute tubular necrosis and evidence of tubular regeneration seem to be indicative of potential recovery of renal function.21 Recovery in patients on renal replacement therapy can be seen up to 2 years after the initial presentation with malignant hypertension.92


25

Prognosis In the 1940’s and 50’s malignant hypertension was relatively frequent and had a poor prognosis with a survival rate of only 20% at 1 year and 1% at 5 years.2 As a result of better treatment and control of hypertension, malignant hypertension is rare nowadays in developed countries,64 and its survival has considerably improved.73 However, in multi-ethnic communities of both developed and lower income countries malignant hypertension is still relatively common.55, 55, 107 In the West Birmingham study there was an excess of Asian and black immigrants presenting with malignant hypertension compared to the population served by that hospital. Moreover, black patients had a poorer outcome with regard to renal function and survival in comparison with other ethnic groups. Late presentation, inadequate anti-hypertensive therapy and poor compliance have been implicated to explain the excess mortality and renal dysfunction in this group.56, 108 The median survival times for the patients who presented before 1970, during 1970-1979 and 1980-1989 were 39.2, 68.6 and 144.0+ months, respectively. The most common causes of death were renal failure (39.7%), stroke (23.8%), myocardial infarction (11.1%) and heart failure (10.3%). Although recent data on the complications and survival of malignant hypertension are lacking, it can be expected that as a result of improved treatment possibilities of malignant hypertension and its complications survival has further improved. Treatment General recommendations Malignant hypertension is considered a hypertensive emergency mandating an immediate and controlled reduction of blood pressure to safe, but not normal blood pressure levels.109 The recommended initial target is a ~25% reduction in mean arterial pressure within minutes to hours. This recommendation is based on the observation that the CA curve is shifted towards higher blood pressure levels in patients with malignant hypertension (i.e. shifted to the right), which makes them prone to cerebral hypoperfusion when blood pressure is lowered abruptly.110 Rapid blood pressure lowering treatment of more than 50% in patients with malignant hypertension has been shown to cause ischemic stroke and death.111-113

Most studies examining the efficacy and safety of oral and parenteral blood pressure lowering agents have included both patients with severe and

Chapter 1

26

malignant hypertension. Therefore it is difficult to establish the optimal therapy for malignant hypertension perse. Initiation of oral agents such as nifedipine and captopril is not preferred because of their unpredictable blood pressure lowering action in hypertensive crises.114, 115 The modes of action and possible side-effects of the various drugs that are commonly used for the parenteral treatment of hypertensive emergencies are summarized in table 2. All agents need close monitoring, preferably with an intra-arterial line in a medium or intensive care setting as the dose required for the desired blood pressure lowering effect cannot be predicted.

Although there is agreement on the initial management of malignant hypertension, there is no consensus on how blood pressure should be treated after the initial episode. One recommendation is to maintain blood pressure at the initial target level for several days and then reduce to normotensive values.109 However, evidence for this strategy is lacking as there have been no studies examining the optimal treatment after the initial blood pressure lowering phase. - Sodium nitroprusside (SNP) is effective in the treatment of malignant hyper-tension,54 and is widely advocated as a first-line agent in the treatment of hypertensive emergencies.109 SNP is an arterial and venous dilator with a short half-life and therefore most suitable for an immediate and controlled reduction of blood pressure. Abrupt blood pressure drops are quickly reversed after SNP is stopped. In spite of its superior pharmacokinetics, SNP has some disadvantages which may hamper its use in the management of malignant hypertension and hypertensive encephalopathy.

First, SNP has been shown to increase intracranial pressure in patients with an intracranial mass.116-118 However, a large decrease in mean arterial pressure induced by SNP decreases cerebral blood flow and reverses the initial increments in intracranial pressure.116-118 Therefore, it is conceivable that a similar fall in systemic blood pressure by SNP compensates for the initial disadvantageous effect on intracranial pressure in patients with malignant hypertension or hypertensive encephalopathy. Second, SNP has been shown to induce a significant reduction in regional coronary blood flow in patients with coronary artery disease and to increase ischemic injury compared to nitroglycerine in patients after myocardial infarction.119, 120 In patients with coronary artery disease, regional blood flow to ischemic areas decreases with SNP, but increases with nitroglycerin.119 This “coronary steal” may be explained by the redistribution of regional blood flow away from ischemic areas as a result of more potent dilatation of resistance vessels with SNP.


27

Table 2 Drugs commonly used for the treatment of a hypertensive emergency Drug Dose Onset

Duration

Adverse effects

Contra-indications

nitroprusside

0.5 µgr/kg/min, increase with 0.5 µgr/kg/ min every 2-3 min (max 10 µgr/kg/min)

Immediate 1-2 min Cyanide intoxication

Liver and renal insufficiency

labetalol (Trandate®)

0.5 mg/kg bolus every 10 min until goal BP (max 200 mg), maintenance dose 5-20 mg/hr

5-10 min 3-6 hrs Bradycardia COPD, 2nd degree AV block, sick sinus, CHF

nicardipine

2.5 mg/hr, increase with 2.5 mg/hr every 5 min (max 15 mg/hr)

5-10 min 2-4 hrs Reflex tachycardia, flushing

enalaprilat 0.625-5 mg bolus, repeat every 6 hrs in responders, switch to other drug in non-responders

15 min 4-6 hrs Precipitous BP falls in volume depleted states

Angioedema with ACE inhibitors

urapidil 12.5-25 mg bolus every 10 min until goal BP, maintenance dose 0.06 mg/kg/hr

5-10 min 3-5 hrs

Reflex tachycardia, drowsiness

fenoldopam 0.1 µgr/kg/min, increase with 0.1 µgr/kg/min every 15 min (max 1.6 µgr/kg/min)

5-15 min 30-60 min

Headache Constriction of cerebral arterioles?

Chapter 1

28

Apart from the regional differences on coronary blood flow by SNP and nitroglycerin, the greater fall in coronary perfusion may also be caused by the greater reductions in mean arterial pressure induced by SNP as compared to nitroglycerin.119 In malignant hypertension, the reduction in pre- and afterload which accompanies the acute blood pressure fall by SNP may compensate for the redistribution of coronary blood flow by decreasing myocardial oxygen demand. Therefore, the significance of the coronary steal phenomenon in the treatment of malignant hypertension is uncertain.

Third, after infusion SNP rapidly dissociates into nitric oxide, responsible for the direct vasodilating effect, and cyanide.121 Cyanide immediately reacts with methemoglobin to form the non-toxic cyano-methemoglobin or is enzymatically converted to thiocyanate in the liver. Thiocyanate is much less toxic than cyanide and is excreted in urine. The conversion to thiocyanate requires the presence of sufficient amounts of thiosulphate. A fasting state, surgery, hepatic failure and diuretic therapy adversely affect whole body thio-sulphate concentrations.122 Thiocyanate toxicity only develops after continuous administration of SNP for more than 7-14 days in concentrations between 2-5 µg/kg/min in patients with normal renal function. Cyanide toxicity has mainly been described in cases of deliberate hypotension during surgery.123-125 Case-reports that described cyanide toxicity related to SNP for the treatment of hypertension were in all but one case after prolonged administration (> 24 hrs) or in amounts exceeding the recommended dose.126 Based on red blood cell cyanide concentrations, treatment with SNP is safe if the average dose does not exceed 2.5 µg/kg/min in 24 hrs. Doses between 5-10 µg/kg/min should not be given for more than 2 hrs and doses higher than 10 µg/kg/min should be avoided.127 Timely institution of oral antihypertensive agents will limit the need for prolonged use of SNP and the risk of cyanide or thiocyanate toxicity. - Labetalol is a combined alpha- and beta-adrenergic blocking agent with completely different pharmacodynamic and -kinetic properties than SNP. Labetalol has been shown effective in the treatment of hypertensive emergencies.128-130 However, therapeutic goals may not be obtained in all patients, especially blacks.130 The hypotensive effect of labetalol is rapid after an intravenous bolus injection, however its long half-life of 4 to 6 hrs limits the ability to promptly correct hypotension with cessation of the drug.128, 131, 132 In contrast to SNP, intracranial pressure does not seem to increase. Cerebral blood flow does not decrease with labetalol in patients having had neurosurgery and healthy volunteers.133, 134 In hypertensive subjects, labetalol has been shown to


29

maintain cardiac output without reflex tachycardia.135 This may be advantageous in patients with a hypertensive emergency accompanied by coronary ischemia or infarction.136-138 Unfortunately there are no comparative trials examining the effect on labetalol and SNP on cerebral blood flow or myocardial ischemia in the treatment of hypertensive emergencies. -Nicardipine, a calcium-channel blocker for intravenous use, has been shown to effectively lower blood pressure in patients with severe and malignant hyper-tension.139, 140 In these patients, nicardipine and SNP are equally effective in lowering blood pressure with a similar increase in heart rate.141 Nicardipine, and other calcium antagonists, exert a profound vasodilatory action on cerebral blood vessels and increase cerebral blood flow.142, 143 However, in patients undergoing neurosurgery, nicardipine does not seem to increase intracranial pressure suggesting that systemic vasodilatation may counterbalance the effect of nicardipine on the cerebral vasculature.144, 145 In patients undergoing coronary artery bypass grafting, treatment with intravenous nicardipine may protect against coronary ischemia compared to SNP.146 Nicardipine has been shown to increase renal blood flow despite a systemic fall in blood pressure in patients with mild to moderate hypertension.147 - Enalaprilat is an intravenous ACE inhibitor inhibiting the formation angiotensin II and thereby angiotensin II induced vasoconstriction. The blood pressure lowering response to enalapril in severe and malignant hypertension is variable and unpredictable. In one study, more than 30% failed to reach the desired goal blood pressure.148 No further blood pressure reductions were observed with dose increments exceeding 10 mg.149 The variable response to enalaprilat is explained by the correlation between blood pressure response and pretreatment angiotensin II levels and plasma renin activity.150 High levels of angiotensin II are associated with risk of severe hypotension following enala-prilat administration, whereas low levels of angiotensin II are associated with therapy failure. Currently, enalaprilat and other ACE-inhibitors are recom-mended only for the treatment of hypertensive crises associated with systemic sclerosis. - Urapidil is a combined alpha-1-adrenoreceptorantagonist with central sympatholytic properties. It has been shown effective and safe in the treatment of severe hypertension and in hypertensive patients with pulmonary edema.115,

151-153 Although the response rate is high,151, 153 the time to reach goal blood

Chapter 1

30

pressure with urapidil is longer compared to SNP.151 There is some evidence that urapidil may temporarily increase intracranial pressure directly after intravenous bolus injections.154 Treatment with urapidil may result in hypotensive episodes which cannot be immediately corrected by cessation of the drug due to its long half-life.155 As with nicardipine, treatment with urapidil improves renal blood flow.156 - Fenoldopam, a short acting selective dopamine-1 agonist, has been shown effective in the treatment of hypertensive crises.157 Fenoldopam has equivalent efficacy compared to SNP and enhances natriuresis.158 A possible beneficial effect of fenoldopam on renal outcome compared to SNP or other parenteral anti-hypertensive agents has not been examined. In healthy volunteers, fenol-dopam has been shown to decrease cerebral blood flow despite normalization of the initial blood pressure fall with phenylephrine likely by constriction of cerebral arterioles.159 Fenoldopam has not been approved for use in the Netherlands. References 1. Volhard F, Fahr T. Die Brightsche Nieren krankheit.Klinik, Pathologie und Atlas.

Berlin: Springer; 1914.

2. Keith NM, Wagener HP, Barker NW. Some different types of essential hyper-tension: their course and prognosis. Am J Med Sci. 1939;196:332-343.

3. Ahmed ME, Walker JM, Beevers DG, Beevers M. Lack of difference between malignant and accelerated hypertension. Br Med J (Clin Res Ed). 1986;292:235-237.

4. McGregor E, Isles CG, Jay JL, Lever AF, Murray GD. Retinal changes in malignant hypertension. Br Med J (Clin Res Ed). 1986;292:233-234.

5. Bakker RC, Verburgh CA, van Buchem MA, Paul LC. Hypertension, cerebral edema and fundoscopy. Nephrol Dial Transplant. 2003;18:2424-2427.

6. Sanerkin NG. Vascular lesions of malignant essential hypertension. J Pathol. 1971; 103:177-184.

7. Jones DB. Arterial and glomerular lesions associated with severe hypertension. Light and electron microscopic studies. Lab Invest. 1974;31:303-313.


31

8. Trostle DC, Bedetti CD, Steen VD, Al-Sabbagh MR, Zee B, Medsger TA, Jr. Renal vascular histology and morphometry in systemic sclerosis. A case-control autopsy study. Arthritis Rheum. 1988;31:393-400.

9. Conlon PJ, Howell DN, Macik G, Kovalik EC, Smith SR. The renal manifestations and outcome of thrombotic thrombocytopenic purpura/hemolytic uremic syndrome in adults. Nephrol Dial Transplant. 1995;10:1189-1193.

10. Nochy D, Daugas E, Droz D, Beaufils H, Grunfeld JP, Piette JC, Bariety J, Hill G. The intrarenal vascular lesions associated with primary antiphospholipid syndrome. J Am Soc Nephrol. 1999;10:507-518.

11. Chester EM, Agamanolis DP, Banker BQ, Victor M. Hypertensive encephalopathy: a clinicopathologic study of 20 cases. Neurology. 1978;28:928-939.

12. Kincaid-Smith P, McMichael J, Murphy EA. The clinical course and pathology of hypertension with papilledema (malignant hypertension). Q J Med. 1958;27:117-153.

13. Garner A, Ashton N, Tripathi R, Kohner EM, Bulpitt CJ, Dollery CT. Pathogenesis of hypertensive retinopathy. An experimental study in the monkey. Br J Ophthal-mol. 1975;59:3-44.

14. Hayreh SS, Servais GE, Virdi PS, Marcus ML, Rojas P, Woolson RF. Fundus lesions in malignant hypertension. III. Arterial blood pressure, biochemical, and fundus changes. Ophthalmology. 1986;93:45-59.

15. McLeod D, Marshall J, Kohner EM, Bird AC. The role of axoplasmic transport in the pathogenesis of retinal cotton-wool spots. Br J Ophthalmol. 1977;61:177-191.

16. Hayreh SS, Servais GE, Virdi PS. Cotton-wool spots (inner retinal ischemic spots) in malignant arterial hypertension. Ophthalmologica. 1989;198:197-215.

17. Hayreh SS, Servais GE, Virdi PS. Fundus lesions in malignant hypertension. V. Hypertensive optic neuropathy. Ophthalmology. 1986;93:74-87.

18. Hayreh SS, Servais GE, Virdi PS. Fundus lesions in malignant hypertension. VI. Hypertensive choroidopathy. Ophthalmology. 1986;93:1383-1400.

19. Kishi S, Tso MO, Hayreh SS. Fundus lesions in malignant hypertension. II. A pathologic study of experimental hypertensive optic neuropathy. Arch Ophthalmol. 1985;103:1198-1206.

20. Schiff D, Lopes MB. Neuropathological correlates of reversible posterior leuko-encephalopathy. Neurocrit Care. 2005;2:303-305.

Chapter 1

32

21. Isles CG, McLay A, Jones JM. Recovery in malignant hypertension presenting as acute renal failure. Q J Med. 1984;53:439-452.

22. McLaren KM, MacDonald MK. Histological and ultrastructural studies of the human juxtaglomerular apparatus in benign and malignant hypertension. J Pathol. 1983;139:41-55.

23. Laragh JH, Angers M, Kelly WG, Lieberman S. Hypotensive agents and pressor substances. The effect of epinephrine, norepinephrine, angiotensin II, and others on the secretory rate of aldosterone in man. JAMA. 1960;174:234-240.

24. Buhler FR, Laragh JH, Vaughan ED, Jr., Brunner HR, Gavras H, Baer L. Antihypertensive action of propranolol. Specific antirenin responses in high and normal renin forms of essential, renal, renovascular and malignant hypertension. Am J Cardiol. 1973;32:511-522.

25. McAllister RG, Jr., Van WC, III, Dayani K, Anderson WJ, Temple E, Michelakis AM, Coppage WS, Jr., Oates JA. Malignant hypertension: effect of therapy on renin and aldosterone. Circ Res. 1971;28:Suppl-74.

26. Gavras H, Brown WC, Brown JJ, Lever AF, Linton AL, MacAdam RF, McNicol GP, Robertson JI, Wardrop C. Microangipathic hemolytic anemia and the development of the malignant phase of hypertension. Circ Res. 1971;28:Suppl-42.

27. Case DB, Atlas SA, Sullivan PA, Laragh JH. Acute and chronic treatment of severe and malignant hypertension with the oral angiotensin-converting enzyme inhibitor captopril. Circulation. 1981;64:765-771.

28. Gavras H, Brunner HR, Laragh JH, Vaughan ED, Jr., Koss M, Cote LJ, Gavras I. Malignant hypertension resulting from deoxycorticosterone acetate and salt excess: role of renin and sodium in vascular changes. Circ Res. 1975;36:300-309.

29. Lohmeier TE, Tillman LJ, Carroll RG, Brown AJ, Guyton AC. Malignant hypertensive crisis induced by chronic intrarenal norepinephrine infusion. Hypertension. 1984;6:I177-I182.

30. Sventek P, Li JS, Grove K, Deschepper CF, Schiffrin EL. Vascular structure and expression of endothelin-1 gene in L-NAME-treated spontaneously hypertensive rats. Hypertension. 1996;27:49-55.

31. Ideishi M, Kishikawa K, Kinoshita A, Sasaguri M, Ikeda M, Takebayashi S, Arakawa K. High-renin malignant hypertension secondary to an aldosterone-producing adenoma. Nephron. 1990;54:259-263.

32. Kaplan NM. Primary aldosteronism with malignant hypertension. N Engl J Med. 1963;269:1282-1286.


33

33. Lenz T, Thiede HM, Nussberger J, Atlas SA, Distler A, Schulte KL. Hyperreni-nemia and secondary hyperaldosteronism in a patient with pheochromocytoma and von Hippel-Lindau disease. Nephron. 1992;62:345-350.

34. Cowley AW, Jr., Lohmeier TE. Changes in renal vascular sensitivity and arterial pressure associated with sodium intake during long-term intrarenal norepinephrine infusion in dogs. Hypertension. 1979;1:549-558.

35. Lohmeier TE, Cowley AW, Jr. Hypertensive and renal effects of chronic low level intrarenal angiotensin infusion in the dog. Circ Res. 1979;44:154-160.

36. Mohring J, Petri M, Szokol M, Haack D, Mohring B. Effects of saline drinking on malignant course of renal hypertension in rats. Am J Physiol. 1976;230:849-857.

37. Mohring J, Mohring B, Petri M, Haack D. Plasma vasopressin concentrations and effects of vasopressin antiserum on blood pressure in rats with malignant two-kidney Goldblatt hypertension. Circ Res. 1978;42:17-22.

38. Munoz-Ramirez H, Chatelain RE, Bumpus FM, Khairallah PA. Development of two-kidney Goldblatt hypertension in rats under dietary sodium restriction. Am J Physiol. 1980;238:H889-H894.

39. McAreavey D, Brown WB, Murray GD, Robertson JI. Exchangeable sodium in Goldblatt one-kidney one-clip hypertension in the rat. Clin Sci (Lond). 1984;66:545-549.

40. Agarwal M, Lynn KL, Richards AM, Nicholls MG. Hyponatremic-hypertensive syndrome with renal ischemia: an underrecognized disorder. Hypertension. 1999;33:1020-1024.

41. Atkinson AB, Davies DL, Leckie B, Morton JJ, Brown JJ, Fraser R, Lever AF, Robertson JI. Hyponatraemic hypertensive syndrome with renal-artery occlusion corrected by captopril. Lancet. 1979;2:606-609.

42. Usberti M, Federico S, Di MG, Ungaro B, Ardillo G, Pecoraro C, Cianciaruso B, Cerbone AM, Cirillo F, Pannain M, . Effects of angiotensin II on plasma ADH, prostaglandin synthesis, and water excretion in normal humans. Am J Physiol. 1985;248:F254-F259.

43. Padfield PL, Brown JJ, Lever AF, Morton JJ, Robertson JI. Blood pressure in acute and chronic vasopressin excess: studies of malignant hypertension and the syndrome of inappropriate antidiuretic hormone secretion. N Engl J Med. 1981;304:1067-1070.

44. Kaloyanides GJ, DiBona GF. Effect of an angiotensin II antagonist on autoregulation in the isolated dog kidney. Am J Physiol. 1976;230:1078-1083.

Chapter 1

34

45. Roman RJ, Cowley AW, Jr., Garcia-Estan J, Lombard JH. Pressure-diuresis in volume-expanded rats. Cortical and medullary hemodynamics. Hypertension. 1988;12:168-176.

46. Mattson DL, Lu S, Roman RJ, Cowley AW, Jr. Relationship between renal perfusion pressure and blood flow in different regions of the kidney. Am J Physiol. 1993;264:R578-R583.

47. Wilson C, Byrom FB. Renal changes in malignant hypertension. Lancet. 1939;I: 136-139.

48. Kincaid-Smith P. Malignant hypertension. J Hypertens. 1991;9:893-899.

49. Mervaala EM, Muller DN, Park JK, Schmidt F, Lohn M, Breu V, Dragun D, Ganten D, Haller H, Luft FC. Monocyte infiltration and adhesion molecules in a rat model of high human renin hypertension. Hypertension. 1999;33:389-395.

50. Muller DN, Dechend R, Mervaala EM, Park JK, Schmidt F, Fiebeler A, Theuer J, Breu V, Ganten D, Haller H, Luft FC. NF-kappaB inhibition ameliorates angiotensin II-induced inflammatory damage in rats. Hypertension. 2000;35:193-201.

51. Hollenberg NK, Epstein M, Basch RI, Couch NP, Hickler RB, Merrill JP. Renin secretion in essential and accelerated hypertension. Am J Med. 1969;47:855-859.

52. Linton AL, Gavras H, Gleadle RI, Hutchison HE, Lawson DH, Lever AF, MacAdam RF, McNicol GP, Robertson JI. Microangiopathic haemolytic anaemia and the pathogenesis of malignant hypertension. Lancet. 1969;1:1277-1282.

53. Beutler JJ, Koomans HA. Malignant hypertension: still a challenge. Nephrol Dial Transplant. 1997;12:2019-2023.

54. Ahearn DJ, Grim CE. Treatment of malignant hypertension with sodium nitroprusside. Arch Intern Med. 1974;133:187-191.

55. Lip GY, Beevers M, Beevers G. The failure of malignant hypertension to decline: a survey of 24 years' experience in a multiracial population in England. J Hypertens. 1994;12:1297-1305.

56. Shea S, Misra D, Ehrlich MH, Field L, Francis CK. Predisposing factors for severe, uncontrolled hypertension in an inner-city minority population. N Engl J Med. 1992;327:776-781.

57. Isles C, Brown JJ, Cumming AM, Lever AF, McAreavey D, Robertson JI, Hawthorne VM, Stewart GM, Robertson JW, Wapshaw J. Excess smoking in malignant-phase hypertension. Br Med J. 1979;1:579-581.


35

58. Bloxham CA, Beevers DG, Walker JM. Malignant hypertension and cigarette smoking. Br Med J. 1979;1:581-583.

59. Edmunds E, Landray MJ, Li-Saw-Hee FL, Hughes BA, Beevers DG, Lip GY. Dyslipidaemia in patients with malignant-phase hypertension. QJM. 2001;94:327-332.

60. Stefansson B, Ricksten A, Rymo L, Aurell M, Herlitz H. Angiotensin-converting enzyme gene I/D polymorphism in malignant hypertension. Blood Press. 2000;9:104-109.

61. Espinel E, Tovar JL, Borrellas J, Piera L, Jardi R, Frias FR, Armadans L, Bachs AG. Angiotensin-converting enzyme i/d polymorphism in patients with malignant hypertension. J Clin Hypertens (Greenwich ). 2005;7:11-15.

62. Guerin C, Gonthier R, Berthoux FC. Long-term prognosis in malignant or accelerated hypertension. Nephrol Dial Transplant. 1988;3:33-37.

63. Scarpelli PT, Livi R, Caselli GM, Di Maria L, Teghini L, Montemurro V, Toti G, Becucci A. Accelerated (malignant) hypertension: a study of 121 cases between 1974 and 1996. J Nephrol. 1997;10:207-215.

64. Gudbrandsson T. Malignant hypertension. A clinical follow-up study with special reference to renal and cardiovascular function and immunogenetic factors. Acta Med Scand Suppl. 1981;650:1-62.

65. Webster J, Petrie JC, Jeffers TA, Lovell HG. Accelerated hypertension--patterns of mortality and clinical factors affecting outcome in treated patients. Q J Med. 1993;86:485-493.

66. Yu SH, Whitworth JA, Kincaid-Smith PS. Malignant hypertension: aetiology and outcome in 83 patients. Clin Exp Hypertens A. 1986;8:1211-1230.

67. Davis BA, Crook JE, Vestal RE, Oates JA. Prevalence of renovascular hypertension in patients with grade III or IV hypertensive retinopathy. N Engl J Med. 1979;301:1273-1276.

68. Jhetam D, Dansey R, Morar C, Milne FJ. The malignant phase of essential hypertension in Johannesburg Blacks. A prospective study. S Afr Med J. 1982;61:899-902.

69. Lim KG, Isles CG, Hodsman GP, Lever AF, Robertson JW. Malignant hypertension in women of childbearing age and its relation to the contraceptive pill. Br Med J (Clin Res Ed). 1987;294:1057-1059.

Chapter 1

36

70. Zech P, Rifle G, Lindner A, Sassard J, Blanc-Brunat N, Traeger J. Malignant hypertension with irreversible renal failure due to oral contraceptives. Br Med J. 1975;4:326-327.

71. Dunn FG, Jones JV, Fife R. Malignant hypertension associated with use of oral contraceptives. Br Heart J. 1975;37:336-338.

72. Petitti DB, Klatsky AL. Malignant hypertension in women aged 15 to 44 years and its relation to cigarette smoking and oral contraceptives. Am J Cardiol. 1983;52:297-298.

73. Lip GY, Beevers M, Beevers DG. Complications and survival of 315 patients with malignant-phase hypertension. J Hypertens. 1995;13:915-924.

74. Isles CG. Malignant Hypertension and Hypertensive Encephalopathy. In: Swales JD, editor. Textbook of Hypertension. Blackwell Scientific Publishers; 1994.

75. Lassen NA. Autoregulation of cerebral blood flow. Circ Res. 1964;15:Suppl-4.

76. Strandgaard S, Olesen J, Skinhoj E, Lassen NA. Autoregulation of brain circulation in severe arterial hypertension. Br Med J. 1973;1:507-510.

77. Griswold WR, Viney J, Mendoza SA, James HE. Intracranial pressure monitoring in severe hypertensive encephalopathy. Crit Care Med. 1981;9:573-576.

78. Hinchey J, Chaves C, Appignani B, Breen J, Pao L, Wang A, Pessin MS, Lamy C, Mas JL, Caplan LR. A reversible posterior leukoencephalopathy syndrome. N Engl J Med. 1996;334:494-500.

79. Finnerty FA, Jr. Hypertensive encephalopathy. Am J Med. 1972;52:672-678.

80. Schwartz RB, Jones KM, Kalina P, Bajakian RL, Mantello MT, Garada B, Holman BL. Hypertensive encephalopathy: findings on CT, MR imaging, and SPECT imaging in 14 cases. AJR Am J Roentgenol. 1992;159:379-383.

81. Casey SO, Truwit CL. Pontine reversible edema: a newly recognized imaging variant of hypertensive encephalopathy? AJNR Am J Neuroradiol. 2000;21:243-245.

82. de SJ, Mastain B, Stojkovic T, Ferriby D, Pruvo JP, Destee A, Vermersch P. Unusual MR findings of the brain stem in arterial hypertension. AJNR Am J Neuroradiol. 2000;21:391-394.

83. Cruz-Flores S, de Assis Aquino GF, Leira EC. Brainstem involvement in hyper-tensive encephalopathy: clinical and radiological findings. Neurology. 2004;62: 1417-1419.


37

84. Edvinsson L, Owman C, Sjoberg NO. Autonomic nerves, mast cells, and amine receptors in human brain vessels. A histochemical and pharmacological study. Brain Res. 1976;115:377-393.

85. Haubrich C, Wendt A, Diehl RR, Klotzsch C. Dynamic autoregulation testing in the posterior cerebral artery. Stroke. 2004;35:848-852.

86. Schwartz RB, Bravo SM, Klufas RA, Hsu L, Barnes PD, Robson CD, Antin JH. Cyclosporine neurotoxicity and its relationship to hypertensive encephalopathy: CT and MR findings in 16 cases. AJR Am J Roentgenol. 1995;165:627-631.

87. Delanty N, Vaughan C, Frucht S, Stubgen P. Erythropoietin-associated hypertensive posterior leukoencephalopathy. Neurology. 1997;49:686-689.

88. Glusker P, Recht L, Lane B. Reversible posterior leukoencephalopathy syndrome and bevacizumab. N Engl J Med. 2006;354:980-982.

89. Ozcan C, Wong SJ, Hari P. Reversible posterior leukoencephalopathy syndrome and bevacizumab. N Engl J Med. 2006;354:980-982.

90. Bakshi R, Shaikh ZA, Bates VE, Kinkel PR. Thrombotic thrombocytopenic purpura: brain CT and MRI findings in 12 patients. Neurology. 1999;52:1285-1288.

91. Ille O, Woimant F, Pruna A, Corabianu O, Idatte JM, Haguenau M. Hypertensive encephalopathy after bilateral carotid endarterectomy. Stroke. 1995;26:488-491.

92. Mamdani BH, Lim VS, Mahurkar SD, Katz AI, Dunea G. Recovery from prolonged renal failure in patients with accelerated hypertension. N Engl J Med. 1974;291:1343-1344.

93. Luft FC, Bloch R, Szwed JJ, Grim CM, Grim CE. Minoxidil treatment of malignant hypertension. Recovery of renal function. JAMA. 1978;240:1985-1987.

94. James SH, Meyers AM, Milne FJ, Reinach SG. Partial recovery of renal function in black patients with apparent end-stage renal failure due to primary malignant hypertension. Nephron. 1995;71:29-34.

95. Luft FC, Mervaala E, Muller DN, Gross V, Schmidt F, Park JK, Schmitz C, Lippoldt A, Breu V, Dechend R, Dragun D, Schneider W, Ganten D, Haller H. Hypertension-induced end-organ damage : A new transgenic approach to an old problem. Hypertension. 1999;33:212-218.

96. Mervaala E, Muller DN, Schmidt F, Park JK, Gross V, Bader M, Breu V, Ganten D, Haller H, Luft FC. Blood pressure-independent effects in rats with human renin and angiotensinogen genes. Hypertension. 2000;35:587-594.

Chapter 1

38

97. Verhaar MC, Beutler JJ, Gaillard CA, Koomans HA, Fijnheer R, Rabelink TJ. Progressive vascular damage in hypertension is associated with increased levels of circulating P-selectin. J Hypertens. 1998;16:45-50.

98. Furlan M, Robles R, Galbusera M, Remuzzi G, Kyrle PA, Brenner B, Krause M, Scharrer I, Aumann V, Mittler U, Solenthaler M, Lammle B. von Willebrand factor-cleaving protease in thrombotic thrombocytopenic purpura and the hemolytic-uremic syndrome. N Engl J Med. 1998;339:1578-1584.

99. Tsai HM, Lian EC. Antibodies to von Willebrand factor-cleaving protease in acute thrombotic thrombocytopenic purpura. N Engl J Med. 1998;339:1585-1594.

100. Sporn LA, Chavin SI, Marder VJ, Wagner DD. Biosynthesis of von Willebrand protein by human megakaryocytes. J Clin Invest. 1985;76:1102-1106.

101. Moake JL, Turner NA, Stathopoulos NA, Nolasco LH, Hellums JD. Involvement of large plasma von Willebrand factor (vWF) multimers and unusually large vWF forms derived from endothelial cells in shear stress-induced platelet aggregation. J Clin Invest. 1986;78:1456-1461.

102. Vesely SK, George JN, Lammle B, Studt JD, Alberio L, El Harake MA, Raskob GE. ADAMTS13 activity in thrombotic thrombocytopenic purpura-hemolytic uremic syndrome: relation to presenting features and clinical outcomes in a prospective cohort of 142 patients. Blood. 2003;102:60-68.

103. Ono T, Mimuro J, Madoiwa S, Soejima K, Kashiwakura Y, Ishiwata A, Takano K, Ohmori T, Sakata Y. Severe secondary deficiency of von Willebrand factor-cleaving protease (ADAMTS13) in patients with sepsis-induced disseminated intravascular coagulation: its correlation with development of renal failure. Blood. 2006;107:528-534.

104. Mitchell HC, Graham RM, Pettinger WA. Renal function during long-term treatment of hypertension with minoxidil: comparison of benign and malignant hypertension. Ann Intern Med. 1980;93:676-681.

105. Cordingley FT, Jones NF, Wing AJ, Hilton PJ. Reversible renal failure in malignant hypertension. Clin Nephrol. 1980;14:98-103.

106. Mourad G, Mimran A, Mion CM. Recovery of renal function in patients with accelerated malignant nephrosclerosis on maintenance dialysis with management of blood pressure by captopril. Nephron. 1985;41:166-169.

107. Kadiri S, Olutade BO. The clinical presentation of malignant hypertension in Nigerians. J Hum Hypertens. 1991;5:339-343.


39

108. Opie LH, Seedat YK. Hypertension in sub-Saharan African populations. Circulation. 2005;112:3562-3568.

109. Calhoun DA, Oparil S. Treatment of hypertensive crises. N Engl J Med. 1990;323:1177-1183.

110. Strandgaard S. Autoregulation of cerebral blood flow in hypertensive patients. The modifying influence of prolonged antihypertensive treatment on the tolerance to acute, drug-induced hypotension. Circulation. 1976;53:720-727.

111. Haas DC, Streeten DH, Kim RC, Naalbandian AN, Obeid AI. Death from cerebral hypoperfusion during nitroprusside treatment of acute angiotensin-dependent hypertension. Am J Med. 1983;75:1071-1076.

112. Ledingham JG, Rajagopalan B. Cerebral complications in the treatment of accelerated hypertension. Q J Med. 1979;48:25-41.

113. Mak W, Chan KH, Cheung RT, Ho SL. Hypertensive encephalopathy: BP lowering complicated by posterior circulation ischemic stroke. Neurology. 2004;63:1131-1132.

114. Damasceno A, Ferreira B, Patel S, Sevene E, Polonia J. Efficacy of captopril and nifedipine in black and white patients with hypertensive crisis. J Hum Hypertens. 1997;11:471-476.

115. Hirschl MM, Seidler D, Zeiner A, Wagner A, Heinz G, Sterz F, Laggner AN. Intravenous urapidil versus sublingual nifedipine in the treatment of hypertensive urgencies. Am J Emerg Med. 1993;11:653-656.

116. Cottrell JE, Patel K, Turndorf H, Ransohoff J. Intracranial pressure changes induced by sodium nitroprusside in patients with intracranial mass lesions. J Neurosurg. 1978;48:329-331.

117. Anile C, Zanghi F, Bracali A, Maira G, Rossi GF. Sodium nitroprusside and intracranial pressure. Acta Neurochir (Wien ). 1981;58:203-211.

118. Turner JM, Powell D, Gibson RM, McDowall DG. Intracranial pressure changes in neurosurgical patients during hypotension induced with sodium nitroprusside or trimetaphan. Br J Anaesth. 1977;49:419-425.

119. Mann T, Cohn PF, Holman LB, Green LH, Markis JE, Phillips DA. Effect of nitroprusside on regional myocardial blood flow in coronary artery disease. Results in 25 patients and comparison with nitroglycerin. Circulation. 1978;57:732-738.

Chapter 1

40

120. Chiariello M, Gold HK, Leinbach RC, Davis MA, Maroko PR. Comparison between the effects of nitroprusside and nitroglycerin on ischemic injury during acute myocardial infarction. Circulation. 1976;54:766-773.

121. Smith RP, Kruszyna H. Nitroprusside produces cyanide poisoning via reaction with hemoglobin. J Pharmacol Exp Ther. 1974;191:557-563.

122. Ivankovich AD, Braverman B, Stephens TS, Shulman M, Heyman HJ. Sodium thiosulfate disposition in humans: relation to sodium nitroprusside toxicity. Anesthesiology. 1983;58:11-17.

123. Pasch T, Schulz V, Hoppelshauser G. Nitroprusside-induced formation of cyanide and its detoxication with thiosulfate during deliberate hypotension. J Cardiovasc Pharmacol. 1983;5:77-85.

124. Vesey CJ, Cole PV, Simpson PJ. Cyanide and thiocyanate concentrations following sodium nitroprusside infusion in man. Br J Anaesth. 1976;48:651-660.

125. Patel CB, Laboy V, Venus B, Mathru M, Wier D. Use of sodium nitroprusside in post-coronary bypass surgery. A plea for conservatism. Chest. 1986;89:663-667.

126. Rindone JP, Sloane EP. Cyanide toxicity from sodium nitroprusside: risks and management. Ann Pharmacother. 1992;26:515-519.

127. Schulz V, Gross R, Pasch T, Busse J, Loeschcke G. Cyanide toxicity of sodium nitroprusside in therapeutic use with and without sodium thiosulphate. Klin Wochenschr. 1982;60:1393-1400.

128. Rosei EA, Trust PM, Brown JJ, Lever AF, Robertson JI. Letter: Intravenous labetalol in severe hypertension. Lancet. 1975;2:1093-1094.

129. Cummings AM, Brown JJ, Fraser R, Lever AF, Morton JJ, Richards DA, Robertson JI. Blood pressure reduction by incremental infusion of labetalol in patients with severe hypertension. Br J Clin Pharmacol. 1979;8:359-364.

130. Wilson DJ, Wallin JD, Vlachakis ND, Freis ED, Vidt DG, Michelson EL, Langford HG, Flamenbaum W, Poland MP. Intravenous labetalol in the treatment of severe hypertension and hypertensive emergencies. Am J Med. 1983;75:95-102.

131. Kanto J, Allonen H, Kleimola T, Mantyla R. Pharmacokinetics of labetalol in healthy volunteers. Int J Clin Pharmacol Ther Toxicol. 1981;19:41-44.

132. Solomons R. Hemiparesis after single minibolus of labetalol for hypertensive encephalopathy. Br Med J. 1979;2:672.


41

133. Olsen KS, Svendsen LB, Larsen FS, Paulson OB. Effect of labetalol on cerebral blood flow, oxygen metabolism and autoregulation in healthy humans. Br J Anaesth. 1995;75:51-54.

134. Orlowski JP, Shiesley D, Vidt DG, Barnett GH, Little JR. Labetalol to control blood pressure after cerebrovascular surgery. Crit Care Med. 1988;16:765-768.

135. Mehta J, Cohn JN. Hemodynamic effects of labetalol, an alpha and beta adrenergic blocking agent, in hypertensive subjects. Circulation. 1977;55:370-375.

136. Marx PG, Reid DS. Labetalol infusion in acute myocardial infarction with systemic hypertension. Br J Clin Pharmacol. 1979;8:233S-238S.

137. Timmis AD, Fowler MB, Jaggarao NS, Vincent R, Chamberlain DA. Role of labetalol in acute myocardial infarction. Br J Clin Pharmacol. 1982;13:111S-114S.

138. Frishman WH, Strom JA, Kirschner M, Poland M, Klein N, Halprin S, LeJemtel TH, Kram M, Sonnenblick EH. Labetalol therapy in patients with systemic hypertension and angina pectoris: effects of combined alpha and beta adrenoceptor blockade. Am J Cardiol. 1981;48:917-928.

139. Wallin JD, Fletcher E, Ram CV, Cook ME, Cheung DG, MacCarthy EP, Townsend R, Saunders E, Davis WR, Langford HG, . Intravenous nicardipine for the treatment of severe hypertension. A double-blind, placebo-controlled multicenter trial. Arch Intern Med. 1989;149:2662-2669.

140. Clifton GG, Cook ME, Bienvenu GS, Wallin JD. Intravenous nicardipine in severe systemic hypertension. Am J Cardiol. 1989;64:16H-18H.

141. Neutel JM, Smith DH, Wallin D, Cook E, Ram CV, Fletcher E, Maher KE, Turlepaty P, Grandy S, Lee R, . A comparison of intravenous nicardipine and sodium nitroprusside in the immediate treatment of severe hypertension. Am J Hypertens. 1994;7:623-628.

142. Gaab MR, Czech T, Korn A. Intracranial effects of nicardipine. Br J Clin Pharma-col. 1985;20 Suppl 1:67S-74S.

143. Bertel O, Conen D, Radu EW, Muller J, Lang C, Dubach UC. Nifedipine in hyper-tensive emergencies. Br Med J (Clin Res Ed). 1983;286:19-21.

144. Nishiyama T, Yokoyama T, Matsukawa T, Hanaoka K. Continuous nicardipine infusion to control blood pressure after evacuation of acute cerebral hemorrhage. Can J Anaesth. 2000;47:1196-1201.

Chapter 1

42

145. Combes P, Durand M. Combined effects of nicardipine and hypocapnic alkalosis on cerebral vasomotor activity and intracranial pressure in man. Eur J Clin Pharmacol. 1991;41:207-210.

146. van Wezel HB, Koolen JJ, Visser CA, Dijkhuis JP, Vergroesen I, Moulijn AC, Deen L. Antihypertensive and anti-ischemic effects of nicardipine and nitroprusside in patients undergoing coronary artery bypass grafting. Am J Cardiol. 1989;64:22H-27H.

147. Baba T, Boku A, Ishizaki T, Sone K, Takebe K. Renal effects of nicardipine in patients with mild-to-moderate essential hypertension. Am Heart J. 1986;111:552-557.

148. Hirschl MM, Binder M, Bur A, Herkner H, Brunner M, Mullner M, Sterz F, Laggner AN. Clinical evaluation of different doses of intravenous enalaprilat in patients with hypertensive crises. Arch Intern Med. 1995;155:2217-2223.

149. DiPette DJ, Ferraro JC, Evans RR, Martin M. Enalaprilat, an intravenous angiotensin-converting enzyme inhibitor, in hypertensive crises. Clin Pharmacol Ther. 1985;38:199-204.

150. Hirschl MM, Binder M, Bur A, Herkner H, Woisetschlager C, Bieglmayer C, Laggner AN. Impact of the renin-angiotensin-aldosterone system on blood pressure response to intravenous enalaprilat in patients with hypertensive crises. J Hum Hypertens. 1997;11:177-183.

151. Hirschl MM, Binder M, Bur A, Herkner H, Mullner M, Woisetschlager C, Laggner AN. Safety and efficacy of urapidil and sodium nitroprusside in the treatment of hypertensive emergencies. Intensive Care Med. 1997;23:885-888.

152. Schreiber W, Woisetschlager C, Binder M, Kaff A, Raab H, Hirschl MM. The nitura study--effect of nitroglycerin or urapidil on hemodynamic, metabolic and respiratory parameters in hypertensive patients with pulmonary edema. Intensive Care Med. 1998;24:557-563.

153. Woisetschlager C, Bur A, Vlcek M, Derhaschnig U, Laggner AN, Hirschl MM. Comparison of intravenous urapidil and oral captopril in patients with hypertensive urgencies. J Hum Hypertens. 2006;20:707-709.

154. Singbartl G, Metzger G. Urapidil-induced increase of the intracranial pressure in head-trauma patients. Intensive Care Med. 1990;16:272-274.

155. van der Stroom JG, van Wezel HB, Langemeijer JJ, Korsten HH, Kooyman J, van der Starre PJ, Kal JE, Porsius M, van den ER, van Zwieten PA. A randomized multicenter double-blind comparison of urapidil and ketanserin in hypertensive


43

patients after coronary artery surgery. J Cardiothorac Vasc Anesth. 1997;11:729-736.

156. de Leeuw PW, van Es PN, de Bruyn HA, Birkenhager WH. Renal haemodynamic and neurohumoral responses to urapidil in hypertensive man. Drugs. 1988;35 Suppl 6:74-77.

157. Shusterman NH, Elliott WJ, White WB. Fenoldopam, but not nitroprusside, improves renal function in severely hypertensive patients with impaired renal function. Am J Med. 1993;95:161-168.

158. Elliott WJ, Weber RR, Nelson KS, Oliner CM, Fumo MT, Gretler DD, McCray GR, Murphy MB. Renal and hemodynamic effects of intravenous fenoldopam versus nitroprusside in severe hypertension. Circulation. 1990;81:970-977.

159. Prielipp RC, Wall MH, Groban L, Tobin JR, Fahey FH, Harkness BA, Stump DA, James RL, Cannon MA, Bennett J, Butterworth J. Reduced regional and global cerebral blood flow during fenoldopam-induced hypotension in volunteers. Anesth Analg. 2001;93:45-52.

Chapter 2

Value of routine funduscopy in patients with hypertension: systematic review Bert-Jan H van den Born, Caroline A A Hulsman, Joost B L Hoekstra, Reinier O Schlingemann, Gert A van Montfrans

BMJ 2005; 331: 73-78.

Chapter 2

48

Abstract Objective To evaluate the additional value of funduscopy in the routine management of patients with hypertension. Design Systematic review. Participants Adults aged 19 or more with hypertensive retinopathy. Data sources Medline, Embase, and the Cochrane Library from 1990. Review methods Studies were included that assessed hypertensive retinopathy with blinding for blood pressure and cardiovascular risk factors. Studies on observer agreement had to be assessed by two or more observers and expressed as a κ statistic. Studies on the association between hypertensive retinopathy and hypertensive organ damage were carried out in patients with hypertension. The association between hypertensive retinopathy and cardiovascular risk was carried out in unselected normotensive and hypertensive people without diabetes mellitus. Results The assessment of microvascular changes in the retina is limited by large variation between observers. The positive and negative predictive values for the association between hypertensive retinopathy and blood pressure were low (47% to 72% and 32% to 67%, respectively). Associations between retinal microvascular changes and cardiovascular risk were inconsistent, except for retinopathy and stroke. The increased risk of stroke, however, was also present in normotensive people with retinopathy. These studies did not adjust for other indicators of hypertensive organ damage. Conclusion Evidence is lacking that routine funduscopy is of additional value in the management of hypertensive patients.

Value of funduscopy

49

Introduction Since 1939 the Keith, Wagener, and Barker classification has been used to assess retinal changes associated with hypertension. From the 1980s onwards, screening for hypertension and antihypertensive treatment have improved considerably and the severity and frequency of retinal changes associated with hypertension seem to have diminished.1 2 In the recent guidelines of the Joint National Committee and the European Society of Hypertension, funduscopy is still recommended as part of the routine examination of hypertensive patients, although its value is being questioned.3 4 We carried out a systematic search to determine the usefulness of routine funduscopy in hypertensive patients. Firstly, we reviewed the literature on the reliability of detecting the various retinal changes associated with hypertension. Then we reviewed articles published since 1990 on the prevalence of retinal changes and their association with blood pressure, other indicators of hypertensive organ damage, and cardiovascular

morbidity and mortality. We excluded studies in patients with hypertensive urgencies and emergencies as this is a separate group in whom funduscopy provides essential information for diagnosis and treatment. Keith, Wagener, and Barker classification of hypertensive retinopathy Grade I—slight or modest narrowing of the retinal arterioles, with an arteriovenous ratio ≥ 1:2

Grade II—modest to severe narrowing of retinal arterioles (focal or generalised), with

an arteriovenous ratio < 1:2 or arteriovenous nicking

Grade III—bilateral soft exudates or flame-shaped haemorrhages

Grade IV—bilateral optic nerve edema

Methods We systematically searched Medline, Embase, and the Cochrane Library for articles on hypertension and retinopathy published since 1990 in adults aged 19 or more. Two investigators (BJHvdB and CAAH) independently carried out the search using the terms retinopathy, retinal arteriolar narrowing, arteriovenous

Chapter 2

50

nicking, AV nicking, arteriovenous ratio, or AV ratio, combined either with observer agreement, observer variability, or kappa, or with hypertension, blood pressure, microalbuminuria, left ventricular hypertrophy, intima-media thick-ness, stroke, myocardial infarction, ischaemic heart disease, or cardiovascular mortality. We use the term hypertensive retinopathy to describe all retinal

changes included in the Keith, Wagener, and Barker classification (box). We use the term retinal microvascular changes to refer to narrowing of the retinal arterioles and arteriovenous nicking (figs 1 and 2). The term retinopathy was originally used to denote soft exudates and flame-shaped haemorrhages only (fig 3), but in the studies reviewed here retinopathy also encompasses micro-aneurysms, blot haemorrhages, and hard exudates. We included studies on the reliability of detecting changes to the retina if retinal photographs were indepen-dently assessed by two or more observers and if variability of detecting retinal

changes associated with hypertension could be expressed as a κ statistic. For the association between retinal changes and hypertensive organ damage we included studies of hypertensive patients only. For the association between hypertensive retinopathy, cardiovascular risk, and blood pressure, we included population based or cohort studies with unselected normotensive and hypertensive people.

In all studies, retinal changes had to be assessed without knowledge of blood pressure and cardiovascular risk factors, and diabetes mellitus had to be exclu-ded, analysed separately, or adjusted for in multivariate analysis. Searches were carried out in the second half of October 2004. Statistical analysis We used κ statistics to express the proportion of agreement for the various retinal changes associated with hypertension, correcting for agreement expected by chance.5 We included both weighted and unweighted κ statistics, where weighted κ provides a scaled probability score so that the sum of the weight equals the number of observations. For the association between blood pressure and retinal changes we considered the various retinal changes as a diagnostic test

for hypertension. We calculated sensitivity, specificity, and positive and negative predictive values. Pooled odds ratios were calculated for the association between hypertensive retinopathy and hypertensive organ damage. We used a random effects model to generate summary measures and confidence intervals. Hetero-geneity was examined using the I2 statistic, which indicates the proportion of variability in the weighted mean differences attributable to heterogeneity.6 To assess the association between hypertensive retinopathy and cardiovascular risk, we took the relative risk after multivariate analysis.

Value of funduscopy

51

Fig. 1 Focal arteriolar narrowing

Fig. 2 Arteriovenous nicking

Fig. 3 Cotton wool spots and haemorrhages

Chapter 2

52

Results

We identified 111 reports with possible data on the association between hypertensive retinopathy and blood pressure, hypertensive organ damage, and cardiovascular risk (fig 4).

Fig. 4 Selection of studies included in review

Potentially relevant papers identified (n=2869)

Duplicates (n=848) Excluded reports on basis of title or abstract in most cases owing to lack of data on hypertensive retinopathy (n=1910)

Papers retrieved for more detailed analysis (n=111)

Excluded (n=93): No comparison between hypertensive retinopathy and blood pressure, hypertensive organ damage, or cardiovascular risk in people without diabetes (n=61) Description of hypertensive retinopathy in selected groups – for example, patients requiring haemodialysis, patients with malignant hypertension (n=18) Did not fulfill inclusion criteria either because assessment of hypertensive retinopathy and blood pressure was not blinded (n=3) or because relevant data were missing (n=3) Double publications (n=8)

Included reports (n=18)

Value of funduscopy

53

Variability in detecting retinal changes Six studies had data on interobserver agreement for hypertensive retinopathy using retinal photographs. We found no such data for direct funduscopy. One study was excluded because we could not calculate κ statistics.7 The most exten-sive study on observer agreement for hypertensive retinopathy was the athero-sclerosis risk in communities study in which three readers graded a random sample of 206 retinal photographs.8 Table 1 lists the κ statistics of this and other studies. In these studies, interobserver agreement was fair or moderate for focal arteriolar narrowing (κ 0.3 to 0.4) and arteriovenous nicking (κ 0.4 to 0.6). Agreement was excellent for haemorrhages and exudates. A computerised grading method was used to assess the arteriovenous ratio.9 With this method, agreement was good (κ 0.7 to 0.8).

Table 1 Kappa statistics for hypertensive retinopathy in studies included in systematic review.

Type of retinal change and study Unweighted κ Weighted κ Any retinopathy: Sharp et al10 Afro-Caribbeans 0.72

white Europeans 0.75 Not reported

Haemorrhages and exudates: Atherosclerosis risk in communities study8

Cardiovascular health study11

Blue Mountains eye study12

0.76 0.88 0.90

0.76 Not reported Not reported

Arteriovenous nicking: Atherosclerosis risk in communities study8 Cardiovascular health study11

0.59 0.43

0.56 Not reported

Focal arteriolar narrowing: Atherosclerosis risk in communities study8

Cardiovascular health study11

0.39 0.31

0.29 Not reported

Arteriovenous ratio*: Atherosclerosis risk in communities study8

Cardiovascular health study11 Blue Mountains eye study13

0.73 0.81 Not reported

0.79 Not reported 0.75

*Computer assisted grading method.

Chapter 2

54

Retinal changes and blood pressure Four large population based studies examined the association between hypertensive retinopathy and blood pressure.1 12 14-16 Hypertension was defined as the use of antihypertensive drugs, or history of hypertension or blood pressure ≥ 140/90 mmHg in the cardiovascular health study and atherosclerosis risk in

communities study or ≥ 160/95 mmHg in the Beaver Dam eye study and Blue Mountains eye study. In one study, age and sex standardised prevalence of hypertensive retinopathy was 12% in white Europeans and 18% in people of

Afro-Caribbean origin.10 Table 2 lists the prevalence of retinal abnormalities for the other studies. The sensitivity for hypertensive retinopathy ranged from 3% to 21%, indicating a low prevalence of retinal abnormalities in hypertensive patients. In contrast, specificity ranged from 88% to 98%, and therefore retinopathy was rarely observed in normotensive patients. The positive predict-tive value, or the chance of hypertension given the presence of hypertensive retinopathy, ranged from 47% to 70% for haemorrhages and exudates, 53% to

66% for arteriovenous nicking, and 49% to 72% for focal arteriolar narrowing (table 3). The negative predictive value, or the chance of normotension in the absence of hypertensive retinopathy, ranged from 43% to 67% for haemorrhages and exudates, 44% to 66% for arteriovenous nicking, and 32% to 59% for focal arteriolar narrowing. Diagnostic accuracy could not be assessed for the arterio-venous ratio as normal and abnormal values were not defined. Retinal changes and other indicators of organ damage Four studies examined the association between hypertensive retinopathy and echocardiographically determined left ventricular hypertrophy.17-20 One study failed to give the number of patients with events; however, the age and sex adjusted odds ratio (95% confidence interval) for left ventricular hypertrophy was 1.92 (1.03 to 3.60) when retinopathy was present.21 Figure 2 gives details of the other studies. The pooled odds for left ventricular hypertrophy in the presence of hypertensive retinopathy was 2.22 (1.36 to 3.62), with little heterogeneity (I2 = 29.1%). Two studies examined the association between hypertensive retinopathy and microalbuminuria (odds ratios 1.51, 0.84 to 2.68 and 4.98, 1.97 to 12.60).18 22 We were unable to pool data because of significant heterogeneity (I2 = 78.3%). Three population based studies reported on the

association between hypertensive retinopathy and intima-media thickness.11 23 24 In the cardiovascular health study and atherosclerosis risk in communities study only the presence of haemorrhages and exudates was associated with increased intima-media thickness. In the Rotterdam eye study a lower arteriovenous ratio

Value of funduscopy

55

Tab

le 2

Cha

ract

eris

tics o

f stu

dies

of r

etin

al c

hang

es in

pat

ient

s with

hyp

erte

nsiv

e re

tinop

athy

. Val

ues a

re n

umbe

rs

(per

cent

ages

) unl

ess s

tate

d ot

herw

ise

Ret

inal

cha

nges

Car

diov

ascu

lar

205

6

6

9-97

142

4 (5

9%)

9

.6

7.6

8.3

he

alth

stud

y14

B

eave

r Dam

4

311

43-

84

1

479

(34%

)

13.5

2

.2

7

.8

eye

stud

y1

Ath

eros

cler

osis

1

0358

48

-73

459

5 (4

4%)

14

.9

14.3

7

.0

risk

in

com

mun

ities

st

udy‡

15

B

lue

Mou

ntai

ns

3614

43

-86

165

6 (4

6%)

7.9

8

.9

N

ot re

porte

d ey

e st

udy16

Blu

e M

ount

ains

327

5

≥

49

14

47 (4

4%)

N

ot re

porte

d

N

ot re

porte

d

9.9

ey

e st

udy12

*Use

of a

ntih

yper

tens

ive

drug

s or a

his

tory

of h

yper

tens

ion

and

bloo

d pr

essu

re ≥

140

/90

mm

Hg

in c

ardi

ovas

cula

r he

alth

stud

y an

d at

hero

scle

rosi

s ris

k in

com

mun

ities

stud

y an

d ≥

160/

95 m

mH

g in

the

Bea

ver D

am e

ye st

udy

and

Blu

e M

ount

ains

eye

stud

y. †

Diff

eren

t def

initi

ons u

sed.

‡In

clud

ed p

atie

nts w

ith d

iabe

tes.

Stu

dy

No

of

Age

ran

ge

Hyp

erte

nsio

n

Foc

al

A

rter

io

H

aem

orrh

ages

subj

ects

(y

ears

)

(%

)*

a

rter

iola

r

ven

ous

and

exu

date

s

nar

row

ing

n

icki

ng

(

%)

(%

)

(%

)†

Chapter 2

56

Table 3 D

iagnostic accuracy (%) of various retinal changes for hypertension included in system

atic review

Cardiovascular 12

94 72 46 9 94 66 44 9 94 70 43 health study

14 B

eaver Dam

19 89 49 68 3

98 53 66 11

94 47 67 eye study

1 A

therosclerosis 21 89 61

59 18

88 55 57 10

95 62 57 risk in com

munities

study*15

Blue 11

95 66 52 12

93 59 56 not not not not

Mountains reported

reported reported reported eye study

16

Blue not not not not not not not not 13 93 58 57

Mountains reported

reported reported reported reported reported reported reported

eye study*12

*Peoplew

ithdiabetesincluded.

Study Generalised or focal arteriolar narrow

ing Arteriovenous nicking H

aemorrhages and E

xudates Sensitivity Specificity Positive N

egative Sensitivity Specificity Positive Negative Sensitivity Specificity Positive N

egative predictive predictive predictive predictive predictive predictive

value value

value value value value

Value of funduscopy

57

Fig.

2

Odd

s rat

ios (

95%

con

fiden

ce in

terv

als)

for a

ssoc

iatio

n be

twee

n hy

perte

nsiv

e re

tinop

athy

and

left

vent

ricul

ar

hype

rtrop

hy

Chapter 2

58

T

able 4 Associations betw

een retinal changes and cardiovascular risk in studies included in systematic review

. Values are relative risks (95%

confidence intervals) unless stated otherw

ise

B

eaver Dam

Case control 10 D

eath from 1611 1.0 (0.5 to 1.9)§ 1.4 (0.8 to 2.4)§ 1.5 (1.1 to 2.1)§ 1.8 (1.2 to 2.7)§

Not reported

eye

coronary heart

study‡27 disease or stroke

Atherosclerosis Population 3

Stroke 10358 1.6 (1.0 to 2.5) 1.1 (0.7 to 1.8) 1.2 (0.7 to 2.3) 2.6 (1.6 to 4.2) N

ot reported risk in

based

comm

unities study‡

15

Atherosclerosis Population 3 C

oronary 9648 Not reported N

ot reported Men M

en Not reported

risk in based heart 1.1 (0.7 to 1.8) 1.1 (0.6 to 2.3)

comm

unities

disease

w

omen

wom

en study‡

28

2.2 (1.0 to 4.6) 1.8 (0.8 to 4.2) C

ardiovascular Cross —

Coronary 2050 1.1 (0.7 to 1.6) 0.9 (0.6 to 1.4) 0.8 (0.6 to 1.0) 1.7 (1.2 to 2.6) N

ot reported health sectional heart

study¶**11 disease

Cardiovascular C

ross — Stroke 2050 1.4 (0.7 to 2.6) 1.2 (0.6 to 2.4) 1.1 (0.7 to 1.8) 2.0 (1.1 to 3.6)

Not reported

health sectional

study¶**11

Shibata Population 15.5 Stroke 2302 N

ot reported Not reported N

ot reported N

ot reported M

en study‡

29 based

3.4 (1.0 to 11.3) *Low

est versus highest quintile of the arteriovenous ratio, using computerised m

ethod. †Keith, W

agener, and Barker classification.

‡Multivariate analysis adjusted for age, ethnicity, blood pressure or use of antihypertensive drugs, diabetes, cholesterol, and sm

oking. §O

dds ratios (95% C

Is).¶Multivariate analysis adjusted for age, ethnicity, and blood pressure or use of antihypertensive drugs.

**People with diabetes excluded.

Study Design Follow

End points N

o of Arteriovenous Focal G

eneralised Haem

orrhages Any retinopathy

up

participants crossings arteriolar arteriolar and exudates grade II

(years)

narrowing narrow

ing*

or higher) †

Value of funduscopy

59

was associated with an increased intima-media thickness; other retinal abnor-malities were not reported. We were unable to compare the studies because

different techniques were used to assess the associations. Retinal changes and cardiovascular risk Hypertensive retinopathy as an independent predictor of cardiovascular risk was examined in six different populations, totalling 16 000 participants. We excluded two studies because blood pressure and patient history were known to the inves-tigators before funduscopy.25 26 Table 4 lists the association between hyper-tensive retinopathy and cardiovascular risk after adjustment in multivariate analysis. Risks of coronary heart disease, stroke, or death from cardiovascular

disease differed with regard to retinal microvascular changes. Data on the association between retinopathy and stroke were more consistent: relative risk of stroke 3.4 (1.0 to 11.3) for men in the Shibata study and 2.6 (1.6 to 4.2) for men and women in the atherosclerosis risk in communities study. In the cardio-vascular health study the odds ratio for stroke was 2.0 (1.1 to 3.6).

In the atherosclerosis risk in communities study and the Beaver Dam eye study, associations between retinal abnormalities and stroke were stratified by hypertension status. The relative risk of stroke in the atherosclerosis risk in communities study was 2.6 (1.2 to 5.6) with retinopathy and hypertension compared with 2.0 (0.5 to 8.4) in normotensive people. In the Beaver Dam eye study, the odds ratio for death from cardiovascular disease in people aged 43-74 with retinopathy and hypertension or diabetes, or both was 2.3 (1.2 to 4.2) compared with 1.5 (0.4 to 5.3) for those without hypertension and diabetes. The

odds ratio for those aged 75-84 was 2.0 (1.0 to 4.0) if retinopathy and hypertension or diabetes, or both were present and 1.8 (0.4 to 8.4) if only retinopathy was present.

Discussion In this systematic review we found limited additional value of funduscopy in the routine management of patients with hypertension, except in emergency cases. The evaluation of hypertensive retinopathy is subject to large variability between observers, especially microvascular changes. Only haemorrhages and exudates can be reliably assessed in retinal photographs. We found no data on observer agreement using direct funduscopy. Recently, a computerised grading method has been developed to increase reproducibility of retinal microvascular changes, but this method is not widely available.

Chapter 2

60

The low sensitivity of retinal abnormalities associated with hypertension indicates that hypertensive retinopathy is not common in hypertensive people. Less than half the retinal changes associated with hypertension cannot be explained by high blood pressure (low positive predictive value). In both the Beaver Dam eye study and the Blue Mountains eye study little difference was found in the presence of haemorrhages and exudates between normotensive and hypertensive people aged over 65.1 30 Various other conditions have been asso-ciated with hypertensive retinopathy, such as ethnicity,10 21 smoking,23 31 32 intima-media thickness,11 23 31 carotid plaque score,31 33 carotid artery stiffness,33 serum cholesterol concentration,23 27 32 diabetes,27 32 and body mass index.27 32 The high specificity indicates that hypertensive retinopathy is rare in patients with normal blood pressure. Half the people without hypertensive retinopathy, however, still have hypertension (low negative predictive value). So funduscopy

cannot reliably determine whether a patient is normotensive or hypertensive. The presence of hypertensive retinopathy doubles the risk of left ventricular hypertrophy. Data on the association between hypertensive retinopathy and microalbuminuria are inconsistent. Patients with haemorrhages and exudates have a higher intima-media thickness; for other retinal abnormalities this asso-ciation is conflicting or absent.

Evidence that hypertensive retinopathy is an independent predictor of cardiovascular risk is based on two large studies in which hypertensive patients with retinal haemorrhages or exudates had a twofold higher risk of stroke. Neither study, however, examined or adjusted for other indicators of hypertensive organ damage (left ventricular hypertrophy, microalbuminuria, and

intima-media thickness). Furthermore, the association between retinopathy and stroke was also observed in normotensive people, suggesting that besides blood pressure, other factors are responsible for these retinal changes. A recent paper reported on the associations between hypertensive retinopathy, blood pressure, and mortality from cardiovascular disease.34 The authors found a strong association between hypertensive retinopathy and hypertension. Although this association was significant in most studies, the low predictive values we calculated indicate that funduscopy cannot reliably determine whether a patient has hypertension. Only for stroke is the association with retinopathy significant and consistent. This association is also observed in people without hypertension and is not corrected for other indicators of hypertensive organ damage. Hence the additional value of funduscopy in the management of patients with hypertension still needs to be determined.

Value of funduscopy

61

What is already known on this topic Funduscopy is recommended in the routine management of hypertensive patients The usefulness of funduscopy is being questioned What this study adds Funduscopy is of limited additional value in hypertensive patients, unless a hypertensive emergency is suspected

The included studies have several limitations. Firstly, in all the studies hypertensive retinopathy was established on a photograph of one eye. As hypertension is a systemic disease, retinal changes would be expected in both eyes. Some retinal abnormalities may thus have been a sign of unilateral eye disease instead of hypertension. Secondly, most of the studies used a broad definition for hypertensive retinopathy, including microaneurysms, hard exudates, and blot haemorrhages. These are not included in the original classification and may have a different association with hypertension. Thirdly,

hypertension may have been misclassified, as blood pressure was measured only once or twice. Fourthly, people with diabetes were included in the athero-sclerosis risk in communities study and Beaver Dam eye study. Diabetic retinopathy is a known risk factor for cardiovascular disease and may have confounded the association between hypertensive retinopathy and cardiovascular

risk. Multivariate analysis may not have totally accounted for this effect. Finally, only middle aged and older people were included in the studies, although the association between hypertensive retinopathy, blood pressure, and risk of cardiovascular disease gets stronger in those under 60.27 Future research should consider a younger population (< 60 years) and other indicators of hypertensive organ damage.

References

1. Klein R, Klein BE, Moss SE, Wang Q. Hypertension and retinopathy, arteriolar narrowing, and arteriovenous nicking in a population. Arch Ophthalmol 1994;112: 92-8.

2. Klein R. Retinopathy in a population-based study. Trans Am Ophthalmol Soc 1992; 90: 561-94.

Chapter 2

62

3. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, et al. The seventh report of the joint national committee on prevention, detection, evaluation, and treatment of high blood pressure: the JNC 7 report. JAMA 2003;289: 2560-72.

4. Cifkova R, Erdine S, Fagard R, Farsang C, Heagerty AM, Kiowski W, et al. Practice guidelines for primary care physicians: 2003 ESH/ESC hypertension guidelines. J Hypertens 2003;21: 1779-86.

5. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33: 159-74.

6. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327: 557-60.

7. Cuspidi C, Salerno M, Salerno DE, Meani S, Valerio C, Esposito A, et al. High prevalence of retinal vascular changes in never-treated essential hypertensives: an inter- and intra-observer reproducibility study with non-mydriatic retinography. Blood Press 2004;13: 25-30.

8. Couper DJ, Klein R, Hubbard LD, Wong TY, Sorlie PD, Cooper LS, et al. Reliability of retinal photography in the assessment of retinal microvascular characteristics: the atherosclerosis risk in communities study. Am J Ophthalmol 2002;133: 78-88.

9. Hubbard LD, Brothers RJ, King WN, Clegg LX, Klein R, Cooper LS, et al. Methods for evaluation of retinal microvascular abnormalities associated with hypertension/sclerosis in the atherosclerosis risk in communities study. Ophthalmology 1999;106: 2269-80.

10. Sharp PS, Chaturvedi N, Wormald R, McKeigue PM, Marmot MG, Young SM. Hypertensive retinopathy in Afro-Caribbeans and Europeans. Prevalence and risk factor relationships. Hypertension 1995;25: 1322-5.

11. Wong TY, Klein R, Sharrett AR, Manolio TA, Hubbard LD, Marino EK, et al. The prevalence and risk factors of retinal microvascular abnormalities in older persons: the cardiovascular health study. Ophthalmology 2003;110: 658-66.

12. Yu T, Mitchell P, Berry G, Li W, Wang JJ. Retinopathy in older persons without diabetes and its relationship to hypertension. Arch Ophthalmol 1998;116: 83-9.

13. Leung H, Wang JJ, Rochtchina E, Tan AG, Wong TY, Klein R, et al. Relationships between age, blood pressure, and retinal vessel diameters in an older population. Invest Ophthalmol Vis Sci 2003;44: 2900-4.

Value of funduscopy

63

14. Wong TY, Hubbard LD, Klein R, Marino EK, Kronmal R, Sharrett AR, et al. Retinal microvascular abnormalities and blood pressure in older people: the cardiovascular health study. Br J Ophthalmol 2002;86: 1007-13.

15. Wong TY, Klein R, Couper DJ, Cooper LS, Shahar E, Hubbard LD, et al. Retinal microvascular abnormalities and incident stroke: the atherosclerosis risk in communities study. Lancet 2001;358: 1134-40.

16. Wang JJ, Mitchell P, Leung H, Rochtchina E, Wong TY, Klein R. Hypertensive retinal vessel wall signs in a general older population: the Blue Mountains eye study. Hypertension 2003;42: 534-41.

17. De Leonardis V, Becucci A, De Scalzi M, Cinelli P. Low incidence of cardiac hypertrophy in essential hypertensives with no retinal changes. Int J Cardiol 1992;35: 95-9.

18. Pontremoli R, Sofia A, Ravera M, Nicolella C, Viazzi F, Tirotta A, et al. Prevalence and clinical correlates of microalbuminuria in essential hypertension: the MAGIC Study. Microalbuminuria: a Genoa investigation on complications. Hypertension 1997;30: 1135-43.

19. Pose-Reino A, Gonzalez-Juanatey JR, Pastor C, Mendez I, Estevez JC, Alvarez D, et al. Clinical implications of white coat hypertension. Blood Press 1996;5: 264-73.

20. Saitoh M, Matsuo K, Nomoto S, Kondoh T, Yanagawa T, Katoh Y, et al. Relationship between left ventricular hypertrophy and renal and retinal damage in untreated patients with essential hypertension. Intern Med 1998;37: 576-80.

21. Wong TY, Klein R, Duncan BB, Nieto FJ, Klein BE, Couper DJ, et al. Racial differences in the prevalence of hypertensive retinopathy. Hypertension 2003;41: 1086-91.

22. Biesenbach G, Zazgornik J. High prevalence of hypertensive retinopathy and coro-nary heart disease in hypertensive patients with persistent microalbuminuria under short intensive antihypertensive therapy. Clin Nephrol 1994;41: 211-8.

23. Ikram MK, de Jong FJ, Vingerling JR, Witteman JC, Hofman A, Breteler MM, et al. Are retinal arteriolar or venular diameters associated with markers for cardiovascular disorders? The Rotterdam Study. Invest Ophthalmol Vis Sci 2004;45: 2129-34.

24. Wong TY, Klein R, Sharrett AR, Couper DJ, Klein BE, Liao DP, et al. Cerebral white matter lesions, retinopathy, and incident clinical stroke. JAMA 2002;288: 67-74.

Chapter 2

64

25. Duncan BB, Wong TY, Tyroler HA, Davis CE, Fuchs FD. Hypertensive retinopathy and incident coronary heart disease in high risk men. Br J Ophthalmol 2002;86: 1002-6.

26. Gillum RF. Retinal arteriolar findings and coronary heart disease. Am Heart J 1991;122: 262-3.

27. Wong TY, Klein R, Nieto FJ, Klein BE, Sharrett AR, Meuer SM, et al. Retinal microvascular abnormalities and 10-year cardiovascular mortality: a population-based case-control study. Ophthalmology 2003;110: 933-40.

28. Wong TY, Klein R, Sharrett AR, Duncan BB, Couper DJ, Tielsch JM, et al. Retinal arteriolar narrowing and risk of coronary heart disease in men and women. The atherosclerosis risk in communities study. JAMA 2002;287: 1153-9.

29. Nakayama T, Date C, Yokoyama T, Yoshiike N, Yamaguchi M, Tanaka H. A 15.5-year follow-up study of stroke in a Japanese provincial city. The Shibata study. Stroke 1997;28: 45-52.

30. Yu T, Mitchell P, Berry G, Li W, Wang JJ. Retinopathy in older persons without diabetes and its relationship to hypertension. Arch Ophthalmol 1998;116: 83-9.

31. Klein R, Sharrett AR, Klein BE, Chambless LE, Cooper LS, Hubbard LD, et al. Are retinal arteriolar abnormalities related to atherosclerosis?: the atherosclerosis risk in communities study. Arterioscler Thromb Vasc Biol 2000;20: 1644-50.

32. Wong TY, Knudtson MD, Klein R, Klein BE, Hubbard LD. A prospective cohort study of retinal arteriolar narrowing and mortality. Am J Epidemiol 2004;159: 819-25.

33. Liao D, Wong TY, Klein R, Jones D, Hubbard L, Sharrett AR. Relationship between carotid artery stiffness and retinal arteriolar narrowing in healthy middle-aged persons. Stroke 2004;35: 837-42.

34. Wong TY, Mitchell P. Hypertensive retinopathy. N Engl J Med 2004;351: 2310-7.

Chapter 3

Ethnic disparities in the incidence, presentation and complications of malignant hypertension Bert-Jan H. van den Born, Richard P. Koopmans, Johan O. Groeneveld and Gert A. van Montfrans

J Hypertens. 2006; 24: 2299-304.

Chapter 3

68

Abstract Background The incidence of malignant hypertension has declined after the introduction of antihypertensive agents. However, previous reports have sugges-ted that malignant hypertension may be relatively common in multi-ethnic populations. The aim of this study was to compare ethnic disparities in the incidence, clinical characteristics and complications of malignant hypertension. Methods A retrospective cohort study on malignant hypertension in a multi-ethnic population in Amsterdam, the Netherlands, between August 1993 and August 2005. Results A total of 122 patients with malignant hypertension were included, mean age 44 years (± 12), 66% were men and 47% were black. The incidence rate remained approximately 2.6 (± 0.9) per 100 000 per year and was higher among blacks. Black individuals had higher systolic blood pressure (234 ± 23 versus 225 ± 22, P = 0.03) and more renal dysfunction compared with white individuals (39% with serum creatinine > 300 µmol/l versus 22%, P = 0.04). Hypertension was previously diagnosed in 58% of all patients, 37% received medication, and 23% stopped their drugs before admission. Health insurance was absent in 25% of black and 2% of white patients (P < 0.01). Secondary causes were identified in 40% of white and 10% of black subjects (P < 0.01). After a mean follow-up of 4.0 ± 3.2 years 10% had died and 19% needed renal replacement therapy. Renal failure was more frequent in black than in white individuals (hazard ratio 2.8; 95% confidence interval 1.1–7.2), but mainly because of higher serum creatinine levels at presentation. Conclusion The incidence of malignant hypertension and related renal complications is higher in black compared with white individuals. These differences may be explained by ethnic disparities in blood pressure control, drug adherence and insurance status.

Ethnic disparities in malignant hypertension

69

Introduction Malignant hypertension is a hypertensive emergency characterized by the presence of severe hypertension and ischaemic retinal changes consistent with grades III or IV hypertensive retinopathy according to the classification of Keith, Wagener and Barker.1 Since the introduction of effective antihyper-tensive agents five decades ago, studies have shown that the survival of malignant hypertension has considerably improved. However, despite better awareness, screening and treatment possibilities for hypertension, some studies have reported that the incidence of malignant hypertension has not declined. In the west Birmingham (UK) cohort, the number of patients admitted annually with malignant hypertension remained practically unchanged between 1970 and 1993, with an estimated incidence of 1–2 cases per 100 000 per year, whereas other studies reported a decline during the same period.2-4 This difference may be explained by differences in the ethnic composition of the study population. The west Birmingham study consisted of a multi-ethnic population, whereas other studies dealt with a predominantly white population. In the west Birmingham study there was an excess of Asian and black immigrants presenting with malignant hypertension compared with the population served by that hospital. Moreover, black patients had a poorer outcome with regard to renal function and survival in comparison with other ethnic groups. Late presen-tation, inadequate antihypertensive therapy, and poor compliance have been implicated to explain the excess mortality and renal dysfunction in this group. The aim of this study was to examine ethnic disparities in the incidence, clinical characteristics and complications of malignant hypertension. Therefore, we retrospectively analysed all patients admitted at two large teaching hospitals catering for a multi-ethnic community in Amsterdam in the past 12 years.

Patients and methods A retrospective analysis was carried out using data from two large teaching hospitals with adjacent catchment areas, the Academic Medical Centre and the Onze Lieve Vrouwe Gasthuis in Amsterdam, the Netherlands. Both hospitals cater for a large multi-ethnic community of 380 000 people, with approximately 65 000 black individuals, and use the same database in which the diagnosis at discharge is recorded according to the International Classification of Diseases (ICD) codes.

Chapter 3

70

All charts of patients admitted between August 1993 and August 2005 with the diagnosis ‘essential malignant hypertension’ (ICD 401.0), ‘hypertensive ence-phalopathy’ (ICD 437.2), ‘secondary malignant hypertension’ (ICD 405.09), ‘hypertension with cardiac disease/ malignant’ (ICD 402.0), and ‘hypertension with kidney disease/malignant’ (ICD 403.0) were reviewed. To identify the presence of registration errors, computer data for all patients discharged with the diagnosis ‘essential hypertension’ (ICD 401.9) were also analysed, which showed the presence of one patient with malignant hypertension. In all cases, the diagnosis of malignant hypertension was verified using the following preset criteria: high blood pressure together with bilateral linear or flame-shaped haemorrhages, or ‘cotton-wool’ exudates with or without papilledema on funduscopic examination. Excluded from analysis were patients less than 18 years of age, pregnant women, patients with papilledema and an intracranial mass (haemorrhage), patients already on dialysis before admission, and patients referred from elsewhere to prevent referral bias.

For follow-up, data were collected from the outpatients department or from general practitioners. All living patients received an invitation to participate in a study on the genetic aspects of malignant hypertension. Information obtained from these interviews (for example with regard to smoking habits or the previous use of antihypertensive therapy) were used to add to or verify any information that was missing or incomplete. Ethnicity was defined as self-reported black or self-reported white. Black individuals were mainly from western Africa (Ghana and Nigeria) and creoles from Surinam or the Dutch Antilles. White individuals were predominantly inborn Dutch; seven patients (6%) were originally from the Asian subcontinent (India and Pakistan), and were, for this study, classified as white. Diabetes was defined as serum glucose greater than 9 mmol/l or receiving glucose-lowering therapy on admission. Previous hypertension was defined as the self-reporting of hypertension assessed by a healthcare professional or receiving antihypertensive drugs. Blood pressure control was defined as a blood pressure less than 140 mmHg systolic and less than 90 mmHg diastolic. Macroalbuminuria was defined as urinary protein excretion greater than 300 mg/l or 2+ for dipstick proteinuria. Thrombotic microangiopathy (TMA) related to malignant hypertension was defined as thrombocytopenia (T < 150 x 109/l) with either an increased lactate dehydrogenase level or the presence of schistocytes, and the resolution of these parameters after blood pressure control. Left ventricular hypertrophy was assessed by electrocardiography using the Sokolow–Lyon criteria. Hypertensive


71

encephalopathy was defined as the presence of clouding of consciousness, convulsions or delirium at admission, with resolution after blood pressure control. Significant renal artery stenosis was defined if narrowing of the renal artery lumen exceeded 50% on computed tomography, magnetic resonance imaging or conventional angiography. Renal failure was defined as the need for (permanent) renal replacement therapy. Statistics To calculate incidence, the number of patients admitted annually with malignant hypertension was divided by the population of the catchment area of both hospitals. Baseline variables were described using mean and standard deviation (SD) or median and range, or interquartile range (IQR) for variables with a skewed distribution. Differences between groups were calculated using an independent samples t-test for parametric and Mann–Whitney U test for non-parametric distributions when appropriate. Categorical variables were calculated using chi-squared statistics with Yates’ correction or Fisher’s exact test for cells with expected values less than 5 (all two-sided). To analyse indicators of renal failure, Cox proportional hazard analysis was used, with secondary causes of malignant hypertension expressed as a dummy variable and creatinine in a logarithmic transformation. First, single variables were used to test an asso-ciation with renal failure. The two strongest single variables that showed a significant partial correlation with renal failure were tested and used for the model with age, sex and ethnicity. Data on secondary malignant hypertension were pooled because of the low frequency of some secondary causes. For statistical analysis, the SPSS software package for windows was used, version 12.0 (SPSS Inc., Chicago, Illinois, USA). P < 0.05 was considered to indicate a statistically significant difference.

Results A total of 140 consecutive patients admitted between August 1993 and August 2005 fulfilled the WHO criteria for malignant hypertension. Eleven patients were excluded because they were referred from elsewhere, three were excluded because they were on renal replacement therapy before presenting with malignant hypertension, four patients were excluded because of an intracranial mass as an alternative explanation for the presence of optic nerve edema, leaving 122 patients for analysis, with a mean age of 44 ± 12 years, 81 (66%) were men and 57 (47%) were black. The average annual incidence rate of malignant

Chapter 3

72

hypertension was 2.6 (± 0.9) per 100 000 per year, with no apparent decline in the past 12 years. During this period the annual distribution of blacks and whites remained essentially unchanged (Fig. 1). Compared with the demographic characteristics of the population, with 65 000 black and 315 000 white inhabitants, the incidence of malignant hypertension was 7.3 (± 4.6) per 100 000 for black and 1.7 (± 0.7) per 100 000 for white individuals. Fig. 1

Absolute number of admissions for malignant hypertension in the past 12 years. Data are stratified for ethnicity. Baseline characteristics according to ethnicity are listed in Table 1. Black individuals had higher systolic blood pressure and a tendency towards higher diastolic blood pressure at presentation. The severity of retinopathy was similar in both groups. Smoking was more common among white than among black individuals. Hypertension was previously recognized in 58% of all patients, and 37% had received antihypertensive therapy at any time before admission. However, 23% of all patients (62% of those who had received antihypertensive


73

treatment) admitted having stopped their antihypertensive drugs before ad-mission. Renal function at presentation was more severely compromised in black compared with white individuals, 39% with serum creatinine greater than 300 µmol/l compared with 22%, (P = 0.04), respectively, and there was a tendency towards more proteinuria. TMA was more frequent in black individuals and left ventricular hypertrophy more common, although the latter did not reach statistical significance. Table 1 Clinical characteristics of all included patients according to their ethnic background defined as black versus white

Values are numbers unless specified otherwise. IQR, Interquartile range. *Data are missing for one in the group defined as white. †Data are missing for 3 in the group defined as white and for 1 in the group defined as black. ‡Data are missing for 2 in the group defined as white and 3 in the group defined as black.

Patient characteristics Black White P value Patients 57 (47%) 65 (53%) Age, yrs (mean ± SD) 42 ± 10 45 ± 14 0.23 Male 41 (72%) 40 (62%) 0.31 Retinopathy, grade IV 28 (49%) 32 (49%) 0.86 Systolic blood pressure, mmHg (mean ± SD)

234 ± 23 225 ± 22 0.03

Diastolic blood pressure, mmHg (mean ± SD)

148 ± 16 143 ± 16 0.11

Current Smoker* 16 (28%) 31 (48%) 0.04 Diabetes mellitus 4 (7%) 6 (9%) 0.75 Previous hypertension 32 (56%) 39 (60%) 0.80 Antihypertensive drugs ≥ 1 22 (39%) 23 (35%) 0.86 Stopped taking anti-hypertensives 18 (32%) 10 (15%) 0.06 No medical insurance 14 (25%) 1 (2%) < 0.01 Left ventricular hypertrophy 51 (90%) 50 (77%) 0.11 Serum creatinine, µmol/l (median, [IQR])

184 [129-504] 122 [84-266] < 0.01

Macroalbuminuria† 38 (68%) 34 (55%) 0.21 Thrombotic microangiopathy‡ 25 (46%) 9 (14%) < 0.01 Cerebral complications cerebral infarction 2 3 posterior leucoencephalopathy 1 4

Chapter 3

74

Imaging studies of the renal arteries were performed in 96 (79%), renal biopsies in 22 (18%), and one or more laboratory tests were performed for pheochromocytoma, Cushing’s disease and Conn’s syndrome in 60 patients (49%). Renal biopsy specimens showed glomerulosclerosis in 10 patients, fibrinoid necrosis in five, glomerulopathy in four (IgA nephropathy in two patients, HIV-related nephropathy and membranous glomerulopathy in one patient), and glomerulonephritis in three patients (systemic lupus erythematosus-related glomerulonephritis in one, and interstitial glomerulonephritis in two patients). Analysis for secondary causes of hypertension was as often carried out in black and white individuals (data not shown). In 26% of all patients a secondary cause of malignant hypertension was established. Table 2 lists the patients with secondary malignant hypertension stratified for ethnicity. The prevalence of secondary causes of malignant hypertension was higher in white than in black individuals, mainly because of more primary renal and renovascular disease. Endocrine disease as a cause of malignant hypertension was present in only one patient, presenting with a malignant adrenal cortical carcinoma. Table 2 Probable secondary causes of malignant hypertension according to ethnic background (black versus white)

Probable cause Black White P value No cause identified 51 (90%) 39 (60%) < 0.01 Renal parenchymal disease 2 (4%) 12 (19%) 0.02 Renal artery stenosis 1 (2%) 7 (11%) 0.07 Other* 3 (5%) 7 (11%) 0.33

Values are numbers and percentages. *Other includes licorice abuse (4), corticoid excess (3), cocaine (2), oral contraceptives (1). The clinical presentation of black and white subjects was similar, although headache was more frequent in black individuals at presentation (Table 3). Either headache or visual disturbances were present in 90 patients (74%). Other common symptoms at presentation included gastrointestinal complaints in 48 patients (39%), mainly nausea and vomiting, sometimes abdominal pain and weight loss. Dyspnoea or peripheral edema as evidence of congestive heart failure was present in 25 patients (20%), and somnolence, delirium or convul-


75

sions consistent with hypertensive encephalopathy was seen in 11 patients (9%). Less frequent complaints were dizziness (11 patients, 9%) and chest pain (eight patients, 7%). Six patients (5%) had no symptoms at presentation. Complete follow-up was available for 109 patients (89%), 13 either moved or did not show up at follow-up visits. Those who were lost to follow-up did not differ with regard to age (42 ± 8 years), sex (85% male, P = 0.21), ethnicity (54% black) or renal function [serum creatinine 215 µmol/l (IQR 116–541, P = 0.24)].

After a mean follow-up time of 4.0 ± 3.2 years, 97 patients (90%) were alive and 12 patients had died (10%). Four died of cardiovascular complications: two of intracerebral bleeding, one of acute myocardial infarction and one after cardiac surgery. Dialysis was needed in 21 out of 109 patients (19%). Blood pressure at the latest visit was higher in black patients, and they were less likely to have their blood pressure controlled than white patients. The characteristics of those with complete follow-up are listed in Table 4. The two strongest predictors of renal failure were baseline serum creatinine and secondary causes of malignant hypertension. Cox proportional hazard analysis showed that black individuals had a greater risk of renal failure than white subjects, with an unadjusted hazard ratio of 2.8 [95% confidence interval (CI) 1.1–7.2]. Figure 2 shows the proportion with renal failure stratified for ethnicity. The adjusted hazard ratio after correction for age, sex and secondary causes of malignant hypertension was 4.3 (95% CI 1.4–13). When correction for baseline creatinine values was carried out, however, the correlation between ethnicity and renal failure was not significant (Table 5). Table 3 Clinical presentation of patients admitted with malignant hypertension.

Clinical presentation Black White P value Headache 40 (70%) 33 (51%) 0.05 Visual disturbance 32 (56%) 36 (55%) 1.0 Gastro-intestinal complaints 25 (44%) 23 (35%) 0.44 Congestive heart failure 12 (21%) 13 (20%) 1.0 Hypertensive encephalopathy 3 (5%) 8 (12 %) 0.30

Values are numbers and percentages.

Chapter 3

76

Table 4 Patient characteristics on follow-up.

Patient characteristics at follow-up Black White P value Lost to follow-up 7 (12%) 6 (9%) 0.82 Follow-up time, years (mean ± SD) 4.5 ± 3.0 3.6 ± 3.4 0.16 Systolic blood pressure, mmHg (mean ± SD)

145 ± 20 134 ± 20 < 0.01

Diastolic blood pressure, mmHg (mean ± SD)

91 ± 13 83 ± 14 < 0.01

Antihypertensive drugs (mean ± SD) 3.3 ± 1.1 2.9 ± 1.3 0.06 Controlled 16 (32%) 33 (56%) 0.02 Dialysis 15 (30%) 6 (10%) 0.02 Died 5 (10%) 7 (12%) 1.0

Values are numbers and percentages unless stated otherwise. Table 5 Cox proportional hazard analysis to predict indicators for renal failure.

A. Variable HR 95% CI P value Age (per year) 0.96 0.91-1.00 0.07 Male 1.15 0.45-2.96 0.77 Black 4.30 1.39-13.3 0.01 Secondary causes 3.18 1.03-9.82 0.04 B. Variable Age (per year) 0.98 0.93-1.03 0.40 Male 0.85 0.29-2.35 0.76 Black 3.10 0.81-11.9 0.10 Secondary causes 3.78 1.01-14.1 0.05 Log (creatinine) 48.3 12.5-187 < 0.01

CI, confidence interval; HR, hazard ratio. Analysis performed on 118 cases with 21 events. A. Model with age, gender, ethnicity and secondary causes. B. Model with indi-cators of model A plus serum creatinine at admission in a logarithmic transformation.


77

Fig. 2

Proportion with renal failure stratified for ethnicity. Analysis performed on 118 cases with complete or partial follow-up and 21 events.

Discussion The present study shows that the incidence of malignant hypertension is similar to a previous report from the west Birmingham study 10 years ago.3 We did not observe specific time trends in either the number of patients or in the ethnic distribution over the past 12 years. The relatively high prevalence of malignant hypertension in black individuals may be related to a higher prevalence of uncontrolled hypertension in black and non-black immigrants. We recently conducted a cross-sectional survey in the source population of this cohort showing that black and South Asian subjects had a higher prevalence of hypertension and were less likely to have their blood pressure controlled.5 No more than 10% of hypertensive black men had their blood pressure controlled compared with 45% of white hypertensive women, which may help to under-stand the predominance of malignant hypertension in male patients of African

Chapter 3

78

descent. Despite differences in blood pressure control, there were no differences in the awareness or prescription of antihypertensive drugs in this survey. Hypertension awareness was 50% for black compared with 55% for white individuals; 33% of black and 24% of white individuals received antihyper-tensive treatment. In the present study, we found similar percentages for hypertension awareness and treatment. However, almost two-thirds of those on antihypertensive treatment admitted that they had stopped taking their medi-cation before admission. The low adherence to antihypertensive treatment, especially among black patients, may be associated with the high number of patients without health insurance in this group. Previous studies have reported on the excess of malignant hypertension and related complications among urban black individuals, both in sub-Saharan Africa and in the United States.6-8 Apart from urban stress and dietary changes, socio-economic differences, including disparities in health insurance status and the use of antihypertensive drugs, have been implicated to explain the excess of malignant hypertension among urban black individuals.7,9 Our results suggest that the same socio-economic factors are involved in the predominance of malignant hypertension among black immigrants in this multi-ethnic cohort. The clinical presentation was similar between black and white subjects and the prevalence of clinical symptoms was identical to previous data from the Glasgow Blood Pressure Clinic.10 However, black individuals had higher systolic blood pressure levels and more hypertensive organ damage than white individuals with malignant hypertension. We previously reported on the predo-minance of TMA in black patients as evidence of more severe vascular damage and its association with renal dysfunction.11 There was no difference in the severity of retinopathy between black and white patients, which is in agreement with previous reports indicating that the clinical characteristics and survival of patients with grades III and IV hypertensive retinopathy are not different.12

We found a relatively low prevalence of secondary malignant hypertension, especially in black individuals, compared with previous reports.2-4,13 One may suggest that the low prevalence of secondary malignant hypertension in the present study was caused by the limited efforts to demonstrate an underlying cause. Renal biopsies were not routinely taken; therefore some primary renal diseases may have been missed. However, renal ultrasounds were performed in most patients with elevated serum creatinine levels at presentation, and macro-albuminuria disappeared during periods of adequate blood pressure control in all


79

patients without secondary malignant hypertension. The low prevalence of renovascular disease seems an adequate reflection of the true prevalence, as imaging studies of the renal arteries were performed in most patients. We observed only one patient with clinical features of Cushing’s syndrome, despite various endocrine tests in half of the study population, which is in agreement with previous studies showing that endocrine disease is rare in malignant hypertension. The high prevalence of renal and renovascular disease reported by others can also be explained by referral bias, as some studies included their patients from renal disease and dialysis units.4, 13 In two other reports unex-plained chronic renal failure accounted for the majority of secondary malignant hypertension.3, 14 However, as renal biopsies were not routinely taken, probably part of the unexplained renal failures must be attributed to hypertension instead of primary renal disease. In the present study, we have excluded patients referred from elsewhere to prevent referral bias, and have attributed unexplained renal failures to hypertension unless proved otherwise.

Finally, similar efforts were undertaken to demonstrate an underlying cause in black and white subjects. Therefore the low prevalence of secondary malignant hypertension in black individuals may best be explained by a higher prevalence of black individuals with uncontrolled hypertension in the population at large. In a population with more vigorous blood pressure control, patients with essential hypertension are more likely to be detected, treated and controlled, thus preventing them from developing malignant hypertension. The poor control of hypertension among black individuals, as recently shown in our cross-sectional survey, may facilitate the development of the malignant phase in some, and therefore result in a relative predominance of primary malignant hypertension in this group. Our study shows that the survival of patients with malignant hypertension has improved in comparison with two previous publications from the UK.15, 16 The proportion of patients surviving after 5 years was 0.78 for those admitted between 1980 and 1989 in the west Birmingham cohort compared with 0.90 in our study. This difference is probably due to the high percentage of patients who died of renal failure at that time. In contrast, the percentage of patients on chronic renal replacement therapy in our study is much higher than in the west Birmingham cohort despite comparable follow-up. The most important indi-cators of future renal failure in our cohort were serum creatinine values at presentation and secondary causes of malignant hypertension. Although data on secondary malignant hypertension were pooled because of the low frequency of

Chapter 3

80

some secondary causes, renal parenchymal disease was the only significant indicator of future renal failure, odds ratio 4.5 (95% CI 1.2–18). The higher prevalence of renal failure among black individuals was mainly because of higher serum creatinine values at presentation.

This study has both strengths and limitations. The strengths include the description of a relatively large cohort of patients with malignant hypertension and the possibility to compare these data with the prevalence, awareness and control of hypertension in the population of origin. Limitations include a reliance on the coding system and possible registration errors. We cannot rule out that some patients were not included; however, we estimate these numbers to be small as a review of all cases with the discharge code ‘hypertension’ revealed only one patient with malignant hypertension. Although it is routine practice to screen patients presenting with severe hypertension for retinal lesions by an ophthalmologist at both hospitals, it cannot be ruled out that some cases with a less severe presentation, who were therefore not admitted, may have been missed. Conclusion We have shown that the incidence of malignant hypertension in this multi-ethnic population has not declined compared with the west Birmingham study 12 years ago. Ethnic differences in blood pressure control may explain a higher prevalence of malignant hypertension, more renal dysfunction, and a lower prevalence of secondary causes. Probably, part of the ethnic disparities in the incidence and complications of malignant hypertension may be clarified by mundane factors such as differences in compliance and insurance status. Although the mortality rate has declined, the high prevalence of renal failure, particularly among black individuals, remains a cause for concern. More vigorous blood pressure control in the population at large will not only prevent excess mortality from stroke and myocardial infarction, but will also lower the incidence of malignant hypertension and related renal complications.


81

References 1. Keith NM, Wagener HP, Barker NW. Some different types of essential hyper-

tension: their course and prognosis. Am J Med Sci 1939; 196:332–343. 2. Gudbrandsson T. Malignant hypertension. A clinical follow-up study with special

reference to renal and cardiovascular function and immunogenetic factors. Acta Med Scand Suppl 1981; 650:1–62.

3. Lip GY, Beevers M, Beevers G. The failure of malignant hypertension to decline: a

survey of 24 years’ experience in a multiracial population in England. J Hypertens 1994; 12:1297–1305.

4. Scarpelli PT, Livi R, Caselli GM, Di Maria L, Teghini L, Montemurro V, et al.

Accelerated (malignant) hypertension: a study of 121 cases between 1974 and 1996. J Nephrol 1997; 10:207–215.

5. Agyemang C, Bindraban N, Mairuhu G, Montfrans G, Koopmans R, Stronks K.

Prevalence, awareness, treatment, and control of hypertension among Black Surinamese, South Asian Surinamese and White Dutch in Amsterdam, the Netherlands: the SUNSET study. J Hypertens 2005; 23:1971–1977.

6. Veriava Y, du TE, Lawley CG, Milne FJ, Reinach SG. Hypertension as a cause of

end-stage renal failure in South Africa. J Hum Hypertens 1990; 4:379–383. 7. Shea S, Misra D, Ehrlich MH, Field L, Francis CK. Predisposing factors for severe,

uncontrolled hypertension in an inner-city minority population. N Engl J Med 1992; 327:776–781.

8. Isaacson C, Milne FJ, van Niekerk I, Kenyon MR, Mzamane DV. The renal histo-

pathology of essential malignant hypertension in black South Africans. S Afr Med J 1991; 80:173–176.

9. Opie LH, Seedat YK. Hypertension in sub-Saharan African populations. Circulation

2005; 112:3562–3568. 10. Isles CG. Malignant hypertension and hypertensive encephalopathy. In: Swales JD,

editor. Textbook of hypertension. Oxford, UK: Blackwell Scientific Publishers; 1994.

11. van den Born BJ, Honnebier UP, Koopmans RP, van Montfrans GA. Microangio-

pathic hemolysis and renal failure in malignant hypertension. Hypertension 2005; 45:246–251.

Chapter 3

82

12. Ahmed ME, Walker JM, Beevers DG, Beevers M. Lack of difference between malignant and accelerated hypertension. BMJ (Clin Res) 1986; 292:235–237.

13. Guerin C, Gonthier R, Berthoux FC. Long-term prognosis in malignant or accele-

rated hypertension. Nephrol Dial Transplant 1988; 3:33–37. 14. Kadiri S, Olutade BO. The clinical presentation of malignant hypertension in Nige-

rians. J Hum Hypertens 1991; 5:339–343. 15. Lip GY, Beevers M, Beevers DG. Complications and survival of 315 patients with

malignant-phase hypertension. J Hypertens 1995; 13:915–924. 16. Webster J, Petrie JC, Jeffers TA, Lovell HG. Accelerated hypertension – patterns of

mortality and clinical factors affecting outcome in treated patients. Q J Med 1993; 86:485–493.

Chapter 4

Thrombotic microangiopathy and renal failure in malignant hypertension Bert Jan H. van den Born, Uwkje P.F. Honnebier, Richard P. Koopmans, Gert A. van Montfrans

Hypertension. 2005; 45: 246-251.

Chapter 4

86

Abstract Background Renal dysfunction is an important cause of morbidity and mortality in patients with malignant hypertension. Thrombotic microangiopathy (TMA) related to malignant hypertension may cause renal insufficiency by obstruction of interlobular arteries. We hypothesized that the presence of TMA is an important indicator of renal dysfunction and recovery in malignant hypertension. Methods We retrospectively analyzed 97 patients admitted between April 1994 and April 2004 with malignant hypertension. TMA was defined as a low platelet count (< 150 x109/L) with either an elevated lactic dehydrogenase (> 220 U/L) or presence of schistocytes. Results TMA was present in 26 of 97 patients (27%). Serum creatinine levels at admission were significantly higher in those with than in those without TMA: median serum creatinine 690 µmol/L (interquartile range [IQR] 394 to 1105) and 120 µmol/L (IQR 82 to 211), respectively (P < 0.01). Macroalbuminuria was present in 88% with versus 41% without TMA (P < 0.01). Patients with TMA were more often black (73%; P < 0.01) and had higher systolic blood pressure (mean 242 mmHg versus 225 mmHg; P < 0.01). Dialysis was needed in 15 patients with TMA (58%) versus 2 patients (3%) without TMA. In 6 patients with TMA, dialysis could be stopped. Cox regression analysis showed that TMA and systolic blood pressure were the most important indicators of renal improvement during follow-up, with a hazard ratio of 0.24 (95% confidence interval [CI], 0.08 to 0.75; P < 0.01) and 1.02 per mmHg increase in systolic blood pressure (95% CI, 1.01 to 1.05; P < 0.01). Conclusion TMA is an important indicator of renal insufficiency and recovery in patients with malignant hypertension.

TMA and renal failure in malignant hypertension

87

Introduction Thrombotic microangiopathy (TMA) is a well-known complication of malignant hypertension. However, little is known about its epidemiology and relationship with renal dysfunction. Although the prognosis of malignant hypertension has improved considerably over the past decades, renal dysfunction remains an important cause of morbidity and mortality.1–4 Yet some patients have a remar-kable recovery of kidney function after adequate control of blood pressure is achieved.5–8 It has been suggested that in patients with malignant hypertension and kidney dysfunction, the presence of TMA may be an indicator of renal recovery.5,7 In most renal biopsy specimens from patients with renal failure related to malignant hypertension, an obliterative vasculopathy is observed with fibrinoid necrosis and sometimes frank thrombosis of interlobular arteries.9,10 These fibrinoid depositions are caused by continued seepage of fibrin en serum proteins through a necrotic vessel wall into surrounding viable tissue.11 The presence of TMA, as evidence of profound endothelial damage, may point toward such an obliterative vasculopathy with subsequent renal dysfunction. Resolution of TMA by blood pressure treatment may, in turn, result in reper-fusion and improvement of kidney function.

We hypothesized that the presence of TMA may be an important predictor of renal dysfunction and recovery in malignant hypertension. Therefore, we performed a retrospective analysis of patients admitted with malignant hypertension to assess whether the presence of TMA was related to renal dys-function at admission and with improvement of renal dysfunction during follow-up. Patients and methods We retrospectively analyzed charts of consecutive patients admitted with malignant hypertension at 2 large teaching hospitals (the Academic Medical Centre and Onze Lieve Vrouwe Gasthuis), catering for a large multiethnic community in Amsterdam, the Netherlands, between April 1994 and April 2004. Both hospitals use the same database in which the diagnosis at discharge is recorded according to the International Classification of Diseases (ICD) codes. The following ICD codes were reviewed: essential malignant hypertension (ICD 401.0), hypertensive encephalopathy (ICD 437.2), secondary malignant hyper-tension (ICD 405.09), hypertension with cardiac disease/malignant (ICD 402.0), and hypertension with kidney disease/malignant (ICD 403.0). To identify the

Chapter 4

88

presence of registration errors, computer data of all patients discharged with the diagnosis essential hypertension (ICD 401.9) were also analyzed.

In all patients, the diagnosis malignant hypertension was verified according to the World Health Organization (WHO) criteria of 1978: high blood pressure together with bilateral linear or flame-shaped hemorrhages or “cotton-wool” exudates with or without papilledema during funduscopic examination (grade III and IV hypertensive retinopathy, respectively, according to the Keith–Wagener and Barker classification). We defined TMA related to malignant hypertension as: (1) the presence of a low platelet count (< 150x109/L) together with either an elevated lactic dehydrogenase (LDH) level (> 220 U/L) or presence of schistocytes, or both; and (2) normalization of platelets and LDH or schistocytes after adequate blood pressure control was achieved. Excluded from analysis were patients < 18 years of age, pregnant women, patients with papilledema and an intracranial mass (hemorrhage), patients already on dialysis before admission, and patients in whom TMA could not be assessed or who had an alternative explanation for a low platelet count. Macroalbuminuria was defined as urinary protein excretion > 300 mg/L or 2+ for dipstick proteinuria. Left ventricular hypertrophy was defined according to the Sokolow–Lyon criteria. Significant renal artery stenosis was defined if narrowing of the renal artery lumen exceeded 50% during computed tomography, MRI, or conventional angiography. Schistocytes were considered present if “several” red cell fragments could be detected on a peripheral blood film. End points were: (1) serum creatinine levels and urinary protein excretion at admission; and (2) improvement of renal function on follow-up, defined as a 50% reduction in serum creatinine compared with baseline. Statistics Baseline variables were described using mean and SD, median and range, or interquartile range for variables with a skewed distribution. Differences between groups were calculated using Yates correction or a t-test for parametric and Mann–Whitney U test for nonparametric distributions where appropriate. Linear regression analysis was used to assess the contribution of platelet count on serum creatinine, with serum creatinine expressed on a logarithmic scale. R2 was used to give an estimate of the variations in serum creatinine explained by platelet count at admission. The confidence interval (CI) of this estimate was calculated using the Fisher z transformation. Univariate ANOVA was performed to test the association between serum creatinine and TMA, adjusted for race,


89

blood pressure at admission, and established secondary causes of malignant hypertension. To analyze indicators of improvement of renal function during follow-up, Cox regression was used. First, single variables were used to test an association with improvement of renal function. Single variables that showed a significant association or could have a possible pathophysiological relationship with improvement of renal function were retained and used for the model. Then, a backward elimination method was used in which the variable with the smallest partial correlation with the dependent variable was removed first. For all tested variables in this model, the hazard ratio (HR) with its 95% CI was calculated. Variables that had a significant association with renal improvement were tested for interaction. For statistical analysis, the SPSS software package for Windows version 12.0 was used. P < 0.05 was considered to indicate a statistically signi-ficant difference. Results A total of 110 patients fulfilled the WHO criteria for malignant hypertension. Thirteen patients were excluded because: (1) they had an alternative explanation for the presence of papilledema attributable to an intracranial mass, 4 (2) the platelet count at admission was unknown, 5 or (3) they had an alternative explanation for the thrombocytopenia, 4 (figure 1). For the remaining 97 patients, imaging studies of the renal arteries were conducted in 79 (81%), ≥ 1 endocrine tests for pheochromocytoma, Cushing’s disease or Conn’s syndrome in 44 (45%), and renal biopsies in 18 (19%) patients. In 19 of 26 patients with thrombocytopenia, schistocytes were present; in 3, schistocytes were absent, and in 4, this test was not performed. The number of patients with primary and secondary malignant hypertension is listed in Table 1. The clinical charac-teristics according to the presence or absence of TMA are summarized in Table 2. The main outcome measures, serum creatinine level and urinary protein excretion at admission, were significantly higher in the group with TMA (figure 2). After correction for blood pressure, ethnicity, and secondary causes, differences in serum creatinine levels between patients with and without TMA remained statistically significant (P < 0.01). Linear regression analysis of serum creatinine and platelet count is shown in figure 3.

Chapter 4

90

Fig. 1

Patient selection. HRP indicates hypertensive retinopathy Table 1 Secondary causes of malignant hypertension according to the presence or absence of TMA

Causes of malignant hypertension TMA present TMA absent P value No cause identified (essential malignant hypertension)

17 (65 %) 48 (68 %) 0.84

Renal parenchymal disease 6 (23 %) 8 (11 %) 0.19 Renal artery stenosis 2 (8 %) 6 (8 %) 1.00 Other (endocrine, drug-induced) 1 (4 %) 9 (13 %) 0.28 Total 26 (100 %) 71 (100 %)

Values are numbers and percentages.

4 alternative explanation for thrombocytopenia

5 platelet count at admission unknown

4 with intracranial mass and papilledema

97 analyzed for the presence or absence of TMA

110 with severe hypertension and grade III/IV HRP

106 with definite malignant hypertension


91

Table 2 Clinical characteristics of all included patients with malignant hypertension and comparison of groups with and without TMA.

Patient characteristics TMA present TMA absent P value* Patients 26 (27 %) 71 (73 %) Age, yrs (mean ± SD) 41 ± 8 44 ± 13 0.13 Male 12 (63 %) 34 (60 %) 1.00 Black 19 (73 %) 25 (35 %) < 0.01 Retinopathy, grade IV 11 (42 %) 37 (52 %) 0.53 Systolic BP, mmHg (mean ± SD) 242 ± 25 225 ± 24 < 0.01 Diastolic BP, mmHg (mean ± SD) 150 ± 18 145 ± 17 0.25 Hemoglobin, mmol/l (median, [IQR])

6.5 [5.4-8.0] 8.8 [7.9-9.6] < 0.01

Platelets, x109/l (median, [IQR]) 90 [63-115] 225 [186-252] < 0.01 LDH, U/l (median, [IQR])† 786 [471-1135] 279 [222-385] < 0.01 Serum creatinine, µmol/l (median, [IQR])

690 [394-1105] 120 [82-211] < 0.01

Serum potassium, mmol/l (median, [IQR])

3.5 [2.8-4.1] 3.4 [3.1-3.9] 0.89

Current Smoker‡ 12 (46 %) 27 (38 %) 0.54 Diabetes mellitus 0 5 (7 %) 0.17 Previous hypertension 12 (46 %) 39 (55 %) 0.49 Left ventricular hypertrophy 23 (88 %) 56 (79 %) 0.24 Macroalbuminuria§ 23 (88 %) 29 (41 %) < 0.01 Cerebral complications

- cerebral infarction - posterior

leucoencephalopathy

1 1

4 4

Values are numbers and percentages unless stated otherwise. IQR indicates interquartile range. *P values are calculated for differences between those with and those without TMA. †Data available for 70 patients (25 missing in those without TMA en 2 in those with TMA). ‡Data are missing for 2 in the group without TMA. §Data are missing for 1 in each group.

Chapter 4

92

Fig. 2 Fig. 3

1.5

2.0

2.5

3.0

3.5

TMA present TMA absent

Log

crea

tinin

e ( µ

mol

/l)

0 100 200 300 400 500 6001.5

2.0

2.5

3.0

3.5

p<0.001R2=0.37

Platelet count (x109/l)

Log

crea

tinin

e ( µ

mol

/l)

Serum creatinine levels at admission on a Log-linear regression analysis of serum logarithmic scale in patients with and creatinine and platelet count at admission without TMA; P < 0.01 for difference for all patients. R2 = 0.37 (95% CI, 0.11 to between groups. 0.51). Dialysis was needed in 15 of 26 (58%) patients with TMA during admission and in 2 patients (3%) without TMA. In 6 patients with TMA, dialysis could be stopped: 4 within 2 months, and in 2 others, after 8 months and 2 years, respectively. Follow-up was complete for 87 of 97 patients (90%). Ten patients either moved or did not show up at the outpatient department. The total number of patients who died or were lost to follow-up were equally distributed in both groups. Nine patients died: 3 from cardiovascular complications, 2 from cerebral hemorrhage, and 1 from myocardial infarction. Characteristics for those at follow-up are listed in Table 3. Improvement of renal function, defined as a reduction of serum creatinine > 50% compared with baseline, was noted in 17 patients during follow-up. Cox regression analysis showed that TMA and systolic blood pressure at admission were the most powerful indicators of renal improvement with an HR of 0.24 (95% CI, 0.08 to 0.75) and 1.02 per mmHg increase in systolic blood pressure (95% CI, 1.01 to 1.05; Table 4). The inter-action between systolic blood pressure and TMA was not significant (P = 0.28). Improvement of kidney function over time in patients with and without TMA is shown in figure 4.


93

Table 3 Patient characteristics on follow-up.

Patient characteristics at follow-up TMA present TMA absent P value Lost to follow-up 3 (11 %) 7 (10 %) Follow-up time, months (mean ± SD) 52 ± 33 40 ± 33 0.09 Systolic BP, mmHg (median, [IQR]) 140 [128-152] 140 [122-150] 0.87 Diastolic BP, mmHg (median, [IQR]) 90 [80-100] 89 [79-99] 0.58 Anti-hypertensive drugs 3.0 ± 1.2 3.0 ± 1.2 0.92 Increase in serum creatinine > 1/3 of baseline*

3 (21 %) 14 (23 %) 0.90

Died 2 (8%) 7 (10 %) TMA, thrombotic microangiopathy; SD, standard deviation; IQR, interquartile range; BP, blood pressure. *Calculated for those not in need of permanent kidney replacement therapy. Table 4 Cox regression analysis using a backward elimination method to predict indicators for improvement of renal dysfunction

Indicator Rank* P HR 95 % CI Age, yrs 3 0.58 0.98 0.93-1.04 Male 2 0.72 0.83 0.29-2.35 Black 5 0.14 2.36 0.75-7.45 Systolic blood pressure, mmHg … 0.01 1.02 1.01-1.05 Diastolic blood pressure, mmHg 4 0.57 0.99 0.94-1.03 Log creatinine 1 0.77 0.76 0.12-4.8 Macroalbuminuria 6 0.11 0.28 0.06-1.35 TMA … 0.01 0.24 0.08-0.75

HR, 95% CI, and P value were determined just before elimination. *Indicates the rank of elimination in stepwise backward Cox regression.

Chapter 4

94

Fig. 4

0 10 20 30 40 50 60

Time (months)

0,0

0,2

0,4

0,6

0,8

1,0

Prop

ortio

n w

ith re

nal i

mpr

ovem

ent

Improvement of renal function over time in patients with and without TMA. Improvement is defined as a > 50% reduction in serum creatinine compared with baseline.

Discussion We found a marked difference between serum creatinine levels and proteinuria in patients with and without TMA in the sense that renal dysfunction was more severe in patients with TMA. Also after adjusting for ethnicity, blood pressure, and secondary causes of malignant hypertension, the association between kidney dysfunction and TMA remained significant. There was a negative log-linear relationship between serum creatinine and platelet count, indicating that a decline in platelet count, as evidence of TMA, was related to a decline in renal function. Furthermore, platelet count could explain 37% of the variations in serum creatinine at admission. The predominance of black patients with TMA in


95

our study is consistent with previous reports in which a high prevalence of TMA in blacks with malignant hypertension was found.4,7,11 There was no difference in severity of retinopathy or left ventricular hypertrophy between the groups with and without TMA, which supports the theory from previous studies in which clinical characteristics and survival are not different for patients with grades III or IV hypertensive retinopathy.12,13 Although recovery from renal failure has been reported previously in patients with malignant hypertension,14–17 few have noted a possible association with TMA.5,6,18 However, in several case reports and case series, a possible link between TMA and recovery from renal failure has been suggested.5–7 In the largest series on renal failure related to malignant hypertension, 12 patients who recovered from dialysis were compared with 42 patients who did not.7 The presence of TMA was found as a possible predictor of renal recovery, although this finding was not significant, possibly because of small group size. In our study, patients with TMA had a 4x higher chance of renal improvement than patients without TMA. Furthermore, dialysis could be stopped in 6 of 15 (40%) patients with TMA who needed kidney replacement therapy during admission. The reason TMA develops in malignant hypertension has not been fully clarified. Most evidence regarding TMA pathogenesis comes from animal studies: rats double transgenic for the human renin and angiotensinogen genes develop severe hypertension and profound endothelial damage similar to TMA seen in humans with malignant hypertension.19 Furthermore, animal studies have shown that high levels of angiotensin II have a direct cytotoxic effect on the vessel wall.20,21 Finally, rats with a unilateral renal artery stenosis develop malignant hypertension when refrained from water and saline, suggesting that pressure natriuresis further enhances renin-angiotensin aldosterone system (RAAS) activation through induction of kidney ischemia.22 The evidence for the importance of RAAS activation in development of malignant hypertension and TMA seems in contrast with the high prevalence of TMA in black hypertensive patients because they tend to have lower plasma renin and angiotensin levels.23 In our population, all blacks were immigrants from western Africa and Dutch Surinam and, in most cases, had never visited a doctor. Delay in presentation could explain a more severe presentation of malignant hypertension with coexisting TMA and renal dysfunction. However, another explanation is also possible: recent reports have suggested a more active role of the intrarenal renin-angiotensin system in blacks.24,25 In accordance with

Chapter 4

96

this finding is a reduced renal plasma flow attributable to renal vasoconstriction in these patients, which is reversible after administration of an angiotensin-converting enzyme (ACE) inhibitor.24 This vasoconstriction could make blacks more prone to development of renal ischemia, with subsequent activation of the RAAS, and to development of TMA, through high levels of angiotensin II. Pre-existent renal arteriolar or glomerular disease, caused by hypertension or other-wise, may contribute to development of TMA by enhancing renal ischemia. In turn, TMA may cause further renal damage by narrowing and thrombosis of renal arteries. In kidney biopsy specimens, this may be reflected by the presence of fibrinoid necrosis and thrombosis of interlobular arteries with subsequent ischemic glomerular and tubular changes.26 However, because kidney biopsies were not routinely taken in this and previous reports, further histological evidence for this association is needed. To differentiate TMA related to malignant hypertension from other causes of microangiopathic hemolysis, particularly thrombotic thrombocytopenic purpura and hemolytic uremic syndrome, can be difficult. Therefore, we included normalization of platelet count and LDH or disappearance of schistocytes after blood pressure control in our definition, making alternative explanations for the thrombocytopenia unlikely. In 3 patients, the presence of a low platelet count together with an elevated LDH was not accompanied by the presence of schistocytes. However, in our experience, the prevalence of schistocytes in a peripheral blood smear is low in malignant hypertension, probably making it a less sensitive criterion for establishing TMA. Using an increased LDH level as an exclusive marker may overestimate the prevalence of TMA because it can also be elevated in several other conditions. Selection bias in this retrospective study may have been introduced by the way patients were referred. Because both hospitals contain a dialysis unit, patients with malignant hypertension and renal insufficiency would be more likely to be included. However, in all patients, renal insufficiency was established only after presentation to the emergency department; no patients were referred from elsewhere for dialysis treatment. Second, although patients with possible malignant hypertension are routinely examined by an ophthalmologist in both hospitals, some patients without symptoms and severe elevation of blood pressure could have been missed. Third, patients with renal parenchymal or renovascular disease were not excluded because TMA may also be related to renal dysfunction in patients with pre-existent renal parenchymal disease. When patients without evidence of renal parenchymal or renovascular disease were


97

analyzed separately, no difference in outcome was observed. Because kidney biopsies were not taken routinely, some patients with renal parenchymal disease could have been missed. However, macroalbuminuria disappeared in all patients said to have essential malignant hypertension after adequate blood pressure control was achieved during follow-up, suggesting hypertensive kidney disease instead of renal parenchymal disease. Furthermore, ultrasonography of kidneys and imaging studies of renal arteries performed in these patients showed no evidence of postrenal obstruction, cystic disease, or renal artery stenosis. Perspectives We have shown that TMA is an important predictor of renal insufficiency and recovery in malignant hypertension. Analysis of platelets, LDH, and schistocytes in patients with malignant hypertension may serve as an important marker for those who are at risk of renal failure. The predominance of black patients with TMA may suggest that a more active intrarenal renin-angiotensin system is primarily involved in the onset of TMA, as is shown recently. Furthermore, TMA as evidence of profound endothelial damage may further limit renal blood flow by narrowing and obstruction of renal arterioles. It would be of interest to examine whether early ACE inhibition, besides standard therapy with sodium nitroprusside or labetalol, in patients with malignant hypertension and TMA could result in a more rapid reduction of microangiopathy and relative preservation of renal function.

References 1. Webster J, Petrie JC, Jeffers TA, Lovell HG. Accelerated hypertension– patterns of

mortality and clinical factors affecting outcome in treated patients. Q J Med. 1993;86:485– 493.

2. Guerin C, Gonthier R, Berthoux FC. Long-term prognosis in malignant or

accelerated hypertension. Nephrol Dial Transplant. 1988;3:33–37. 3. Lip GY, Beevers M, Beevers G. The failure of malignant hypertension to decline: a

survey of 24 years’ experience in a multiracial population in England. J Hypertens. 1994;12:1297–1305.

4. Milne FJ, James SH, Veriava Y. Malignant hypertension and its renal complications

in black South Africans. S Afr Med J. 1989;76:164 –167. 5. Isles CG, McLay A, Jones JM. Recovery in malignant hypertension presenting as

acute renal failure. Q J Med. 1984;53:439–452.

Chapter 4

98

6. Bakir AA, Bazilinski N, Dunea G. Transient and sustained recovery from renal

shutdown in accelerated hypertension. Am J Med. 1986;80:172–176. 7. James SH, Meyers AM, Milne FJ, Reinach SG. Partial recovery of renal function in

black patients with apparent end-stage renal failure due to primary malignant hypertension. Nephron. 1995;71:29 –34.

8. Mitchell HC, Graham RM, Pettinger WA. Renal function during long-term

treatment of hypertension with minoxidil: comparison of benign and malignant hypertension. Ann Intern Med. 1980;93:676–681.

9. Sanerkin NG. Vascular lesions of malignant essential hypertension. J Pathol. 1971;

103:177–184. 10. Jones DB. Arterial and glomerular lesions associated with severe hypertension.

Light and electron microscopic studies. Lab Invest. 1974;31: 303–313. 11. Kadiri S, Olutade BO. The clinical presentation of malignant hypertension in Nige-

rians. J Hum Hypertens. 1991;5:339 –343. 12. Ahmed ME, Walker JM, Beevers DG, Beevers M. Lack of difference between

malignant and accelerated hypertension. Br Med J (Clin Res Ed).1986;292:235–237. 13. McGregor E, Isles CG, Jay JL, Lever AF, Murray GD. Retinal changes in malignant

hypertension. Br Med J (Clin Res Ed). 1986;292:233–234. 14. Cordingley FT, Jones NF, Wing AJ, Hilton PJ. Reversible renal failure in malignant

hypertension. Clin Nephrol. 1980;14:98 –103. 15. Luft FC, Bloch R, Szwed JJ, Grim CM, Grim CE. Minoxidil treatment of malignant

hypertension. Recovery of renal function. J Am Med Assoc. 1978;240:1985–1987. 16. Mitchell HC, Pettinger WA. Renal function in long-term minoxidiltreated patients. J

Cardiovasc Pharmacol. 1980;2(suppl 2):S163–S172. 17. Mourad G, Mimran A, Mion CM. Recovery of renal function in patients with

accelerated malignant nephrosclerosis on maintenance dialysis with management of blood pressure by captopril. Nephron. 1985;41:166 –169.

18. Sevitt LH, Evans DJ, Wrong OM. Acute oliguric renal failure due to accelerated

(malignant) hypertension. Q J Med. 1971;40:127–144. 19. Ganten D, Wagner J, Zeh K, Bader M, Michel JB, Paul M, Zimmermann F, Ruf P,

Hilgenfeldt U, Ganten U, Kaling M, Bachmann S, Fukamizu A, Mullins JJ, Murakami K. Species specificity of renin kinetics in transgenic rats harboring the


99

human renin and angiotensinogen genes. Proc Natl Acad Sci U S A. 1992;89:7806 –7810.

20. Muller DN, Dechend R, Mervaala EM, Park JK, Schmidt F, Fiebeler A, Theuer J,

Breu V, Ganten D, Haller H, Luft FC. NF-kappaB inhibition ameliorates angio-tensin II-induced inflammatory damage in rats. Hypertension. 2000;35:193–201.

21. Funakoshi Y, Ichiki T, Ito K, Takeshita A. Induction of interleukin-6 expression by

angiotensin II in rat vascular smooth muscle cells. Hypertension. 1999;34:118 –125. 22. Mohring J, Petri M, Szokol M, Haack D, Mohring B. Effects of saline drinking on

malignant course of renal hypertension in rats. Am J Physiol. 1976;230:849–857. 23. He FJ, Markandu ND, Sagnella GA, MacGregor GA. Importance of the renin

system in determining blood pressure fall with salt restriction in black and white hypertensives. Hypertension. 1998;32:820–824.

24. Price DA, Fisher ND, Osei SY, Lansang MC, Hollenberg NK. Renal perfusion and

function in healthy African Americans. Kidney Int. 2001; 59:1037–1043. 25. Price DA, Fisher ND, Lansang MC, Stevanovic R, Williams GH, Hollenberg NK.

Renal perfusion in blacks: alterations caused by insuppressibility of intrarenal renin with salt. Hypertension. 2002;40:186 –189.

26. Kincaid-Smith P. Renal pathology in hypertension and the effects of treatment. Br J

Clin. Pharmacol. 1982;13:107–115.

Chapter 5

The M235T polymorphism in the angiotensinogen gene is associated with the risk of malignant hypertension in white patients Bert-Jan H. van den Born, Gert A. van Montfrans, Andre G. Uitterlinden, Aeilko H. Zwinderman and Richard P. Koopmans

J. Hypertens. 2007; in press

Chapter 5

104

Abstract Background Malignant hypertension can be considered an extreme phenotype of renin-mediated hypertension. Therefore, we compared the allelic frequencies of the angiotensinogen (AGT) M235T, angiotensin-converting enzyme inser-tion/deletion (ACE I/D) and angiotensin II-type I receptor (AT1R) A1166C polymorphisms in malignant hypertensive patients with hypertensive and normotensive controls. Methods A total of 101 consecutive patients between 1995 and 2005 admitted at a large university hospital fulfilled the criteria for malignant hypertension. Seventy-five patients (74%) were compared with 150 hypertensive and 150 normotensive controls, randomly selected from a population study and indivi-dually matched on age, sex and ethnicity. Results The odds of malignant hypertension in white subjects with the TT genotype of the AGT M235T polymorphism was 14.3 (5.5–37) compared to hypertensive controls, and 9.4 (3.8–23.2) compared to normotensive controls. Adjustment for age, sex, smoking and antihypertensive therapy did not affect this association. The association of AGT M235T with malignant hypertension was not significant in blacks. In patients with malignant hypertension, the TT genotype was associated with more severe renal dysfunction and thrombotic microangiopathy. No differences were found in allele frequencies of the ACE I/D or the AT1R A1166C polymorphisms between study groups. Conclusions The TT genotype of AGT M235T is associated with malignant hypertension in whites, carriers having an odds of approximately 10 to 1 com-pared to hypertensive and normotensive controls. These observations may provide a better understanding of the pathophysiology of malignant hypertension and offer possibilities for identifying patients at risk. Larger association or linkage studies are needed for a more detailed risk assessment.

The M235T polymorphism and malignant hypertension

105

Introduction Malignant hypertension can be considered an extreme phenotype of renin-mediated hypertension as evidenced by a marked activation of the renin–angiotensin system (RAS) in both humans and animals.1-3 Rats that are double transgenic for the human renin and angiotensinogen (AGT) genes develop severe hypertension and profound endothelial damage similar to that observed in humans with malignant hypertension.4 Treatment with a human renin inhibitor, ACE-inhibitor or angiotensin-receptor blocker reverses these vascular abnor-malities.5, 6 As a result of better blood pressure control in the population at large, malignant hypertension is rare nowadays. In multiethnic populations, however, malignant hypertension is still relatively frequent, with an estimated annual incidence of 1–3 per 100 000 inhabitants, mainly because of a predominance of black patients with this disease.7, 8 Although differences in hypertension control and adherence to antihypertensive drugs have been implicated to clarify the increased prevalence of malignant hypertension,8, 9 this may not fully explain why some (uncontrolled) hypertensive patients progress to the malignant phase while others do not.

Two previous studies have found a possible association between the ACE I/D polymorphism and the risk of malignant hypertension in white subjects;10, 11 however, both studies lacked comprehensive information on the clinical cha-racteristics and selection of patients. The aim of the present study was to compare the allelic frequencies of three polymorphisms in the RAS – the M235T polymorphism in the AGT gene, the angiotensin-converting enzyme insertion/deletion (ACE I/D) polymorphism and the A1166C polymorphism in the angiotensin-1 receptor (AT1R) – of conse-cutive patients with malignant hypertension and to compare them with hyper-tensive and normotensive controls from the source population matched for age, sex and ethnicity. Patients and methods The records were reviewed of all patients admitted between August 1995 and August 2005 with malignant hypertension at a large university hospital (the Academic Medical Centre) catering for a large multiethnic population in Amsterdam, The Netherlands. This hospital uses a database recording the diagnoses at discharge based on ICD (International Classification of Diseases)

Chapter 5

106

codes. The following ICD codes were used: ‘essential malignant hypertension’ (ICD 401.0), ‘hypertensive encephalopathy’ (ICD 437.2), ‘secondary malignant hypertension’ (ICD 405.09), ‘hypertension with cardiac disease/malignant’ (ICD 402.0), ‘hypertension with kidney disease/malignant’ (ICD 403.0) and ‘essential hypertension’ (ICD 401.9). After chart retrieval the diagnosis ‘malignant hyper-tension’ was verified using the WHO criteria for malignant hypertension: high blood pressure together with bilateral linear or flame-shaped haemorrhages or ‘cotton-wool’ exudates with or without papilledema on funduscopic exami-nation. Excluded were patients < 18 years of age, pregnant women, patients with papilledema and an intracranial mass (haemorrhage), patients already on dialysis prior to presentation at the emergency department and patients referred from elsewhere to prevent referral bias. A total of 101 patients fulfilled the criteria for malignant hypertension, mean age 44.2 ± 12.7 years, 63% male and 46% black. Five patients died, 13 patients moved outside the area. All living patients received an invitation to participate. Eight patients declined participation, leaving 75 patients (74%) for analysis. Their mean age was 47.4 ± 15.6 years, 73% were male and 31% were black. Systolic blood pressure was 229 ± 24 mmHg, diastolic blood pressure 144 ± 18 mmHg and serum creatinine 183 µmol/l [interquartile range (IQR) 101–454]. The characteristics of patients who did not consent or could not participate were no different than those who were included in the present study, P > 0.10 for all variables. Participating patients received a standard questionnaire to add or verify any information that was missing or incomplete and a blood sample was taken for the extraction of genomic DNA. Patients were classified as having primary or secondary malignant hypertension according to the results of imaging studies of the renal arteries, hormone tests or renal biopsies. Imaging studies of the renal arteries were performed in 63 (84%) of the 75 patients with significant renal artery stenosis, defined as narrowing of the renal artery lumen more than 50% on computed tomography, magnetic resonance imaging or conventional angio-graphy. Renal biopsies were performed in 16 patients (21%) and one or more laboratory tests for phaeochromocytoma, Cushing’s disease and Conn’s syn-drome in 36 patients (48%). Informed consent was obtained from all study participants. The study was approved by the Medical Ethics Committee of the Academic Medical Centre (project number 03/053).


107

Controls Hypertensive and normotensive controls were randomly selected in a ratio of 1 : 2 to cases and individually matched for age, sex and ethnicity with cases of the present study. Matching for ethnicity was carried out for persons of West African descent, white Europeans and Asians from the Indian subcontinent. Controls were selected randomly from a large multiethnic population study (the SUNSET study) on cardiovascular risk conducted in the catchment area of the Academic Medical Centre. The details of the SUNSET study have been described elsewhere.12 The SUNSET study was approved by the Medical Ethics Committee of the Academic Medical Centre (project number 00/161).

Measurement of phenotypes Ethnic background was defined according to self-reported ethnicity. Blacks were mainly of West African descent, whites were predominantly inborn Dutch. Only four of the 75 patients (5%) with malignant hypertension came originally from the Indian subcontinent and were, for this study, classified as white. Cardiovascular risk factors and self-defined ethnicity were assessed through a questionnaire. Physical examination included height, weight and blood pressure. Blood pressure was measured in a quiet room while seated after 5 min rest, with an automated oscillometric device (Omron M4, Omron Matsusaka Co. Ltd, Japan). Using appropriate cuff sizes, two readings were taken on the right arm, the mean of the two readings was used in the analyses. Hypertension was defined as a systolic blood pressure > 140 mmHg, or diastolic blood pressure > 90 mmHg, or being on antihypertensive therapy. Awareness of hypertension was defined as self-reporting of any prior diagnosis of hypertension by a health-care professional. Treatment of hypertension was defined as receiving prescribed antihypertensive medication for the management of high blood pressure at the time of the interview. Previous hypertension was defined as self-reporting of hypertension assessed by a health-care professional or receiving antihyperten-sive drugs.

Measurement of genotypes Genomic DNA was extracted from whole blood and prepared on an AutopureLS (Gentra Systems, Minneapolis, USA) following standard procedures. Geno-typing was not possible for 10 patients with ACE I/D, 16 with AT1R A1166C and 8 with the AGT M235T polymorphism.

Chapter 5

108

Analysis of the AT1R A1166C and ACE I/D polymorphisms The angiotensin II type I receptor A1166C was determined using polymerase chain reaction (PCR) primers 5'-TTCCCCCAAAAGCCAAATCCCAC-3' and 5'-CAGGCTAGGGAGATTGCATTTCTGTCAG-3'. The PCR reaction consis-ted of 50 ng DNA, 2 pmol of each primer and 1× Amplitaq gold PCR master mix (Applied Biosystems Inc., Foster City, California, USA). The following PCR protocol was used on a ABI9700 PCR machine: initial denaturation for 10 min at 95°C and 37 cycles using a touchdown approach – the annealing temperature during the 1 min annealing step decreased from 72°C to 64°C over the first 12 cycles and remained 64°C for the remaining cycles – and extension for 60 s at 72°C. The PCR product was subjected to restriction digestion with 5 U of enzyme HpyF3I (MBI Fermentas, Burlington, Canada) at 37°C for 3 h and subsequent gel electrophoresis. For the A allele, HpyF3I cleaved the 428 bp PCR fragment into a 64-bp and a 364-bp fragment, whereas three fragments were generated for the 1166C allele: 64, 143 and 221 bp. The angiotensin-I-converting enzyme insertion/deletion polymorphism was determined using the following primers: 5'-CTGGAGACCACTCCCATCCTTT CT-3' and 5'-GATGTGGCCATCACATTCGTCACGAT-3'. Amplification with this primer pair results in 490 bp and 190 bp amplification products corre-sponding to the I and D alleles, respectively. The PCR reactions consisted of 50 ng DNA, 2 pmol of each primer, and 1× Amplitaq gold PCR master mix (Applied Biosystems Inc.). The following PCR protocol was used on a ABI9700 PCR machine: initial denaturation for 10 min at 95°C followed by 35 cycles of 1 min at 94°C, 1 min at 58°C and 1 min at 72°C. PCR products were detected on a 2% agarose-gel containing ethidium bromide. Mistyping of ID heterozygotes as D homozygotes may occur due to the preferential amplification of the D allele and inefficiency in the amplification of the I allele. To increase the specificity of DD genotyping, PCR amplifications were also performed with an insertion-specific primer pair (5'-TGGGACCACA GCGCCCGCCACTAC-3' and 5'-TCGCCAGCCCTCCCATGCCCATAA-3') in all samples that were found to be DD after amplification with the flanking primers. Briefly, insertion-specific amplification was performed using 50 ng DNA, 2 pmol of each primer, and 1× Amplitaq gold PCR master mix (Applied Biosystems Inc.). The following PCR protocol was used on a ABI9700 PCR machine: initial denaturation at 95°C for 10 min followed by 35 cycles of 1 min


109

at 94°C, 1 min at 66°C and 1 min at 72°C. PCR products were detected on a 2% agarose-gel containing ethidium bromide. Under these conditions, only the I allele produced a 335 bp amplicon. The 335 bp fragment was identified on 2% agarose-gel containing ethidium bromide. The reaction yields no products in samples of DD genotype.

Analysis of the AGT M235T polymorphism For the AGT M235T polymorphism, the Assay-by-Design service (www. appliedbiosystems.com) was used to set up Taqman genotyping assays. The Taqman reaction consisted of 5 ng of genomic DNA in a 2 µl volume and the following reagents: FAM and VIC probes (400 nmol/l), primers (1.8 µmol/l), 2× Taqman PCR master mix (ABgene). PCR cycling reactions were performed in 384-well PCR plates in an ABI 9700 PCR system (Applied Biosystems Inc.) and consisted of initial denaturation for 15 min at 95°C, and 40 cycles with denaturation of 15 s at 95°C and annealing and extension for 60 s at 60°C. Results were analysed by the ABI Taqman 7900HT using the sequence detection system 2.22 software (Applied Biosystems Inc.). Statistical analysis Continuous data were expressed as mean and SD or median and interquartile range for variables with a skewed distribution. Categorical data were expressed as number and percentages. Differences between groups were assessed by an independent samples t-test or Mann–Whitney U test for variables with a skewed distribution, chi-square statistics were used for categorical variables. Genotype frequencies of the different groups were compared by chi-squared analysis. Hardy–Weinberg equilibrium was tested to quantify differences between observed and expected genotypes by ethnic group (black and white). Multinominal logistic regression analysis was used to explore the risk of malignant hypertension compared to hypertensive and normotensive controls by ethnicity corrected for age, sex and smoking (model 1). In addition to these variables, previous hypertension and antihypertensive therapy were added to further explore the risk of malignant hypertension compared to hypertensive controls (model 2). Genotypes were expressed as dummy variables in this model, either with a recessive or dominant mode of inheritance. Odds ratios are given with their 95% confidence interval. For statistical analysis the SPSS software package for windows was used, version 14.0 (SPSS Inc., Chicago, Illinois, USA). A P value < 0.05 was considered statistically significant.

Chapter 5

110

Results

General characteristics A total of 75 patients with malignant hypertension were compared with 150 hypertensive and 150 normotensive controls, matched for age, sex and ethnicity. Their baseline characteristics are listed in Table 1. Age, sex, ethnicity and smoking were not different between the three blood pressure groups. Previous hypertension was as frequently diagnosed in patients with malignant hyper-tension as in hypertensive controls (P = 0.28). Twenty-four per cent of hypertensive and 31% of malignant hypertensive patients (72% and 64% of those previously diagnosed with hypertension, respectively) received anti-hypertensive medication (P = 0.22 for difference between groups).

Table 1 Characteristics of malignant hypertensive patients and hypertensive and normotensive controls.

Values are numbers and percentages unless stated otherwise. IQR, interquartile range; SD, standard deviation. * Current smoker is defined as smoking prior to presentation in patients with malignant hypertension. † Few normotensive persons were previously diagnosed with hypertension, but had nor-mal readings during physical examination without antihypertensive therapy.

Patient characteristics Normotensive Hypertensive Malignant Hypertensive

Patients 150 150 75 Age, yrs (mean ± SD) 43.5 ± 7.7 44.2 ± 6.1 43.1 ± 11.4 Male 90 (60%) 90 (60%) 44 (59%) Black 74 (49%) 74 (49%) 37 (49%) Systolic blood pressure, mmHg (mean ± SD) 116 ± 12 144 ± 16 228 ± 21

Diastolic blood pressure, mmHg (mean ± SD) 75 ± 7 93 ± 9 144 ± 16

Current Smoker* 80 (53%) 67 (45%) 32 (43%) Serum creatinine, µmol/l (median, [IQR]) 75 [66-87] 76 [68-89] 162 [103-399]

Previous Hypertension† 13 (9%) 50 (40%) 36 (49%) Antihypertensives - 36 (24%) 23 (31%)


111

In patients with malignant hypertension, 37 patients (49%) had grade IV retinopathy, 60 (80%) had evidence of left ventricular hypertrophy and 25 (33%) had evidence of thrombotic microangiopathy (defined as the presence of thrombocytopenia < 150 × 109 with either an elevated lactic dehydrogenase level or presence of fragmented red blood cells in a peripheral blood smear). Serum creatinine values were significantly higher in patients with malignant hypertension compared to hypertensive or normotensive controls (P < 0.01). A secondary cause could be established in 18 patients (24%) with malignant hypertension. They had either primary renal disease (8), renal artery stenosis (5) or other causes (5). The latter included oral contraceptive use (2), liquorice abuse (2) and exogenous corticoid excess (1). No underlying cause could be established in 56 (75%) despite extensive work-up.

Association between genotypes and malignant hypertension Genotypes of the tested polymorphisms in the different blood pressure groups were all in Hardy–Weinberg equilibrium (P > 0.20). Genotypes and allelic frequencies of the ACE I/D and AT1R A1166C polymorphisms are listed in Table 2 according to blood pressure category. The frequency distributions of the ACE I/D and the AT1R A1166C polymorphism showed no significant difference between patients with malignant hypertension and hypertensive and normotensive controls. As a result of differences in allelic frequencies of the M235T polymorphism in the AGT gene between white Europeans and blacks from West African descent [13], analysis was performed after stratification for ethnicity. There was a large difference in the allelic frequency distribution of the AGT M235T polymorphism in white subjects with malignant hypertension compared to hypertensive and normotensive controls. In white subjects, the TT genotype was present in 68.4% with malignant hypertension compared to 13.2% of hypertensive (χ2 = 37.3, P < 0.001) and 18.7% of normotensive controls (χ2 = 28.2, P < 0.001). The MT genotype was not associated with an increased risk of malignant hypertension, being present in 23.7% of malignant hypertensive patients and in 43.4% of hypertensive and 45.3% of normotensive controls. In black subjects the TT genotype was detected in 88.2% with malignant hyper-tension compared to 82.2% in hypertensive (χ2 = 0.91, P = 0.64) and 84.3% in normotensive controls (χ2 = 0.64, P = 0.73). The genotype frequencies of the AGT M235T polymorphism stratified for ethnicity are shown in figure 1.

Chapter 5

112

Table 2 Genotypes and allelic frequencies [number (%)] of the ACE I/D polymorphism and the A1166C polymorphism in the AT1R in patients with malignant hypertension compared to hypertensive and normotensive controls.

P is for difference between blood pressure groups with malignant hypertension as the reference group. Fig. 1.

White subjects Black subjects

AGT Met [M] 235 Thr [T]

Genotype frequencies of the M235T polymorphism of the AGT gene stratified by ethnicity and blood pressure group. Data are for n = 72 malignant hypertensive patients (34 black and 38 white), n = 148 hypertensive controls (73 black and 76 white) and n = 145 normotensive controls (70 black and 75 white).

ACE I→D II ID DD P normotensives 31 (21.2) 71 (48.6) 44 (30.1) 0.53 hypertensives 31 (20.8) 79 (53.0) 39 (26.2) 0.79 malignant hypertensive 17 (24.3) 37 (52.9) 16 (22.9) - A1166→C AA AC CC normotensive 97 (69.8) 35 (25.2) 7 (5.0) 0.40 hypertensive 102 (69.4) 42 (28.6) 3 (2.0) 0.76 malignant hypertensive 54 (74.0) 18 (24.7) 1 (1.4) -


113

There was no difference in genotypic distribution of AGT M235T by sex, with identical allelic frequencies in females and males in both black and white subjects. The odds ratios of malignant hypertension stratified for ethnicity are listed in Table 3. For whites with the TT genotype the risk of malignant hyper-tension was 14.3 (5.5–37.1) compared to hypertensive controls, and 9.4 (3.8–23.2) compared to normotensive controls. The association of AGT M235T with malignant hypertension was not significant in blacks. After adjustment for age, sex and smoking the TT genotype remained associated with malignant hyper-tension in whites, odds ratio 14.9 (5.6–39.7) compared to hypertensive and an odds ratio of 10.6 (4.2–27) for normotensive controls. Table 3 Odds ratios and 95% confidence intervals for malignant hypertension in subjects having the TT genotype versus the MT or MM genotype.

Comparison of malignant hypertensive patients with hypertensive and normotensive controls. * Model 1 = Adjusted for age, sex and smoking. † Model 2 = Additional adjustment for previous hypertension and anti-hypertensive therapy. Additional adjustment for previous hypertension and antihypertensive therapy did not change the odds of malignant hypertension compared to hypertensive controls. As the AGT M235T polymorphism may also be associated with renal insufficiency in patients with hypertension [14], we also corrected for serum creatinine in addition to age, sex and ethnicity, knowing that renal insufficiency often coincides with malignant hypertension. After additional correction for serum creatinine, there was a small decrease in the odds of malignant hyper-tension with an odds ratio of 9.1 (2.7–30.2) in white Europeans compared to hypertensive controls.

Crude odds Model 1* Model 2† Hypertensive whites 14.3 (5.5-37.1) 14.9 (5.6-39.7) 15.2 (5.3-43.6) blacks 1.6 (0.5-5.4) 1.6 (0.5-5.4) 2.4 (0.7-8.9) Normotensive whites 9.4 (3.8-23.2) 10.6 (4.2-27) - blacks 1.4 (0.4-4.8) 1.4 (0.4-5.1) -

Chapter 5

114

Within the group of malignant hypertensive patients, there was no difference in age, sex, ethnicity and blood pressure between patients with the TT genotype compared to those with the MT or MM genotype, although carriers of the TT genotype were more likely to be of white European origin (54% versus 25%, P = 0.08). In contrast, patients with the TT genotype had more renal dysfunction and thrombotic microangiopathy than patients with the MT or MM genotype. Median serum creatinine was 205 µmol/l (IQR 111–451) for the TT genotype compared to 95 µmol/l (IQR 70–225) for MT and MM genotypes (P < 0.01). Thrombotic microangiopathy was present in 39% of patients with the TT genotype compared to 6% of patients with MT or MM genotypes (P = 0.03). The frequency of secondary causes was similar between TT genotypes (25%) and MT or MM genotypes (21%). Discussion In this multiethnic population with malignant hypertension, we observed a marked difference in the genotype frequency distribution of the AGT M235T polymorphism in white patients with malignant hypertension compared to matched hypertensive and normotensive controls. The odds of malignant hyper-tension was approximately 10 to 1 in carriers of the TT genotype compared to hypertensive or normotensive MT and MM carriers. Within the group of malignant hypertensive patients, the TT genotype was associated with a more severe phenotype, as evidenced by increased renal dysfunction and a higher frequency of thrombotic microangiopathy compared to MT and MM carriers, despite similar age, sex, ethnicity and blood pressure. The frequency distribu-tions of the ACE I/D and the AT1R A1166C polymorphisms were not different between blood pressure categories or ethnic groups. Compared to other reported frequencies of the AGT M235T polymorphism, the frequency of the 235T allele in our sample was somewhat lower for white hypertensive subjects (34.9%) than that reported by others.15 This may be related to the predominance of patients with mild hypertension in the hypertensive control group of our study. In contrast, the 235T allele frequency for normotensive controls in our sample (41.3%) is similar to previous studies, which reported allele frequencies between 31 and 49% for white normotensive subjects.15 In black subjects, the frequency of the 235T allele was 91.4% among hypertensive and 90.4% among normotensive controls, which is similar with those reported in populations of West Africa (93%) and somewhat higher than in


115

African-Americans (82%).16 This difference may point towards limited admixture of our group with populations of Western European origin. The high frequency of the 235T allele in normotensive and hypertensive West Africans makes it difficult to establish an association with malignant hypertension. So, we cannot exclude that an association of the AGT M235T polymorphism with malignant hypertension is present in blacks; however, due to the high frequency of the 235T allele our study had insufficient power to detect this difference. It is conceivable that the predominance of 235T homozygosity in blacks contributes to a higher susceptibility of malignant hypertension, explaining the higher incidence of malignant hypertension in this population.9 We previously estimated, based on the demographic characteristics of this population, that malignant hypertension is 3–4 times more common in blacks than in whites, in part as a result of differences in the control of hypertension and adherence to antihypertensive treatment.8 Apart from these socio-economic differences, however, the predominance of homozygosity for AGT 235T in this group may be an additional risk factor for malignant hypertension. A previous study in white patients with malignant hypertension found no significant association of AGT M235T compared to hypertensive controls, although the TT genotype was 10% higher in malignant hypertensive patients;10 however, selection of patients and controls may have influenced this association. Another study in a large cohort of white patients with severe (but not malignant) hypertension also failed to demonstrate an association of the M235T polymorphism compared to normotensive controls.17 A likely explanation for this finding is that, in general, the RAS is not activated in subjects with mild to severe essential hypertension.18 This is in contrast with malignant hypertensive patients who, in most cases, have an extremely activated RAS.1, 19 Our results are in agreement with a study conducted in patients with pre-eclampsia.20 Pre-eclampsia and malignant hypertension share similar clinical and pathophysiological features, including the presence of an activated RAS. Homozygosity for 235T was significantly associated with the risk of pre-eclampsia in white patients compared to controls who were previously pregnant, but had no history of pre-eclampsia. As the hepatic transcription of AGT is regulated by oestrogen, we performed a separate analysis with stratification by sex; however, the allelic frequency distribution of AGT M235T between males and females was similar in all blood pressure categories.

Chapter 5

116

Functionally, the TT genotype of AGT M235T is associated with an average rise of 11% in plasma AGT compared to MT or MM genotypes in white subjects.21 Such an association between genotype and phenotype has not been demon-strated in blacks; however, the studies conducted in black populations may have been too small to detect differences in plasma AGT.22, 23 The cleavage of AGT by renin is the rate-limiting step in activation of the renin–angiotensin system.24,

25 Small increases in either renin or AGT levels may increase angiotensin II production and alter blood pressure.26 Possibly, increases in basal plasma AGT, as observed in white patients with the TT genotype, facilitate the onset of the renin-mediated blood pressure acceleration observed in malignant hypertension. The question is whether AGT M235T itself is associated with increased levels of plasma AGT. Several studies suggest that the promoter region that regulates transcriptional activity is in tight disequilibrium with AGT M235T and that therefore AGT M235T may serve as a genetic marker for the active variant.27, 28

The lack of association between the ACE I/D polymorphism and malignant hypertension in our study is in contrast to a previous report which showed an association between the ACE DD polymorphism and the risk of malignant hypertension compared to hypertensive controls.10 This finding could not be reproduced in a later study, although a significant increase in the frequency of the ACE DD genotype was found in malignant hypertensive patients compared to normotensive controls.11 Therefore, the association of the ACE I/D polymorphism and malignant hypertension remains undetermined. We had the possibility of analysing the genotypes of consecutive patients admitted with malignant hypertension and to compare them to allelic frequen-cies within a large cohort matched for age, sex and ethnicity in the source population served by our hospital. To prevent selection of surviving patients we limited ourselves to the inclusion of patients admitted in the past 10 years. As a result, our study is limited by a fairly small sample size and therefore insufficient to detect small differences in allelic frequencies or to compare the interaction of separate polymorphisms on the risk of malignant hypertension. Finally, we cannot exclude the possibility that variations in patterns of linkage disequilibrium with other polymorphisms in the AGT gene, between white European compared to blacks of West African descent, might interfere with our observations.


117

Conclusion In a multiethnic cohort of consecutive patients with malignant hypertension, homozygosity for the 235T polymorphism in the AGT gene is associated with an increased risk of malignant hypertension with an odds of approximately 10 to 1 compared to hypertensive and normotensive controls, but only in white subjects. The high frequency of the 235T allele observed in populations of West African descent may help explain the higher frequency of malignant hyper-tension in this group. In patients with malignant hypertension, the TT genotype of the AGT M235T polymorphism was associated with a more severe phenotype, as evidenced by increased frequency of renal dysfunction and thrombotic microangiopathy, compared to MT or MM carriers, despite similar age, sex, ethnicity and blood pressure. These observations may provide a better understanding of the pathophysiology of malignant hypertension; however, larger association or linkage studies are needed for a more detailed risk assessment.

References

1. Laragh JH, Ulick S, Januszewicz V, Deming QB, Kelly WG, Lieberman S. Aldosterone secretion and primary and malignant hypertension. J Clin Invest 1960; 39:1091-1106.

2. Lohmeier TE, Tillman LJ, Carroll RG, Brown AJ, Guyton AC. Malignant hyper-tensive crisis induced by chronic intrarenal norepinephrine infusion. Hypertension 1984; 6(2 Pt 2):I177-I182.

3. Mohring J, Petri M, Szokol M, Haack D, Mohring B. Effects of saline drinking on malignant course of renal hypertension in rats. Am J Physiol 1976; 230(3):849-857.

4. Ganten D, Wagner J, Zeh K, Bader M, Michel JB, Paul M, et al. Species specificity of renin kinetics in transgenic rats harboring the human renin and angiotensinogen genes. Proc Natl Acad Sci U S A 1992; 89(16):7806-7810.

5. Montgomery HE, Kiernan LA, Whitworth CE, Fleming S, Unger T, Gohlke P, et al. Inhibition of tissue angiotensin converting enzyme activity prevents malignant hypertension in TGR(mREN2)27. J Hypertens 1998; 16(5):635-643.

6. Mervaala EM, Muller DN, Park JK, Schmidt F, Lohn M, Breu V, et al. Monocyte infiltration and adhesion molecules in a rat model of high human renin hyper-tension. Hypertension 1999; 33(1 Pt 2):389-395.

Chapter 5

118

7. Lip GY, Beevers M, Beevers G. The failure of malignant hypertension to decline: a survey of 24 years' experience in a multiracial population in England. J Hypertens 1994; 12(11):1297-1305.

8. van den Born BJ, Koopmans RP, Groeneveld JO, van Montfrans GA. Ethnic dispa-rities in the incidence, presentation and complications of malignant hypertension. J Hypertens 2006; 24(11):2299-2304.

9. Lip GY, Beevers M, Beevers DG. Complications and survival of 315 patients with malignant-phase hypertension. J Hypertens 1995; 13(8):915-924.

10. Stefansson B, Ricksten A, Rymo L, Aurell M, Herlitz H. Angiotensin-converting enzyme gene I/D polymorphism in malignant hypertension. Blood Press 2000; 9(2-3):104-109.

11. Espinel E, Tovar JL, Borrellas J, Piera L, Jardi R, Frias FR, et al. Angiotensin-converting enzyme i/d polymorphism in patients with malignant hypertension. J Clin Hypertens (Greenwich ) 2005; 7(1):11-15.

12. Agyemang C, Bindraban N, Mairuhu G, Montfrans G, Koopmans R, Stronks K. Prevalence, awareness, treatment, and control of hypertension among Black Surinamese, South Asian Surinamese and White Dutch in Amsterdam, The Netherlands: the SUNSET study. J Hypertens 2005; 23(11):1971-1977.

13. Rotimi C, Puras A, Cooper R, Farlane-Anderson N, Forrester T, Ogunbiyi O, et al. Polymorphisms of renin-angiotensin genes among Nigerians, Jamaicans, and African Americans. Hypertension 1996; 27(3 Pt 2):558-563.

14. Fabris B, Bortoletto M, Candido R, Barbone F, Cattin MR, Calci M, et al. Genetic polymorphisms of the renin-angiotensin-aldosterone system and renal insufficiency in essential hypertension. J Hypertens 2005; 23(2):309-316.

15. Kunz R, Kreutz R, Beige J, Distler A, Sharma AM. Association between the angio-tensinogen 235T-variant and essential hypertension in whites: a systematic review and methodological appraisal. Hypertension 1997; 30(6):1331-1337.

16. Rotimi C, Morrison L, Cooper R, Oyejide C, Effiong E, Ladipo M, et al. Angioten-sinogen gene in human hypertension. Lack of an association of the 235T allele among African Americans. Hypertension 1994; 24(5):591-594.

17. Sunder-Plassmann G, Kittler H, Eberle C, Hirschl MM, Woisetschlager C, Derha-schnig U, et al. Angiotensin converting enzyme DD genotype is associated with hypertensive crisis. Crit Care Med 2002; 30(10):2236-2241.


119

18. Parfrey PS, Markandu ND, Roulston JE, Jones BE, Jones JC, MacGregor GA. Relation between arterial pressure, dietary sodium intake, and renin system in essential hypertension. Br Med J (Clin Res Ed) 1981; 283(6284):94-97.

19. Linton AL, Gavras H, Gleadle RI, Hutchison HE, Lawson DH, Lever AF, et al. Microangiopathic haemolytic anaemia and the pathogenesis of malignant hypertension. Lancet 1969; 1(7609):1277-1282.

20. Ward K, Hata A, Jeunemaitre X, Helin C, Nelson L, Namikawa C, et al. A molecu-lar variant of angiotensinogen associated with preeclampsia. Nat Genet 1993; 4(1):59-61.

21. Sethi AA, Nordestgaard BG, Tybjaerg-Hansen A. Angiotensinogen gene poly-morphism, plasma angiotensinogen, and risk of hypertension and ischemic heart disease: a meta-analysis. Arterioscler Thromb Vasc Biol 2003; 23(7):1269-1275.

22. Bloem LJ, Manatunga AK, Tewksbury DA, Pratt JH. The serum angiotensinogen concentration and variants of the angiotensinogen gene in white and black children. J Clin Invest 1995; 95(3):948-953.

23. Forrester T, Farlane-Anderson N, Bennet F, Wilks R, Puras A, Cooper R, et al. Angiotensinogen and blood pressure among blacks: findings from a community survey in Jamaica. J Hypertens 1996; 14(3):315-321.

24. Cumin F, Le-Nguyen D, Castro B, Menard J, Corvol P. Comparative enzymatic studies of human renin acting on pure natural or synthetic substrates. Biochim Biophys Acta 1987; 913(1):10-19.

25. Gould AB, DeWolf R, Goodman S, Onesti G, Swartz C. Kinetic studies of the human renin and human substrate reaction. Biochem Med 1980; 24(3):321-326.

26. Gould AB, Green D. Kinetics of the human renin and human substrate reaction. Cardiovasc Res 1971; 5(1):86-89.

27. Inoue I, Nakajima T, Williams CS, Quackenbush J, Puryear R, Powers M, et al. A nucleotide substitution in the promoter of human angiotensinogen is associated with essential hypertension and affects basal transcription in vitro. J Clin Invest 1997; 99(7):1786-1797.

28. Dickson ME, Zimmerman MB, Rahmouni K, Sigmund CD. The -20 and -217 promoter variants dominate differential angiotensinogen haplotype regulation in angiotensinogen-expressing cells. Hypertension 2007; 49(3):631-639.

Chapter 6

The renin-angiotensin system in malignant hypertension revisited - plasma renin activity, thrombotic microangiopathy and renal failure in malignant hypertension Bert-Jan H van den Born , Richard P Koopmans, Gert A van Montfrans

Am. J. Hypertens. 2007; 20: 900-906.

Chapter 6

124

Abstract Background Malignant hypertension is a renin-dependent form of hypertension. However, the variations in renin-angiotensin system (RAS) activation in malignant hypertension are incompletely understood. A proposed mechanism for ongoing RAS activation is the presence of thrombotic microangiopathy resulting in renovascular ischemia. Methods We prospectively examined the association between plasma renin activity (PRA), thrombotic microangiopathy and renal dysfunction in 30 consecutive patients with malignant hypertension (n = 18) and severe hyper-tension (n = 12). PRA and aldosterone were measured in supine position and before initiating therapy. Results PRA was 8.8 ng AI/ml/h (interquartile range [IQR] 4.8 - 20) in malignant hypertensive patients and 2.8 ng AI/ml/h (IQR 0.6 - 6.3) in patients with severe hypertension (P < 0.01). Aldosterone was 1.30 ± 1.02 nmol/l in patients with malignant hypertension compared to 0.44 ± 0.37 nmol/l in those with severe hypertension (P < 0.01). In malignant hypertension, PRA highly correlated with lactic dehydrogenase (LDH) (r = 0.76, P < 0.001), meaning that 58% of the variations in PRA could be explained by LDH. PRA positively correlated with serum creatinine values at presentation (r = 0.50, P = 0.007), but adjustment for LDH abolished the effect of PRA on creatinine (P = 0.24). Conclusion PRA and aldosterone were markedly elevated in patients with malignant hypertension, but not in severely hypertensive patients despite small differences in blood pressure. The strong logarithmic correlation between PRA, microangiopathic markers and renal dysfunction suggests a renin mediated acceleration of vascular damage and renal dysfunction in patients with malignant hypertension.

The renin-angiotensin system in malignant hypertension

125

Introduction Malignant hypertension is a hypertensive emergency characterized by a severe elevation of blood pressure and ischemic retinal lesions including the bilateral presence of cotton wool spots, flame shaped hemorrhages and papilledema. In the acute phase the renin-angiotensin system is activated in most patients with malignant hypertension.1, 2 However, the extent of renin-angiotensin system activation varies greatly. Some patients with malignant hypertension present with normal renin activity, whilst others have profound activation of the renin-angiotensin system.2-6 These variations in renin-angiotensin activation are incompletely understood. A possible pathophysiological mechanism for the development of malignant hypertension originates from the idea that microvascular damage and pressure natriuresis, induced by a critical elevation of blood pressure, result in ischemia of the renovascular bed and a paradoxical activation of the renin-angiotensin system.7 Stimulation of the renin-angiotensin system, in turn, may further promote blood pressure increase by high levels of circulating angiotensin II and lead to more vascular damage, thereby inducing a vicious circle (figure 1).8, 9 The detrimental effects of high levels of renin and angiotensin have been demonstrated in transgenic rats harboring the human renin and angiotensin genes. They develop severe elevation of blood pressure and profound endothelial damage with evidence of a thrombotic microangiopathy early in life.10 Treatment with anti-hypertensive drugs not blocking the renin-angiotensin system does not ameliorate renal and vascular damage, despite normalization of blood pressure in these animals.11 However, administration of a non-blood pressure lowering dose of ACE-inhibition regresses these vascular abnormalities suggesting that high levels of renin and angiotensin may have a direct vasculotoxic effect.12 In humans, the presence of thrombotic microangiopathy seems associated with ischemic glomerular changes and fibrinoid necrosis of renal arterioles.3 We previously showed that thrombotic microangiopathy, evidenced by a decreased platelet count and either an increased lactic dehydrogenase (LDH) or presence of schistocytes, is common and strongly associated with renal dysfunction in a retrospective cohort of patients with malignant hypertension.13 In the present study we hypothesized, based on the above, that the degree of renin-angiotensin activation is associated with the severity of thrombotic microangiopathy in patients with malignant hypertension. Therefore we prospectively examined the association between plasma renin

Chapter 6

126

activity (PRA), microangiopathic markers (LDH and platelets) and renal function in consecutive patients presenting at the emergency department with a suspected hypertensive emergency. Fig. 1 Simplified presentation of the vicious circle as previously proposed by Kincaid-Smith.7

Patients and methods PRA and aldosterone were measured in 30 consecutive patients with a suspected hypertensive emergency presenting at the emergency department of a large teaching hospital serving a multi-ethnic community in Amsterdam, the Netherlands. After presentation blood pressure was measured using an aneroid sphygmomanometer in the recumbent position. All retinal examinations were carried out in mydriasis by an ophthalmologist. Blood samples for PRA and aldosterone were taken after at least 30 minutes rest with the patient supine and prior to the initiation of therapy. Malignant hypertension was defined according to the WHO criteria with bilateral linear or flame shaped hemorrhages or ‘cotton-wool’ exudates with or without papilledema on funduscopic exami-nation. Left ventricular hypertrophy was assessed by electrocardiography using the Sokolow-Lyon criteria. Macroalbuminuria was defined as urinary protein excretion > 300 mgr/l or 2+ for dipstick proteinuria. Kidney replacement therapy

Severe elevation of blood pressure

Endothelial damage

RAAS activation

Renovascular ischemia Volume depletion

Pressure natriuresis


127

was defined as the need for hemodialysis or peritoneal dialysis. Additional tests for secondary hypertension were performed on indication. Renovascular imaging by CT, MR or conventional angiography were performed in 22/30 patients (73%), one or more hormonal tests for Conn, Cushing or pheochromo-cytoma in 11 patients (37%) and kidney biopsies in 4 patients. PRA samples were immediately frozen at -20°C after collection. PRA was measured by an enzyme kinetic assay as described previously.14 With this method the incubation product of renin, angiotensin I, is generated from angiotensinogen and measured by radioimmunoassay (RIA). 125I labelled angiotensin I was acquired from Amersham Pharmacia Biosciences UK. The detection limit is 0.3 ng AI/ml/h, the inter-assay coefficient of variation (CV) 6.0 to 11.0% and the intra-assay CV 4.0 to 6.0% (at 10.6 to 1.4 ng AI/ml/h). Normal values on a normal salt diet are < 0.3-3.2 ng A1/ml/h in the recumbent position and < 0.3-4.9 ng AI/ml/h while standing. Plasma aldosterone was measured by commercial RIA (coat-a-count, Diagnostic Products Corporation, Los Angeles, CA). The detection limit is 0.03 nmol/l, the interassay CV 4.7 to 12.0% and the intra-assay CV 3.6 to 8.4% (at 1.53 to 0.13 nmol/l). Normal values for aldosterone are 0.03-0.35 nmol/l. Serum creatinine and urea nitrogen were measured by means of enzymatic methods, serum potassium was determined using ion selective electrodes. LDH was assessed by spectrophotometry (normal < 220 U/l). All measurements were performed using standard clinical chemical methods and reagents on a Hitachi 747 analyser (Roche Diagnostics, Almere, the Netherlands). Statistical analysis Baseline variables were described using mean and standard deviation (SD) or median and range, or interquartile range (IQR) for variables with a skewed distribution. Differences between groups were calculated using an independent samples t-test for parametric and Mann-Whitney U test for non-parametric distributions where appropriate. Categorical variables were calculated using χ2 statistics with Yates’ correction or Fisher’s exact test for cells with expected values < 5 (all 2-sided). Variables with a skewed distribution (PRA, LDH and creatinine) were log transformed before statistical analysis. Linear regression analysis was used to assess correlations between variables. R2 was used to give an estimate of the variations in PRA explained by LDH. The Fisher’s z-transformation was used to give a confidence interval (CI) of this estimate. Values below the detection limit of the renin assay were included in the analysis as having a value of 50% of the lower limit of detection. To assess the contribution of PRA on future kidney replacement therapy, Cox-regression

Chapter 6

128

analysis was used with PRA and with PRA and LDH in the model. For statistical analysis the SPSS software package for windows was used, version 12.0 (Chicago, Illinois, USA). P < 0.05 was considered to indicate a statistically significant difference. Results Eighteen patients had malignant hypertension. Twelve patients had severe hypertension (BP > 180/110 mmHg) without retinal abnormalities (10 patients) or unilateral retinal defects (2 patients). Thirteen patients were from sub-Saharan African descent, 2 from the Asian subcontinent, the remaining patients were inborn Dutch. Systolic blood pressure was 232 ± 20 mmHg in patients with malignant hypertension and 219 ± 30 mmHg in those with severe hypertension (P = 0.20). Patients with malignant hypertension had higher diastolic blood pressure 148 ± 15 mmHg than those with severe hypertension 133 ± 16 mmHg (P = 0.02). Left ventricular hypertrophy was present in 19 patients, 4 (33%) with severe hypertension and 15 (83%) with malignant hypertension (P < 0.01). Hypertension was previously diagnosed in 12 patients and treated in 7. Five patients admitted having stopped their antihypertensive medication several months to years before presentation, leaving two patients on possible anti-hypertensive treatment at the time of admission (one in each group). A secondary cause could be established in 5 patients (2 with severe and 3 with malignant hypertension): unilateral renal artery stenosis, high dose corticosteroid treatment, cortisol producing adrenal carcinoma and 2 with IgA nephropathy. In the remaining patients no secondary cause could be identified. Five patients, all with malignant hypertension, were in need of kidney replacement therapy (hemodialysis or peritoneal dialysis) within 6 months after presentation. Platelet count was lower and LDH higher in patients with malignant hypertension compared to those with severe hypertension. The correlation between microangiopathic markers, platelet count and (log) LDH, was r = 0.77, P < 0.001. Clinical and laboratory characteristics are fully described in table 1.


129

Table 1 Clinical characteristics of patients with malignant hypertension and severe hypertension

Values are numbers and percentages unless stated otherwise. BP=blood pressure; SD=standard deviation; IQR=interquartile range; KRT=Kidney replacement therapy * Data missing for 2 patients (one in each group).

Patient characteristics Malignant hypertension (n = 18)

Severe hypertension (n = 12)

P

Age, yrs (mean ± SD) 41 ± 10 45 ± 17 0.49 Male 14 (78%) 6 (50%) 0.14 Black 7 (39%) 6 (50%) 0.82 Systolic BP, mmHg (mean ± SD) 232 ± 20 219 ± 30 0.20

Diastolic BP, mmHg (mean ± SD) 148 ± 15 133 ± 16 0.02

Current Smoker 6 (33%) 2 (17%) 0.42 Diabetes mellitus 1 (6%) 2 (17%) 0.55 Previous hypertension 6 (33%) 6 (50%) 0.46 On active treatment 1 (6%) 1 (8%) 1.00 Left ventricular hypertrophy 15 (83%) 4 (33%) < 0.01 Serum creatinine, mg/dl (median, [IQR]) 2.30 [1.19 – 7.13] 1.08 [0.82 – 1.60] < 0.01

BUN, mg/dl (mean ± SD) 46.5 (16.2 – 69.5) 17.4 (14.0 – 22.7) 0.02 Macroalbuminuria 12 (67%) 5 (42%) 0.33 Need for KRT < 6 mnths 5 (28%) 0 Lactic dehydrogenase, U/l (median, [IQR]) 457 [45 – 790] 227 [186 – 377] 0.02

Platelet count x109/l (median, [IQR]) 181 [83 – 243] 251 [207 – 331] 0.03

Fragmented red blood cells 6 (33%) 1 (8%) 0.19 Serum potassium, mmol/l (mean ± SD) 3.5 ± 0.7 3.8 ± 0.6 0.35

PRA, ng AI/ml/h* (median, [IQR]) 8.8 [4.8 – 20] 2.8 [0.6 – 6.3] < 0.01

Serum aldosterone, nmol/l (mean ± SD) 1.30 ± 1.02 0.44 ± 0.37 < 0.01

Chapter 6

130

Patients with malignant hypertension had higher PRA and aldosterone values at presentation than patients with severe hypertension (figure 2). Five patients (45%) with severe hypertension had normal PRA (< 4.9 ng AI/ml/h), in 2 patients with severe hypertension PRA was below the lower limit of detection (< 0.3 ng AI/ml/h). One patient received high dose corticosteroids and one patient with diabetic nephropathy was on medication at the time of admission. Aldosterone levels in these patients were 0.06 nmol/l and 0.26 nmol/l. Four of 17 patients (24%) with malignant hypertension had normal PRA, in none PRA was below the detection limit. PRA could not be measured in two patients because samples were incorrectly collected. Two patients were on possible anti-hypertensive treatment at presentation: one patient with malignant hypertension was prescribed an ACE-inhibitor at the time of admission and had a PRA of 2.0 ng/ml/h and an aldosterone of 0.3 nmol/l. One patient with severe hypertension was prescribed a combination of a loop-diuretic, ACE-inhibitor, calcium-antagonist and beta-blocker. In this patient, PRA was suppressed (< 0.3 ng AI/ml/h) and his serum aldosterone concentration 0.26 nmol/l. Fig. 2

0

10

20

30

40

50

Group 1 Group 2

Plas

ma

reni

n ac

tivity

(ngA

I/ml/h

)

PRA in patients with malignant hypertension (group 1) and severe hypertension (group 2).


131

Correlation between PRA and intravascular hemolysis There was a strong correlation between PRA and markers for intravascular hemolysis. PRA and LDH correlated best when both were analyzed after logarithmic transformation, r = 0.52, P = 0.007 (figure 3a). There was a nega-tive correlation between PRA and platelet count (r = -0.45, P = 0.02) meaning that higher PRA values were associated with lower platelets. Patients with malignant hypertension had the strongest correlation between PRA and LDH (r = 0.76, P < 0.001), meaning that PRA could explain 58% (95% CI 19% - 83%) of the variations in LDH. The correlation between LDH and PRA was not significant in patients with severe hypertension (r = 0.34, P = 0.35). Correlation between PRA and renal function A higher PRA was associated with more severe renal dysfunction, r = 0.50, P = 0.007 (figure 3b), but only in patients with malignant hypertension (r = 0.50, P = 0.04). There was no correlation between PRA and renal function in patients with severe hypertension (r = 0.07, P = 0.84). Higher LDH and lower platelets were associated with increased serum creatinine values at presentation. The correlation between LDH and serum creatinine was r = 0.60, P = 0.001 after logarithmic transformation for both variables. There was a negative correlation between platelet count and (log) serum creatinine (r = -0.65, P < 0.001). Adjustment for LDH on serum creatinine corrected the effect of PRA on serum creatinine (B = 0.14, P = 0.24). The interaction between LDH and PRA was not significant (P = 0.68) suggesting that LDH and PRA are independently asso-ciated with renal function. Cox-regression analysis showed that PRA values at admission could serve as an indicator for future renal replacement therapy, although this finding did not reach statistical significance (B = 1.43, P = 0.13). Adjustment for LDH levels at presentation also abolished the effect of PRA on future renal failure (B = -0.07, P = 0.95).

Chapter 6

132

Fig. 3a

2.0 2.4 2.8 3.2

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Log LDH (U/l)

Log

PRA

(ngA

I/ml/h

)

Fig. 3b

2.0 2.5 3.0 3.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Log creatinine (µmol/l)

Log

PRA

(ngA

I/ml/h

)

Scatter plots showing the correlation between: (a) plasma renin activity (PRA) and lactic dehydrogenase (LDH) after logarithmic transformation, r = 0.52 (P = 0.007) and (b) PRA and creatinine after logarithmic transformation, r = 0.50 (P = 0.007). Unfilled squares represent patients with severe hypertension and filled circles patients with malignant hypertension. The closed line represents the regression line with the 95% confidence interval (dotted line) for all patients (patients with malignant hypertension and severe hypertension).


133

Correlation between PRA, aldosterone and potassium Aldosterone was 1.30 ± 1.02 nmol/l in patients with malignant hypertension compared to 0.44 ± 0.37 nmol/l in patients with severe hypertension (P < 0.01). Aldosterone correlated strongly with PRA (r = 0.65, P < 0.001), however this association was linear for patients with severe hypertension and loglinear for patients with malignant hypertension: r = 0.54 (P = 0.03) for linear association between PRA and aldosterone and r = 0.66 (P = 0.004) after logarithmic trans-formation of PRA. Higher aldosterone levels were associated with lower potassium concentrations, although this did not reach statistical significance (r = 0.32, P = 0.10), likely because of more severe renal insufficiency in those with higher aldosterone levels. Discussion The main finding of our study is that PRA and aldosterone, strongly elevated in most patients with malignant hypertension, highly correlate with markers for intravascular hemolysis and renal dysfunction. Despite relatively small diffe-rences in systolic and diastolic blood pressure between patients with severe and malignant hypertension, large differences were found with regard to renal function, thrombotic microangiopathy and activation of the renin-angiotensin-aldosterone system. This could support the notion that not the absolute blood pressure level, but rather the rate of change in blood pressure is more important for the development of malignant hypertension.15 In that sense the ischemic retinal changes that corroborate malignant hypertension may be considered a reliable indicator for ischemic vascular lesions elsewhere, particularly in the renovascular bed. The paradoxical activation of the renin-angiotensin system in malignant hypertension has long been established both from human data,1 and from experimental studies in animals.16 However, previous reports have shown that not all patients with malignant hypertension have an activated renin-angiotensin system,17, 18 which is consistent with our finding of a normal (but not suppressed) PRA in some patients with malignant hypertension. One of the possible mechanisms responsible for renin-angiotensin system activation may be the induction of renovascular ischemia by damage of small renal arterioles as a result of severe hypertension.9 Activation of the renin-angiotensin system in turn may result in further vascular damage as has been demonstrated in rats transgenic for the human renin and angiotensinogen genes.10, 11 Previous studies

Chapter 6

134

in stroke prone hypertensive rats (SHRSP) have demonstrated that aldosterone may have an important role in the development of malignant vascular injury.19 Adrenalectomy or administration of a selective aldosterone antagonist markedly reduces renal vascular damage and thrombotic microangiopathy in SHRSP without significantly altering blood pressure.20, 21 These findings may explain why some patients with primary aldosteronism due to an aldosterone secreting adrenal adenoma may present with malignant hypertension and the hallmarks of malignant vascular injury despite being initially protected from the vicious circle of renin release as a result of their underlying disease.22-24 However, some of these patients may present with an elevated PRA in the acute phase.25, 26 This may be explained by stimulation of renin release as a result of ischemic reno-vascular injury.26 The association between renin hypersecretion, ischemia and vascular damage in malignant hypertension is supported by the combination of a hyperplastic appearance of the juxtaglomerular apparatus and ischemic glomerular changes with myointimal hyperplasia and fibrinoid necrosis in renal biopsies of these patients.27 In an angiographic study of patients with essential and malignant hypertension the association between vascular damage and renin secretion has also been demonstrated.28 Patients with severe angiographic abnormalities of the arcuate and distal interlobular renal arteries had substantially higher plasma renin secretion and PRA than patients with less severe or normal angiographic findings. More severe small artery changes (and increased levels of PRA) coincided with a proportionate decrease in renal blood flow and loss of renal function. We previously showed that the degree of thrombocytopenia and LDH as markers for thrombotic microangiopathy correlate with the severity of renal insufficiency at presentation and on follow-up in patients with malignant hypertension.13 In the present prospective study both LDH and platelet count strongly correlated with serum creatinine and PRA at presentation. LDH levels correlated better with serum creatinine values at presentation and future renal replacement therapy than PRA. Adjusting for LDH as the most sensitive marker of thrombotic microangiopathy abolished the association between PRA and renal function, supporting the concept that LDH reflects the degree of renin release resulting from renovascular damage. The loglinear correlation between PRA and LDH may point towards the proposed vicious circle by which stimulation of renin leads to more severe vascular damage, increased renin secretion and so on.


135

As expected, PRA significantly correlated with aldosterone both in patients with severe and malignant hypertension. In patients with malignant hypertension, the association between PRA and aldosterone was best described when PRA was log transformed. A likely explanation for this phenomenon is that decreasing levels of potassium as observed in patients with higher aldosterone values inhibits further aldosterone secretion. In other words, the inhibition of aldosterone caused by decreasing levels of potassium may further limit the rise of serum aldosterone despite increases in PRA. There are some limitations that need comment. First, LDH is not an exclusive marker for thrombotic microangiopathy, but can also be elevated in several other conditions. However, the close negative correlation between platelet count and LDH suggests that the increases in LDH are, indeed, evidence of intravascular hemolysis. Liver disease and myocardial infarction, as a possible cause of an elevated LDH, were excluded by standard laboratory tests and electro-cardiogram at the emergency department. Second, in at least 4 patients PRA could have been influenced by concomitant pathology besides malignant hypertension (primary renal disease, renal artery stenosis, corticoid excess). We believe however that the presence of secondary causes may dilute rather than enhance the observed associations. Third, PRA and aldosterone values were not assessed on follow-up in most patients as cessation of (some) of the anti-hypertensive drugs, required for reliable measurement of PRA and aldosterone, would not be ethical in these patients. Finally, although we did not find any added value of PRA on future renal dysfunction after correction for LDH, we cannot exclude that such an association may be present in a larger sample of patients with a hypertensive emergency. Conclusion PRA and aldosterone are markedly elevated in patients with malignant hypertension compared to severely hypertensive patients, despite relatively small differences in blood pressure. The variations in PRA observed in patients with malignant hypertension can, in part, be explained by the severity of thrombotic microangiopathy and vice versa. In the present study 58% of the variations of PRA could be explained by LDH as the most sensitive marker for intravascular hemolysis. The positive correlation between PRA and serum creatinine in patients with malignant hypertension disappeared after adjustment for LDH, suggesting that the degree of renin release and thrombotic micro-angiopathy are interdependent in their effect on renal function.

Chapter 6

136

References 1. Laragh JH, Angers M, Kelly WG, Lieberman S: Hypotensive agents and pressor

substances. The effect of epinephrine, norepinephrine, angiotensin II, and others on the secretory rate of aldosterone in man. JAMA 1960;174:234-240.

2. Buhler FR, Laragh JH, Vaughan ED, Jr., Brunner HR, Gavras H, Baer L: Antihypertensive action of propranolol. Specific antirenin responses in high and normal renin forms of essential, renal, renovascular and malignant hypertension. Am J Cardiol 1973;32:511-522.

3. Linton AL, Gavras H, Gleadle RI, Hutchison HE, Lawson DH, Lever AF, MacAdam RF, McNicol GP, Robertson JI: Microangiopathic haemolytic anaemia and the pathogenesis of malignant hypertension. Lancet 1969;1(7609):1277-1282.

4. Laragh JH, Ulick S, Januszewicz V, Deming QB, Kelly WG, Lieberman S: Aldosterone secretion and primary and malignant hypertension. J Clin Invest 1960;39:1091-1106.

5. Brown JJ, Davies DL, Lever AF, Robertson JI: Plasma renin concentration in human hypertension. II. Renin in relation to aetiology. Br Med J 1965;5472:1215-1219.

6. Creditor MC, Loschky UK: Plasma renin activity in hypertension. Am J Med 1967; 43:371-382.

7. Kincaid-Smith P: Malignant hypertension: mechanisms and management. Pharma-col Ther 1980;9:245-269.

8. Wilson C, Byrom FB: Renal changes in malignant hypertension. Lancet 1939;I:136-139.

9. Kincaid-Smith P: Malignant hypertension. J Hypertens 1991;9:893-899.

10. Ganten D, Wagner J, Zeh K, Bader M, Michel JB, Paul M, Zimmermann F, Ruf P, Hilgenfeldt U, Ganten U: Species specificity of renin kinetics in transgenic rats harboring the human renin and angiotensinogen genes. Proc Natl Acad Sci U S A 1992;89:7806-7810.

11. Mervaala E, Muller DN, Schmidt F, Park JK, Gross V, Bader M, Breu V, Ganten D, Haller H, Luft FC: Blood pressure-independent effects in rats with human renin and angiotensinogen genes. Hypertension 2000;35:587-594.

12. Montgomery HE, Kiernan LA, Whitworth CE, Fleming S, Unger T, Gohlke P, Mullins JJ, McEwan JR: Inhibition of tissue angiotensin converting enzyme activity


137

prevents malignant hypertension in TGR(mREN2)27. J Hypertens 1998;16:635-643.

13. van den Born BJ, Honnebier UP, Koopmans RP, van Montfrans GA: Microangio-pathic hemolysis and renal failure in malignant hypertension. Hypertension 2005;45:246-251.

14. Hollemans HJ, van der MJ, Kloosterziel W: Measurement of renin activity in plasma by a radioimmunoassay for angiotensin I. Clin Chim Acta 1969;24:353-357.

15. Kincaid-Smith P, McMichael J, Murphy EA: The clinical course and pathology of hypertension with papilledema (malignant hypertension). Q J Med 1958;27:117-153.

16. Bali T, Goldblatt H: On the pathogenesis of the vascular lesions of malignant hypertension in the rat; the role of elevated blood pressure and renal functional failure. Exp Med Surg 1954;12:460-479.

17. Case DB, Atlas SA, Sullivan PA, Laragh JH: Acute and chronic treatment of severe and malignant hypertension with the oral angiotensin-converting enzyme inhibitor captopril. Circulation 1981;64:765-771.

18. Gavras H, Brown WC, Brown JJ, Lever AF, Linton AL, MacAdam RF, McNicol GP, Robertson JI, Wardrop C: Microangiopathic hemolytic anemia and the development of the malignant phase of hypertension. Circ Res 1971;28:Suppl.II: 127-141.

19. Rocha R, Chander PN, Zuckerman A, Stier CT, Jr.: Role of aldosterone in renal vascular injury in stroke-prone hypertensive rats. Hypertension 1999;33:232-237.

20. Rocha R, Chander PN, Khanna K, Zuckerman A, Stier CT, Jr.: Mineralocorticoid blockade reduces vascular injury in stroke-prone hypertensive rats. Hypertension 1998 ;31:451-458.

21. Chander PN, Rocha R, Ranaudo J, Singh G, Zuckerman A, Stier CT, Jr.: Aldosterone plays a pivotal role in the pathogenesis of thrombotic microangiopathy in SHRSP. J Am Soc Nephrol 2003;14:1990-1997.

22. Kaplan NM: Primary aldosteronism with malignant hypertension. N Engl J Med 1963;269:1282-1286.

23. Aloia JF, Beutow G: Malignant hypertension with aldosteronoma producing adenoma. Am J Med Sci 1974;268:241-245.

24. Zarifis J, Lip GY, Leatherdale B, Beevers G: Malignant hypertension in association with primary aldosteronism. Blood Press 1996;5:250-254.

Chapter 6

138

25. Ideishi M, Kishikawa K, Kinoshita A, Sasaguri M, Ikeda M, Takebayashi S, Arakawa K: High-renin malignant hypertension secondary to an aldosterone-producing adenoma. Nephron 1990;54:259-263.

26. Oka K, Hayashi K, Nakazato T, Suzawa T, Fujiwara K, Saruta T: Malignant hypertension in a patient with primary aldosteronism with elevated active renin concentration. Intern Med 1997;36:700-704.

27. McLaren KM, MacDonald MK: Histological and ultrastructural studies of the human juxtaglomerular apparatus in benign and malignant hypertension. J Pathol 1983;139:41-55.

28. Hollenberg NK, Epstein M, Basch RI, Couch NP, Hickler RB, Merrill JP: Renin secretion in essential and accelerated hypertension. Am J Med 1969;47:855-859.

Chapter 7

The severity of thrombotic microangiopathy in malignant hypertension is associated with reduced activity of the von Willebrand factor cleaving protease (ADAMTS13) Bert-Jan H van den Born, Niels V van der Hoeven, Evelyn Groot, Peter J Lenting, Joost CM Meijers, Marcel Levi and Gert A van Montfrans

Submitted

Chapter 7

142

Abstract Background The thrombotic microangiopathy (TMA) observed in malignant hypertension is similar to that of thrombotic thrombocytopenic purpura (TTP). TTP is associated with a deficiency of the von Willebrand factor (VWF) cleaving protease ADAMTS13, that cleaves large prothrombogenic multimers. We hypothesized that ADAMTS13 is deficient in malignant hypertension and that the severity of TMA is associated with decreased ADAMTS13 activity. We assessed active VWF, VWF and free hemoglobin (Hb) to explore predictors of ADAMTS13 activity. Methods We included 20 patients with malignant and 20 patients with severe hypertension, 20 matched normotensive individuals served as controls. Results Patients with malignant hypertension had lower ADAMTS13 activity (80% [interquartile range, IQR 53%-130%]) compared to controls (99% [IQR 82%-129%], P < 0.01), but not compared to patients with severe hypertension (P = 0.14). ADAMTS13 activity significantly and negatively correlated with logLDH (r = -0.65, P < 0.001) explaining 42% of its variations. ADAMTS13 activity was also associated with platelet count (r = 0.34, P = 0.04) and the presence of schistocytes (r = -0.37, P = 0.02). Apart from the association with TMA parameters, ADAMTS13 was negatively associated with creatinine (r = -0.42, p = 0.008). Increasing levels of VWF were associated with a decrease in ADAMTS13 activity (r = -0.34, P = 0.03). There was no significant association between ADAMTS13 activity and other parameters, including blood pressure. Conclusion ADAMTS13 is deficient in malignant hypertension and associated with TMA severity, likely because of the release of VWF after endothelium stimulation. A severe deficiency such as in TTP was not demonstrated. More studies are needed to identify the role of ADAMTS13 in the TMA and ischemic complications that corroborate malignant hypertension.

ADAMTS13 in malignant hypertension

143

Introduction Malignant hypertension is a condition characterized by severe hypertension and acute ischemic complications and frequently complicated by a thrombotic microangiopathy (TMA). TMA is characterized by thrombosis of small vessels, intravascular hemolysis with fragmentation of red blood cells (schistocytes), elevated lactic dehydrogenase (LDH) levels, and consumption of platelets. TMA was observed in 27% of patients with malignant hypertension presenting at the emergency room of our hospital and was associated with renal impairment at presentation, but an increased recovery of renal function during follow-up.1 The pathogenesis of TMA in malignant hypertension is incompletely understood. Several mechanisms may lead to the TMA associated with malignant hypertension. Malignant hypertension bears resemblance with thrombotic thrombocytopenic purpura (TTP). TTP is associated with a congenital or acquired deficiency of ADAMTS13, a zinc-containing metalloprotease that cleaves large von Willebrand factor (VWF) multimers, thereby decreasing their prothrombogenic properties.2, 3 Deficiency of ADAMTS13 (activity) leads to the appearance of unusually large prothrombogenic multimers (UL-VWF ) in the circulation resulting in thrombocytopenia, intravascular hemolysis and ischemic microvascular complications because of thrombus formation.4 First, a secondary deficiency of ADAMTS13 activity has been demonstrated in situations where high levels of circulating VWF are present as a result of endothelial stimulation or damage.5, 6 This may suggest that in a state of marked endothelial activation or damage, such as malignant hypertension, a deficiency of ADAMTS13 activity may develop as a result of binding of ADAMTS13 to the A2 domain of the VWF molecule. Second, apart from the association with VWF, high fluid shear rates may result in conformational changes of the VWF molecule and promote binding of ADAMTS13, thereby reducing its activity.7 Recently, an antibody fragment has been developed which allows quantification of this conformational change. This conformational change of the VWF molecule is known as “active VWF”.8 Finally, the degree of extra-cellular free hemoglobin (Hb), which may accompany the destruction of red blood cells, has been shown to inhibit ADAMTS13 in vitro and may also be responsible for a secondary deficiency of ADAMTS13 in vivo.9

We hypothesized that ADAMTS13 is deficient in malignant hypertension and associated with TMA severity. In addition, we explored possible mecha-nisms influencing ADAMTS13 activity by measuring VWF, active VWF, free Hb and multimer size.

Chapter 7

144

Patients and methods Study population We used a prospective case-control design to compare patients with malignant hypertension, defined according to the WHO criteria as severe hypertension (BP > 120 mmHg diastolic) with bilateral linear or flame shaped hemorrhages or ‘cotton-wool’ exudates with or without papilledema on funduscopic examination, to patients having severe hypertension (BP > 120 mmHg diastolic) without these retinal lesions. Normotensive controls were included to differentiate between the ischemic complications corroborating malignant hypertension and the influence of blood pressure. To assess TMA severity, we assessed LDH levels, platelet count and schistocytes at presentation. The primary outcome measure was the difference in ADAMTS13 activity between blood pressure groups. The second outcome measure was the association between ADAMTS13 activity and TMA severity. To explore the mechanisms leading to a deficiency of ADAMTS 13, we assessed VWF, active VWF, free Hb and VWF multimer size and assessed the influence of these parameters on ADAMTS13 activity. We included 20 patients with malignant hypertension and 20 patients with severe hypertension, presenting at a large teaching hospital serving a multi-ethnic community in Amsterdam, the Netherlands. Twenty age, sex and ethnicity matched normotensive persons served as controls. Excluded were patients < 18 years of age, pregnant women and patients on dialysis before admission. Blood pressure was measured using an aneroid sphygmomanometer in the recumbent position. Retinal examinations were carried out in mydriasis by an ophthalmologist. Schistocytes were considered present if “several” red cell fragments could be detected on a peripheral blood film. Macroalbuminuria was defined as urinary protein excretion > 300 mg/L or 2+ for dipstick proteinuria. Left ventricular hypertrophy was defined according to the Sokolow–Lyon criteria. Additional screening for secondary hypertension consisted of imaging studies of the renal arteries in 25 patients (66%), one or more endocrine tests in 16 (42%) and renal biopsy in 7 (18%) patients. All blood samples were taken just before initiating anti-hypertensive treatment and immediately analyzed or frozen as plasma at -80 degrees Celsius. The study was approved of by the local medical ethics committee. Informed consent was obtained from all participants.


145

Laboratory procedures ADAMTS13 activity was determined using the rapid fluorescence resonance energy transfer (FRET) assay.10 The normal range was 54-150% in 60 healthy individuals (SD 26.5%), the intra-assay variation was 1-6% (3-6% in the lower range), and the inter-assay variation was 2-5% (3-6% in the lower range). VWF antigen was determined by ELISA (Dako, Denmark). Active VWF was determined by immunosorbent assay using an antibody fragment (AU/VWFa-11) that specifically recognizes the GpIb-alpha binding configuration as previously described.8 This antibody fragment allows for the detection of active VWF in plasma of patients characterized by spontaneous VWF–platelet interactions and discriminates active from resting VWF. The ratio of the slope for different plasma samples over the slope for normal pooled plasma was calculated and represents the relative amount of active VWF. In previous experiments the intra-assay variation was 7.1%, the inter-assay variation was 13.7%. Free hemoglobin was assessed in plasma by absorption spectrophoto-metry using a cyanide free HgB reagent (Cell Dyn 4000, Abbott Diagnostics, Abbott Park, IL). Reference values are 5.0 – 28.0 µmol/L. Compared to a standardized serial dilution, linearity was R2=0.996 in the normal range (0-1.0 mmol/L) and R2 = 0.984 in the lower range (0-30.0 µmol/L). The coefficient of variation was 10.9% at a range of 17.0 µmol/L (SD 1.8 µmol/L). False-positive results can be obtained with lipemic or hemolytic plasma and were considered if free haemoglobin concentration ≥ 40 µmol/L. The multimeric pattern of VWF was analyzed using 2.5% agarose gel electrophoresis, followed by immuno-blotting according to previous described methods.11 Statistical analysis A sample size of 20 patients in each group was calculated to allow detection of a difference in ADAMTS13 activity of at least 1 SD (26.5%) between blood pressure groups with a 80% power and α-level of 0.05. Baseline variables were described using mean and standard deviation (SD) or median and range, or interquartile range (IQR) for variables with a skewed distribution. Baseline differences were calculated to compare patients with severe and malignant hypertension using an independent samples t-test for parametric and Mann-Whitney U test for non-parametric distributions where appropriate. Categorical variables were calculated using χ2 with Yates’ correction. We assessed differences in ADAMTS13 activity between blood pressure groups using 1-way ANOVA followed by Tukey’s test for multiple comparisons to assess differences between blood pressure groups. Linear

Chapter 7

146

regression analysis was used to assess the association between ADAMTS13 activity and LDH, platelet count and schistocytes expressing TMA severity and renal insufficiency. To assess predictors of ADAMTS13 the univariate correlations with ADAMTS13 were tested. The bivariate interaction between predictors of ADAMTS13 was also determined. Finally, cross-correlations were assessed to explore the associations between ADAMTS13 activity, VWF, active VWF and free Hb. P < 0.05 was considered to indicate a statistically significant difference. For statistical analysis the SPSS software package for Windows was used, version 14.0 (Chicago, Illinois, USA). Results The clinical characteristics of patients with malignant hypertension, severe hypertension and normotensive controls are shown in table 1. Two patients with severe hypertension had TMA as evidenced by a low platelet count, elevated LDH and fragmented red blood cells in a peripheral blood smear. A secondary cause for the hypertension was established in 5 patients (25%) having malignant hypertension and in 3 patients (15%) with severe hypertension. Secondary causes of hypertension included IgA nephropathy (1), renal artery stenosis due to polyarteritis nodosa (1 patient), bilateral hydronephrosis (1), a cortisol producing adrenal carcinoma (1) and corticoid therapy (1) in the malignant hypertensive group and IgA nephropathy (2) and SLE nephritis (1) in the group with severe hypertension. The individual data for ADAMTS13 activity by blood pressure group are de-picted in figure 1. Patients with malignant hypertension had lower ADAMTS13 activity (80% [interquartile range, IQR 53%-130%]) compared to controls (99% [IQR 82%-129%], P < 0.01), but not compared to patients with severe hypertension (P = 0.14). The association between ADAMTS13 activity and logLDH is depicted in figure 2. ADAMTS13 activity significantly and nega-tively correlated with logLDH (r = -0.65, P < 0.001), explaining 42% of its variations. ADAMTS13 activity was also associated with platelet count (r = 0.34, P = 0.04) and the presence of schistocytes (r = -0.37, P = 0.02). Apart from the association with these TMA parameters, ADAMTS13 was negatively associated with creatinine (r = -0.42, P = 0.008).


147

Table 1 Patient characteristics by blood pressure category

Values are numbers and percentages unless stated otherwise. † Data missing for 2 in the group with severe hypertension. ‡ Data missing for 1 patient in the group with malignant hypertension. ¶ Data missing for 2 patients in the severe hypertension group.

Normotensive controls (n = 20)

Severe hypertension (n = 20)

Malignant hypertension (n = 20)

P (MHT vs SHT)

Age, yrs 43 ± 9 46 ± 15 42 ± 11 0.34 Male 11 (55%) 11 (55%) 15 (75%) 0.32 Black 10 (50%) 12 (60%) 10 (50%) 0.75 Systolic BP, mmHg (mean ± SD)

125 ± 11 226 ± 20 229 ± 23 0.74

Diastolic BP, mmHg (mean ± SD)

80 ± 9 133 ± 14 150 ± 14 < 0.01

Current Smoker 2 (10%) 5 (25%) 5 (25%) 1.00 Diabetes 0 2 (10%) 1 (5%) 1.00 Previous hypertension 14 (70%) 7 (35%) 0.06 On active treatment 4 (20%) 0 0.11 Serum creatinine, µmol/l (median [range])

92 [72-142] 149 [74-1545] < 0.01

Macroalbuminuria† 5 (28%) 17 (85%) < 0.01 Left ventricular hypertrophy

8 (40%) 17 (85%) < 0.01

Hemoglobin, mmol/l (median [range])

9.3 [7.3-10.9] 8.6 [4.0-10.3] 0.17

Platelets, x109/l (median [range])‡

234 [102-363] 246 [45-345] 0.76

Lactic dehydrogenase, U/l (median [range])¶

226 [126-471] 325 [161-1467] 0.01

Schistocytes 2 (10%) 6 (30%) 0.24

Chapter 7

148

The only variable that had a significant association with ADAMTS13 activity was VWF, increasing VWF levels were associated with a decrease in ADAMTS13 activity (r = -0.34, P = 0.03). Associations between ADAMTS13 and active VWF (r = -0.22, P = 0.17), free Hb (r = 0.12, P = 0.44), diastolic blood pressure (r = -0.26, P = 0.11) and systolic blood pressure (r = -0.19, P = 0.24) were not significant. The cross-correlations between ADAMTS13, VWF, active VWF and free Hb are depicted in table 2. The multimeric pattern of VWF was analyzed in 8 patients with the lowest ADAMTS13 activity and in a patient with the highest active VWF. However, no difference in multimer size could be demonstrated in these patients. Seven free Hb samples showed abnormal high values (≥ 40 µmol/L), 3 of which were macroscopically lipemic. The results were not different after excluding these samples. Fig. 1

50

100

150

Control Severe Malignant

AD

AM

TS13

act

ivity

(%)

Plasma levels of ADAMTS13 activity in the three blood pressure groups. The solid line represents median values


149

Fig. 2

Correlation between ADAMTS13 activity and log (LDH). The solid line represents the regression line with the 95% confidence interval (R2 = 0.42, P < 0.001). Table 2 Cross-correlations between ADAMTS13 activity, VWF, active VWF and free hemoglobin in patients with malignant and severe hypertension.

Values represent regression coefficients with P-values between brackets. VWF = von Willebrand factor.

ADAMTS13 VWF active VWF free hemoglobin

ADAMTS13 1.00 -0.34 (P = 0.03)

-0.22 (P = 0.17)

0.12 (P = 0.44)

VWF 1.00 0.63 (P < 0.001)

-0.02 (P = 0.91)

active VWF 1.00 -0.10 (P = 0.51)

free hemoglobin 1.00

Chapter 7

150

Discussion We showed that ADAMTS13 is deficient in patients with malignant hyper-tension compared to normotensive controls and associated with TMA severity: a more pronounced TMA was associated with lower ADAMTS13 activity. Differences in ADAMTS13 activity between patients with malignant and severe hypertension were not significantly different, likely because some of the patients with severe hypertension also had TMA. A severe deficiency of ADAMTS13, such as that observed in TTP, could not be demonstrated. Therefore, measurement of ADAMTS13 activity may help to discriminate between TTP and malignant hypertension. To explore the mechanisms leading to a decrease in ADAMTS13 activity, we assessed several parameters which may decrease ADAMTS13 or inhibit its activity. Of these parameters only VWF showed a significant association with ADAMTS13. Increased levels of VWF were associated with a decreased activity of ADAMTS13. This supports the idea that an acute increase of VWF, as a result of endothelial stimulation or damage, may provoke an acute decrease in ADAMTS13 activity. Previous studies have demonstrated that either admini-stration of DDAVP or endotoxin to healthy volunteers leads to an increase in VWF levels and concomitant decreases of ADAMTS13 activity.5, 12 Although a severe ADAMTS13 deficiency was not demonstrated in our study, recent experiments suggest that the balance between ADAMTS13 and VWF is more important than the separate concentrations of these parameters. Mice that lack ADAMTS13 also require stimulation with epinephrine or collagen and appropriate flow conditions to develop TMA.13 In humans, patients with a clinical diagnosis of TTP-HUS syndrome often have additional disorders at the time of diagnosis and infections may provoke TTP relapses.14, 15 Vice versa, in patients with severe sepsis a deficiency of ADAMTS13 is associated with the presence of UL-VWF multimers.6 Therefore the interaction between VWF and ADAMTS13 seems a two-way association: expression of large amounts of VWF reduces ADAMTS13 activity, vice versa, a lower ADAMTS13 activity may lead to the expression of large prothrombogenic VWF multimers which stimulate platelet binding, thrombus formation and the development of tissue ischemia and necrosis. We could not demonstrate the presence of UL-VWF multimers in patients with the lowest ADAMTS13 activity in our study. However, a relative predominance of larger VWF multimers can not be excluded as our technique was inappropriate to detect more subtle increases in VWF multimer size.


151

The association between ADAMTS13 activity and renal insufficiency, as observed in our study and in a previous study of patients with severe sepsis,6 may support the hypothesis that even small deficiencies of ADAMTS13 may be relevant in conditions of marked endothelial activation. In addition to the association between ADAMTS13 and VWF, we also exa-mined the possibility that high fluid shear rates may be responsible for the relative expression of a more active VWF. High fluid shear rates promote binding of platelets at the A1 domain of the von Willebrand subunit, changing the VWF molecule to the “active” GpIb-alpha binding configuration.7, 16 This conformational change stimulates cleavage by ADAMTS13 as a result of exposure of the adjacent A2 domain resulting in a decrease in multimer size.17, 18 In our study, a more active VWF was positively associated with VWF, showing that with increasing VWF the amount of circulating VWF was also more active. However, we found no association between active VWF and ADAMTS13 activity. In addition, a decrease in multimer size could not be demonstrated in 8 patients with the lowest ADAMTS13 activity or highest active VWF level. These findings suggest that conformational changes in the VWF factor molecule are not a principal determinant of ADAMTS13 activity in malignant hyper-tension. Finally, since a previous study suggested that free hemoglobin, which is released during hemolysis, may in itself affect ADAMTS13 activity,19 we also examined the association between free Hb and ADAMTS13 activity. However, no association between free Hb and ADAMTS13 activity could be demonstrated suggesting that the decrease in ADAMTS13 activity cannot be explained by an inhibitory effect of free Hb. There are some limitations to our study. First, the association between ADAMTS13 and TMA severity, and between ADAMTS13 and renal dysfunction does not prove causality. It is conceivable that the high arterial pressures associated with severe and malignant hypertension directly result in endothelial damage and TMA. The degree of VWF release as a result of the high arterial pressures and subsequent reductions in ADAMTS13 activity may therefore be just a marker for the degree of endothelial damage rather than being causally related to the TMA and renal insufficiency of malignant hypertension. Although their was a lack of association between ADAMTS13 and either systolic or diastolic blood pressure, the acceleration rather than the absolute blood pressure level may be a more important determinant of endothelial damage. A previous study showed that both endothelial microparticles and

Chapter 7

152

VWF, as evidence of endothelial damage, are increased in patients with severe hypertension.20 Although the association between VWF and endothelial microparticles did not reach significance in that study, it is possible that some of the VWF released in the circulation is bound to these microparticles and contributes to a lower ADAMTS13 concentration. Therefore ADAMTS13 may either be a cause or a consequence, or both, of the TMA associated with malignant hypertension.

Second, underlying diseases and ABO blood group may have influenced the measurements performed in this study. Inflammatory conditions such as polyarteritis nodosa and lupus erythematosus are associated with endothelial activation and may stimulate VWF release and lower ADAMTS13 activity. In contrast, blood group O is generally associated with lower VWF levels and somewhat higher ADAMTS13 activity.21 The frequency of the inflammatory conditions underlying malignant and severe hypertension were equally distributed between groups and accounted for only a small proportion of the total study population. When these secondary causes were removed no significant changes occurred in the association between ADAMTS13 activity and blood pressure groups or in the association between ADAMTS13 and VWF. Blood groups were not assessed in this study and may therefore have accounted for some of the differences in ADAMTS13 activity between blood pressure groups. However, we estimate that the effect of blood group on ADAMTS13 activity is small and consider it unlikely that blood group would have altered the association between VWF and ADAMTS13 activity. Perspectives We have shown that in patients with malignant hypertension ADAMTS13 is deficient and associated with the severity of TMA, likely because of the release of large amounts of VWF as a result of endothelium stimulation. A severe deficiency of ADAMTS13 such as that observed in TTP could not be demonstrated, suggesting that measurement of ADAMTS13 may be a useful tool to discriminate between these conditions. The close correlation between ADAMTS13 activity and renal insufficiency in patients with malignant hypertension may suggest that treatment directed at increasing ADAMTS13 activity may beneficially affect renal function. To further delineate the mechanisms leading to the TMA of malignant hypertension and to examine whether ADAMTS13 activity is involved in the ischemic complications associated with malignant hypertension, further studies are needed that examine


153

whether administration of ADAMTS13 or agents that stimulate its function leads to amelioration of the ischemic renal lesions observed in this condition. References 1. van den Born BJ, Honnebier UP, Koopmans RP, van Montfrans GA. Micro-

angiopathic hemolysis and renal failure in malignant hypertension. Hypertension. 2005;45:246-251.

2. Furlan M, Robles R, Galbusera M, Remuzzi G, Kyrle PA, Brenner B, Krause M, Scharrer I, Aumann V, Mittler U, Solenthaler M, Lammle B. von Willebrand factor-cleaving protease in thrombotic thrombocytopenic purpura and the hemolytic-uremic syndrome. N Engl J Med. 1998;339:1578-1584.

3. Tsai HM, Lian EC. Antibodies to von Willebrand factor-cleaving protease in acute thrombotic thrombocytopenic purpura. N Engl J Med. 1998;339:1585-1594.

4. Moake JL, Turner NA, Stathopoulos NA, Nolasco LH, Hellums JD. Involvement of large plasma von Willebrand factor (vWF) multimers and unusually large vWF forms derived from endothelial cells in shear stress-induced platelet aggregation. J Clin Invest. 1986;78:1456-1461.

5. Reiter RA, Knobl P, Varadi K, Turecek PL. Changes in von Willebrand factor-cleaving protease (ADAMTS13) activity after infusion of desmopressin. Blood. 2003;101:946-948.

6. Ono T, Mimuro J, Madoiwa S, Soejima K, Kashiwakura Y, Ishiwata A, Takano K, Ohmori T, Sakata Y. Severe secondary deficiency of von Willebrand factor-cleaving protease (ADAMTS13) in patients with sepsis-induced disseminated intravascular coagulation: its correlation with development of renal failure. Blood. 2006;107:528-534.

7. Dong JF, Moake JL, Nolasco L, Bernardo A, Arceneaux W, Shrimpton CN, Schade AJ, McIntire LV, Fujikawa K, Lopez JA. ADAMTS-13 rapidly cleaves newly secreted ultralarge von Willebrand factor multimers on the endothelial surface under flowing conditions. Blood. 2002;100:4033-4039.

8. Hulstein JJ, de Groot PG, Silence K, Veyradier A, Fijnheer R, Lenting PJ. A novel nanobody that detects the gain-of-function phenotype of von Willebrand factor in ADAMTS13 deficiency and von Willebrand disease type 2B. Blood. 2005;106:3035-3042.

9. Studt JD, Hovinga JA, Antoine G, Hermann M, Rieger M, Scheiflinger F, Lammle B. Fatal congenital thrombotic thrombocytopenic purpura with apparent

Chapter 7

154

ADAMTS13 inhibitor: in vitro inhibition of ADAMTS13 activity by hemoglobin. Blood. 2005;105:542-544.

10. Kokame K, Nobe Y, Kokubo Y, Okayama A, Miyata T. FRETS-VWF73, a first fluorogenic substrate for ADAMTS13 assay. Br J Haematol. 2005;129:93-100.

11. Raines G, Aumann H, Sykes S, Street A. Multimeric analysis of von Willebrand factor by molecular sieving electrophoresis in sodium dodecyl sulphate agarose gel. Thromb Res. 1990;60:201-212.

12. Reiter RA, Varadi K, Turecek PL, Jilma B, Knobl P. Changes in ADAMTS13 (von-Willebrand-factor-cleaving protease) activity after induced release of von Willebrand factor during acute systemic inflammation. Thromb Haemost. 2005;93:554-558.

13. Banno F, Kokame K, Okuda T, Honda S, Miyata S, Kato H, Tomiyama Y, Miyata T. Complete deficiency in ADAMTS13 is prothrombotic, but it alone is not sufficient to cause thrombotic thrombocytopenic purpura. Blood. 2006;107:3161-3166.

14. Vesely SK, George JN, Lammle B, Studt JD, Alberio L, El Harake MA, Raskob GE. ADAMTS13 activity in thrombotic thrombocytopenic purpura-hemolytic uremic syndrome: relation to presenting features and clinical outcomes in a prospective cohort of 142 patients. Blood. 2003;102:60-68.

15. Cserti CM, Landaw S, Uhl L. Do infections provoke exacerbations and relapses of thrombotic thrombocytopenic purpura? J Clin Apher. 2007;22:21-25.

16. Ruggeri ZM. von Willebrand factor. J Clin Invest. 1997;99:559-564.

17. Nishio K, Anderson PJ, Zheng XL, Sadler JE. Binding of platelet glycoprotein Ibalpha to von Willebrand factor domain A1 stimulates the cleavage of the adjacent domain A2 by ADAMTS13. Proc Natl Acad Sci U S A. 2004;101:10578-10583.

18. Donadelli R, Orje JN, Capoferri C, Remuzzi G, Ruggeri ZM. Size regulation of von Willebrand factor-mediated platelet thrombi by ADAMTS13 in flowing blood. Blood. 2006;107:1943-1950.

19. Studt JD, Hovinga JA, Antoine G, Hermann M, Rieger M, Scheiflinger F, Lammle B. Fatal congenital thrombotic thrombocytopenic purpura with apparent ADAMTS13 inhibitor: in vitro inhibition of ADAMTS13 activity by hemoglobin. Blood. 2005;105:542-544.

20. Preston RA, Jy W, Jimenez JJ, Mauro LM, Horstman LL, Valle M, Aime G, Ahn YS. Effects of severe hypertension on endothelial and platelet microparticles. Hypertension. 2003;41:211-217.


155

21. Mannucci PM, Capoferri C, Canciani MT. Plasma levels of von Willebrand factor regulate ADAMTS-13, its major cleaving protease. Br J Haematol. 2004;126:213-218.

Chapter 8

Impaired cerebral autoregulation in patients with malignant hypertension

Rogier V. Immink, Bert-Jan H. van den Born, Gert A. van Montfrans, Richard P. Koopmans, John M. Karemaker, Johannes J. van Lieshout

Circulation 2004; 110: 2241-2245.

Chapter 8

160

Abstract

Background In patients with a malignant hypertension, immediate parenteral treatment with blood pressure–lowering agents such as intravenous sodium nitroprusside (SNP) is indicated. In this study, we evaluated static and dynamic cerebral autoregulation (CA) during acute blood pressure lowering with SNP in these patients.

Methods In 8 patients with mean arterial pressure (MAP) > 140 mmHg and grade III or IV hypertensive retinopathy at hospital admission, middle cerebral artery blood velocity (MCA V) and blood pressure were monitored. Dynamic

CA was expressed as the 0.1-Hz MCA Vmean to MAP phase lead and static CA as the MCA Vmean to MAP relationship during SNP treatment. Eight normotensive subjects served as a reference group.

Results In the patients, the MCA Vmean to MAP phase lead was lower (30 ± 8°

versus 58 ± 5°, mean ± SEM; P < 0.05), whereas the transfer gain tended to be higher. During SNP treatment, target MAP was reached within 90 minutes in all patients. The MCA Vmean decrease was 22 ± 4%, along with a 27 ± 3% reduction in MAP (from 166 ± 4 to 121 ± 6 mmHg; P < 0.05) in a linear fashion (averaged slope, 0.82 ± 0.15% cm · s–1 · % mmHg–1; r = 0.70 ± 0.07).

Conclusion In patients with malignant hypertension, dynamic CA is impaired. An MCA Vmean plateau was not detected during the whole SNP treatment, indicating loss of static CA as well. This study showed that during the whole rapid reduction in blood pressure with SNP, MCA Vmean decreases almost one on

one with MAP.

Cerebral autoregulation in malignant hypertension

161

Introduction

In patients with malignant hypertension, immediate treatment with blood pressure (BP) reducing agents such as intravenous sodium nitroprusside (SNP) or labetalol is indicated to limit cerebral, myocardial, and renal damage.1 In these patients with a mean arterial pressure (MAP) > 140 mmHg, the presence of grade III to IV retinopathy and occasionally hypertensive encephalopathy, reflecting cerebral hyperperfusion and edema, is considered to indicate compromised cerebral autoregulatory capacity.2,3 From this assumption, it has become generally accepted that the initial reduction in BP should not exceed ~25% of the presenting level within the first hours of treatment to prevent cerebral

hypoperfusion.1,3–6

Cerebral autoregulation (CA) is defined as the intrinsic capacity of cerebral vasculature to maintain constant cerebral blood flow (CBF).7–9 In normotensive subjects, when MAP decreases below ~60 mmHg, considered the lower limit of CA, CBF decreases proportionally with BP. Most patients suffering from malignant hypertension have a history of chronic hypertension,10 and in those patients, the lower limit of CA has been shifted in proportion toward higher pressures.11,12 For obvious reasons, the upper limit of CA has not been determined in normotensive or hypertensive humans. It was located between 120 and 150 mmHg 13 in normotensive baboons and between 155 and 170 mmHg in chronic hypertensive baboons.14

A moderate reduction in BP with SNP in normotensive subjects does not influence CBF,15 but to the best of our knowledge, no human studies specifically tested the assumption that CA is impaired in patients with malignant hypertension. Therefore, we determined CA in this group of patients before and during a decline in BP elicited by SNP.

Patients and methods

Subjects Eight consecutive patients fulfilling the World Health Organization criteria for malignant hypertension, namely severely elevated blood pressure plus grade III (bilateral retinal hemorrhages or cotton wool exudates; n = 5) or IV (III plus papilledema; n = 3) hypertensive retinopathy according to the Keith, Wagener, and Barker classification, were included in the study (Table 1).

Chapter 8

162

Table 1 Characteristics of reference subjects and patients

Values are mean (range). *P < 0.01 vs reference subjects.

Before hospital admission, 3 of these 8 patients were known to have had moderate hypertension treated with 1 or 2 antihypertensive drugs. Two of these 3 patients withdrew medication without consulting their general practitioners. One had untreated hypertension, 1 was normotensive, and the other 3 had undocumented BP.

At hospital admission, 3 of 8 patients presented with symptoms of hypertensive encephalopathy (eg, convulsions before hospital admission). Six patients had left ventricular hypertrophy, 6 had moderately elevated plasma creatinine levels (between 100 and 200 µmol · L–1), and 1 had considerably elevated (1050 µmol · L–1) plasma creatinine level. A cause of hypertension was identified in 3 patients: high-dose corticosteroid treatment, a cortisol-producing adrenal carci-noma, and deterioration of renal function with volume overload related to IgA nephropathy. In 5 patients, extended testing, including renal artery imaging, did not identify a definite cause of the hypertension.

To decrease BP, intravenous SNP infusion was started at 0.3 µg · kg–1 · min–1. After 5 minutes, it was increased to 0.5 µg · kg–1 · min–1 and, from then on, by 0.5 µg · kg–1 · min–1 every 5 minutes until MAP had stabilized at ~25% below the

Patient characteristics Reference subjects (n = 8) Patients (n = 8) Age (yrs) 46 (25–64) 44 (24–54) Sex (M/F) 5/3 6/2 Height, cm 177 (160–186) 175 (165–188) Weight, kg 76 (62–92) 76 (63–96) BP, mmHg Systolic 124 (101–146) 225 (180–260)* Mean 82 (73–94) 166 (147–187)* Diastolic 67 (55–85) 137 (130–150)* HR, bpm 62 (54–69) 89 (55–106)* MCA V, cm · s–1 Maximum 96 (82–112) 97 (65–126) Mean 63 (54–74) 64 (37–85) Minimum 42 (35–50) 43 (30–56)


163

presenting value. Eight normotensive healthy subjects matched for age, weight, and height (Table 1) were included to estimate normal CA parameters.

In the first 6 months of follow-up, the patient with corticosteroids-related

malignant hypertension was well controlled (BP, 110/80 mmHg) with 2 antihypertensive drugs and tapering of corticosteroid dosage. Four patients were well controlled (BP < 140/90 mmHg) with 3 or 4 antihypertensive drugs. In 2 others, BP was moderately elevated (BP, 150/90 mmHg and 160/100 mmHg) despite the use of 3 and 5 drugs, respectively.

All subjects, or a direct relative in case of encephalopathy, received verbal and written explanation of the objectives and techniques of measurements and risks and benefits associated with the study. They provided written informed consent in accordance with the Helsinki Declaration. This study was approved by the

Medical Ethics Committee of the Academic Medical Center, University of Amsterdam (the Netherlands; MEC 02/194).

Measurements Patients and reference subjects were instrumented with ECG electrodes while in the supine position. In the patients, intra-arterial pressure (IAP) was monitored through a catheter (1.1-mm ID, 20 gauge) placed in the radial artery of the nondominant arm. For comparison between patients and reference subjects, finger BP (FinAP) was measured by photoelectric plethysmography (Portapres;

Netherlands Organization for Applied Scientific Research, Biomedical

Instrumentation, TNO-BMI) in both groups. Changes in FinAP accurately track changes in IAP during normotension,16–18 and this study confirms this observation in extreme hypertension. The cuff was applied to the middle finger of the dominant hand placed at heart level. Stroke volume (SV) was determined by a 3-element model of arterial input impedance (Modelflow).19 Modelflow computes a flow wave from the FinAP wave that is integrated to obtain the SV of the heart.

When calibrated against a "gold standard" method to determine cardiac output (CO) such as thermodilution or the Fick method, this methodology provides accurate estimates of changes in SV in patients with septic shock20 and cardio-vascular disease19,21 and during orthostatic stress.22 The middle cerebral artery

blood velocity (MCA V) was measured in the proximal segment of the right MCA (Multidop X4). Once the optimal signal-to-noise ratio was obtained, the

Chapter 8

164

probe was secured with a headband (Mark 600, Spencer Technologies). SNP was administered with a perfusor (B. Braun Medical Inc). Data analysis The signals of IAP, FinAP, spectral envelope of MCA V, and a marker signal were A/D converted at 100 Hz and stored on a hard disk for offline analysis. In the patients, all signals were monitored from 20 minutes before SNP infusion to 1 hour after the last increment of SNP dose. Measurements were obtained in the reference subjects in the supine position for 30 minutes. Mean MCA V (MCA Vmean), mean IAP, and mean FinAP were integral over 1 heartbeat. Heart rate (HR) was the inverse of the interbeat pressure interval; cardiac output (CO) was the product of SV and HR; and systemic vascular resistance (SVR) was mean IAP divided by CO. Because Modelflow was not calibrated, SV, CO, and SVR were expressed as percentage of the presenting value. To assess the effect of SNP on cerebrovascular conductance, a cerebrovascular conductance index was calculated as MCA Vmean divided by mean IAP.18 Changes in systemic and cerebral hemodynamics elicited by SNP were expressed as averages of 3-minute episodes of mean IAP, HR, SV, CO, SVR, MCA Vmean, and cerebrovascular conductance index determined at 3 moments: before SNP treatment, midway through treatment, and when target BP was reached. For maintaining CBF, both fast- and slow-acting regulatory mechanisms are required to span the prevailing demands on CBF in everyday life.23

Static CA (sCA) reflects overall efficiency of the autoregulation system and is assessed by monitoring the CBF during different levels of BP. Because changes in CBF are tracked by changes in MCA Vmean,24 sCA is considered intact when MCA Vmean barely changes during a decrease in IAP. To assess sCA for individual subjects in this study, the continuous signals of MCA Vmean

and IAP were first averaged to 30-second episodes and then linearly related to each other.

Dynamic CA (dCA) refers to the ability to restore CBF in the face of a sudden change in perfusion pressure and reflects the latency of the system.25 It is quantified by the counterregulatory capacity to maintain constant MCA V during an induced or spontaneous abrupt changes in BP. In healthy subjects, MCA Vmean leads MAP with ~2.5 seconds in the time domain,25 and ~60° in the frequency domain around the 0.1-Hz spontaneous BP variability.26 In this study, dCA was determined in the frequency domain from 3-minute episodes of beat-to-beat data of mean FinAP and MCA Vmean before SNP treatment and after


165

target BP was reached. These data were compared with dCA in reference subjects determined in the supine resting state. To quantify the variability of pressure and velocity, power spectra were computed for mean FinAP and MCA Vmean with discrete Fourier transform after spline interpolation and resampling at 4 Hz of the beat-to-beat data sets. Results were expressed as the integrated area in the low-frequency range (0.07 to 0.15 Hz). To examine the strength between low-frequency mean FinAP and MCA Vmean, coherence was used to signify that the 2 cardiovascular signals covary significantly at a given frequency. The squared coherence function reflects the fraction of output power (MCA Vmean) that can be linearly related to the input power (mean FinAP) at each frequency. Like a correlation coefficient, it varies between 0 and 1. From the mean FinAP to MCA Vmean cross spectrum, the transfer function gain (cm · s–1 · mmHg–1) and the MCA Vmean to mean FinAP phase lead (degrees) were obtained in the low-frequency area.27 The transfer function gain was normalized for mean FinAP and MCA Vmean to account for the intersubject variability and expressed as percent change in cm · s–1 per percent change in mmHg.28 Phase was defined as positive when MCA Vmean leads mean FinAP. Statistical Analysis Data are expressed as mean and range in tables and as mean ± SEM in figures. Changes in systemic and cerebral hemodynamics during SNP treatment were examined by Friedman ANOVA on ranks. sCA was analyzed by nonparametric regression technique (Spearman rank order). Differences in dCA between reference subjects and hypertensive patients (unpaired) and before and after treatment (paired) were examined with the Wilcoxon rank-sum test and the Wilcoxon signed-rank test, respectively. A value of P < 0.05 was considered to indicate a statically significant difference. Characteristics of Patients and Reference Subjects Patients did not differ from normotensive reference subjects in age, height, and weight. BP and HR at admission were higher in the patient group (P < 0.01), although MCA V was comparable (Table 1). CA Before Treatment Low-frequency variability of mean FinAP and MCA Vmean were comparable for patients and reference subjects with a smaller phase lead (P < 0.05) and a tendency toward a higher transfer gain in the patients (Table 2).

Chapter 8

166

CA During Treatment Target mean IAP was reached within 90 minutes in all patients (figure 1). With increasing dose of SNP, SVR declined (– 43 ± 5%), whereas HR (16 ± 4 bpm), SV (9 ± 4%), and thus CO (31 ± 8%) increased (figure 2). Systolic IAP decreased from 225 ± 8 to 182 ± 6 mmHg; mean IAP, from 166 ± 4 to 122 ± 6 mmHg, and diastolic IAP, from 137 ± 6 to 102 ± 6 mmHg. The cerebrovascular conductance index did not increase significantly (9.5 ± 5.3%).

MCA Vmean decreased 22 ± 4%, along with a 27 ± 3% reduction in MAP in a linear fashion (averaged slope, 0.82 ± 0.15; averaged r = 0.70 ± 0.07; figures 3 and 4). With mean IAP stabilized, variability of mean FinAP and MCA Vmean and the MCA Vmean to mean FinAP phase lead and transfer gain were not signi-ficantly different from values before SNP treatment or from reference subjects (Table 2).

Table 2 dCA in reference subjects and patients

Patients (n = 8)

Reference subjects (n = 8)

Before SNP During SNP MAP power, mmHg2 · Hz–1 4.5 (1.5–10.4) 5.8 (1.4–15.8) 11.3 (2.6–22.8)

Vmean power, (cm · s–1)2 · Hz–1 3.6 (1.1–9.8) 1.8 (1.1–4.9) 4.7 (0.9–9.1)

Coherence 0.75 (0.54–0.84) 0.64 (0.49–0.76) 0.73 (0.48–0.91)

Phase,° 58 (41–82) 30 (6–67)* 42 (4–91)

Gain, % · %–1 1.02 (0.48–1.37) 1.32 (0.93–1.87) 1.29 (1.00–1.70)

MAP power indicates low-frequency mean MAP variability; Vmean power, MCA Vmean variability. Data are presented as mean (range). *P < 0.05 vs reference subjects.


167

Fig. 1

BP response to SNP. Averaged decrease in mean IAP during SNP treatment of 8 subjects with malignant hypertension. Values are mean ± SEM. Fig. 2

Systemic and cerebral hemodynamic responses to SNP. Averaged changes in HR, SV, CO, and SVR of 8 subjects with malignant hypertension during decrease in mean IAP induced by SNP. Values are mean ± SEM. All parameters differ significantly from starting values.

Chapter 8

168

Fig. 3

Representative example of sCA. Representative example of decrease in MCA Vmean during decrease in mean IAP induced by SNP in patient with malignant hypertension. Circles represent average of 30 seconds. Regression coefficient of this patient is as follows: MCA Vmean = 0.81xIAP+19.7; r = 0.87. Fig. 4

sCA. Averaged decrease in MCA Vmean during decrease in mean IAP as induced by SNP in 8 patients with malignant hypertension. Values are mean ± SEM.


169

Discussion The main findings of this study are that both dCA and sCA are impaired in patients with malignant hypertension before and during treatment with SNP. The data of this study show that during a rapid reduction in BP with SNP, MCA Vmean decreases almost one on one with MAP. Study Limitations The technique to monitor changes in arterial CBF may raise discussion for several reasons. First, MCA Vmean is calculated from the frequency distribution of the Doppler shifts, and it is assumed to represent flow velocity in the center of the vessel. Changes in MCA Vmean, however, reflect changes in flow,24 only when the diameter of the MCA remains constant during SNP treatment. Direct observations made during craniotomy have revealed that SNP did not affect the vessel diameter of the M1 segment of the MCA;29 therefore, we consider that changes in MCA Vmean in this study are proportional to those in flow.

Second, it is uncertain at which level of MAP the upper and lower limits of sCA are located in patients (admitted to hospital) with malignant hypertension. In normotensive humans, the lower limit of sCA is located ~60 mmHg.12,30,31 For obvious reasons, the upper limit of sCA cannot be studied in (normotensive) humans, but in normotensive baboons, it ranges between 120 and 150 mmHg.13 In humans with untreated severe chronic hypertension, the lower limit of sCA is shifted rightward to ~115 mmHg.11,12 Data on the upper limit of sCA are limited in that it was determined in only 3 hypertensive baboons and was found to be shifted rightward (between 155 and 169 mmHg).14 We assume, with MAP ~165 mmHg at admission in this study, that the pressure level was located around the upper limit of sCA with dCA impaired before treatment with SNP. The finding that, during treatment with IAP in the autoregulatory range, MCA Vmean decreased almost one on one with MAP conforms with impaired CA.

Third, in the reference subjects, we did not determine MCA V during a substantial decrease in BP induced by SNP infusion; thus, sCA was not evaluated. We consider that, in healthy subjects, a decline in MAP induced by SNP of ~20% barely affects MCA Vmean,32 and CBF,15 whereas a reduction in mean IAP in the patients was associated with a significant decline in MCA Vmean. In addition, in healthy subjects, data on dCA reflect the integrity of sCA,33 and we assume sCA to be preserved in the reference subjects.

Chapter 8

170

sCA and dCA Autoregulation implies that blood flow is maintained at a normal level of ~60 mL/ 100 g brain tissue per minute despite changes in perfusion pressure.7 Impaired sCA, with loss of the more-or-less zero-slope MAP–MCA Vmean relationship, has been reported in ischemic stroke,34 in severe head injury,35 and after cardiac arrest.36 Our findings suggest that sCA also is impaired in patients with malignant hypertension.

dCA is quantified by the counterregulatory capacity to maintain MCA V during abrupt changes in BP induced by thigh cuff deflation,25 or as evaluated by the transfer gain and phase lead of MCA Vmean to MAP during imposed,37 or spontaneous BP oscillations.38,39 During both hypotension8 and hypertension,26 this phase lead remains unaltered compared with the normal situation. In the present study, the MCA Vmean to MAP phase lead in patients with malignant hypertension before SNP treatment was significantly smaller than in the reference subjects and comparable to the phase lead found in patients with carotid artery obstruction.37 The tendency toward a larger transfer gain further supports a deteriorated dampening of BP oscillations in patients with malignant hypertension. SNP and Cerebral Hemodynamics Kety et al. observed that, in chronically hypertensive patients, global resting CBF and cerebral oxygen consumption are not different from those in normo-tensive subjects but that cerebrovascular resistance is elevated.40 SNP reduces SVR, but when infused directly in the carotid artery, it does not modify cerebrovascular resistance in healthy subjects.41 These findings, together with recent evidence that in normotensive subjects a reduction in MAP by SNP does not affect MCA Vmean,32 suggest that when BP is reduced pharmacologically, MCA V is secured by CA-mediated cerebral vasodilatation rather than a direct SNP-induced pharmacological effect on the cerebral vasculature. The present data demonstrate systemic vasodilatation during SNP treatment with a considerable increase in CO but no change in the cerebrovascular conductance index. One may speculate that the linear relation between MAP and MCA Vmean during treatment with SNP reflects a preferential blood flow to the (low-resistance) systemic vascular bed versus the (high-resistance) cerebrovascular bed.


171

Conclusion In patients with malignant hypertension, dCA is impaired. Decreases in cerebral artery blood velocity and BP in a linear fashion during SNP treatment also indicate impairment of sCA. This study confirms the contention that CA is compromised in patients with malignant hypertension. References

1. The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Arch Intern Med. 1997; 157: 2413–2446.

2. Hinchey J, Chaves C, Appignani B, et al. A reversible posterior leukoencephalo-pathy syndrome. N Engl J Med. 1996; 334: 494–500.

3. Vaughan CJ, Delanty N. Hypertensive emergencies. Lancet. 2000; 356: 411–417.

4. Ledingham JG, Rajagopalan B. Cerebral complications in the treatment of accele-rated hypertension. Q J Med. 1979; 48: 25–41.

5. Haas DC, Streeten DH, Kim RC, et al. Death from cerebral hypoperfusion during nitroprusside treatment of acute angiotensin-dependent hypertension. Am J Med. 1983; 75: 1071–1076.

6. Kaplan NM. Hypertensive crises. In: Kaplan NM, ed. Clinical Hypertension. Baltimore, Md: Williams & Wilkins; 1998: 265–280.

7. Lassen NA. Cerebral blood flow and oxygen consumption in man. Physiol Rev. 1959; 39: 183–238.

8. Schondorf R, Stein R, Roberts R, et al. Dynamic cerebral autoregulation is preserved in neurally mediated syncope. J Appl Physiol. 2001; 91: 2493–2502.

9. Van Lieshout JJ, Wieling W, Karemaker JM, et al. Syncope, cerebral perfusion, and oxygenation. J Appl Physiol. 2003; 94: 833–848.

10. Zampaglione B, Pascale C, Marchisio M, et al. Hypertensive urgencies and emer-gencies: prevalence and clinical presentation. Hypertension. 1996; 27: 144–147.

11. Strandgaard S. Autoregulation of cerebral blood flow in hypertensive patients: the modifying influence of prolonged antihypertensive treatment on the tolerance to acute, drug-induced hypotension. Circulation. 1976; 53: 720–727.

Chapter 8

172

12. Schmidt JF, Waldemar G, Vorstrup S, et al. Computerized analysis of cerebral blood flow autoregulation in humans: validation of a method for pharmacologic studies. J Cardiovasc Pharmacol. 1990; 15: 983–988.

13. Strandgaard S, MacKenzie ET, Sengupta D, et al. Upper limit of autoregulation of cerebral blood flow in the baboon. Circ Res. 1974; 34: 435–440.

14. Strandgaard S, Jones JV, MacKenzie ET, et al. Upper limit of cerebral blood flow autoregulation in experimental renovascular hypertension in the baboon. Circ Res. 1975; 37: 164–167.

15. Henriksen L, Paulson OB. The effects of sodium nitroprusside on cerebral blood flow and cerebral venous blood gases, II: observations in awake man during successive blood pressure reduction. Eur J Clin Invest. 1982; 12: 389–393.

16. Friedman DB, Jensen FB, Matzen S, et al. Non-invasive blood pressure monitoring during head-up tilt using the Penaz principle. Acta Anaesthesiol Scand. 1990; 34: 519–522.

17. Eckert S, Horstkotte D. Comparison of Portapres non-invasive blood pressure measurement in the finger with intra-aortic pressure measurement during incremental bicycle exercise. Blood Press Monit. 2002; 7: 179–183.

18. Zhang R, Crandall CG, Levine BD. Cerebral hemodynamics during the Valsalva maneuver: insights from ganglionic blockade. Stroke. 2004; 35: 843–847.

19. Wesseling KH, Jansen JRC, Settels JJ, et al. Computation of aortic flow from pressure in humans using a nonlinear, three-element model. J Appl Physiol. 1993; 74: 2566–2573.

20. Jellema WT, Wesseling KH, Groeneveld AB, et al. Continuous cardiac output in septic shock by simulating a model of the aortic input impedance: a comparison with bolus injection thermodilution. Anesthesiology. 1999; 90: 1317–1328.

21. Jansen JRC, Schreuder JJ, Mulier JP, et al. A comparison of cardiac output derived from the arterial pressure wave against thermodilution in cardiac surgery patients. Br J Anaesth. 2001; 87: 212–222.

22. Harms MPM, Wesseling KH, Pott F, et al. Continuous stroke volume monitoring by modeling flow from non-invasive measurement of arterial pressure in humans under orthostatic stress. Clin Sci. 1999; 97: 291–301.

23. Aaslid R. Cerebral hemodynamics. In: Newell DW, Aaslid R, eds. Transcranial Doppler. New York, NY: Raven Press, Ltd; 1992: 49–55.


173

24. Bishop CC, Powell S, Rutt D, et al. Transcranial Doppler measurement of middle cerebral artery blood flow velocity: a validation study. Stroke. 1986; 17: 913–915.

25. Aaslid R, Lindegaard KF, Sorteberg W, et al. Cerebral autoregulation dynamics in humans. Stroke. 1989; 20: 45–52.

26. Lipsitz LA, Mukai S, Hamner J, et al. Dynamic regulation of middle cerebral artery blood flow velocity in aging and hypertension. Stroke. 2000; 31: 1897–1903.

27. Zhang R, Zuckerman JH, Levine BD. Deterioration of cerebral autoregulation during orthostatic stress. J Appl Physiol. 1998; 85: 1113–1122.

28. Panerai RB, Dawson SL, Potter JF. Linear and nonlinear analysis of human dynamic cerebral autoregulation. Am J Physiol Heart Circ Physiol. 1999; 277: H1089–H1099.

29. Giller CA, Bowman G, Dyer H, et al. Cerebral arterial diameters during changes in blood pressure and carbon dioxide during craniotomy. Neurosurgery. 1993; 32: 737–741.

30. Larsen FS, Olsen KS, Hansen BA, et al. Transcranial Doppler is valid for determination of the lower limit of cerebral blood flow autoregulation. Stroke. 1994; 25: 1985–1988.

31. Olsen KS, Svendsen LB, Larsen FS. Validation of transcranial near-infrared spectroscopy for evaluation of cerebral blood flow autoregulation. J Neurosurg Anesthesiol. 1996; 8: 280–285.

32. Lavi S, Egbarya R, Lavi R, et al. Role of nitric oxide in the regulation of cerebral blood flow in humans: chemoregulation versus mechanoregulation. Circulation. 2003; 107: 1901–1905.

33. Tiecks FP, Lam AM, Aaslid R, et al. Comparison of static and dynamic cerebral autoregulation measurements. Stroke. 1995; 26: 1014–1019.

34. Schwarz S, Georgiadis D, Aschoff A, et al. Effects of induced hypertension on intracranial pressure and flow velocities of the middle cerebral arteries in patients with large hemispheric stroke. Stroke. 2002; 33: 998–1004.

35. Czosnyka M, Smielewski P, Piechnik S, et al. Cerebral autoregulation following head injury. J Neurosurg. 2001; 95: 756–763.

36. Sundgreen C, Larsen FS, Herzog TM, et al. Autoregulation of cerebral blood flow in patients resuscitated from cardiac arrest. Stroke. 2001; 32: 128–132.

Chapter 8

174

37. Diehl RR, Linden D, Lucke D, et al. Phase relationship between cerebral blood flow velocity and blood pressure: a clinical test of autoregulation. Stroke. 1995; 26: 1801–1804.

38. Zhang R, Zuckerman JH, Giller CA, et al. Transfer function analysis of dynamic cerebral autoregulation in humans. Am J Physiol Heart Circ Physiol. 1998; 274: H233–H241.

39. Panerai RB, Hudson V, Fan L, et al. Assessment of dynamic cerebral autoregulation based on spontaneous fluctuations in arterial blood pressure and intracranial pressure. Physiol Meas. 2002; 23: 59–72.

40. Kety SS, Hafkenschiel JH, Jeffers WA, et al. The blood flow, vascular resistance and oxygen consumption of the brain in essential hypertension. J Clin Invest. 1948; 27: 511–514.

41. Joshi S, Young WL, Duong H, et al. Intracarotid nitroprusside does not augment cerebral blood flow in human subjects. Anesthesiology. 2002; 96: 60–66.

Chapter 9

Cerebral and systemic hemodynamics during acute blood pressure reduction with sodium nitroprusside or labetalol in patients with malignant hypertension Rogier V. Immink, Bert-Jan H. van den Born, Gert A. van Montfrans, Yu Sok Kim, Markus W. Hollmann, Johannes J. van Lieshout

Submitted

Chapter 9

178

Abstract Background In patients with malignant hypertension both dynamic (dCA) and static cerebral autoregulation (sCA) are impaired. Therefore, a pharmacological induced decline of BP reduces cerebral blood flow (CBF) with the danger of cerebral hypoperfusion. We compared the reduction in CBF velocity during BP lowering with sodium nitroprusside (SNP) to that of labetalol in patients with malignant hypertension. Methods In fifteen patients fulfilling WHO criteria for malignant hypertension beat-to-beat mean arterial pressure (MAP), systemic vascular resistance (SVR; Modelflow), mean middle cerebral artery blood velocity (MCA Vmean) and cerebrovascular resistance (CVR; MAP to MCA Vmean ratio) were monitored during treatment with SNP (n = 8) or labetalol (n = 7).

Results The reduction in MAP with SNP (28 ± 3%; mean and SEM) and labetalol (28 ± 4%) was identical. Treatment with SNP resulted in a larger decline in systemic vascular resistance (-53 ± 4) compared to labetalol (-13 ± 10%; P < 0.005), but not in cerebrovascular resistance (SNP 10 ± 5%, labetalol -17 ± 5%). This resulted in a larger reduction in MCA Vmean for a given decrease in MAP during treatment with SNP (-22 ± 4%) than with labetalol (-13 ± 3%; P < 0.001). Conclusion In malignant hypertension, SNP reduced systemic rather than cerebral vascular resistance resulting in a considerable reduction in MCA Vmean per unit pressure compared to labetalol. Labetalol may be a safer alternative than nitroprusside in the treatment of patients with malignant hypertension and an impaired CA.

SNP versus labetalol in malignant hypertension

179

Introduction

Malignant hypertension is a life threatening condition requiring an immediate blood pressure (BP) reduction to limit or prevent target organ damage.1, 2 In patients with malignant hypertension, cerebral autoregulation (CA) is impaired meaning that cerebral blood flow decreases more or less linearly with BP.3 The initial reduction in BP should therefore not exceed ~25% of the presenting level to prevent symptomatic hypoperfusion of the brain.1, 2 Of the therapeutic agents available, sodium nitroprusside (SNP) and labetalol are commonly used for the initial treatment of malignant hypertension.2, 4 SNP, an arteriolar and venous vasodilator, is widely advocated as a first-line agent in the treatment of hypertensive emergencies.1, 2 SNP has a short half-life making it most suitable for an immediate and controlled reduction of blood pressure. In spite of its superior pharmacokinetics, SNP has some disadvantages which may hamper its use in the management of malignant hypertension and hypertensive encephalopathy such as a possible increase in intracranial pressure and risk of cyanide toxicity.5, 6 Labetalol, a combined alpha- and beta-adrenergic blocking agent, has a slower onset of action with a maximal hypotensive effect 5 to 15 min. after an intravenous bolus injection, however its long half-life of 4 to 6 hrs limits the ability to promptly correct hypotension with cessation of the drug.7-9 In contrast to SNP however, intracranial pressure does not seem to increase and the toxicity of labetalol is limited.10, 11 Both agents reduce BP effectively in patients with malignant hypertension,12, 13 but their specific effects on the cerebral and systemic circulation have not been investigated.

We considered that in patients with malignant hypertension and failing CA mechanisms, an immediate reduction of blood pressure is to be achieved with the smallest reduction of cerebral perfusion possible. In this study we therefore set out to determine the cerebral and hemodynamic effects of an immediate ~25% reduction in blood pressure with SNP and labetalol in patients with malignant hypertension.

Chapter 9

180

Subjects and Methods

Subjects We included 15 patients fulfilling the WHO criteria for malignant hypertension, severely elevated blood pressure together with grade III (bilateral retinal hemorrhages or cotton wool exudates) or IV (grade III plus papilledema) hypertensive retinopathy according to the Keith, Wagener and Barker classification. All subjects were admitted at a medium care facility for continuous BP monitoring. After admission an intravenous line was placed for administration of SNP (n = 8) or labetalol (n = 7) with the aim to reduce mean arterial pressure (MAP) by ~25% below the MAP just prior to the initiation of therapy. Both SNP and labetalol were administered using standardized protocols. SNP infusion was started at 0.3 µg·kg-1·min-1, increased to 0.5 µg·kg-

1·min-1 after 5 min and from then on, by 0.5 µg·kg-1·min-1 every 5 min (to a maximum dose of 5.0 µg·kg-1·min-1) until the desired blood pressure reduction was achieved. Data on these eight subjects have been previously reported.3 Labetalol was administered as boluses of 0.5 mg·kg-1 every 8 minutes (to a maximal total dose of 200 mg) until the desired 25% reduction in MAP was achieved followed by a continuous infusion of 20 mg/hr. Written informed consent was provided in accordance with the Helsinki declaration. The study protocol was approved by the Medical Ethical Committee of the Academic Medical Center, University of Amsterdam, the Netherlands.

Measurements Patients were instrumented with electrocardiogram electrodes in the supine position. A 1.1 mm ID, 20 gauge catheter was placed in the radial artery for continuous BP monitoring. Heart rate (HR) was the inverse of the interbeat interval. Stroke volume (SV) was determined by a three-element model of arterial input impedance (Modelflow).14 Modelflow computes a flow wave from the BP wave that is integrated to obtain the SV of the heart.14-17 SV was expressed as percentage change of the presenting value, CO was HR times SV and systemic vascular resistance (SVR) was the ratio of mean BP and CO. The middle cerebral artery blood velocity (MCA V) was measured in the proximal segment of the right middle cerebral artery (Multidop X4, Sipplingen, Germany). Once the optimal signal-to-noise ratio was obtained, the probe was secured with a headband (Mark 600, Spencer Technologies, Seattle, USA). To assess the effect of SNP and labetalol on cerebrovascular resistance, a


181

cerebrovascular resistance index (CVR) was calculated as the ratio of MAP and MCA Vmean.18

Data and statistical analysis Data were expressed as mean ± SEM. Changes in cerebral blood flow were tracked by MCA Vmean 19 and integrity of CA was reflected by constancy of MCA Vmean despite changes in MAP. For assessment of CA the signals of MCA V and BP were first averaged to 30 seconds episodes and then linearly related to each other. To compare CA between groups MAP and MCA Vmean were expressed as percentage change of pretreatment values. Dynamic CA was determined by calculating power spectra of pressure and velocity in the frequency domain from a 3-minute episode of beat-to-beat data of MAP and MCA Vmean prior to blood pressure lowering treatment with discrete Fourier transform, after spline interpolation and resampling at 4 Hz. Results were expressed as the integrated area in the low frequency range (0.07 to 0.15 Hz). To examine the strength between low frequency MAP and MCA Vmean, coherence was used to signify that the 2 cardiovascular signals covary significantly. The squared coherence function reflects the fraction of output power (MCA Vmean) that can be linearly related to the input power (MAP). From the MAP to MCA Vmean cross spectrum, the MCA Vmean to MAP phase lead (degrees) were obtained. Changes in systemic and cerebral hemodynamics during treatment were examined by Friedman ANOVA on ranks. Differences in CA between labetalol and SNP treatment (unpaired) and before and after treatment (paired) were examined with Wilcoxon rank sum test and Wilcoxon signed rank test, respectively. The slopes of the MAP-MCA Vmean relationship during treatment with SNP or labetalol were identified by regression analysis and Student’s t-test. P < 0.05 was considered to indicate a statically significant difference.

Results The characteristics of patients with malignant hypertension are listed in table 1. The phase difference between MCA Vmean and MAP in the lower frequency range was equally affected for SNP (30 ± 8º) and labetalol (26 ± 9º) with compa-rable coherences of 0.64 ± 0.04 and 0.59 ± 0.05. Target blood pressure was reached within 60 min in all patients. Changes in systemic and cerebral hemo-dynamics before and after treatment with SNP and labetalol are listed in table 2.

Chapter 9

182

The reduction in MAP with SNP (28 ± 3%) and labetalol (28 ± 4%) was identical, whereas the decline in systemic vascular resistance with SNP was larger (-13 ± 10% vs. -53 ± 4%; P < 0.005). SV did not change with SNP or labetalol (figure 1). The change in HR (increase with SNP and reduction with labetalol) determined the changes in CO. A decrease in CVR with labetalol (-17 ± 5%) but not with SNP (10 ± 5%; figure 2) resulted in a less significant percentage decline in MCA Vmean with labetalol compared to SNP for a given decrease in MAP (-13 ± 3% vs. -22 ± 4%; P < 0.001; figure 3).

Table 1 Baseline characteristics of patients with malignant hypertension

Mean ± SD. MCA = middle cerebral artery. Table 2 Hemodynamic variables at hospital admission and after treatment

Mean arterial blood pressure (MAP), heart rate (HR) and percentage change in stroke volume (SV), cardiac output (CO), systemic vascular resistance (SVR), mean middle cerebral artery blood velocity and cerebral vascular resistance (CVR) before and after treatment with sodium nitroprusside (SNP) or labetalol. Mean ± SD * P < 0.05 vs. before treatment, † P < 0.05 vs. after treatment with SNP.

Patient characteristics Nitroprusside Labetalol Patients, no. 8 7 Age (yrs) 44 ± 5 40 ± 5 Sex (M/F) 6:2 5:2 Height, cm 175 ± 5 172 ± 5 Weight, kg 76 ± 5 73 ± 5 Systolic BP, mmHg 225 ± 5 227 ± 5 Diastolic BP, mmHg 137 ± 3 133 ± 4 Mean MCA V, cm · s–1 64 ± 6 58 ± 8

Nitroprusside Labetalol Before After Before After

MAP (mmHg) 155 ± 6 112 ± 6* 166 ± 7 118 ± 5* HR (min-1) 87 ± 7 102 ± 6* 99 ± 8 76 ± 6*† SV (%) 100 109 ± 4 100 109 ± 9 CO (%) 100 131 ± 8* 100 86 ± 9*†

SVR (%) 100 57 ± 5* 100 87 ± 10*† MCA Vmean (cm·s-1) 64 ± 6 50 ± 5* 58 ± 8 49 ± 5*

CVR (%) 100 110 ± 5 100 83 ± 5*†


183

Fig. 1 Systemic circulatory responses

PRE POST

MA

P (m

mH

g)

0

50

100

150

200

PRE POST

HR

(bpm

)

0

50

100

150

PRE POST

SV ( ∆

%)

0

50

100

150

PRE POST

CO

( ∆%

)

0

50

100

150

**

** *

***

Mean arterial blood pressure (MAP), heart rate (HR) and percentage change in stroke volume (SV) and cardiac output (CO) before (PRE) and after (POST) treatment with sodium nitroprusside (black bars; n = 8) and labetalol (gray bars; n = 7). Mean ± SEM * P < 0.05.

Chapter 9

184

Fig. 2 Systemic versus. cerebral vascular resistance responses

PRE POST

SVR

( ∆%

)

0

50

100

150

PRE POST

CVR

( ∆%

)

0

50

100

150

* *

***

Percentage change in systemic vascular resistance (SVR) and cerebral vascular resistance (CVR) before (PRE) and after (POST) treatment with sodium nitroprusside (black bars; n = 8) and labetalol (gray bars; n = 7). Mean ± SEM, * P < 0.05. Fig. 3 Mean arterial pressure - middle cerebral artery blood velocity relationship

MAP (%)

50 75 100 125

MC

A V m

ean (

%)

50

75

100

125

*

Reduction in mean middle cerebral artery blood velocity (MCA Vmean) during a decrease in mean blood pressure (mean BP) by sodium nitroprusside (closed circles; n = 8) vs. labetalol (open circles; n = 7) Slope: 0.209 vs. 0.295 (P < 0.001)


185

Discussion In patients with malignant hypertension the therapeutic challenge is to reduce blood pressure without jeopardizing the cerebral circulation against the background of an impaired CA. The present study shows that the decline in cerebral blood flow velocity was smaller for a given decrease in MAP during treatment with labetalol compared to treatment with SNP. Treatment with labetalol reduced arterial pressure through a proportional decrease in systemic and cerebral vascular resistance, whereas SNP induced a greater decline in systemic rather than cerebral vascular resistance. Taken together, these findings indicate that against the background of impaired CA labetalol preserves cerebral blood flow better for a given decrease in arterial pressure than SNP. When either SNP or labetalol is administered to normotensive subjects, cerebral blood flow remains unaltered conforming maintained integrity of CA.10, 20 There is a multitude of data which suggests that sympathetic vasoconstriction in conjunction with CA protects the brain from regional overperfusion during hypertension.21, 22 As soon as mean BP consistently exceeds ~140 mmHg, CA becomes impaired and cerebral blood flow will passively follow changes in BP.3, 23 SNP is often recommended as the first choice treatment because of its rapid onset of action and ease of BP titration.1, 2 However, due its rapid action BP may incidentally be reduced too far below the target level. With labetalol the onset of blood pressure decline is considerably slower. This means that, together with the findings in this study, the risk of symptomatic reductions in cerebral blood flow should be smaller with labetalol compared to SNP. The larger reduction in cerebral blood flow with SNP compared to labetalol may be due to a deviation of blood flow to the systemic circulation as evidenced by the larger decline in systemic compared to cerebral vascular resistance. This deviation of blood flow with SNP has also been described for the coronary circulation in patients with coronary artery disease. Here, SNP treatment results in a deviation of blood flow away from the ischemic myocardium to coronary resistance vessels.24, 25 This coronary steal phenomenon associated with SNP in patients with coronary artery disease shows similarities with the “cerebral steal” phenomenon observed in the patients with malignant hypertension of the current

Chapter 9

186

study where blood flow is deviated away from the cerebral circulation to the systemic circulation. This study has some limitations that need comment. First, critical for the interpretation of the data are to what extent MCA Vmean reflects volume flow. The MCA Vmean is calculated from the frequency distribution of the Doppler shifts and is assumed to represent flow velocity in the center of the vessel. Changes in MCA Vmean however reflect changes in flow only when the diameter of the MCA remains constant during SNP or labetalol treatment.19 Direct observations made during craniotomy have revealed that SNP did not affect the vessel diameter of the MCA.26 Also, constancy of MCA diameter has been demonstrated for a range of pressures.27 Therefore we consider that in this study changes in MCA Vmean are proportional to those in flow.

Second, the order of the open label administration of the two drugs was not randomized. The incidence of malignant hypertension in the Netherlands is low and for practical reasons a sequential drug protocol was used because the pharmacokinetic properties of SNP and labetalol are entirely different. In spite of this study design patient groups were fully comparable for anthropomorphic data, cardiovascular variables, cerebral autoregulatory capacity and magnitude of blood pressure reduction, leaving the main findings of this study unchallenged. In summary, both SNP and labetalol reduce blood pressure adequately in patients with malignant hypertension. However, the hemodynamic changes elicited by these agents are completely different. The use of labetalol results in a proportional reduction in systemic and cerebral vascular resistances whereas SNP reduces systemic rather than cerebral vascular resistance with a preferential blood flow to the low resistance systemic vascular bed and a considerable reduction in MCA Vmean per unit pressure. Therefore labetalol may be a safer alternative than nitroprusside in the treatment of patients with malignant hypertension and an impaired CA. References 1. Calhoun DA, Oparil S. Treatment of hypertensive crises. N Engl J Med. 1990;323:

1177-1183.

2. Vaughan CJ, Delanty N. Hypertensive emergencies. Lancet. 2000;356:411-417.


187

3. Immink RV, van den Born BJ, van Montfrans GA, Koopmans RP, Karemaker JM, van Lieshout JJ. Impaired cerebral autoregulation in patients with malignant hypertension. Circulation. 2004;110:2241-2245.

4. Kaplan NM. Management of hypertensive emergencies. Lancet. 1994;344:1335-1338.

5. Turner JM, Powell D, Gibson RM, McDowall DG. Intracranial pressure changes in neurosurgical patients during hypotension induced with sodium nitroprusside or trimetaphan. Br J Anaesth. 1977;49:419-425.

6. Rindone JP, Sloane EP. Cyanide toxicity from sodium nitroprusside: risks and management. Ann Pharmacother. 1992;26:515-519.

7. Kanto J, Allonen H, Kleimola T, Mantyla R. Pharmacokinetics of labetalol in healthy volunteers. Int J Clin Pharmacol Ther Toxicol. 1981;19:41-44.

8. Rosei EA, Trust PM, Brown JJ, Lever AF, Robertson JI. Letter: Intravenous labetalol in severe hypertension. Lancet. 1975;2:1093-1094.

9. Solomons R. Hemiparesis after single minibolus of labetalol for hypertensive encephalopathy. Br Med J. 1979;2:672.

10. Olsen KS, Svendsen LB, Larsen FS, Paulson OB. Effect of labetalol on cerebral blood flow, oxygen metabolism and autoregulation in healthy humans. Br J Anaesth. 1995;75:51-54.

11. Orlowski JP, Shiesley D, Vidt DG, Barnett GH, Little JR. Labetalol to control blood pressure after cerebrovascular surgery. Crit Care Med. 1988;16:765-768.

12. Ahearn DJ, Grim CE. Treatment of malignant hypertension with sodium nitroprusside. Arch Intern Med. 1974;133:187-191.

13. Wilson DJ, Wallin JD, Vlachakis ND, Freis ED, Vidt DG, Michelson EL, Langford HG, Flamenbaum W, Poland MP. Intravenous labetalol in the treatment of severe hypertension and hypertensive emergencies. Am J Med. 1983;75:95-102.

14. Wesseling KH, Jansen JR, Settels JJ, Schreuder JJ. Computation of aortic flow from pressure in humans using a nonlinear, three-element model. J Appl Physiol. 1993;74:2566-2573.

15. Jellema WT, Wesseling KH, Groeneveld AB, Stoutenbeek CP, Thijs LG, van Lieshout JJ. Continuous cardiac output in septic shock by simulating a model of the aortic input impedance: a comparison with bolus injection thermodilution. Anesthesiology. 1999;90:1317-1328.

Chapter 9

188

16. Jansen JR, Schreuder JJ, Mulier JP, Smith NT, Settels JJ, Wesseling KH. A comparison of cardiac output derived from the arterial pressure wave against thermodilution in cardiac surgery patients. Br J Anaesth. 2001;87:212-222.

17. Harms MP, Wesseling KH, Pott F, Jenstrup M, Van GJ, Secher NH, van Lieshout JJ. Continuous stroke volume monitoring by modelling flow from non-invasive measurement of arterial pressure in humans under orthostatic stress. Clin Sci (Lond). 1999;97:291-301.

18. Zhang R, Crandall CG, Levine BD. Cerebral hemodynamics during the Valsalva maneuver: insights from ganglionic blockade. Stroke. 2004;35:843-847.

19. Bishop CC, Powell S, Rutt D, Browse NL. Transcranial Doppler measurement of middle cerebral artery blood flow velocity: a validation study. Stroke. 1986;17:913-915.

20. Lavi S, Egbarya R, Lavi R, Jacob G. Role of nitric oxide in the regulation of cerebral blood flow in humans: chemoregulation versus mechanoregulation. Circulation. 2003;107:1901-1905.

21. Bill A, Linder J. Sympathetic control of cerebral blood flow in acute arterial hypertension. Acta Physiol Scand. 1976;96:114-121.

22. Heistad DD, Marcus ML, Abboud FM. Effects of sympathetic nerve stimulation on cerebral blood flow. Acta Neurol Scand Suppl. 1977;64:306-307.

23. Strandgaard S, MacKenzie ET, Sengupta D, Rowan JO, Lassen NA, Harper AM. Upper limit of autoregulation of cerebral blood flow in the baboon. Circ Res. 1974;34:435-440.

24. Mann T, Cohn PF, Holman LB, Green LH, Markis JE, Phillips DA. Effect of nitroprusside on regional myocardial blood flow in coronary artery disease. Results in 25 patients and comparison with nitroglycerin. Circulation. 1978;57:732-738.

25. Chiariello M, Gold HK, Leinbach RC, Davis MA, Maroko PR. Comparison between the effects of nitroprusside and nitroglycerin on ischemic injury during acute myocardial infarction. Circulation. 1976;54:766-773.

26. Giller CA, Bowman G, Dyer H, Mootz L, Krippner W. Cerebral arterial diameters during changes in blood pressure and carbon dioxide during craniotomy. Neurosurgery. 1993;32:737-741.

27. Serrador JM, Picot PA, Rutt BK, Shoemaker JK, Bondar RL. MRI measures of middle cerebral artery diameter in conscious humans during simulated orthostasis. Stroke. 2000;31:1672-1678.

Chapter 10

Summary and discussion

Chapter 10

192

Summary The motivation for this thesis originated from the observation that malignant hypertension was relatively frequent in the catchment area of the hospital where the studies described in this thesis were conducted. Part I reviews the value of funduscopy in the management of patients with hypertension and contains studies on the epidemiology and genetics of malignant hypertension. Part II contains several mechanistic studies on the pathophysiology of malignant hypertension. Part I. In patients with a hypertensive crisis a retinal examination provides useful information to discriminate between hypertensive urgencies (situations which do not require immediate blood pressure reduction) and emergencies (situations requiring prompt blood pressure reduction). However, in the routine care of patients with hypertension the value of a retinal examination is disputed. In chapter 2 we performed a systematic search to answer the question whether a retinal examination is useful in the routine care of the hypertensive patient. We reviewed literature regarding the reliability of detecting the various retinal changes associated with hypertension and assessed the association of these retinal changes with blood pressure, other indicators of hypertensive organ damage, and cardiovascular morbidity and mortality. We found that of the various retinal abnormalities only hemorrhages and exudates could be reliably assessed. However, the positive and negative predictive values for the association between these retinal abnormalities and hypertension were low. The presence of hemorrhages and exudates increased the risk of stroke, but this association was also present in normotensive persons. We concluded that current evidence is lacking that funduscopy is of additional value in the routine management of hypertensive patients. Cohort studies have shown that patients from sub Saharan African descent are at increased risk for developing malignant hypertension. Differences in socio-economic factors such as health insurance and compliance have been implicated to explain the excess of malignant hypertension observed in this group. Chapter 3 contains a retrospective analysis on ethnic disparities in the incidence, presentation and complication of malignant hypertension in relation to hypertension awareness, treatment and compliance in a multi-ethnic population. The incidence of malignant hypertension was 4 times higher in blacks compared

Summary and Discussion

193

to white individuals and they had more renal dysfunction at presentation and during follow-up. Hypertension awareness was not different between ethnic groups, but only few black patients received anti-hypertensive treatment at the time of admission and they were more often uninsured compared to white patients. We concluded that ethnic disparities in the incidence and complications of malignant hypertension exist and may, at least in part, be explained by differences in anti-hypertensive treatment and insurance status. Previous studies demonstrated that recovery from renal failure may occur in patients admitted with malignant hypertension and renal insufficiency. In kidney biopsies of these patients, fibrinoid necrosis and thrombosis of small and medium sized renal arteries may be observed. In chapter 4 we tested the hypothesis that recovery from renal failure may be associated with the presence of a thrombotic microangiopathy (TMA). We found that patients with TMA as evidenced by a low platelet count and either an elevated LDH or presence of schistocytes (fragmented red blood cells) had a markedly increased serum creatinine at presentation compared to patients without TMA, but were more likely to have recovery of renal function during follow-up. Assessment of these parameters at presentation may provide useful information regarding the prognosis of renal function. Malignant hypertension can be considered an extreme phenotype of renin-mediated hypertension. In chapter 5 we assessed the frequencies of three polymorphisms of the renin-angiotensin system in consecutively admitted patients with malignant hypertension and compared them with age, sex and ethnicity matched hypertensive and normotensive controls from the source population. The TT genotype of the M235T polymorphism in the angio-tensinogen (AGT) gene was associated with an increased risk of malignant hypertension in white subjects with an odds of approximately 10 to 1 compared to hypertensive and normotensive controls. The high frequency of the TT genotype observed in populations of West-African descent may help explain the higher frequency of malignant hypertension in this group. Within the group with malignant hypertension, the TT genotype of the AGT M235T polymorphism was also associated with an increased frequency of renal dysfunction and TMA compared to MT or MM carriers despite similar age, sex, ethnicity and blood pressure.

Chapter 10

194

Part II. In Chapter 6 we tested the hypothesis whether the variations in activation of the renin-angiotensin aldosterone system (RAAS) could be explained by the degree of microvascular damage, as evidenced by the presence of TMA, in patients with malignant hypertension. We prospectively examined the association between RAAS activation, as evidenced by plasma renin activity (PRA) and aldosterone, with TMA parameters, including LDH, in consecutive patients presenting at the emergency department with a suspected hypertensive emergency. As expected PRA was markedly higher in patients with malignant hypertension compared to severely hypertensive patients. However, 24% of malignant hypertensive patients presented with normal PRA, suggesting that RAAS activation is not an obligatory feature of malignant hypertension. The degree of TMA, according to LDH levels, could explain most of the variations in PRA, suggesting that the degree of RAAS activation and the severity of microvascular damage strongly influence each other. The TMA associated with malignant hypertension is virtually indistinguishable from that in thrombotic thrombocytopenic purpura (TTP). TTP is associated with an acquired or congenital deficiency of the von Willebrand factor cleaving protease (ADAMTS13) that cleaves large prothrombogenic multimers. To explore the pathogenesis of the TMA associated with malignant hypertension, we assessed ADAMTS13 activity in patients with malignant hypertension, severe hypertension and normotensive controls (chapter 7). Patients with malignant hypertension had a lower ADAMTS13 activity compared to normotensive controls and the severity of TMA negatively correlated with ADAMTS13 activity. A severe deficiency of ADAMTS13, such as that obser-ved in TTP, could not be demonstrated. The association between ADAMTS13 activity and VWF as observed in this study suggests that deficiency of ADAMTS13 may be secondary to the release of large amounts of VWF upon endothelial stimulation. Cerebral autoregulation (CA) keeps cerebral blood flow constant despite variations in blood pressure. CA is assumed to be impaired in malignant hypertension and hypertensive encephalopathy. In chapter 8 we tested static and dynamic CA in 8 patients with malignant hypertension during an acute blood pressure reduction with SNP using a transcranial Doppler technique. Middle cerebral artery blood flow velocity decreased almost one on one with mean arterial pressure. A plateau phase, indicating restoration of static CA, could not


195

be detected during treatment with SNP. Dynamic CA was disturbed before and during treatment with SNP. In line with previous assumptions, we concluded that both static and dynamic CA are, indeed, impaired in patients with malignant hypertension. In chapter 9 we compared the effects of treatment with SNP and labetalol on cerebral blood flow and systemic hemodynamics in an open label trial in fifteen patients with malignant hypertension. The reduction in mean arterial pressure with SNP was identical to that of labetalol. Treatment with SNP resulted in a larger decrease in cerebral blood flow compared to labetalol, likely as a result of a more pronounced reduction in systemic rather than cerebral vascular resistance. The preservation of cerebral blood flow with labetalol may suggest that blood pressure can be lowered more safely with labetalol than with SNP against the background of an impaired CA. Discussion In this thesis a number of questions concerning the epidemiology and pathophysiology of malignant hypertension have been addressed. However, several issues regarding the pathogenesis, diagnosis and treatment of malignant hypertension are debated, these will be further discussed. Pathogenesis The pathogenesis of malignant hypertension has been a subject of research and debate for some time beginning with the vicious circle theory from Wilson and Byrom,1 and the discussion on the role of blood pressure and the renin-angiotensin system in the pathogenesis of this condition.2, 3 Activation of the renin-angiotensin system is invariably associated with the development of malignant hypertension (chapter 1). The rate limiting step in the generation of angiotensin II, the vasoactive effector molecule, is the formation of angiotensin I from AGT by renin. Therefore, factors leading to either increased generation of AGT or renin may increase the risk of malignant hypertension. In chapter 5 we showed that the TT genotype of the M235T polymorphism in the AGT gene was associated with an increased risk of malignant hypertension and the development of TMA. Previous studies have shown that this polymorphism is associated with higher levels of AGT, thereby facilitating the generation of angiotensin II. It is conceivable that other functional polymorphisms in the renin-angiotensin system may also predispose to malignant hypertension.

Chapter 10

196

Linkage studies in families of probands with malignant hypertension may provide more insight into the pathogenesis of this condition and may help to identify patients at risk of malignant hypertension. Although RAAS activation is associated with malignant hypertension, clinical observations and data from animal experiments suggest that RAAS activation is a secondary event (chapter 1). For the generation of angiotensin II, renin has to be released first. The renin release from juxtaglomerular cells in the kidney may be stimulated by ischemia as a result of vascular injury resulting from severe uncontrolled hypertension. In chapter 6 we showed that the TMA associated with malignant hypertension, as evidenced by increasing LDH levels, correlates well with PRA in a log-log association, suggesting that the degree of vascular damage and activation of the RAAS may positively enhance each other. The association of the AGT polymorphism with TMA (chapter 5) suggests that a facilitated activation of the renin-angiotensin system may predispose to this vicious circle of RAAS activation and vascular injury. To determine the mechanisms involved in the development of the vascular injury of malignant hypertension we assessed several prohemostatic factors possibly implicated in the development of TMA (chapter 7). A decrease in ADAMTS13 activity was associated with a more severe TMA, likely as a result of increased release of VWF upon endothelial stimulation. This indicates that factors negatively influencing ADAMTS 13 activity may be important in aggravating the TMA of hypertensive crises. However, the role of ADAMTS13 activity in the vascular injury of malignant hypertension has to be explored further in experimental models of malignant hypertension with TMA. Diagnosis The WHO definition of malignant hypertension implies that patients with “severe hypertension” and grade III or IV retinopathy according to Keith, Wagener and Barker (KWB) classification should be admitted and receive parenteral anti-hypertensive therapy to lower blood pressure to “safe”, but not normal levels.4 The term “severe” was chosen to indicate that the acceleration of blood pressure is more important than the absolute blood pressure values in the development of malignant hypertension.4 However, “severe hypertension” may denote any blood pressure above 180 mmHg systolic or 110 mmHg diastolic according to current guidelines.5 In previously normotensive persons, a hypertensive emergency may develop at blood pressure levels as low as 110 mmHg diastolic.6 In contrast, some patients may present with blood pressures


197

above 220 mmHg systolic and 130 mmHg diastolic without signs of a hypertensive emergency.7 To improve diagnostic performance experts have advocated to include a blood pressure threshold of “usually above 120 mmHg diastolic”. However, the large overlap of blood pressure values limits the use of blood pressure as a criterion to discriminate between hypertensive urgencies and emergencies. Therefore, identification of patients according to the WHO criteria for malignant hypertension largely depends on the discriminative value of the retinal abnormalities associated with this condition. In chapter 2 we showed that the retinopathy of malignant hypertension (hemorrhages and exudates) can be reliably assessed with an excellent agreement between observers. Although agreement on the detection of papilledema is moderate,8 papilledema is seldom observed without hemorhages or exudates and will therefore yield a low number of false-negative and -positive results.9 Therefore a retinal examination may, in theory, be useful to discriminate hypertensive urgencies from emergencies. The question is therefore whether the retinal abnormalities observed in patients with severe hypertension can sufficiently discriminate between patients who should be admitted for parenteral therapy and patients who can receive oral medication. In general, there are two immediate risks mandating an immediate but controlled blood pressure reduction in patients with a hypertensive crises: cerebral hypo- or hyperperfusion resulting from an impaired CA and acute renal failure. In chapter 8 we demonstrated that patients with severe hypertension and grade III or IV retinopathy have an impaired CA indicating that, in theory, cerebral ischemia may develop even at small reductions in blood pressure. To prevent cerebral hypo- or hyperperfusion parenteral anti-hypertensive therapy should be pre-ferred as oral agents have either an unpredictable or slow onset of action. Although the retinopathy of malignant hypertension is associated with an impaired CA, the question remains whether funduscopy can sufficiently discriminate patients with impaired and patients with intact CA. Patients with hypertensive encephalopathy for example are expected to have an impaired CA, but may not always have retinal abnormalities.10 Vice versa, not all patients with retinal hemorrhages and exudates have an impaired CA. In our sample, severe impairment of CA, defined as a phase lead < 30%,11 was noted in 10 of 15 (67%) patients with malignant hypertension, but was completely normal (phase lead > 50%) in 4 of 15 patients. There are no data on CA in patients with severe hypertension without retinal abnormalities. Therefore, the negative predictive value of grade III and IV retinopathy and disturbances in CA remains to be determined. Assessment of the clinical signs of hypertensive encephalopathy

Chapter 10

198

remains paramount in patients with (suspected) hypertensive crises, irrespective of the retinal findings. Future studies may be directed at examining CA in severely hypertensive patients without retinal abnormalities to better assess the value of a retinal examination in patients with malignant hypertension. The survival of malignant hypertension has considerably improved. In our cohort the survival was 90% after 4 years and death was in most cases from non-cardiovascular causes (chapter 3). However, almost 20% needed kidney replacement therapy indicating that renal failure is an important complication of malignant hypertension. In chapter 4 we showed that TMA is frequently observed in malignant hypertension and associated with impaired renal function at presentation. However, recovery from renal failure occurred more often in these patients during follow-up, even if temporary kidney replacement therapy was needed. Prompt blood pressure lowering treatment in these patients may preserve renal function. However, a large blood pressure reduction may further compromise blood flow to an ischemic kidney. Therefore, an immediate but controlled reduction of blood pressure by parenteral anti-hypertensive agents seems indicated in these patients. As with CA, the predictive value of the retinal lesions associated with malignant hypertension in the assessment of TMA is undetermined. In chapter 6 and 7 we showed that patients with grade III or IV retinopathy had a more severe TMA, more pronounced activation of the renin-angiotensin system and greater renal impairment compared to patients without retinal abnormalities. Still, some of the patients without retinopathy had TMA (chapter 7). If one considers the retinopathy of malignant hypertension as a diagnostic test for TMA, as defined in chapter 5, the sensitivity is 71%. Therefore, 29% of those with TMA do not have grade III or IV retinopathy (false-negative rate). Although these numbers are derived from a small sample of 40 consecutive patients, this figure indicates that a retinal examination cannot sufficiently discriminate patients with and without TMA.

To improve identification of patients with a true hypertensive emergency, one may consider to extent the current working definition with the clinical criteria for hypertensive encephalopathy and the laboratory criteria for TMA (table 1).


199

Table 1 Proposed working definition of malignant hypertension. Severe hypertension (Diastolic BP usually above 120 mmHg) and 1 or more of the following: 1. Clinical symptoms consistent with cerebral complications: unexplained lowering of consciousness or delirium, lethargy, confusion, generalized or focal seizures or visual impairment. 2. Retinal abnormalities consistent with grade III or IV retinopathy according to the Keith, Wagener and Barker classification. 3. Low platelet count with either an elevated LDH or presence of schistocytes.

Treatment The treatment of malignant hypertension is based on the idea that lowering blood pressure by ~25% in minutes to hours results in a safe area where cerebral blood flow can be held constant despite variations in blood pressure, the so called plateau phase of CA.4 However, in the measurements of chapter 8 and 9 a plateau phase could not be detected during treatment with either SNP or labetalol. This means that the consensus of a ~25% reduction in mean arterial pressure is based on an incorrect assumption. Nevertheless, a blood pressure reduction not exceeding ~25% in the first minutes to hours (depending on the clinical severity) appears to be well tolerated and seems an appropriate recommendation: a rapid blood pressure reduction of more than 50% has been shown to cause ischemic stroke and even death;12-14 theoretically, smaller blood pressure reductions may lead to aggravation of cerebral hyperperfusion and cerebral edema. The question is for what time period the ~25% reduction in blood pressure should be maintained. This may depend on the time frame CA regains its functional capacity. Preliminary data obtained in patients admitted with malignant hypertension suggest that the initial impairment of CA is improved within 1 week after admission (unpublished observations). Therefore, the current recommendation to maintain blood pressure at the initial target level for “several days” and then reduce to normotensive values seems appropriate.15 Finally, there are no trials comparing parenteral anti-hypertensive agents in malignant hypertension on cardiovascular endpoints and, as a result of the low incidence of malignant hypertension, it is not expected that these trials will appear in the near future. We therefore compared the effects of SNP and

Chapter 10

200

labetalol on cerebral blood flow and systemic hemodynamics. Although both labetalol and SNP were equally effective in lowering blood pressure, the greater decline in middle cerebral artery blood flow and systemic vascular resistance following treatment with SNP suggests that blood flow may be directed away from the cerebral vasculature. This “cerebral steal phenomenon”, which has also been described for the coronary circulation (chapter 1), may increase the risk of cerebral hypoperfusion. The 10% difference in middle cerebral artery flow between SNP and labetalol may not be clinically significant for patients with an otherwise intact cerebral vasculature, but may be sufficient to cause hypo-perfusion in situations where the cerebral vasculature is compromised such as in patients with atherosclerosis and elderly patients. The addition of the disadvantageous effect of SNP on cerebral blood flow to the other hazards of this drug may render labetalol a better alternative for the treatment of malignant hypertension. References


2. Giese J. Renin, angiotensin and hypertensive vascular damage: a review. Am J Med. 1973;55:315-332.

3. Beilin LJ, Goldby FS, Mohring J. High arterial pressure versus humoral factors in the pathogenesis of the vascular lesions of malignant hypertension. Clin Sci Mol Med. 1977;52:111-117.


5. Cifkova R, Erdine S, Fagard R, Farsang C, Heagerty AM, Kiowski W, Kjeldsen S, Luscher T, Mallion JM, Mancia G, Poulter N, Rahn KH, Rodicio JL, Ruilope LM, van Zwieten P, Waeber B, Williams B, Zanchetti A. Practice guidelines for primary care physicians: 2003 ESH/ESC hypertension guidelines. J Hypertens. 2003;21: 1779-1786.


7. Zampaglione B, Pascale C, Marchisio M, Cavallo-Perin P. Hypertensive urgencies and emergencies. Prevalence and clinical presentation. Hypertension. 1996;27:144-147.


201


9. Lip GY, Beevers M, Dodson PM, Beevers DG. Severe hypertension with lone bilateral papilledema: a variant of malignant hypertension. Blood Press. 1995;4: 339-342.


11. Immink RV, van Montfrans GA, Stam J, Karemaker JM, Diamant M, van Lieshout JJ. Dynamic cerebral autoregulation in acute lacunar and middle cerebral artery territory ischemic stroke. Stroke. 2005;36:2595-2600.




15. Calhoun DA, Oparil S. Treatment of hypertensive crises. N Engl J Med. 1990;323: 1177-1183.

Chapter 11

Samenvatting en discussie

Chapter 11

206

Samenvatting De motivatie voor dit proefschrift komt voort uit de observatie dat maligne hypertensie relatief vaak werd gezien in de regio van het ziekenhuis waar de studies die in dit proefschrift zijn beschreven later werden uitgevoerd. Deel I van dit proefschrift gaat over de waarde van het routinematig onderzoeken van het netvlies (retina) bij patiënten met hypertensie en bevat studies over de epidemio-logie en genetica van maligne hypertensie. In deel II worden een aantal mecha-nistische studies over de pathofysiologie van maligne hypertensie beschreven. Deel I. Het onderzoek van de retina geeft belangrijke informatie om te discrimineren tussen een hypertensief spoedgeval (situatie waarbij de bloeddruk meteen moet worden verlaagd) en een hypertensieve urgentie (situatie waarbij de bloeddruk niet onmiddellijk hoeft te worden verlaagd). De waarde van een routinematig onderzoek van de retina bij patiënten met hypertensie wordt echter betwist. In hoofdstuk 2 van dit proefschrift voerden wij een systematisch literatuuronder-zoek uit met als doel de vraag te beantwoorden of onderzoek van de retina nuttig is bij de routine behandeling van hypertensie. Wij keken hierbij naar de betrouwbaarheid van het vaststellen van de verschillende retina-afwijkingen die met hypertensie zijn geassocieerd, en stelden de associatie van de verschillende retina-afwijkingen met hypertensie, hypertensieve orgaanschade en cardio-vasculaire morbiditeit en mortaliteit vast. De conclusie was dat van de verschil-lende retina-afwijkingen alleen retinabloedinkjes en exsudaten betrouwbaar konden worden vastgesteld. De positief en negatief voorspellende waarde van deze retina-afwijkingen voor het vaststellen van hypertensie was echter laag. De aanwezigheid van retinabloedinkjes en exsudaten verhoogden het risico op een cerebrovasculair accident, maar deze associatie was ook aanwezig bij personen met een normale bloeddruk. Wij concludeerden dat er onvoldoende bewijs is dat het routinematig onderzoeken van de retina van toegevoegde waarde is bij de behandeling van patiënten met hypertensie. Cohort-studies laten zien dat patiënten die oorspronkelijk afkomstig zijn uit West-Afrika een verhoogd risico hebben op het krijgen van maligne hyper-tensie. Verschillen in sociaal-economische factoren zoals verzekeringsstatus en compliantie zijn als oorzaak aangevoerd om het verhoogde voorkomen van maligne hypertensie in deze groep te verklaren. Hoofdstuk 3 bevat een analyse van de etnische verschillen in het voorkomen, de presentatie en de complicaties


207

van maligne hypertensie in relatie tot de behandeling van hypertensie in een multi-etnische populatie. De incidentie van maligne hypertensie bleek 4 keer hoger bij patiënten van Afrikaanse afkomst dan bij van oorsprong Nederlandse inwoners. Bij hen kwam ook meer nierinsufficiëntie voor, zowel bij presentatie als bij vervolgcontroles. De kennis van het hebben van hypertensie was niet verschillend tussen de twee groepen, maar slechts weinig patiënten van Afrikaanse afkomst kregen bloeddrukverlagende behandeling voor opname en waren vaker onverzekerd. Wij concludeerden dat etnische verschillen in de incidentie en complicaties van maligne hypertensie bestaan en, deels, verklaard kunnen worden door verschillen in behandeling en verzekeringsstatus. Eerdere studies hebben laten zien dat herstel van nierfunctie kan optreden bij patiënten die zich presenteren met maligne hypertensie en nierinsufficiëntie. In nierbiopten van deze patiënten wordt vaker fibrinoïde necrose en trombose gezien van de middelgrote en kleine arteriën. In hoofdstuk 4 onderzochten wij of herstel van nierfunctie geassocieerd zou kunnen zijn met de aanwezigheid van een trombotische microangiopathie (TMA). Wij vonden dat patiënten met een TMA, zich uitend in een laag bloedplaatjesgetal met een verhoogd lactaat-dehydrogenase (LDH) of aanwezigheid van fragmentocyten (kapotte rode bloed cellen) in het perifere bloed, een sterk verhoogde serum kreatinine concentratie hadden vergeleken met patiënten zonder TMA, maar dat herstel van nierfunctie tijdens vervolgafspraken vaker optrad. Het vaststellen van deze TMA para-meters bij opname kan bruikbare informatie geven ten aanzien van de prognose van de nierfunctie. Maligne hypertensie kan als een extreem fenotype van renine-gemedieerde hypertensie worden beschouwd. In hoofdstuk 5 onderzochten wij de frequenties van 3 polymorfismen in het renine-angiotensine systeem bij patiënten met maligne hypertensie en vergeleken die met leeftijd, geslacht en etniciteit gematchte controles uit het verzorgingsgebied. In dit multi-etnische cohort bleek dat het TT genotype van het M235T polymorfisme in het angiotensinogeen (AGT) gen geassocieerd was met een sterk verhoogd risico op maligne hypertensie bij de oorspronkelijk Nederlandse populatie met een odds ratio van ongeveer 10 op 1 vergeleken met hypertensieve en normotensieve controles. Bij de van oorsprong West-Afrikaanse populatie werd een veel hogere frequentie van het TT genotype gevonden, ook bij de hypertensieve en normotensieve controles. Dit zou een verklaring kunnen zijn waarom maligne hypertensie vaker bij deze groep wordt gezien. Binnen de groep patiënten met maligne hypertensie

Chapter 11

208

werd ook een associatie gevonden tussen het TT genotype van het AGT polymorfisme en een verhoogde frequentie van nierinsufficiëntie en TMA vergeleken met patiënten die het MT of MM gen droegen zonder dat er verschillen waren in leeftijd, geslacht, etniciteit en bloeddruk. Deel II. In hoofdstuk 6 onderzochten wij of de variaties in activatie van het renine-angiotensine aldosteron systeem (RAAS) bij maligne hypertensie, verklaard zouden kunnen worden door de mate van microvasculaire schade zich uitend in een TMA. In dit onderzoek vergeleken wij de plasma renine activiteit (PRA) en de aldosteron concentratie als uitingen van RAAS activatie met het LDH als uiting voor de ernst van de TMA, bij opeenvolgende patiënten die zich presenteerden met een hypertensieve crise. De PRA bij patiënten met maligne hypertensie was duidelijk hoger vergeleken met patiënten die een ernstige hypertensie hadden zonder retinale afwijkingen. Desondanks had 24% van de patiënten met maligne hypertensie een normale PRA, wat suggereert dat RAAS activatie niet een noodzakelijk verschijnsel is bij maligne hypertensie. De mate van TMA, zich uitend in hogere LDH waarden, kon het grootste deel van de variaties in PRA verklaren, wat suggereert dat de mate van RAAS activatie en de ernst van de microvasculaire schade elkaar sterk beïnvloeden. De TMA die wordt gezien bij maligne hypertensie is bijna niet te onderscheiden van de TMA bij trombotische trombocytopenische purpura (TTP). TTP is geassocieerd met een verworven of aangeboren tekort van het von Willebrand factor cleaving protease (ADAMTS13), dat grote protrombogene multimeren splitst. Om de oorzaak van de TMA bij maligne hypertensie te achterhalen, onderzochten wij de ADAMTS13 activiteit bij patiënten met maligne hyper-tensie, ernstige hypertensie en normotensieve controles. Patiënten met maligne hypertensie bleken een lagere ADAMTS13 activiteit te hebben vergeleken met normotensieve controles. Een ernstiger TMA was geassocieerd met een lagere ADAMTS13 activiteit. Een ernstig tekort van ADAMTS13, zoals die wordt gezien bij TTP, kon niet worden vastgesteld. De associatie tussen de ADAMTS13 activiteit en de hoogte van het VWF die wij in deze studie vonden, suggereert dat de verminderde ADAMTS13 activiteit mogelijk het gevolg is van het vrijkomen van grote hoeveelheden VWF door endotheelstimulatie. De cerebrale autoregulatie (CA) zorgt ervoor dat de bloedstroom in de hersenen constant wordt gehouden ondanks grote variaties in bloeddruk. Verondersteld


209

wordt dat de CA bij maligne hypertensie en hypertensieve encephalopathie is aangedaan. In hoofdstuk 8 onderzochten we het effect van nitroprusside (SNP) op de bloedstroom in de middelste hersenslagader met behulp van transcraniëel Doppler onderzoek tijdens de acute behandeling van patiënten met maligne hypertensie. We onderzochten zowel de korte- (dynamische) als langetermijn (statische) regulatie van de cerebrale bloedstroom. Wij vonden dat de bloed-stroom in de middelste hersenslagader bijna 1 op 1 daalde met de gemiddelde bloeddruk tijdens de behandeling met nitroprusside. Een plateau fase, duidend op herstel van de statische CA, werd tijdens deze meting niet gevonden. De dynamische CA was zowel voor als tijdens de behandeling met SNP aangedaan. In overeenstemming met eerdere veronderstellingen, vonden wij dat zowel de statische als dynamische CA is aangedaan bij patiënten met maligne hypertensie. In hoofdstuk 9 vergeleken we het effect van SNP en labetalol op de bloedstroom in de middelste hersenslagader en systemische hemodynamiek in een open label onderzoek bij 15 patiënten met maligne hypertensie. De gemiddelde bloeddruk-daling tijdens de behandeling met SNP was gelijk aan die van labetalol. Behandeling met SNP resulteerde echter in een grotere vermindering van de cerebrale bloedstroom dan de behandeling met labetalol, waarschijnlijk door een meer uitgesproken daling van de systemische weerstand vergeleken met de cerebrale vaatweerstand. Het relatieve behoud van de cerebrale bloedstroom met labetalol suggereert dat de bloeddruk veiliger kan worden verlaagd met labetalol dan met SNP in situaties waarbij de CA gestoord is. Discussie In dit proefschrift wordt een aantal vragen behandeld over de epidemiologie en pathofysiologie van maligne hypertensie. Niet alle vragen over de pathogenese, diagnose en behandeling van maligne hypertensie kunnen worden beantwoord. Op een aantal discussiepunten zal nog verder worden ingegaan. Pathogenese De pathogenese van maligne hypertensie is sinds geruime tijd onderwerp van discussie, die begon met de vicieuze cirkel theorie van Wilson en Byrom,1 en het debat over de rol van bloeddruk en het renin-angiotensine systeem in het ontstaan van deze aandoening.2, 3 Activatie van het renine-angiotensine systeem is sterk geassocieerd met de ontwikkeling van maligne hypertensie (hoofdstuk 1). De snelheidsbeperkende stap in de productie van angiotensine II, het

Chapter 11

210

vasoactieve molecuul van het RAAS, is de generatie van angiotensine I uit AGT door renine. Factoren die leiden tot een verhoogde productie van AGT of renine kunnen daardoor het risico op maligne hypertensie vergroten. In hoofdstuk 5 lieten we zien dat het TT genotype van het M235T polymorfisme in het AGT gen sterk geassocieerd is met het risico op maligne hypertensie en het ontwikkelen van TMA. Eerdere studies hebben laten zien dat dit polymorfisme is geassocieerd met hogere AGT waarden, waardoor de aanmaak van angio-tensine II makkelijker verloopt. Het is denkbaar dat ook andere polymorfismen in het renine-angiotensine systeem het risico op het krijgen van maligne hypertensie vergroten. Linkage studies in families van patiënten met maligne hypertensie zouden meer inzicht kunnen geven in het ontstaan van deze aandoening en het identificeren van personen met een verhoogd risico op maligne hypertensie. Hoewel activatie van het RAAS vaak bij maligne hypertensie wordt gezien, suggereren klinische observaties en dierexperimenten dat RAAS activatie niet primair verantwoordelijk is voor het ontstaan van maligne hypertensie (hoofdstuk 1). Voor de generatie van angiotensine II is eerst secretie van renine nodig. De renine afgifte uit juxtaglomerulaire cellen in de nier zou gestimuleerd kunnen worden door ischemie als gevolg van vaatschade bij de ernstige verhogingen van de bloeddruk. In hoofdstuk 6 lieten we zien dat de TMA, die bij maligne hypertensie wordt gezien, goed correleert met de PRA in een log-log associatie, wat suggereert dat de mate van vasculaire schade en RAAS activatie elkaar kunnen versterken. De associatie van het AGT polymorfisme met TMA (hoofdstuk 5) suggereert dat de gefaciliteerde activatie van het renine-angio-tensine systeem zou kunnen predisponeren tot de vicieuze cirkel van RAAS activatie en vaatschade. Om de mechanismen die betrokken zijn bij deze vaatschade te ontrafelen bepaalden we enkele prohemostatische factoren die betrokken zouden kunnen zijn bij het ontstaan van TMA (hoofdstuk 7). Een vermindering van de ADAMTS13 activiteit bleek geassocieerd te zijn met een ernstiger manifestatie van TMA, waarschijnlijk door het vrijkomen van grote hoeveelheden VWF door stimulatie van het endotheel. Dit betekent dat factoren die de ADAMTS13 activiteit negatief beïnvloeden belangrijk kunnen zijn in het verergeren van de TMA van hypertensieve crisen. De rol van ADAMTS13 bij de vaatschade van maligne hypertensie zou verder kunnen worden uitgediept in experimentele modellen van maligne hypertensie met TMA.


211

Diagnose De huidige WHO definitie van maligne hypertensie houdt in dat patiënten met “ernstige hypertensie” en graad III of IV retinopathie volgens de Keith, Wagener en Barker classificatie opgenomen moeten worden voor intraveneuze behande-ling om de bloeddruk naar veilige, maar nog niet normale, waarden te verlagen.4 De term “ernstig” is destijds gekozen om aan te geven dat de snelheid van het stijgen van de bloeddruk belangrijker is bij het ontstaan van maligne hypertensie dan de absolute hoogte.4 “Ernstige hypertensie” kan volgens de huidige richt-lijnen iedere bloeddruk boven de 180 mmHg systolisch en 110 mmHg diastolisch betekenen.5 In voorheen normotensieve patiënten kan een hyper-tensief spoedgeval optreden bij diastolische bloeddrukwaarden tot 110 mmHg.6 Daarentegen kunnen sommige patiënten zich met een bloeddruk presenteren boven de 220 mmHg systolisch en 130 mmHg diastolisch zonder aanwijzingen voor een hypertensief spoedgeval.7 Om de diagnostische accuratesse te vergro-ten is een drempel gepropageerd van “meestal boven de 120 mmHg diastolisch”. De grote overlap in bloeddrukwaarden waarbij een hypertensief spoedgeval kan optreden beperkt echter het gebruik van bloeddruk als criterium om een onderscheid te maken tussen hypertensieve spoedgevallen en hypertensieve urgenties. Daardoor hangt de identificatie van patiënten met maligne hypertensie volgens de WHO definitie vooral af van de onderscheidende waarde van de retinale afwijkingen die geassocieerd zijn met deze aandoening. In hoofdstuk 2 lieten we zien dat de bloedingen en exsudaten van maligne hypertensie betrouwbaar kunnen worden vastgesteld met een uitstekende overeenkomst tussen verschillende onderzoekers. Hoewel het vaststellen van papiloedeem een matige overeenkomst heeft tussen verschillende onderzoekers,8 wordt papil-oedeem zelden gezien zonder bloedingen en exsudaten met als gevolg dat het aantal foutnegatieve en foutpositieve uitslagen beperkt blijft.9 Omdat de retinale afwijkingen bij maligne hypertensie betrouwbaar kunnen worden vastgesteld zouden ze daarom, in theorie, bruikbaar kunnen zijn om een hypertensieve urgentie van een hypertensief spoedgeval te onderscheiden. De vraag is echter of de retinale afwijkingen een onderscheid kunnen maken tussen patiënten die moeten worden opgenomen voor intraveneuze behandeling en patiënten waarbij kan worden volstaan met orale medicatie. In het algemeen zijn er twee onmiddellijke risico’s die een snelle maar gecontro-leerde verlaging van de bloeddruk noodzakelijk maken: cerebrale hypo- of hyperperfusie veroorzaakt door een falende CA en acute nierinsufficiëntie. In hoofdstuk 8 lieten we zien dat patiënten met een ernstige hypertensie en graad

Chapter 11

212

III of IV retinopathie een gestoorde CA hebben, er op duidend dat in theorie zelfs bij een geringe verlaging van de bloeddruk cerebrale ischemie zou kunnen ontstaan. Om cerebrale hypo- of hyperperfusie te voorkomen is intraveneuze behandeling te prefereren omdat orale bloeddrukverlagende medicatie een onvoorspelbare of te langzame werking heeft. Hoewel de retinopathie van maligne hypertensie geassocieerd is met een gestoorde CA, is het de vraag of onderzoek van de retina voldoende kan discrimineren tussen een falende en een intacte CA. Bij patiënten met een hypertensieve encephalopathie kan worden verwacht dat de CA gestoord is, echter bij deze patiënten zijn niet altijd retinale afwijkingen aanwezig.10 Andersom hebben niet alle patiënten met bloedingen en exsudaten een falende CA. Bij de groep patiënten uit hoofdstuk 8 en 9 bleek een ernstig gestoorde CA, gedefinieerd als een fasevoorloop < 30%,11 in 10 van de 15 (67%) patiënten met maligne hypertensie aanwezig, maar was de CA normaal (fasevoorloop > 50%) bij 4 van de 15 patiënten. Gegevens over de CA bij patiënten met ernstige hypertensie, maar zonder retinale afwijkingen ontbreken. Daarom moet de negatief voorspellende waarde van deze retina-afwijkingen nog worden vastgesteld. Het zorgvuldig vaststellen van de klinische verschijnselen die passen bij een hypertensieve encephalopathie blijft dus onmisbaar bij patiënten met een (verdenking op) een hypertensieve crise, onafhankelijk van de retinale bevindingen. In toekomstige studies zouden gegevens over de integriteit van de CA in ernstig hypertensieve patiënten zonder retinale afwijkingen verder onderzocht moeten worden om de onderscheidende waarde van de retina-afwijkingen bij maligne hypertensie vast te stellen. De overleving van maligne hypertensie is sterk verbeterd. In de groep patiënten uit hoofdstuk 3 was de overleving meer dan 90% na 4 jaar en werd het over-lijden in de meeste gevallen veroorzaakt door niet-cardiovasculaire oorzaken. Daarentegen had bijna 20% nierfunctievervangende therapie nodig, erop duidend dat nierinsufficiëntie een belangrijke complicatie van maligne hypertensie is. In hoofdstuk 4 lieten we zien dat een TMA vaak bij maligne hypertensie wordt gezien en sterk geassocieerd is met een slechtere nierfunctie bij presentatie, maar dat herstel van nierfunctie tijdens bloeddrukbehandeling kan optreden, ook als bij opname tijdelijk nierfunctievervangende therapie nodig was. Een snelle verlaging van de bloeddruk zou de kans op het herstel van nierfunctie kunnen vergroten, echter een te grote bloeddrukdaling zou de door-bloeding van een al beschadigde nier verder kunnen compromitteren. Een onmiddellijke en gecontroleerde verlaging van de bloeddruk met intraveneuze therapie lijkt dus bij patiënten met een TMA aangewezen. Net als bij de CA, is


213

de voorspellende waarde van de retinale afwijkingen die met maligne hyper-tensie zijn geassocieerd voor het vaststellen van TMA niet bekend. In hoofdstuk 6 en 7 lieten we zien dat bij achtereenvolgende patiënten die werden opgenomen met een hypertensieve crise, patiënten met een graad III of IV retinopathie een ernstigere TMA en nierfunctiestoornissen hadden dan patiënten zonder retinale afwijkingen. Niettemin, hadden enkele patiënten zonder retinopathie aanwij-zingen voor een TMA (hoofdstuk 7). Als men de retinopathie van maligne hypertensie als een diagnostische test voor TMA beschouwt volgens de definitie die wij hanteerden in hoofdstuk 5, is de sensitiviteit 71% voor het vaststellen van een TMA. Ofwel 29% van de patiënten met een TMA had geen graad III of IV retinopathie (aantal foutnegatieven). Hoewel deze gegevens zijn afgeleid uit een relatief klein cohort van 40 patiënten, laat het zien dat het retina-onderzoek onvoldoende kan discrimineren tussen de aan- of afwezigheid van TMA.

Om de identificatie van patiënten met een hypertensief spoedgeval te verbeteren, zou men kunnen overwegen om de huidige definitie te vervangen door een definitie waarin de klinische criteria voor hypertensieve ence-phalopathie en laboratorium criteria voor TMA zijn opgenomen (zie tabel 1). Tabel 1 Voorgestelde definitie van maligne hypertensie. Ernstige hypertensie (Diastolische bloeddruk meestal boven de 120 mmHg) en 1 of meer van de volgende: 1. klinische symptomen passend bij cerebrale complicaties: onverklaarde verlaging van het bewustzijn, delirium, verwardheid, gegeneraliseerde of focale epilepsie of visusdaling. 2. retinale afwijkingen passend bij graad III of IV retinopathie volgens de Keith, Wagener en Barker classificatie. 3. verlaagd aantal bloedplaatjes met ofwel een verhoogd LDH ofwel aanwezigheid van fragmentocyten in een perifere bloeduitstrijk. Behandeling De behandeling van maligne hypertensie is gebaseerd op het idee dat een bloeddrukverlaging van ~25% in minuten tot uren resulteert in een veilig gebied waarin de cerebrale bloedstroom constant kan worden gehouden, de zoge-noemde plateau fase van de CA.4 In hoofdstuk 8 lieten we zien dat zowel de

Chapter 11

214

statische als dynamische CA gestoord waren bij patiënten met maligne hypertensie zonder dat er aanwijzingen waren voor een plateau fase. Dit betekent dat de consensus van een ~25% reductie van de gemiddelde arteriële bloeddruk gebaseerd is op een incorrecte veronderstelling. Eerdere gevals-beschrijvingen hebben laten zien dat snelle bloeddrukverlagingen van meer dan 50% aanleiding kunnen geven tot een ischemisch CVA en overlijden.12-14 In theorie zouden kleinere bloeddrukdalingen kunnen leiden tot een verergering van de cerebrale hyperperfusie en cerebraal oedeem. Een initiële bloeddruk-verlaging van niet meer dan ~25% in de eerste minuten tot uren (afhankelijk van de ernst) wordt daarentegen goed getolereerd en lijkt daarom een terechte aanbeveling. De vraag is voor welke periode een dergelijke reductie van de bloeddruk moet worden gehandhaafd. Dit zou kunnen afhangen van de tijd die nodig is voor de CA om te herstellen. Voorlopige data suggereren dat de initiële afwijkingen in de CA een week na opname weer grotendeels zijn hersteld (ongepubliceerde observaties). Daarom lijkt het huidige advies om de bloeddruk gedurende “enkele dagen” op het initiële niveau van 25% onder de bloeddruk bij presentatie te handhaven een terechte aanbeveling.15 Tenslotte zijn er geen studies die het effect van verschillende intraveneuze bloeddrukverlagende geneesmiddelen op cardiovasculaire eindpunten bij maligne hypertensie hebben onderzocht. Door de lage incidentie van maligne hypertensie is het niet te verwachten dat deze studies in de nabije toekomst zullen verschijnen. We beperkten ons daarom tot vergelijking van het effect van SNP en labetalol op de cerebrale bloedstroom en de systemische hemodynamiek (hoofdstuk 9). Hoewel labetalol en SNP even effectief waren in het verlagen van de bloeddruk, resulteerde behandeling met SNP in een grotere daling van de bloedstroom in de middelste hersenslagader en systemische weerstand, wat suggereert dat bloed van de hersencirculatie wordt weggeleid. Een dergelijke deviatie van de bloedstroom tijdens behandeling met SNP is ook beschreven voor de kransvaten van het hart (hoofdstuk 1) en zou het risico op cerebrale hypoperfusie kunnen vergroten. Het gevonden verschil van 10% tussen SNP en labetalol met betrekking tot de stroomsnelheid in de middelste hersenslagader hoeft geen klinisch belangrijk verschil te zijn, maar zou voldoende kunnen zijn om cerebrale hypoperfusie te veroorzaken in situaties waarbij de cerebrale circulatie gecompromitteerd is, zoals bij ouderen en bij patiënten met athero-sclerose. Het nadelige effect van SNP op de cerebrale bloedstroom boven de reeds bekende nadelen van dit middel maken dat labetalol misschien een beter alternatief is voor de behandeling van maligne hypertensie.


215

Literatuur


2. Giese J. Renin, angiotensin and hypertensive vascular damage: a review. Am J Med. 1973;55:315-332.

3. Beilin LJ, Goldby FS, Mohring J. High arterial pressure versus humoral factors in the pathogenesis of the vascular lesions of malignant hypertension. Clin Sci Mol Med. 1977;52:111-117.


5. Cifkova R, Erdine S, Fagard R, Farsang C, Heagerty AM, Kiowski W, Kjeldsen S, Luscher T, Mallion JM, Mancia G, Poulter N, Rahn KH, Rodicio JL, Ruilope LM, van Zwieten P, Waeber B, Williams B, Zanchetti A. Practice guidelines for primary care physicians: 2003 ESH/ESC hypertension guidelines. J Hypertens. 2003;21: 1779-1786.


7. Zampaglione B, Pascale C, Marchisio M, Cavallo-Perin P. Hypertensive urgencies and emergencies. Prevalence and clinical presentation. Hypertension. 1996;27:144-147.


9. Lip GY, Beevers M, Dodson PM, Beevers DG. Severe hypertension with lone bilateral papilledema: a variant of malignant hypertension. Blood Press. 1995;4: 339-342.


11. Immink RV, van Montfrans GA, Stam J, Karemaker JM, Diamant M, van Lieshout JJ. Dynamic cerebral autoregulation in acute lacunar and middle cerebral artery territory ischemic stroke. Stroke. 2005;36:2595-2600.


Chapter 11

216



15. Calhoun DA, Oparil S. Treatment of hypertensive crises. N Engl J Med. 1990;323: 1177-1183.

218

Dankwoord

Bij de totstandkoming van dit proefschrift hebben velen mij van raad voorzien. Zonder mensen tekort te doen, wil ik enkelen hiervoor in het bijzonder bedanken. Allereerst wil ik mijn co-promotoren bedanken: dr. van Montfrans, beste Gert, zonder jouw steun en linguïstische overredingskracht was niets van dit proefschrift tot stand gekomen. Nu het proefschrift is afgerond kan ik dan ook rustig bekennen dat ik, zeker in de beginfase, er wel eens over heb gedacht mijn aandacht weer te beperken tot de dagelijkse handelingen en vrije avonduren. Dankzij de goede samenwerking en grote (wetenschappelijke) vrijheid heb ik dit proefschrift met veel plezier tot een goed einde kunnen brengen. Prof. dr. Koopmans, beste Richard, hoewel het schrijven voor de rubriek “klinisch denken” met Peter Hart al een leuke tijdsbesteding was, dank ik je voor de veronderstelling dat ik me misschien af en toe verveelde. Je aanstekelijke enthousiasme gaf enorm veel extra energie. Niet alle vragen die je aan het begin van deze onderneming had opgeschreven zijn beantwoord, maar de meerderheid is toch min of meer aan bod gekomen. Professor Levi, beste Marcel, dank dat je mijn promotor hebt willen zijn. Je snelheid, kennis en oog voor nog onopgeloste wetenschappelijke vragen is fantastisch. Professoren Büller en Kastelein, beste Harry en John, jullie diplomatieke vermogens, doortastendheid en enthousiasme maken het enorm prettig om bij jullie te werken. Rogier Immink en dr. J.J. van Lieshout wil ik danken voor hun grote enthousiasme, vaardigheden en kennis over alles wat met het niet-invasief meten van de cerebrale bloedstroom te maken heeft. Rogier je bereidheid om midden in de nacht met een opgewekt humeur deze lastige metingen te doen heeft grote indruk gemaakt. Prof. dr. Uitterlinden, beste André, en Pascal Arp: dank voor jullie bereidheid de verschillende polymorfismen voor ons te bepalen. Dr. J.O. Groeneveld, beste

219

Johan, dank voor je medewerking met het verzamelen van de patiënten data vanuit het OLVG. Dr. J.C.M. Meijers, beste Joost, dank voor je goede adviezen en ideeën met betrekking tot de stollingsbepalingen. Prof. dr. A.H. Zwinderman, beste Coos, dank voor je hulp met de statistiek. Je gaf me altijd het idee zelf met de oplossing te komen, hoewel ik twijfel hoe vaak dat nu werkelijk het geval is geweest. Dr. R.O. Schlingemann, beste Reinier dank voor je hulp bij de oogheelkundige aspecten en goede suggesties. Prof. dr. J.B. Hoekstra, beste Joost dank voor de prettige discussies en prima samenwerking. Uwkje Honnebier, Niels van der Hoeven en Evelyn Groot dank ik voor hun inzet en enthousiasme bij het verzamelen van de data en hulp met de bepalingen. Lieve Caroline en lieve Lucas, zonder jullie hulp was het nooit gelukt. Bij jullie ben ik gelukkig.

uva-dare (digital academic repository) epidemiology and ... · revisited - plasma renin activity,...

Documents