Kidney International, Vol. 43 (1993), PP. 1363—1370

Clinicopathology of kidneys from brain-dead patients treatedwith vasopressin and epinephrine

T0M0FuMI NAGAREDA, Y0sHIHIR0 KIN0sHITA, AKIRA TANAKA, MASASHI TAKEDA,TSUTOMU SAKANO, K0HEI YAWATA, TsuYosHI SuGIM0T0, YASUKO NIsHIzAwA,

and NoBuYuiu TERADA

Department of Laboratory, National South Wakayama Hospital, Wakayama; Departments of Traumatology and Pathology, Osaka UniversityMedical School, The Center For Adult Diseases, and Department of Traumatology, Osaka Preftctural Hospital, Osaka, Japan

Clinicopathology of kidneys from brain-dead patients treated withvasopressin and epinephrine. Studies were made on the biochemical andpathological conditions of kidneys of 20 brain-dead patients who weremaintained for 0 to 48 days after brain death by administration ofvasopressin and epinephrine. Twenty specimens were obtained bypercutaneous biopsy or at autopsy. The biochemical and pathologicaldegrees were compared with those on the day of brain death (day 0).Biochemical tests on day 0 indicated that they showed the diureticphase of prerenal failure, and then glomerular hyperemia was exten-sive. Renal function recovered on day 1 and remained almost normalduring the 14 day period. Their urine retained high levels of sodium andosmolarity for days 0 to 14, with mild hyponatremia and hypo-osmolarity of the plasma. Tubulointerstitial nephritis gradually becameextensive. There was no significant change in the degrees of mesangialwidening, mesangial cell proliferation or hyalinosis. Arterial intimalproliferation was gradually extensive after day 3 and glomerular endo-thelial proliferation was gradually extensive after a week. Brain-deadpatients have been mostly reported to develop diabetes insipidus, butour brain-dead patients did not show any manifestation of this disease.We suggest that constant natriurisis and continuing high level of urineosmolarity might have been caused by prerenal renal failure, braindeath followed by neurogenic impairment, high level of serum vaso-pressin, or interstitial nephritis.

Brain-dead patients who have cerebral lesions, such as intra-cranial hemorrhage, neoplasia or infection, usually developcardiac arrest within a few days. Recently, the prolongedmaintenance of these patients has been achieved by continuousinfusion of vasopressin with epinephrine [1]. We have reportedclinicopathological studies on the livers of brain-dead patientsafter prolonged maintenance [2]. We observed that duringprolonged hemodynamic maintenance of brain-dead patients,pathological lesions did not spread or diminished, and thatbiochemical indices did not become worse, or improved, in thefirst two weeks, except for increases in cholangitis and theserum alkaline phosphatase level. These disorders may havebeen caused by functional cholestasis due to dysfunction of the

Received for publication September 24, 1992and in revised form January 4, 1993Accepted for publication January 8, 1993

© 1993 by the International Society of Nephrology

vagus after brain death. Here we made clinicopathologicalstudy of kidneys from brain-dead patients.

Brain-dead patients have been reported by Howlett et altoprimarily develop clinical diabetes insipidus a few days afterbrain death [3]; this disease has manifestations of polyuria,urinal hypo-osmolarity, and normal sodium excretion [4]. Pa-tients with intracranial disease also are reported to frequentlyshow the syndrome of inappropriate secretion of antidiuretichormone (SIADH) [5—71. In this syndrome, continued secretionof antidiuretic hormone results in natriuresis, low level of serumosmolarity and dilutional hyponatremia [8—10]. Cerebral dis-eases may also result in cerebral salt-losing syndrome [11—161.Most features of this syndrome are same as those of SIADH,but this syndrome is usually accompanied with extracellularvolume depletion which is not usually found in SIADH [17, 18].Originally, the primary defect of this syndrome is thought to bethe inability of the kidney to conserve sodium [17, 18].

Clinical renal data in brain-dead patients has been reportedby Kinoshita et al [19]. However, pathological renal data inbrain-dead patients have not been reported. We report investi-gations on the clinicopathological changes of the kidneys ofbrain-dead patients during prolonged hemodynamic mainte-nance by administration of vasopressin and epinephrine.

The brain-dead patients were not kept alive simply for thepurpose of the study. In Japan, death is legally pronounced atthe point of cardiac arrest. Brain death per se is not a legaldeterminant. Therefore, physicians are obligated to sustain lifedespite brain death until cardiac arrest occurs. Accordingly,these studies were conducted in acceptable ethical principles ofJapanese law.

Methods

Patients

Brain death was defined according to the criteria of the BrainDeath Study Group sponsored by the Ministry of Health andWelfare, which requires all of the following criteria: deep coma,apnea, lack of a brain-stem reflex, and a silent electroenceph-alogram for over six hours. Details of the 20 brain-dead patientsexamined in this study are given in Table 1. These 20 brain-deadpatients ranged from 22 to 70 years old. Their primary diseaseswere as follows: head injury in 12, subarachnoidal hemorrhage

1363

1364 Nagareda et at: Kidneys from brain-dead patients

Table 1. Data on brain-dead patients and times when kidney specimens were obtained

Patient Age Sex

Day ofcardiacarrest Cause of brain death

Kidneyspecimen

hoursa

1 22 Male 0 Malignant lymphoma Autopsy (1)2 46 Female 0 Brain tumor Autopsy (1)3 47 Male 0 Brain tumor Autopsy (1.5)4 37 Female 0 Subarachnoidal hemorrhage Autopsy (10)5 44 Male 0 Subarachnoidal hemorrhage Autopsy (7)6 46 Male 3 Thoracoabdominal injury Autopsy (15)7 58 Male 3 Head injury Autopsy (1.5)8 33 Male 4 Head injury Necropsy9 45 Male 7 Head injury Autopsy (4)

10 37 Male 9 Head injury Necropsy11 50 Male 9 Head injury Necropsy12 42 Male 10 Head injury Autopsy (11)13 43 Female 12 Head injury Autopsy (2.5)14 58 Female 12 Subarachnoidal hemorrhage Autopsy (14)15 61 Female 13 Head injury Autopsy (14)16 45 Male 15 Cyanide poisoning Autopsy (4)17 29 Male 15 Head injury Autopsy (2)18 70 Male 16 Head injury Autopsy (5.5)19 43 Female 40 Head injury Autopsy (7)20 22 Male 48 Head injury Autopsy (2)

Patients 1-5 received no mechanical ventilation or infusion of vasopressin. Patients 6 to 20 received vasopressin with epinephrine, and alsorespiratory and fluid management. The day of brain death was defined as day 0.

a Hours between cardiac arrest and autopsy.

in three, brain tumor in two, malignant lymphoma in one,cyanide poisoning in one, and thoracoabdominal injury in one.Fifteen of the 20 brain-dead patients received vasopressin at aconstant rate of one or two units/hr (285 45 tU/kg/min) withepinephrine (total amount: 34 to 984 mg) to maintain a systolicblood pressure of more than 90 mm Hg. They also receivedrespiratory control and fluid management, with 2400 to 4800 mlof fluid per day. The serum anti-diuretic hormone level in thepatients receiving vasopressin was 10 to 40 pg/mI (normalrange, 0.3 to 4.2 pg/mi). These procedures were carried out atthe Department of Traumatology, Osaka University Hospital.The direct causes of cardiac arrest of the patients were heartfailure in two, pulmonary failure in two and renal failure in one,and were unknown in 10. All 15 patients received antiulcerativedrugs, antibiotics, and various electrolyte solutions. They hadnot received any irradiation therapy and had no episode ofimmunological disorder. Of the other five patients, two wereexamined at autopsy at the Tokyo Medical Examiner's Office,two at Osaka Prefectural Hospital, and one at Osaka UniversityMedical School. These patients had not received mechanicalventilation or infusion of vasopressin.

Biochemical tests on kidney function

We measured the urine volume, creatinine clearance, serumcreatinine, blood urea nitrogen, serum potassium, sodium in-take, urinary sodium output, plasma osmolarity and urinaryosmolarity. The urine volume and creatinine clearance weremeasured by collecting urine via a urethral catheter for 24hours. Serum creatinine, blood urea nitrogen, serum potassiumand serum sodium were examined by standard methods with anautomated chemical analyzer. Sodium intake was calculatedbased on the total volume of electrolyte solution given intrave-nously, and urinary sodium output was calculated from the

urinary sodium level and urine volume. The osmolarities ofplasma and urine were measured with an osmometer.

The normal ranges of these parameters are as follows: urinevolume 600 to 1600 ml/day; creatinine clearance 60 to 145mI/mm; serum creatinine level 0.9 to 1.6 mg/dl; blood ureanitrogen level 9 to 20 mgldl; serum potassium level 3.5 to 5.0mEq/liter; urinary sodium output 40 to 185 mEq/day; serumsodium level 134 to 144 mEq/liter; and plasma osmolarity 285 to295 mOsm/liter.

Pathological examinationSeventeen kidney specimens were obtained at autopsy and

three by percutaneous biopsy with a 15-gauge needle. Thespecimens were fixed in 10% buffered formalin (pH 7.2), andprocessed routinely for embedding in paraffin. Sections werecut at 3 to 5 jm thickness, and stained with hematoxylin andeosin, periodic acid Shiff reaction (PAS), periodic acid-methe-namine silver stain (PAM), and Azan-Mallory stain.

The degrees of pathological lesions of the tubular lesions(vacuolar degeneration, acidophilic degeneration, atrophy andnecrosis of the lining cells of the proximal or distal tubules),interstitial scar, arterial intimal proliferation, and glomerularlesions (glomerular hyperemia, glomerulitis, intracapillary en-dothelial proliferation, mesangial widening, mesangial cell pro-liferation, hyalinosis, periglomerular fibrosis, and periglomeru-litis) were evaluated. Tubular vacuolar degeneration wasdefined as vacuolated or vesicular degeneration of the tubularlining cells in the cortex. Tubular acidophilic degeneration wasdefined as the presence of tubular lining cells containing acido-philic granules. Tubular atrophy was defined as a dilatedtubulus or flattened tubular epithelium. Tubular necrosis wasdefined as pyknosis, karyolysis or karyorrhexis of tubular liningcells. Distal tubules consisted of distal convoluted tubules and

Nagareda et al: Kidneys from brain-dead patients 1365

the cortical collecting duct. Interstitial scar was defined as afibrous lesion containing necrosed or destroyed tubules in thecortex detected in sections stained by the Azan-Mallorymethod. Arterial intimal proliferation was defined as intimaledematous cellular thickening detected in sections stained bythe Azan-Mallory method. Glomerular hyperemia was definedas the presence of segmental or global capillary loops com-pacted with erythrocytes. Glomerulitis was defined as thepresence of five or more white blood cells such as lymphocytes,neutrophils or monocytes in a glomerulus. Endothelial prolifer-ation was defined as the presence of two or more endothelialcells in more than one capillary loop of a glomerulus. Mesangialcell proliferation was defined as the presence of four moremesangial cells in a mesangium of a glomerulus [20]. Mesangialwidening and hyalinosis were evaluated by examining sectionsstained with PAS, and periglomerular fibrosis was evaluated byexamining sections stained with Azan-Mallory stain. Penglom-erulitis was defined as periglomerular lymphocytic infiltration.

The degrees of tubular lesions were evaluated as the numbersof involved fields among 100 randomly selected high powerfields (x 400 fields). Interstitial scars were evaluated as meannumbers of lesions in 100 glomerulo-cortical areas. The degreeof arterial intimal proliferation was evaluated as the percentageof arteries involved; totals of 100 to 140 (mean SE; 114 3)arteries in autopsy specimens and 11 to 29 (mean SE; 17 6)arteries in necropsy specimens were examined. The degrees ofglomerular lesions were evaluated from the percentages ofglomeruli involved; totals of 104 to 701 (mean SE: 327 22)glomeruli in autopsy specimens and 21 to 64 (mean SE: 336) glomeruli in necropsy specimens were examined.

Statistical evaluation

The significances of difference were evaluated by Student'st-test.

Results

Bioche,nical findings

The time courses of changes in the mean urine volume,creatinine clearance, serum creatinine, blood urea nitrogen,sodium intake, urinary sodium output, serum sodium, plasmaosmolarity and urinary osmolarity are shown in Figures 1 and 2.The urine volume was markedly above the upper normal rangeon days 0 to 14. The mean value for creatinine clearance wasbelow the normal range on day 0, but improved, being withinthe normal range on days 1 to 14. The mean values of serumcreatinine, blood urea nitrogen and serum potassium werewithin almost normal ranges on days 0 to 14. However, themean value of serum potassium on day 1 decreased to below thenormal range and then increased slightly.

The mean urinary sodium output was markedly increasedthroughout the 14-day period. Sodium intake significantly ex-ceeded urinary sodium output on days 0 and 1, and then wasalmost equal with the urinary sodium output. The mean valuesof serum sodium and plasma osmolarity were above the normalrange on days 0 and 1, but then decreased gradually. The meanurinary osmolarity was 523 to 622 mOsm/liter, and was abovethe mean plasma osmolarity (270 to 316 mOsm/liter) throughoutthe 14-day period.

3,000

1,500

110

90

60

40

11 11 12 8

0

Serum potassium*

5

4

wE 3

122 •

01 3 5 7 10 14

Time, days after brain death

Fig. 1. Time courses of change in urine volume, creatinine clearance,serum creatinine, blood urea nitrogen, and serum potassium of 15brain-dead patients. All the patients received vasopressin at a constantrate of 1 or 2 U/hr in conjunction with epinephrine to maintain theirsystolic blood pressure at above 90 mm Hg. They also receivedrespiratory control and fluid management (2400—4800 ml of fluid/day).The normal ranges of urine volume, creatinine clearance, serum creat-mine, blood urea nitrogen and serum potassium are 600—1600 ml/day,60—145 mI/mm, 0.9—1.6 mg/dl, 9—20 mgldl and 3.5—5.0 mEq/liter, respec-tively. Points are means SE of values for the numbers of patientsshown below the points. * P < 0.05, vs. value on day 0. § mean valueabove or below the normal range.

Pathological findingsKidney specimens were divided into five groups depending

on the number of days after brain death on which they wereobtained: day 0, days 3 to 6, days 7 to 10, days 11 to 14, and

Urine volume

§§

§>,

E

EE

E

E

Serum creatinine

Blood urea nitrogen40

30

20

10

1366 Nagareda et a!: Kidneys from brain-dead patients

Time, days after brain death

Fig. 2. Time courses of change in sodium intake, urinary sodiumoutput, serum sodium level, plasma osmolarity and urinary osmolarityof 15 brain-dead patients. Explanations of management of the patientsare as for Figure 1. The normal ranges of urinary sodium output, serumsodium and plasma osmolanty are 40—186 mEq/day, 134—144 mEq/literand 285—295 mOsm/liter, respectively. Points are means SE of valuesfor the numbers of patients shown below the points. * P < 0.05, vs.value on day 0. § mean value above or below the normal range. p <0.05, vs. value of urinary sodium output.

days 15 to 48. The degrees of pathological lesions in thesegroups were examined (Tables 2 and 3).

Degeneration of tubular lining cells, such as vacuolar degen-eration, acidophilic degeneration, atrophy and necrosis, waslimited on day 0 (Fig. 3A), but increased on days 3 to 48. These

degenerative changes were more extensive in the distal tubulithan in the proximal tubuli on days 0 to 48. Arterial intimalproliferation was found in 7% of the arteries on day 0, but itincreased markedly on days 3 to 48, spreading to 42 to 85% ofthe arteries.

Glomerular hyperemia was found in 92% of the glomeruli onday 0 (Fig. 3B), but its incidence decreased gradually after day3. This lesion was found in less than 10% of the glomeruli ondays 7 to 14 (Fig. 4). The degrees of glomerulitis, endothelialproliferation and periglomerulitis did not change until day 6, butincreased slightly on days 7 to 48. There were no significantchanges in the degrees of mesangial widening, mesangial cellproliferation, hyalinosis or interstitial scar in the 48-day period.Periglomerular fibrosis and periglomerulitis were found in 16%and 4%, respectively, of the glomeruli on day 0 (Fig. 3B).However, these lesions increased gradually on days 3 to 14(Fig. 4) and markedly on days 15 to 48, spreading to 92% and65%, respectively, of the glomeruli (Fig. 5). There was nopurulent lesion of glomeruli or tubuli, which is usually found inacute pyelonephritis.

Discussion

Brain-dead patients mostly develop clinical diabetes inspidusafter brain death [3j. In our brain-dead patients, urine volumeincreased and creatinine clearance decreased on day 0. Thelevel of serum potassium fell on day 1. Creatinine clearanceimproved on day 1 and remained within normal range for daysIto 14. The levels of their serum sodium and plasma osmolaritywere above the normal ranges on days 0 and 1 when sodiumintake was markedly more than its output. These levels de-creased after day 3, when the level of urinary sodium outputwas almost the same as that of sodium intake. The levels ofurinary sodium output and urinary osmolarity remained at highlevels for days 0 to 14. These abnormalities are not symptoms ofdiabetes insipidus [4]. The conditions of our patients are foundin obstructive nephropathy [211, adrenal insufficiency, renalfailure, SIADH, and the cerebral salt-losing syndrome [11—18].Our patients showed no episode of urinary obstruction. Theyprobably had normal adrenal function, because the hypothala-mus and pituitary of the brain-dead patients are reported tomaintain some functional activity, the level of cortisol beingwithin the normal range [22, 23]. We suspect that four processes

14 might have contributed to these conditions of our patients.The first process might have been the diuretic phase of acute

prerenal renal failure [241. The large dose of sodium on day 0may have been helpful to show a rapid recovery of renalfunction, since increased ingestion of sodium chloride is re-ported to protect against the impairment of renal function afterrenal failure [25]. In the pathological examination, glomerularhyperemia was extensive at day 0, but decreased markedlythereafter. Degenerations were more extensive in the distaltubules than in the proximal tubules from day 0. Glomerularhyperemia is usually observed in circulatory disturbances suchas heart insufficiency, infarct and renal vein thrombus [26].Epithelial cell necrosis of distal tubule is commonly found inprerenal renal failure [24]. At the brain death (day 0), whencirculatory disturbance may occur, the blood pressure dropsrapidly, causing circulatory disturbances. Our patients afterbrain death received infusion of vasopressin, infusion of fluid,and respiratory control to maintain their blood pressure. Their

1050 — Sodium intake and urine sodium output

900

>a)

600wE

300

150

Serum sodium155

a)

140-wE

125

325

300

275

250

850

ci)

(I)0E

250 I I - I0 1 3 5 7 10

Nagareda et a!: Kidneys from brain-dead patients 1367

Table 2. Degrees of tubular, interstitial and arterial lesions in the kidneys of brain-dead patientsa

Degree of pathological lesions %b

Day 0 Days 3—6 Days 7—10 Days 11—14 Days 15—48Pathological lesions (N = 5) (N = 3) (N = 4) (N = 4) (N = 4)

Proximal tubuli; vacuolation 3 1 51 1" 47 15 66 170 61 200

acidophilic degeneration 1 0 11 20 10 10 15 5 23 2C

atrophy 0 0 5 20 4 1 1 0" 1 1

necrosis 4±2 5±5 14±8 16±5 28±50Distal tubuli; vacuolation 8 2 57 60 34 9 83 90 67 15"

acidophilic degeneration 6 2 40 8"" 45 9"" 42 12 35 80

atrophy 0 0 29 le" 32 12 49 100(1 35 30necrosis 17 4" 20 3 24 7 64 7"" 70 7"'

Interstitial scar 0 0 1 1 7 2 3 1 3 2

Arterial intimal proliferation 7 2 76 80 42 80 64 9" 85 50

a The kidney specimens were divided into 5 groups. The degrees of degeneration of the lining cells of the tubuli were evaluated as percentagesof lesions in the high-power fields examined. The degree of interstitial scar was evaluated as the mean number of lesions in 100 glomerulo-corticalareas examined. The degree of arterial intimal proliferation were evaluated as the percentage of involved arteries.

b Data are mean SEP< 0.05 vs. value on day 0 by Student's t-test

"p < 0.05 vs. the value for the same lesion in proximal tubuli

Table 3. Degrees of pathological glomerular lesions in the kidneys of brain-dead patientsa

Degree of pathological lesions (%)"

Day 0 Days 3—6 Days 7—10 Days 11—14 Days 15—48Pathological lesion (N = 5) (N = 3) (N = 4) (N = 4) (N = 4)

Glomerular hyperemia 92 4 18 5" 8 2C 7 3" 32 80

Glomerulitis 3 1 6 2 12 IC 13 20 41 13

Endothelial proliferation 14 1 38 6 43 60 49 6C 67 30

Mesangial widening 7 2 3 0 11 3 8 2 13 4

Mesangial cell proliferation 1 0 3 2 3 1 3 1 3 1

Hyalinosis 0 0 1 1 4 2 1 0 4 4

Periglomerular fibrosis 16 4 43 5" 67 70 76 60 92 40

Periglomerulitis 4 2 6 1 27 60 35 70 65 I 1C

a The kidney specimens were divided into 5 groups. The degrees of glomerular hyperemia, glomerulitis, endothelial proliferation, mesangial

widening, mesangial cell proliferation, hyalinosis, penglomerular fibrosis, and periglomerulitis were evaluated as percentages of involved

glomeruli.b Data are means SEC P < 0.05 vs. value for group on day 0 by Student's t-test

rapid recovery may be caused by maintenance of the bloodpressure, and may also have been caused by a large dose ofsodium.

The second process might have been neurogenic impairment.Brain death is defined as irreversible cessation of activity of thebrain, including the brain stem [27]. In this state the sympa-thetic nervous system is dysfunctional since noradrenergicneurons are restricted to the pontine and medullary segmentalregions, ramifying projections to the brain and spinal cord [28].The renal cortex is under the control of its sympathetic fibers[29], and renal sodium excretion is controled via sympatheticnerves [30—34]. The renal sympathetic nerves are important forrapid control of sodium excretion when sodium intake is altered[321. Denervation of the peripheral renal sympathetic nerves isreported to cause increase of urine flow and sodium excretion,despite normal renal function [30, 34], and a central alpha-2-adrenergic receptor antagonist is also reported to inhibit renalsympathetic nerve activity and increase urinary sodium excre-tion [34]. In our patients, although the sodium intake was in alarge dose on days 0 and 1, their urinary sodium output was notreflected by the intake. This feature is also shown in the

"cerebral salt-losing syndrome" [14]. Thus, dysfunction of thecentral sympathetic nervous system may have been one of thecauses of losing control of the serum sodium level of ourpatients.

The third process might have been the high level of the serumvasopressin. After brain death, our patients received vasopres-sin continuously, and its level was maintained at above thenormal range. The high level of serum vasopressin causesSIADH in which hyponatremia and hypo-osmolarity of theserum, with normal renal function, are shown [8—10]. Thus,continuous infusion of vasopressin may have caused gradualhyponatremia and lower osmolarity of the plasma of our pa-tients.

The fourth process might have been the pathological condi-tion of the kidney. The major pathological changes of thekidneys of our patients were changes of tubules, predominantlydistal tubules, soon after brain death. Then interstitial fibrosisgradually became extensive, and two weeks after brain death,extensive lymphocytic interstitial inflammation was also ob-served. However, they scarcely had hyalinosis or interstitial

1368 Nagareda et a!: Kidneys from brain-dead patients

FIg. 3. Histological features of tubuli andglomeruli of the kidney from brain-deadpatients on day 0. (A) Acidophiicdegeneration and pyknotic nuclei wereobserved focally in the lining cells of distaltubuli in the kidney of Patient 2 in Table 1 onthe day of brain death (day 0). (H&E; x400),(B) Glomerular hyperemia, observed globallyin the kidney of Patient 1 in Table I on theday of brain death (day 0). (H&E; x200).

scars, which are usually found to increase in chronic pyelone-phritis [35]. These findings conform well with those in tubu-lointerstitial nephritis [35, 36]. Glomerulitis, which is oftenfound in tubulointerstitial nephritis [35], was extensive after day15. There are reports that the kidneys show the pathologicalfeatures of interstitial nephritis in patients with hyponatremiaand salt loss [37, 38]. Mild hyponatremia and the constantlyhigh level of urinary sodium output may have been concernedwith tubulointerstitial nephritis. The kidneys in the diureticphase of acute renal failure may show tubulointestitial nephritis[24]. Our patients showed this phase. Other causes may havebeen also thought to be hypokalemia, irradiation, immunologicreaction, drugs [36, 39, 40], and ischemia [41]. Our patients hadnot received irradiation. They had not had the feature of

hypokalemia after day 2 nor immunologic reaction. They re-ceived various antiulcerative drugs (cimetidine, ranitidine hy-drochloride and secretin) and various antibiotics (cefametasolesodium, ribostamycin sulphate, ampidillin, piperacillin sodium,cefalotin sodium, fosfomycin, latamoxef sodium, sulbeniciffinsodium and amikacin sulphate) until cardiac arrest after braindeath. Except for secretin these drugs have possibility of beingnephrotoxic [42]. Our patients had received mechanical venti-lation after brain death, so that they might have been inischemia. Thus the administration of these various drugs orischemia, as well as the diuretic phase of acute renal failure,might have caused tubulointerstitial nephritis in the presentbrain-dead patients.

Arterial intimal proliferation was observed extensive after

"4

C a

A

we

St5

•

3#1)

V

Vt.

Nagareda et a!: Kidneys from brain-dead patients 1369

Fig. 4. Histological features of theglomerulus of the kidney from Patient 13 inTable 1 on day 12 after brain death. There isno evidence of glomerular hyperemia, butmild periglomerular fibrosis is seen. (H&E;x 200).

Fig. 5. Histological features of theperiglomerular interstinum of the kidney ofPatient 18 on day 16 after brain death.Lymphocytic infiltration in the periglomerularand peritubular interstinum, periglomerularfibrosis, and necrosis of the lining cells oftubuli are marked. (Azan-Mallory stain;xlOO).

day 3. This intimal proliferation is found in the chronic stage ofinterstitial nephritis as adaptive hypertensive vasculopathy[361, but the present patients were not hypertensive. However,repeated administration of epinephrine causes arterial intimalproliferation [43], and as our patients were given epinephrinerepeatedly to increase their blood pressure, this may havecaused extensive arterial intimal proliferation.

Glomerular endothelial proliferation of our patients increasedgradually after day 7. The glomerular endothelial proliferation isthought to be observed in the kidney with overload [44]. Ourpatients received large infusions and their urine volume wererelatively great. Thus they may have been in the state of fluidoverload.

Our brain-dead patients, after rapid recovery from prerenalfailure at brain death, kept a high level of the sodium output andurinary osmolarity, with mild hyponatremia and hypo-osmolar-ity of plasma. We suggest that their conditions might have beencaused by prerenal renal failure, brain death in itself, a highlevel of serum vasopressin or interstitial nephritis.

Acknowledgments

This work was supported by a grant from the Osaka Kidney Foun-dation (OKF9O-0018). We thank Dr. Kiichiro Itoh (Osaka PrefecturalHospital), Dr. Etsuko Iwasaki and Dr. Taeko Nagareda for advice. Weare also grateful to Dr. Miho Masui and Choei Wakasugi (Osaka

1370 Nagareda et a!: Kidneys from brain-dead patients

University Medical School), and to Dr. Masahiko Fujimoto, Shin-zaburo Noguchi and Tetsuya Taguchi (Osaka University MedicalSchool), Dr. Kiyoshi Kotoh (Osaka Prefectural Hospital) and Dr.Yoshihiro Matsuo (Tokyo Examiner's Hospital) for providing autopsyspecimens.

Reprint requests to Dr. Tomofumi Nagareda, Department of Labo-ratory, National South Wakayama Hospital, Takinai 2974, Shinjo-cho,Tanabe, Wakayama, 646, Japan.

References

1. Yosi-iiosc& T, SUGIMOTO H, UENIsHI M, SAKAMOTO T, SAD-AMITSu D, SAKANO T, SuGIM0T0 T: Prolonged hemodynamicmaintenance by the combined administration of vasopressin andepinephrine in brain death: Clinical study. Neurosurgery 18:565—567, 1986

2. NAGAREDA T, KINOSHITA Y, TANAKA A, HASUIKE Y, TERADA N,NISHIZAWA Y, FUJITA M, KURODA H, YAWATA K, AOZASA K,SAKANO T, SUGIMOTO T, KOTOH K: Clinicopathological study oflivers from brain-dead patients treated with a combination ofvasopressin and epinephrine. Transplantation 47:792—797, 1989

3. HOWLETT TA, KEOGH AM, PERRY L, TOUZEL R, REES LH:Anterior and posterior pituitary function in brain-stem-dead do-nors. A possible role for hormonal replacement therapy. Transplan-tation 47:828—834, 1989

4. LABHART A, THo1 GW: Clinical endocrinology, in The Hypo-thalamo-neurophyseal System, edited by LARHART A, LANDOLTAM, Berlin, Springer-Verlag, 1986, pp. 54—66

5. Doczi T, BENDE J, HUSZKA E, Kiss J: Syndrome of inappropriatesecretion of antidiuretic hormone after subarachnoid hemorrhage.Neurosurgery 9:394—397, 1981

6. JOYNTRI, AFIFI A, HARRISON J: Hyponatremia in subarachnoidhemorrhage. Arch Neurol 13:633—638, 1965

7. Wiwicis EFM, VERMEULEN M, HuDRA A, VAN GUN J: Hy-ponatremia and cerebral infarction in patients with ruptured intra-cranial aneurysms: Is fluid restriction harmful? Ann Neurol 17:137—140, 1985

8. JAENIKE J, WATERHOUSE C: The renal response to sustainedadministration of vasopressin and water in man. J Gun Endocrinol

21:231—242, 19619. BARTTER FC, SCHWARTZ WB: The syndrome of inappropriate

secretion of antidiuretic hormone. Am J Med 42:790—806, 196710. MENDOZA SA: Syndrome of inappropriate antidiuretic hormone

secretion (SIADH). Pediatr Gun N Am 23:681—690, 197611. WELT LG, SELDIN DW, NELSON WP, GERMAN WJ, PETERS JP:

Role of the central nervous system in metabolism of electrolytesand water. Arch Intern Med 90:355—378, 1952

12. EPSTEIN FH, LEVITIN H, GLASER G, LAVIETES P: Cerebral hy-ponatremia. N Engi J Med 265:513—518, 1961

13. CORT JH: Cerebral salt wasting. Lancet i:752—754, 195414. O5TER JR. PEREZ GO, LARIOS 0, EMERY WE, BOURGOIGNIE JJ:

Cerebral salt wasting in a man with carcinomatous meningitis. ArchIntern Med 143:2187—2188, 1983

15. AL-MUFTI H, ARIEFF Al: Hyponatremia due to cerebral salt-wasting syndrome. Combined cerebral and distal tubular lesion.Am J Med 77:740—746, 1984

16. LABHART A, THORN GW: Clinical endocrinology, in The Hypo-thalamo-neurophyseal System, edited by LABHART A, LANDOLTAM, Berlin, Springer-Verlag, 1986, pp. 66—68

17. NELSON PB, SEll' SM, MAROON JC, ROBINSON AG: Hyponatremiain intracranial disease: Perhaps not the syndrome of inappropriatesecretion of antidiuretic hormone (SIADH). J Neurosurg 55:938—

941, 198118. WUDICKS EFM, VERMEULEN M, TEN HAAF JA, HUDRA A, BAR-

KER WH, VAN GUN J: Volume depletion and natriuresis in patientswith a ruptured intracranial aneurysm. Ann Neurol 18:211—216,1985

19. KINOSHITA Y, YAHATA K, Y0sHI0KA T, ONISHI 5, SUGIMOTO T:Long-term renal preservation after brain death maintained withvasopressin and epinephrine. Transplant mt 3:15—18, 1990

20. SAKAGUCHI H, KITAMOTO K: Pathology of renal biopsy: Atlas of

renal disease, in Proliferative Glomerulonephritis, Tokyo, Shindanto Chiryo Corp., 1992, pp. 37—39

21. ABU-ASHA H, IBRAHIM A, SALIH AM, BELEIL 0: Sodium-losingnephropathy. J Urol 121:8—9, 1979

22. SCHRADER H, KROGNESS K, AAKVAAG A, SORTLAND 0, PtJRVISK: Changes of pituitary hormones in brain death. Acta Neurochir52:239—248, 1980

23. HALL GM, MASHITER K, LUMLEY J, ROBSON JG: Hypothalamic-pituitary function in the "Brain-Dead" patient. Lancet ii: 1259, 1980

24. BRENNER BM, RECTOR FC: The kidney, in Acute Renal Failure,

edited by LEVINSKY NG, ALEXANDER ED, Philadelphia, W.B.Saunders, 1976, pp. 806—825

25. RAMMER L, GERDIN B: Protection against the impairment of renalfunction after intravascular coagulation in the rat kidney by in-creased ingestion of sodium chloride. Nephron 14:433—441, 1975

26. ZOLLINGER HU, MIHATSH MJ: Renal pathology in biopsy: His-topathology of the glomerulus under high power magnification,Berlin, Springer-Verlag, 1978, p. 58

27. REPORT OF THE MEDICAL CONSULTANTS ON THE DIAGNOSIS OFDEATH TO THE PRESIDENT'S COMMISSION FOR STUDY OF ETHICALPROBLEMS IN MEDICINE AND BIOCHEMICAL AND BEHAVIORALRESEARCH: Guidelines for determination of death. JAMA 246:2184,

198128. NIEUWENHUYS R, V000D J, HUUZEN C: The Human Central

Nervous System: Monoamine Neuron Systems. Berlin, Springer-Verlag, 1979, pp. 217—226

29. BARAJAS L: Innervation of the renal cortex, Fed Proc 37:1192—1201, 1978

30. CICCONE CD, ZAMBRASKI EJ: Effects of acute renal denervation onkidney function in deoxycorticosterone acetate-hypertensiveswine. Hypertension 8:925—931, 1986

31. BEERS ET, CARROLL RG, YOUNG DB, GUYTON AC: Effects ofgraded changes in reflex renal activity on renal function. Am J

Physiol 250:F559—F565, 198632. OSBORN JL: Relation between sodium intake, renal function, and

the regulation of arterial pressure. Hypertension 17(Suppl I):I-9l—1-96, 1991

33. DIBONA GF: Neural control of renal function: Cardiovascularimplications. Hypertension 13:539—548, 1989

34. KOEPKE JP, JONES S, DIBONA GF: Sodium responsiveness ofcentral alpha-2-adrenergic receptors in spontaneously hypertensiverats. Hypertension 11:326-333, 1988

35. ZOLLINGER HU, MIHATSH MJ: Renal pathology in biopsy: 20, inInflammatory Interstitial Renal Lesions, Berlin, Springer-Verlag,1978, pp. 407—442

36. ROBBINS SL, KUMAR V: Basic pathology, in The Kidney and ItsCollecting System (4th ed), edited by COTRAN RS, Philadelphia,W.B. Saunders, 1987, pp. 457—491

37. Uiuw.RRi J, OH MS. CARROLL HJ: Salt-losing nephropathy. Clin-ical presentation and mechanisms. Am J Nephrol 3:193—198, 1983

38. ENTICKNAP JB: The condition of the kidneys in salt-losing nephri-tis. Lancet ii:458—461, 1952

39. COGAN MC, ARIEFF A!: Sodium wasting, acidosis and hyperkale-mia induced by methicillin interstitial nephritis. Evidence for selec-tive distal tubular dysfunction. Am J Med 64:500—507, 1978

40. BLACKSHEAR JL, NAPIER JS, DAVIDMAN M, STILLMAN MT: Renalcomplications of nonsteroidal antiinflammatory drugs: Identifica-tion and monitoring of those at risk. Semin Arthr Rheum 14:163—175, 1985

41. ROBBINs SL, KUMAR V: Basic pathology, in Cell Injury andAdaptation (4th ed), edited by COTRAN RS, Philadelphia, W.B.Saunders, 1987, pp. 3—27

42. JAPAN PHARMACEUTICAL INFORMATION CENTER: Drugs in Japan:

Ethical Drugs. Tokyo, Yakugyo Jiho Co., LTD, 1992, pp. 47—48,95—96, 509—510, 555—556, 563—564, 578, 606—607, 925—926, 1160—1161, 1282—1283, 1284—1285, 1302—1303

43. FRIEDMAN B, OESTER YT, DAVIS OF: The effect of arterenol andepinephrine on experimental arteriopathy. Arch lnt Pharmacodyn

1-2:226—234, 195544. ZOLLINGER HU, MIHATSH MJ: Renal Pathology in Biopsy: 6, in

Histopathology of the Glomerulus Under High Power Magnfica-tion, Berlin, Springer-Verlag, 1978, pp. 54—115