A controversial renovascular case.JP Traynor, J Pryce, M Hand and D Kingsmore

A 62-year old female patient with documented history of atheromatous renovasular disease requiring bilateral renal artery stents developed acute renal failure and pulmonary oedema after a 1 week history of nausea, vomiting and mild fever. She was admitted directly to ITU for ventilation and dialysis. Her past medical history included bilateral renal artery stenting performed in 2005, a possible sub-endothelial MI in 2000, hypertension and central obesity. Prior to this admission, her baseline creatinine was 118 umol/L and she was 2 anti-hypertensives although BP control was sub-optimal.

After initial investigations including Doppler ultrasound proved to be either negative or unhelpful, formal renal angiography was performed (18 days after she presented). This revealed a patent right renal artery supplying a small kidney, and an occluded renal artery stent on the left supplying a larger (10.4cm) kidney. The cause of the occlusion was not clear but was felt to be either neo-intimal hyperplasia or in-situ thrombosis. The renal artery lesion was successfully angiolastied and 2 further stents placed within the original stent. She started passing large volumes of urine almost immediately, and other than one dialysis session immediately after the procedure to minimise contrast nephropathy, required no further dialysis. She was started on a statin and warfarin and was able to be discharged shortly after. At follow up 1 month later, her serum creatinine had fallen to 85 umol/L with a BP 120/76 on atenolol only.

In January 2007, she re-presented with acute pulmonary oedema and again required ITU admission and dialysis. This time she had been completely well until 12 hours prior to admission. Further angiography revealed that the left renal artery stents had collapsed although a small amount of contrast filling normal vessels distally. It was felt that this was providing some renal perfusion although not enough to allow adequate clearance of small solutes and free water.. Attempted angioplasty from the groin was unsuccessful on 2 occasions. Her long-term dialysis options were limited. Temporary dialysis access had been extremely difficult to achieve and we were pessimistic about chances of either a tunnelled semi-permanent catheter or AV fistula. Also, due to her central obesity PD was not a realistic option. After 10 days of being dialysis-dependent she was therefore referred for consideration of exploration of the right kidney, and if viable, revascularisation using either ilio-renal bypass or auto-transplantation to the right iliac fossa, with reconstruction of the renal artery with either the internal iliac or long saphenous vein, given that the long-term future on dialysis would be a transplant.

RENAL FUNCTION AND VASCULAR STIFFNESS IN CORONARY ARTERY DISEASE

Christian Delles1 Lukas U. Zimmerli1, Kenneth J. MacArthur2, Tracey Steedman1, Henry J. Dargie1, and Anna F. Dominiczak1

1 BHF Glasgow Cardiovascular Research Centre, University of Glasgow, Glasgow2 Department of Cardiothoracic Surgery, Western Infirmary, Glasgow

Objective. We have previously demonstrated increased vascular stiffness in end-stage renal disease and in coronary artery disease (CAD). Here we examine whether mild to moderate renal impairment further increases vascular stiffness in patients with severe CAD.

Design and method: In 72 patients with severe three-vesel CAD (age, 62±9 years) we measured carotid-femoral pulse wave velocity (PWV; n=52) using the SphygmoCor® Vx system and compliance of the ascending aorta by cardiac MRI (1.5 T Siemens Sonata; n=49). Glomerular filtration rate (eGFR) and creatinine clearance (ClCrea) were estimated by 4-variable MDRD formula and the formula of Cockroft and Gault, respectively.

Results. eGFR ranged from 27 to 103 (mean, 63±14) mL/min and ClCrea from 34 to 129 (mean, 74±23) mL/min. Patients with eGFR <60 mL/min (n=27) had higher PWV (9.2±1.9 vs 7.6±1.8 m/s; P=0.003) and lower aortic compliance (6.7±3.8 vs 13.0±6.4 µL/mmHg; P<0.001) than patients with eGFR ≥60 mL/min (n=45). In line with these findings, PWV (r=-0.301; P=0.030) and aortic compliance (r=0.601; P<0.001) were correlated with ClCrea. eGFR <60 mL/min was associated with an unfavourable constellation of factors involved in the pathogenesis of atherosclerosis (Table).

Conclusions. In patients with CAD mild to moderate impairment of renal function is associated with increased vascular stiffness. This may be explained by unfavourable effects of chronic renal failure on inflammatory and pro-atherogenic factors. Our data support aggressive management of co-existing conditions such as chronic renal failure in patients with CAD.

CRP(mg/L)

Leptin(µg/L)

oxLDL/LDL(mmol/L AU-1)

eGFR<60 mL/min 7.0±13.6 30.7±25.3 28.0±9.7eGFR ≥60 mL/min 5.7±16.0 15.0±9.0 22.7±8.9P value 0.038 0.004 0.027

Table.oxLDL/LDL denotes ratio between oxidised LDL cholesterol and LDL cholesterol

Source of funding: British Heart FoundationConflict of interest: none

Fabry Disease: An overview and case presentation

Dana Kidder and William SmithRenal Unit, Monklands Hospital, Airdrie

Fabry disease is an under-recognised X- linked lysosomal storage disorder characterised by the deficiency of alpha-galactosidase A (α-GAL A). This deficiency leads to the progressive accumulation of globotriaosylceramide (GL-3/Gb3) in the vascular endothelium and visceral tissues culminating in end organ damage.

The clinical manifestations are variable, from early symptoms in childhood and adolescence to late life threatening complications such as renal failure, stroke and cardiovascular events in early and mid adulthood.Early classic symptoms include acroparathesias, ocular opacities, hypohidrosis and angiokeratomas. Renal manifestations commonly include proteinuria, progressive glomerulosclerosis with GL-3 deposition and end stage renal disease.

Recent advances in this condition include enzyme replacement therapy in the form of agalsidase which is administered by iv infusion every 2 weeks. Results of recent stage III and IV clinical trials show significant risk reduction for renal, cardiac and cerebrovascular events with enzyme replacement therapy.

We present a case of proteinuria and haematuria, with classical zebra inclusion bodies on renal biopsy and negligible α-GAL A activity. The clinical profile, progress, NSCAG approval and response to enzyme replacement therapy will be discussed.

Cardiac Troponins in Established Renal Failure

Asif Ansari, Graham Smith* and William SmithRenal Unit, Monklands Hospital, Airdrie; *Department of Statistics and Modelling Science, University of Strathclyde, Glasgow

Background: Troponin (Tn) proteins are the gold standard of biomarkers in myocardial injury/infarction. Both TnT and TnI are sensitive and specific in general but can be altered in non cardiac conditions such as renal failure, sepsis and acute stroke. Cardiovascular disease, left ventricular hypertrophy, inflammation, sepsis, abnormal protein metabolism and clearance are common in end stage renal disease. Many hypotheses have been put forward for the elevation of Tn, in the absence of myocardial injury in dialysis patients. How do we interpret raised Tn, does it have a diagnostic or prognostic value?

Methods: We studied an unselected haemodialysis population. Analyses were performed on predialysis monthly bloods for TnT, TnI, CRP, PTH, and cholesterol. All case records were reviewed for ECG’s and echocardiograms. Bloods for TnT, CRP, and PTH were repeated 6 months later and patients were followed up for 12 months. Statistical analysis was performed using SPSS.

Results: The study started with 141 patients, mean age 60, mean duration of dialysis 39 months. ECG’s were reviewed in 89% of patients and 74% had Echo’s. TnT(1) was raised (normal <0.03) in 63 patients (45%), TnT(2) was raised in 56 of 118 patients (48%) and TnI was elevated (normal <0.2) in 5 of 135 patients (3.7%). TnT, TnI, CRP and PTH were all skewed in distribution and results are median and (interquartile range). TnT(1) = 0.028 (0.07), TnT(2) = 0.025 (0.09), TnI = 0.019 (0.04), CRP(1) = 12 (28), CRP(2) = 9.5 (24), PTH(1) = 26 (30), PTH(2) = 24 (28). Correlations were done with Spearman rho analysis. Strong correlations were found between Tn measurements, correlation coefficient (cc) >0.6, p at 0.01 level. Moderate correlations were found between TnT and CRP, age and outcome cc >0.3, p at 0.01. Weaker correlations were detected between TnT and ECG and Echo results. No significant correlations were derived between TnT and PTH, cholesterol or duration of dialysis. Mortality at 12 months was 26.2% of which 18.4% had raised TnT.

Conclusion: These preliminary data need cautious interpretation. TnT was very marginally elevated in almost half of the haemodialysis patients whereas TnI was normal. TnT appears less specific but a higher accepted cut-off for renal failure patients would correct this apparent false elevation. The association between TnT and CRP is suggestive that inflammation/infection may be an influential factor. Elevated TnT appears to be a negative prognostic factor after 12 months follow-up.

The evolution of end stage renal failure in Ayrshire and Arran – the influence of referral patterns, diabetes and vascular disease.

Andrew Innes, Mark S MacGregor, Nestor Velasco, Pamela M Mackenzie and Ian G Mackay.

Renal Unit, Crosshouse Hospital, Kilmarnock.

The Renal Unit in Ayrshire was established in 1990 and since then almost 650 patients have received RRT for chronic renal failure at the Unit. The take-on rate has risen from 53 per million (1993) to 157 per million in 2006. In the early years between 42% and 58% of patients needed dialysis less than one month after first seeing a nephrologist but by 2006 this had fallen to 10%. Similarly, the median creatinine at first presentation to a nephrologist was consistently over 500µmol/l in the early 1990s but was 248µmol/l (eGFR =22 ml/min corrected) in 2006.

The median age at first dialysis has risen from 52 (1990) to 64 (2006) with a peak of 70 years in 2004. Currently, 61% of new RRT patients have either diabetes or pre-existing vascular disease (32% have diabetes; 44% have vascular antecedents). As in the rest of Scotland, increasing numbers on RRT have produced a major demand on hospital HD provision (39 patients on hospital HD in 1994; 134 in 2006). Crosshouse has a flourishing PD programme which has also increased from 27 in 1994 to 42 in 2006; now with 86% on APD. In the prevalent dialysis population (HD and PD) 21% are diabetic and 38% have vascular antecedents; almost half (48%) have at least one of these comorbidities. Of the 59 patients starting RRT in 2006, 12 started PD and 47 HD (10 with functioning fistulas, 37 with catheters).

As in other parts of Scotland, the relentless increase in numbers of patients starting haemodialysis places demands on HD spaces. These patients are increasingly elderly and now a sizeable majority enter dialysis with significant comorbidity in terms of vascular disease and diabetes. The problem of late referral (and consequent “acute” need for dialysis) appears to be addressed and diminished by the establishment of a renal unit in an area. Nevertheless, creatinines at the time of referral (of those who subsequently go on to require dialysis) remain stubbornly around 250µmol/l.

Nephrotic syndrome presenting as venous thromboembolism

Vik Selvarajah, Bill Ambler, Chris IslesRenal Unit, Dumfries and Galloway Royal Infirmary, Dumfries, DG1 4AP.

A patient presenting with a swollen left leg and pleuritic chest pain was shown to have deep vein thrombosis by doppler studies. He was anticoagulated but represented six weeks later with swelling of both legs while on warfarin. No fresh thrombus was seen and warfarin was continued. Three weeks later he was readmitted with swelling that extended to both upper thighs. A degree of redness led to a diagnosis of cellulitis and treatment with antibiotics. Urinalysis was not recorded during one of these admissions, while a positive dipstick result for proteinuria was overlooked twice. Following his third discharge from hospital, the GP noted heavy proteinuria and referred the patient directly to the Renal Clinic where urine protein was quantified at 12.3 g/24 hours and a diagnosis of nephrotic syndrome was confirmed. Renal biopsy showed that this was due to membranous nephropathy.

We subsequently conducted two audits. The first was of patients with diagnostic discharge codes for nephrotic syndrome and venous thromboembolism in south west Scotland (population 147,000) from 1996 to 2006 in order to determine the frequency with which DVT or PE had been the presenting feature of nephrotic syndrome. We were able to confirm a diagnosis of nephrotic syndrome in 32 patients, all of whom had oedema, serum albumin < 30 g/l and proteinuria > 3 g/24 hours. No fewer than four (12.5%) of these including the index case had presented with DVT (2) or PE (2). Three had membranous nephropathy and one had minimal change nephropathy. Renal vein doppler studies were normal in both patients with PE. The diagnosis of nephrotic syndrome was made during the first admission to hospital in all patients except the index case. A second audit of 98 consecutive patients with doppler positive lower limb DVT presenting to A&E in Dumfries from July 2005 to July 2006 showed that the urine had been tested for protein in one case only.

These results support the view that nephrotic syndrome is complicated by venous thromboembolism sufficiently frequently for the diagnosis to be considered in all patients with DVT or PE. The move towards managing patients with DVT in the community may mean that patients are even less likely to have urinalysis performed. The take home message for patients with DVT or PE must simply be – don’t forget to dip the urine.

Word count: 385

Abstract

Renal Failure in Patients with Multiple Myeloma

Shona Methven, Nicola Joss, Jonathan FoxRenal Unit, Glasgow Royal Infirmary

Background and aims: Renal disease in multiple myeloma can present in a variety of ways and predicts outcome. We present data on the severity of renal failure at referral to renal services, frequency of renal biopsy, frequency of renal replacement therapy (RRT) and survival in patients diagnosed with multiple myeloma in a single centre.

Methods: Patients attending the Renal Unit, Glasgow Royal Infirmary between 1989 and March 2006 with a diagnosis of multiple myeloma were identified from the electronic patient record.

Results: One hundred and twenty two patients were identified (60.7% males). At referral, the mean age was 65.5 years (SD 11), median serum creatinine was 400µmol/L (IQR 200,693), and median estimated glomerular filtration rate (eGFR) by the MDRD formula was 11.5 ml/min (IQR 6,25). 1% of patients were referred at CKD stage 1, 6% at CKD stage 2, 15% at CKD stage 3, 23% at CKD stage 4 and 55% at CKD stage 5. Other baseline data were mean serum albumin 32.7 g/L (SD 7.3), adjusted calcium 2.4 mmol/L (SD 0.3) and haemoglobin 10 g/dL (SD 2.2). 35.3% had serum albumin less than 30 g/L. Of those who had proteinuria measured (82 patients), 32% had nephrotic range proteinuria.

Thirty-four patients (27.9%) had a renal biopsy. There was no statistically significant difference in age, serum creatinine, eGFR, or serum albumin between the biopsy group and the non-biopsy group. The diagnoses made by renal biopsy were: cast nephropathy n=14, AL amyloidosis n=9, light chain disease n=5, heavy chain disease n=1, acute interstitial nephritis n=4, acute tubular necrosis n=2 and ischaemic/hypertensive nephropathy n=2 (some patients had >1 diagnosis).

Fifty-three patients (43.4%) received RRT with 28 patients (52.8%) starting RRT within 1 week of referral. Fifteen patients who started RRT recovered renal function with a median time to recovery of 4.0 days. No patients fully recovered renal function (to eGFR > 60ml/min) during follow up. Of those who did not recover renal function, median survival on RRT was 22.5 months (95%CI 7.5,37.5). Overall, median patient survival from time of referral was 18.7 months (95%CI 10,27.4) with a 1-year survival of 58.6% and 5-year survival of 25.9%.

Conclusion: Renal failure in patients with multiple myeloma often presents late, frequently requires renal replacement therapy, and has a poor prognosis. Indications for renal biopsy in patients with multiple myeloma remain unclear.

Diurnal variation of phosphate is maintained in end-stage renal disease.

Elaine M Spalding & Ken Farrington. Lister Hospital, Stevenage.

Introduction. Marked diurnal variation exists in plasma phosphate concentration in normal individuals but it is not known if this, or diurnal variation in regulating hormones is maintained in CRF.

Subjects and Methods. Eight subjects, four with normal renal function and four with CRF were studied for 24 hours with hourly measurements of phosphate, calcium, bicarbonate, potassium, albumin and glucose and two hourly measurements of PTH, GH, cortisol and insulin. Analysis of diurnal variation was by rhythm biometry using cosinor methodology and relationships between variables were investigated by cross-correlation.

Results. Diurnal variation in phosphate, calcium, phosphate excretion, bicarbonate, PTH and cortisol is maintained and is well described by harmonic periodic regression. The relationship seen between phosphate and PTH in normal renal function is maintained in advanced CRF but there are marked differences in the relationship between calcium and PTH and in acid-base status between the two groups.

Conclusion. The diurnal variation that is maintained in non-oliguric CRF has implications for the timing of blood samples when initiating treatment for bone disease or monitoring response to treatment. There are potential long-term implications for the planning of dialysis schedules.

Conflict of interest: noneSource of funding: unit funds

The provision of accommodation by Roche Pharmaceuticals Clinical Pharmacology Unit, Welwyn Garden City for the duration of the above study is gratefully acknowledged.

Endothelial cell protection from complement activation – relevance to the pathogenesis of atypical

Haemolytic Uraemic Syndrome

Anna Richards1,3, David Kavanagh2,3, Kathy Liszewski3, John Atkinson3

1Edinburgh Royal Infirmary, Little France, Edinburgh2Department of Chemistry, Kings Building, University of Edinburgh3Washington University School of Medicine, St Louis, MO63110, USA

IntroductionHaemolytic uraemic syndrome, the clinical triad of microangiopathic haemolytic anaemia, thrombocytopenia and acute renal failure, is characterised by damage to microvascular endothelial cells, particularly in the glomerular and renal arteriolar endothelium. Endothelial cells (EC) are highly specialised cells capable of producing both anti and pro-inflammatory mediators. The macrovascular EC line, HUVEC, express the membrane bound complement regulators CD46, CD55 and CD59. They also synthesise the soluble complement regulators factor H, factor I and the activating components factor B and C3. Mutations in the alternative pathway complement regulators factor H, factor I and CD46 have previously been described in atypical HUS (aHUS). More recently, mutations in Factor B and C3 have been reported. Our aim was to characterise the complement regulatory profile and response to complement activation of microvascular and glomerular ECs.

MethodsCharacterisation of an immortalised human microvascular endothelial cell line (HMEC) and a primary glomerular endothelial cell (GEC) line was undertaken. Expression of the complement regulators CD46, CD55 and CD59 was examined using FACS, ELISA and Western blotting. Synthesis of Factor H, C3 and Factor I was assessed by Western blotting. Binding of factor H and factor I to HMEC was assessed by FACS. A ‘Complement Challenge’ experiment, whereby antibody was used in the presence of human serum to deposit complement on the endothelial cell surface was developed and the effects of blocking complement regulators singly, and in combination, examined.

ResultsLike HUVEC, HMEC and GEC also express the membrane bound complement regulatory proteins CD46, CD55 and CD59 on their surface. They do not express CR1. These cells are typical endothelial cells, expressing a wide range of endothelial cell marker antigens. However differences between HMEC and GEC can be found, suggesting an important role for specialisation in different endothelial cell beds. Using EDTA to activate the classical pathway of complement in the Challenge assay, it was possible to show differential function and effectiveness of the soluble and membrane bound complement regulatory proteins in down-regulating complement on the HMEC surface. These effects were additive. Using MgEGTA as a buffer to preferentially activate the alternative pathway it was possible to show a significant difference in the ease with which the two pathways of complement could be activated on HMEC.

ConclusionsWe have characterised complement regulator protein expression in both a microvascular and renal endothelial cell systems. We have developed a sensitive and highly modifiable system for assessing the effects of complement activation on the surface of endothelial cells. Future work will look at Factor B and C3 synthesis and secretion by HMEC and GEC, which have recently been implicated in the pathogenesis of aHUS. This system could also be used to look at the effectiveness of pharmacological methods of protecting endothelial cells from complement-mediated attack.

Characterization of Mutations in Complement Factor I (CFI) Associated with Haemolytic Uraemic Syndrome

David Kavanagh1, Anna Richards1, Marina Noris2, Judith Goodship3, Veronique Fremeaux-Bacchi4, Giuseppe Remuzzi2, Timothy H. J. Goodship3, John P. Atkinson1*

1 Washington University School of Medicine, St. Louis, U.S.A.2 Mario Negri Institute,Bergamo, Italy3 Institute of Human Genetics, University of Newcastle upon Tyne.4 Hôpitaux Européen Georges Pompidou, Paris, France

Recent studies have identified mutations in the complement regulatory gene factor-I (CFI) that predispose to atypical haemolytic uraemic syndrome (aHUS).

CFI is a two chain serine protease in which the light chain carries the catalytic domain while the heavy chain’s function is unclear. CFI downregulates the alternative and classical complement pathways by cleaving the α chains of C3b and C4b (cofactor activity). Many CFI mutations in aHUS result in low CFI levels with a consequent quantitative defect in complement regulation. In others, the mutant protein is present in normal amounts but the presumed functional deficiency has not been defined. In this report we examine the nature of the functional defect in aHUS-associated CFI mutations.

A representative subset of aHUS associated CFI mutants was chosen for study from our prior genetic analysis. The aHUS-mutations were introduced into CFI cDNA containing a 6X histidine tag using Quikchange®. Transient transfections of 293T cells were performed. The recombinant CFI was purified using an AKTA purifier® with a HisTrap column. C3b and C4b endpoint and kinetic cofactor assays were then performed.

Three mutations (D506V, D501N, R299W) demonstrated a loss of both C3b and C4b cofactor activity. In two heavy chain mutants (M120I, G243D) cofactor activity was normal. The L466V+Q467G+W468X mutant was retained intracellularly consistent with the low serum level of CFI seen in this patient.

Modelling of CFI predicts that both D501 and D506 form part of the walls of the specificity pocket of the serine protease domain of CFI. Our results provide experimental evidence supporting this modelling prediction and demonstrate that this region is critically important to the serine protease function of CFI.

R299W resides in part of an unidentified domain region of unknown function between the LDLRa domain and the serine protease domain. Our experiments demonstrate that the R299W mutation results in a decrease in C3b and C4b cofactor activity, establishing that this novel region plays a role in C3b and C4b cleavage.

Mutations in the heavy chain domains CD5 and LDLr did not affect fluid phase cofactor activity. These mutations may point to an as yet undiscovered function of the non-catalytic domain in the regulation of complement.

In summary we demonstrate that 4 out of the 6 CFI mutations studied result in a loss of both alternative and classical pathway cofactor activity.

THE RELATIONSHIP BETWEEN HAEMODIALYSIS CATHETER CHOICE, CLINICAL OUTCOME AND RECOMMENDED PRACTICE.

Peter C Thomson*, Catherine Stirling**, Scott Morris**,Robert A Mactier**.* Research Fellow in Renal Medicine, Renal Unit, Glasgow Royal Infirmary.** Consultant Nephrologist, Renal Unit, Glasgow Royal Infirmary.

Correspondence to:Dr Peter Thomson, Renal Unit, 3rd Floor Walton Building, Glasgow Royal Infirmary, 84 Castle Street, Glasgow, G4 OSFTel: (+44) 141 211 0570 Fax: (+44) 141 211 4843

Introduction

Haemodialysis catheter use can be associated with significant complications. In cases of suspected catheter thrombosis or low-grade catheter-related bacteraemia in the absence of subcutaneous tunnel infection, NKF-KDOQI advocate catheter exchange over a guide-wire. This study compared rates of catheter-related bacteraemia following various types of catheter insertion, including how catheter-exchange over a guidewire compares with catheter replacement at a different site.

Methods

We performed a prospective analysis of all incident vascular access haemodialysis catheter insertions over the period starting 05/08/2005 and ending 05/08/2006. Laboratory and clinical variables were recorded at catheter insertion and the clinical course was followed up to the point of catheter removal. The primary outcome measure was catheter-related bacteraemia as defined by the presence of positive blood cultures, clinical and biochemical evidence of a raised systemic inflammatory response and the absence of evidence of infection arising from a different site. Univariate analysis was used to test for association between clinical and laboratory variables and subsequent outcome. Significant univariates were then put forward for inclusion in a multivariate model to test for independent association.

Results

15,834 catheter days were accumulated over the study period during which time a total of 206 patients underwent insertion of 131 tunnelled central venous catheters (TCVCs) and 271 non-tunnelled central venous catheters (NTCVCs).

Rates of catheter-related bacteraemia per 1000 catheter days were 1.88 in the TCVC group, 7.86 in the internal jugular vein NTCVC group [Hazard Ratio (HR) 2.8, p=0.004], 10.99 in the femoral vein NTCVC group [HR 5.2, p=0.009], 14.89 in the internal jugular NTCVC guidewire group [HR 6.9, p<0.001] and 45.45 in the femoral vein NTCVC guidewire group [HR 21.7, p<0.001]. These associations were independent of age, sex, diabetes and length of time on renal replacement therapy.

Patients who had undergone NTCVC exchange over a guidewire were found to have significantly higher rates of catheter-related bacteraemia (17.9 v 7.3 per 1000 catheter days, p=0.025) when compared with those who had undergone NTCVC replacement at a de novo site.

Conclusion

Haemodialysis catheter-exchange over a guidewire was the catheter insertion strategy most strongly associated with subsequent catheter-related bacteraemia. Guidelines should be amended to suggest that catheter-exchange over a guidewire only be conducted when catheter replacement at a de novo site is unachievable. Vascular access catheters remain a significant contributor to the burden of infection in the haemodialysis population.

Paying The Price For A Transplant - An Unusual Case Of Urinary Sepsis.

Kate Preece1, David Walbaum1, Ian Laurenson2, John Govan2, and Paddy Gibson1

1Renal Unit and 2Department of Microbiology, Royal Infirmary Edinburgh.

A thirty-five year old man who had been on hospital haemodialysis in Edinburgh for two years due to end stage renal failure secondary to membranous glomerulonephritis, returned from Pakistan having received a live unrelated renal transplant.

On return to the UK initial graft function was excellent, however he had an episode of acute cellular rejection confirmed by allograft biopsy on day 15, and was treated with intravenous corticosteroid and switched from cyclosporine to tacrolimus.

On day 50 he presented with a fever, general malaise and night sweats. He had no urinary symptoms. Subsequent investigations confirmed a urinary tract infection with the gram negative bacillus Burkholderia Cepacia.

This unusual pathogen is most frequently found in cystic fibrosis or lung transplant patients, and is notoriously difficult to eradicate. The only previous reported case of B. Cepacia in a renal transplant recipient resulted in graft nephrectomy. Our patient was treated with removal of his ureteric stent and one month of intravenous antibiotics. Two months later his graft function is satisfactory and his urine free of infection.

No conflict of interest or funding.

Unsuspected Mycotic Aneurysm of Renal Transplant Artery1D Henderson, 1A Pall, 2S Chakravarty1Departments of Renal Medicine & 2Radiology, Ninewells Hospital & Medical School, Dundee

A 31year old female with ESRD secondary to mesangiocapillary glomerulonephritis type 2 received a cadaveric renal transplant. Cause of donor death was hypothermia and extradural hamatoma. Mismatch 210 and crossmatch negative. Protocol immunosuppression was used and consisted of loading dose IV Methyl Prednisolone followed by maintenance tacrolimus, azathioprine and prednisolone. Transplant function was satisfactory with serum creatinine 95 umol/L. However the immediate post-operative period was complicated by a febrile illness but with initial blood and urine cultures and USS/Doppler of transplant kidney negative. By day 10 she had developed obvious oropharangeal thrush and blood cultures grew candida albicans. She was initiated on fluconazole with apparent complete clinical response and negative blood cultures. During the course of treatment for the candidaemia the serum creatinine increased to 150 umol/L, USS/Doppler again normal but transplant biopsy confirmed an acute cellular rejection. This was treated with pulse IV Methyl Prednisolone 250 mg x3 and switch from azathioprine to mycophenolate. Her subsequent condition remained stable with excellent renal allograft function. However four months post-transplant she was admitted with subacute bowel obstruction thought to be due to fibrous adhesions related to abdominal surgery pre-transplant. At this time she was afebrile with stable allograft function (serum creatinine 100 umol/L), normal CRP and controlled BP. USS abdomen noted the incidental finding of a transplant artery aneurysm (figure 1). This was confirmed on the MRA examination which reported a presumed mycotic aneurysm of the transplant artery measuring 4.5 cm diameter and extending to the external iliac artery (figure 2). She underwent transplant nephrectomy and excision of the aneurysm with saphenous vein patch to the external iliac artery. She was treated empirically at this time with IV flucytosine and amphoteracin although fungal cultures from the aneurysm sac and stitches from iliac vessels returned negative. Subsequent progress was complicated by bleeding from deep circumflex branch of the external iliac artery controlled by endovascular embolisation and subsequent to this an infected haematoma and false aneurysm with compromise to left leg circulation, which required surgical repair. One year later she is well on maintenance haemodialysis and is under consideration for a second renal transplant. Mycotic aneurysm complicating renal transplantation is rare and often diagnosed at post-mortem after rupture. A low threshold of suspicion is needed for early detection and should be considered whenever renal transplantation is complicated by serious fungal infection.

Evolution of B-Lymphocyte Infiltration in Acute Cellular Renal Transplant Rejection (ACR) – an illuminating case report.

1A Pall, 1V Sanu, 2S Fleming, 1G Stewart, 1I HendersonDepartments of Renal Medicine1 & Pathology2

Ninewells Hospital & Medical School, Dundee

B-lymphocyte infiltration in ACR of the renal allograft has been recently reported and associated with steroid resistance and worse graft survival. This is seen without evidence of an allohumoral response (negative C4d staining and DSA). The underlying mechanism and treatment strategy have yet to be elucidated. We describe a renal transplant recipient, who provided an opportunity to detail the evolution of a significant B-lymphocyte infiltration, the clinical and histological response to rituximab. A 32-y/o male with ESRD secondary to IgA nephropathy on maintenance HD for 2 years received a LRT from his mother. PRA 0%, luminex & XM negative, MM 101, EBV D+/R+, CMV D-/R-. He received protocol loading dose MePred, tacrolimus, azathioprine and prednisolone. The transplant was complicated by DGF and the best se.creatinine achieved was 200 umol/L presumed because of donor/ recipient size disparity. Progress from day 20 and over the next eight months was complicated by recurrent ACR. This was of varying intensity but appeared to be steroid responsive on follow-up biopsies. C4d staining and DSA were negative. During the course of follow-up maintenance IST was adjusted to SRL/MMF/Pred. He had declined anti-lymphocyte antibody treatment. However after the fourth course of pulse MePred he suffered a further episode of acute rejection and the biopsy was stained for B-Lymphocytes. These were found to account for 60% of the lymphocyte infiltrate. A lymphoid aggregate had been noted on a previous biopsy and stained negative for EBV with low proliferation rate (evidence against PTLD). Because of the high proportion of B-lymphocytes it was decided to treat with rituximab 175mg/m2 weekly x2. This resulted in a sustained B cell depletion and repeat biopsy showed complete resolution of the infiltrate with no CD20. He was switched back to tacrolimus and maintained on this with MMF and prednisolone. Allograft function stabilised to serum creatinine 270 umol/L. Retrospective staining for CD20 of all previous biopsies identified a progressive B-lymphocyte infiltrate also apparent on the biopsies suggesting resolving ACR after pulse steroids. B-lymphocyte infiltration should be considered with recurrent episodes of ACR when steroid-resistant but also if follow-up biopsy is suggestive of resolution. Banff has recently been updated in the context of C4d staining and AMR. In the future the classification may also need to consider the role of B-lymphocytes. We have shown that rituximab can clear B-cells from the renal allograft when these constitute a significant proportion of the lymphocyte infiltrate in ACR. Studies are needed to confirm that this translates into improved clinical outcome.

Pancreas Transplantation in Scotland.

David Walbaum, Christine Jansen, Murat Akyol and Caroline Whitworth.

Transplant Unit, Royal Infirmary of Edinburgh, Old Dalkeith Road, Edinburgh, EH16 4SA.

The first pancreatic allograft transplant in Scotland took place in April 2000, since when a total of 77 have been performed. This includes 68 simultaneous pancreas-kidney transplants, 6 pancreas after kidney transplants and 3 pancreas alone transplants.

Patients have been referred from all ten adult renal units in Scotland, with a significant number of patients being transplanted pre-emptively, before the need for renal replacement therapy.

The median waiting time for pancreatic transplantation is significantly shorter than for cadaveric kidney transplantation, and the donors are younger.

One year patient survival is 96.8%, renal allograft survival is 94% and insulin-free pancreatic allograft survival is 80%. The rate of biopsy proven renal allograft rejection is 24%. The median in-patient hospital stay following a pancreas transplant is 16 days.

We present a review of pancreatic transplant activity in Scotland.

No conflict of interest or funding.