Abstracts from

THE SCOTTISH RENAL ASSOCIATION MEETING

Friday 25th March & Saturday 26th March 2011

Drumossie Hotel, Old Perth Road,

Inverness,IV2 5BE

A1. Recipient M235T polymorphism of the angiotensinogen gene predicts long- term graft survival in kidney transplantation

Patrick B Mark1, Jamie P Traynor2, Kevin McLaughlin3, Kathryn K Stevens1, Rajan K Patel1, Catherine J Clark1, Alan G Jardine1,

1. University of Glasgow, Glasgow1. Renal Unit, Monklands Hospital2. Department of Medicine, University of Calgary, Canada.

Background Polymorphisms of genes encoding the renin angiotensin system and - adducin 1(AAD1) have been associated with progression of renal disease and increased cardiovascular risk. These encode for factors regulating blood pressure. The relationship between these polymorphisms and outcome in renal transplant recipients is less clear. Methods We undertook a longitudinal follow up study of 422 first renal transplant patients (median age 42.6 years; 56.6% male; 10% live donor transplants). Patients were surveyed in 1994 and followed up after 16 years. Genotyping was performed for the M235T polymorphism of the angiotensinogen (AGT) gene, D polymorphism of the gene encoding angiotensin converting enzyme and AAD1.Results There were 179 (42.2%) deaths and 155 (36.7%) death-censored graft failures. Kaplan-Meier analysis demonstrated decreased patient survival in patients with TT genotype of AAD1 (p<0.001) and a trend towards poorer survival in TT genotype of AGT (p=0.07). Multivariate analyses showed reduced estimated glomerular filtration rate (eGFR) at 1 year (hazard ratio (HR) 0.97, p=0.004), recipient age (HR 1.07, p<0.001), diabetes (HR 7.058, p=0.003) and AAD1 genotype TT (HR 3.33, p=0.01) to be independent predictors of mortality. Graft survival was poorer in patients with the TT genotype of the AGT gene (see figure) but the other genetic polymorphisms studied did not influence graft survival. Multivariate analysis demonstrated reduced eGFR at 1 year (HR 0.97, p=0.006), deceased donor transplant (HR 14.87, p=0.013) and TT genotype of the AGT gene (HR 4.43, p=0.003) to be independent predictors of long-term graft failure.

Conclusions Polymorphisms of the AGT gene may identify renal transplant patients at increased risk of long-term graft loss whilst AAD1 polymorphisms identify patients at increased risk of mortality. In this cohort, these relationships were independent of blood pressure.

Funding: None Conflict of interest: None

A2. The Role of Surveillance Biopsies in Clinical Renal Transplantation

L.K. Henderson1,2 B.J. Nankivell1 J.R. Chapman1

1Department of Renal Medicine, Westmead Hospital, University of Sydney, Australia 2

Department of Renal Medicine, Royal Infirmary of Edinburgh

Aim: To determine the frequency of events found with a surveillance biopsy strategy using current immunosuppressive protocols in a low or intermediate risk population.

Background: Surveillance biopsies have shown that early detection and treatment of subclinical rejection (SCR), prior to clinical evidence of graft dysfunction, can result in improved graft outcomes [1-3]. It has been argued that clinical utility has fallen with the decreased prevalence of subclinical rejection following the introduction of more potent immunosuppression and that better individualised dosing of calcineurin inhibitors (CNI) has reduced CNI nephrotoxicity.

Methods: We retrospectively examined the histological findings in 280 surveillance biopsies and detailed clinical action taken following biopsy in 100 kidney and combined kidney pancreas transplant recipients transplanted between January 2008 until July 2009 at Westmead Hospital, Sydney. Only patients biopsied with stable graft function were considered as having surveillance biopsies. Biopsies were classified and adequacy assessed using Banff 97 criteria [4] by pathologists blinded to clinical data. Kidney biopsies were carried out at implantation, 1, 3 and 12 months. Decision to alter immunosuppression was dependant on combined clinical and histologic criteria.

Results: 50% of all surveillance biopsies were abnormal. The major findings were SCR (6%), borderline changes (12.5%), interstitial fibrosis and tubular atrophy (IF/TA) (20%), acute tubular necrosis (7.5%), CNI toxicity (5%), BKVAN (1.4%) and recurrent disease (3.2%). Subclinical rejection was detected in 19.8%, 14.9% and 18.8% of biopsies and IF/TA detected in 9%, 17.2% and 33.8% of biopsies at respective time points. 15.7% of all surveillance biopsies led to an increase, 8% a switch and 32% a decrease in immunosuppression.Following treatment of SCR, creatinine clearance (CrCl) improved from 64.6 ± 25.9 ml/min to 72.8 ± 33 ml/min at one year from transplant with a mean ∆ CrCl of +10.7 ±22.1 ml/min (p= not significant, NS). Mean i,t scores fell from 2.65 ± 0.24 to 0.88 ± 0.29 (mean ± SEM, p < 0.001). Chronic scores stabilised: mean ci,ct score 1.1 ± 0.24 vs 1.3 ±0.27. Three patients treated for BKVAN had a mean creatinine clearance of 49.4 ±31.6ml/min and two had no ongoing evidence of BKVAN at 1 year.

Conclusions: Early routine surveillance biopsies reveal a high frequency of findings undetected by clinical surveillance and often unveil pathology that, if left undetected and untreated, may lead to worse graft outcome.

1. Kee, T.Y., et al., Treatment of subclinical rejection diagnosed by protocol biopsy of kidney transplants. Transplantation, 2006. 82(1): p. 36-42.

2. Kurtkoti, J., et al., The utility of 1- and 3-month protocol biopsies on renal allograft function: a randomized controlled study. Am J Transplant, 2008. 8(2): p. 317-23.

3. Choi, B.S., et al., Clinical significance of an early protocol biopsy in living-donor renal transplantation: ten-year experience at a single center. Am J Transplant,2005. 5(6): p. 1354-60.

4. Racusen, L.C., et al., The Banff 97 working classification of renal allograft pathology. Kidney Int, 1999. 55(2): p. 713-23.

A3. Late-presenting Scleroderma masquerading as Scleroderma sine scleroderma

Fraser Millar*, Neeraj Dhaun†, Phil Masson†, Anna Richard†s, Wendy Metcalfe†, JohnNeary†

(* 4th yr Student, University of Edinburgh, † Department of Renal Medicine, RoyalInfirmary of Edinburgh)

Introduction: We present two cases in which patients initially presented with a Thrombotic microangiopathy (TMA) of uncertain aetiology. Both patients subsequently developed signs of scleroderma. Endothelin antagonists were used in addition to renin- angiotensin blockade.

Case 1: A 61 year old man presented with acute renal failure. He was normotensive. Renal biopsy showed thrombotic microangiopathy (TMA) of uncertain aetiology.. Stool cultures were negative for E. coli 0157. No malignancy was found, PSA was normal. A genetic screen revealed no mutations in genes of alternate pathway of complement. He had a chronic small joint arthritis and his TMA was presumed secondary to connective tissue disease. Three and an half years later he presented with rapidly progressive sclerodactyly, digital ulceration, soft tissue calcification and telangiectasia. He was successfully treated successfully with Bosentan (a mixed (ETa/ETb) Endothelin receptor antagonist) with full resolution of hand signs.

Case 2: A 60 year old female presented with acute renal failure with TMA presumed secondary to malignant hypertension. She developed polymyalgia 12 months later and then presented with telangiectasia, sclerodactyly, worsening Raynauds and Pulmonary hypertension 6 months after that. She was treated with Ambrisentan (a selective ETa Endothelin receptor antagonist) as part of management of pulmonary hypertension, but this no discernable effect on her cutaneous signs.

Discussion: The initial presentation of scleroderma / systemic sclerosis can be variable. Our patients initially presented with acute renal failure and had TMA of uncertain aetiology, rather than the classic signs such as Reynaud’s or skin manifestations. The visceral and cutaneous manifestations of scleroderma developed later (mean 30 months).

Renal TMA has a broad differential diagnosis and several well recognised associations were excluded. Scleroderma should therefore be considered in diagnosis of TMA, even in the absence of more classic features of the disease. ACE-I is accepted first line therapy for Scleroderma renal crisis. An additional role for ET receptor antagonism has yet to be established but accumulating data appears to supports a potential role for ET-1 in the pathogenesis.

References:Molina JF, J Rheum, 1995, 22(3): 557-60Bussone G. et al, Curr Rheumatol Rep (2011) 13:37-43

Conflict of interest: None.

Cum

ulat

ive

Pro

porti

on S

urvi

ving

A4. An analysis of factors affecting dialysis modality choice and the outcome of each different dialysis modality in Waikato Regional Renal Service, New Zealand

Claire Parke: Raigmore Hospital, Inverness; Kim Wong: Hamilton Hospital, New Zealand; Stewart Lambie: Raigmore Hospital, Inverness.

Introduction: Patient’s dialysis modality choice is a due to an intricate interaction of economics, clinician and patient factors. Outcome of each dialysis modality may be affected by these factors.Objective: The aim of this study is to assess the factors affecting dialysis modality choice and the outcomes of dialysis for Waikato Midland Regional Renal Service (WMRRS).Method: The method used included conducting a structured patient interview, sending questionnaires to clinicians, examining WMRRS databases and sending questionnaires to major New Zealand Units to assess demand and supply of dialysis and reviewing ANZDATA patient mortality between 1999-2008.Results: 282 patients were interviewed. 57.8% reported having a choice of modality; for peritoneal dialysis (PD) patients, lifestyle/convenience (43%), travel (23%); whilst for haemodialysis (HD) lifestyle/convenience (30.2%), physical factors (20.6%), recommended by friends (15.9%) and perceived less complication (14.3%) are the important factors. Of those with no choice, 93.3 % of PD patients and 36.4 % of HD patients reported physician directive was the reason. Only 4% of PD and 36.4% of HD patients reported medical reasons prevented free choice. There is limited HD capacity in dependent HD but not in HHD and PD. All clinicians strongly recommended home based dialysis.Comparable outcomes were found in all dialysis modalities. Of the 823 patients in the ANZDATA audit, PD and HHD shows a significant difference in age. Overall 5year survival was 40% but HHD showed the best survival outcome (69%), after controlling for age.

1.0

0.9

0.8

0.7

0.6

Cumulative Proportion Surviving (Kaplan-Meier) Complete Censored

Kaplan-Meier cumulative survival by modality of treatment.

0.5

0.4

0.3

0.2

0.1

0.0

-0.10 20 40 60 80 100 120 140

Time

IHD SHD HHD PD

Conclusion: Home based therapy is preferred by patients and clinicians. Limitation in IHD and SHD capacity has indirectly increased patient allocation to PD and HHD. All modalities have comparable outcomes, except HHD which has the best survival.Future Recommendations: A similar study based on a large UK unit may provide interesting contrast and comparison between UK and New Zealand. UK has lesser restrictions on dependent HD capacity and may have impacted the development of HHD and home PD. Whilst it may be cheaper in New Zealand to run HHD and PD, this may not be true in UK if there is already exiting space in the dependent HD units. Comparably it may prove more expensive to promote HHD. We are planning to run the same study in Raigmore Hospital to see whether these outcomes for HHD may benefit UK patients.

A5. Selective endothelin-A receptor antagonism reduces proteinuria, blood pressure & arterial stiffness in chronic proteinuric kidney disease

N Dhaun, IM MacIntyre, D Kerr, V Melville, NR Johnston, J Goddard, DJ WebbQueen’s Medical Research Institute, University of Edinburgh

Introduction: Proteinuria is associated with adverse cardiovascular and renal outcomes that are not prevented by current treatments. Endothelin-1 (ET-1) promotes the development and progression of chronic kidney disease (CKD), and associated cardiovascular disease (CVD). We therefore studied the effects of selective endothelin-A receptor antagonism in proteinuric CKD patients, assessing proteinuria, blood pressure (BP) and arterial stiffness (AS) - key independent, surrogate markers of CKD progression and CVD risk.

Methods: In a randomised double-blind, 3-way crossover study, 27 subjects receiving recommended renoprotective treatment, received 6 weeks of placebo, sitaxsentan100mg od and nifedipine LA 30mg od. 24h proteinuria, protein:creatinine ratio (PCR),24h ambulatory BP, and pulse wave velocity (PWV; as a measure of arterial stiffness), were measured at baseline and week 6 of each treatment period. In 13 subjects, renal blood flow and glomerular filtration rate were assessed at baseline and week 6 of each period.

Results: Compared to placebo, sitaxsentan significantly reduced proteinuria (24h proteinuria by -0.62 0.11 g/d, p < 0.01; PCR by -43 8 mg/mmol, p = 0.01), BP (24h mean arterial BP by -4 6 mmHg, p < 0.01), and PWV by -5 9 % (p < 0.01). Nifedipine matched the BP and PWV reductions seen with sitaxsentan. However, whereas sitaxsentan reduced proteinuria, nifedipine did not. Sitaxsentan alone reduced both glomerular filtration rate and filtration fraction. It caused no clinically significant side effects.

Conclusions: Selective endothelin-A receptor antagonism may provide additional cardiovascular and renal protection by reducing proteinuria, BP and arterial stiffness in CKD subjects already receiving optimal recommended treatment. These antiproteinuric effects likely relate to changes in BP and renal haemodynamics.

A6. Macrophages regulate pressor responses to endothelin-1

EL Owen, N Dhaun, M Bailey, DJ Webb, DC KluthQueen’s Medical Research Institute, University of Edinburgh

Introduction: Endothelin-1 (ET-1) is a potent vasoconstrictor. Its effects are mediated by two receptors, the endothelin-A (ETAR) and the endothelin-B (ETBR) receptors. Beneficial effects of ETAR antagonism are observed in experimental and clinical hypertension. ET-1 also has important pro-inflammatory effects. We assessed the effects of ET-1 on human macrophage (Mφ) activation and chemokinesis. We also determined whether Mφ are involved ET-1 uptake and responses to ET-1 in vivo using a murine model of selective Mφ depletion.

Methods: Human peripheral blood monocytes were matured to Mφ for 7d in vitro. Expression of ETAR and ETBR protein was assessed by immunofluoresence, Western blot, and RT-PCR. Mφ were exposed to LPS/IFNγ (positive control) or exogenous human ET-1 (0-50ng/ml), and expression of IL-6, IL-8, IL-10, and TNFα production were assessed by ELISA, and ET-1 by radio-immunoassay. Expression of cell surface antigen presentation molecules MHC II and CD80/86 were assessed by FACS analysis. Mφ migration towards ET-1 or MCP-1 (positive control) was assessed using a transwell system. ET-1 removal by Mφ was investigated by co- incubation with chloroquine (CQ, inhibitor of lysosomal acidification), protease inhibitors (PI), or selective ETAR (BQ123) and ETBR (BQ788) antagonists, respectively. Finally, we assessed whether selective Mφ depletion in CD11b-DTR transgenic mice altered pressor responses to ET-1 (0.01-1 nmol/kg).

Results: Mφ expressed both ETAR and ETBR protein and mRNA. ET-1 did not stimulate Mφ production of IL-6, IL-8, IL-10, and TNFα, or increase expression of MHC II or CD80/86 at any dose. Mφ demonstrated chemokinesis in response to ET-1 as demonstrated by increased transwell migration over 24h: no ET-1 vs ET-1 1ng/ml vs MCP-1: 68 ± 8 vs 390 ± 103* vs 158 ±12 (mean ± SD, Mφ/hpf, n=4, *p < 0.05). Interestingly, Mφ removed ET-1 from the media by 70 ±14% - 100 ± 0% (mean ± SD, n=12) at 24h across the dose range. Mφ mediated ET-1 removal was inhibited by both PI (ET-1 1ng/ml vs ET-1 1ng/ml + PI: 65 ± 4 vs 34 ± 11%) and BQ788, but not BQ123 (ET-1 vs ET-1+BQ788 vs ET-1+BQ123: 38 ± 10% vs 27 ± 6%* vs 46 ± 3% mean ± SD, n=4, *p < 0.05). ET-1 removal was not affected by CQ.

In vivo, ET-1 caused a pressor response in control (saline treated) and sham (DT treated strain control) mice at the highest ET-1 dose (see Table). By contrast, DT treated CD11b-DTR mice were more sensitive to the pressor effect of ET-1 (responded at lower doses), and displayed a greater peak change in BP that lasted for a greater duration (expressed as area under curve) than control and sham-operated. Responses to angiotensin II were not altered.

ET-1(nmol/kg) Maximal BP (mmHg) Duration of Pressor Response (AUC)

Control Sham DT Control Sham DT0.01 -1 ± 4 1±1 +1 ± 4 +40 ± 69 105±148 +150 ± 1310.1 -1 ± 3 1±1 +10 ± 5 +294 ± 509 79±34 +2665 ± 21081 +25 ± 5 18±3 +41 ± 15*† +12287 ± 4234 6059±1 +49706

30796§†±

CD11b-DTR mice BP response to ET-1. Data are mean SD (n=3, *p<0.01, § p<0.05 vs.control,†p <0.01 vs. sham)

ConclusionsHuman Mφ possess ETAR and ETBR. Mφ are not classically activated by ET-1 but do migrate to an ET-1 chemotactic gradient. In addition, Mφ remove ET-1 from their environment by ETB

receptor mediated uptake and proteolytic degradation. In vivo, murine Mφ depletion augments the pressor effects of ET-1. Thus, these data suggest a potential role for Mφ in regulating the effects of ET-1 mediated hypertension.

A7. FGF-23 is an independent predictor of LVH in patients with CKD

K.K. Stevens1,2, E.P.McQuarrie1,2, R.K. Patel1,2, Mark P.B.1,2 and A.G. Jardine1,2

1BHF Cardiovascular Research Centre, University of Glasgow2Renal Unit, Western Infirmary, Glasgow

Aim: CKD is associated with increased risk of cardiovascular disease (CVD). ‘Traditional’ risk factors do not explain this. Serum phosphate is an independent risk factor for CVD. Phosphate concentration is tightly regulated by several factors including fibroblast growth factor 23 (FGF-23), a novel molecule, which plays an important but ill-defined role in phosphate regulation. In CKD, levels of FGF-23 are increased. Elevated FGF-23 has been associated with left ventricular hypertrophy (LVH) in CKD. We assessed the relationship between FGF-23 and LVH in CKD using cardiac MRI.

Methods: 75 patients were recruited; 48 patients with CKD stages 3/4 and 27 ‘control’ patients with hypertension. Fasting venous blood samples were drawn for analysis and patients completed a 24 hour urine collection. FGF-23 levels were measured on plasma samples using a C-terminal human enzyme-linked immunoabsorbent assay kit. Left ventricular mass index (LVMI) was determined using a 1.5T MRI scanner. LVH wasdefined as LVMI ≥84.1g m-2 (male) or 76.4g m-2 (female).

Results: Mean age was 58.3 (±11.4) years. 76% were male. FGF-23 and phosphate levels were significantly higher in patients with CKD (238RU/ml (IQR 109.6-393.3) ‘v’12.5RU/ml (IQR1.5-36) and 1.2mMol/L ±0.2 ‘v’ 1.0mMol/L ±0.22 respectively). eGFR was significantly lower (31ml/min/1.73m2± 11 ‘v’ 92.7ml/min/1.73m2±15.3 p<0.001). LVMI was not significantly different between those with CKD and ‘control’ patients (84.5g/m2± 19.6‘v’ 87.7g/m2± 22.7). Stratifying patients according to the presence or absence of CKD and LVH revealed significant differences in the levels of FGF-23 (p<0.001). Those with CKD and LVH had the highest levels and those without CKD or LVH the lowest (277.9 (IQR158-414) ‘v’ 13.8 (IQR 1.5-38)). There was a strong negative correlation between FGF-23 level and eGFR (r=-0.6 p<0.001) and a positive correlation between LVMI and FGF-23 level (r=0.4 p=0.005). FGF-23 level also correlated positively with phosphate and negatively with vitamin D level.

In CKD, significant predictors of LVH included eGFR, FGF-23 level, systolic BP and 24 hour urinary protein excretion. On multivariate regression analysis, the significant independent predictors of LVH were renal function, systolic BP and FGF-23 level.

Conclusions: FGF-23 appears to be an independent predictor of LVH, defined by cardiac MRI, in patients with CKD. It may exert direct effects on left ventricular mass offering a potential mechanism for LVH in patients with CKD or it may simply be a marker of LVH. This was a small study which was intended to be hypothesis generating. It adds to the evidence of a link between FGF-23, CKD and LVH. Further larger studies are necessary to establish the exact nature of this relationship.

Funding: Darlinda’s Charity for Renal Research

A8. USING PCR OR ACR TO PREDICT PATIENT OUTCOMES: IMPACT OF MUSCLE MASS

Methven, S1, Traynor, JP2, Deighan, CJ3, O'Reilly, DStJ4, MacGregor, MS1

1John Stevenson Lynch Renal Unit, Crosshouse Hospital, Kilmarnock2Renal Unit, Monklands Hospital, Airdrie3Renal Unit and 4Dept of Biochemistry, Glasgow Royal Infirmary

Introduction: Protein:creatinine ratio (PCR) and albumin:creatinine ratio (ACR) are widely used to quantify proteinuria. Their ability to predict 24h urine protein excretion varies with age and gender. This may relate to differing creatinine excretion as a consequence of muscle mass. We have previously shown that adjusting PCR/ACR for estimated creatinine excretion (ECE) improves the ability to predict 24h urine protein. Our aim was to examine the impact of adjusting PCR and ACR for ECE, on the prediction of patient outcomes.

Methods: Retrospective data were available from 5586 patients attending general nephrology clinics at a city hospital. Creatinine excretion was estimated by the Cockcroft and Gault formula, using ideal, lean or actual body weight (ABW). To adjust for muscle mass, PCR/ACR were multiplied by ECE using each weight measure, thus producing 4 values for PCR and ACR. The performance of these variables was assessed using receiver operator characteristic (ROC) curves to predict outcome: all cause mortality, commencement of renal replacement therapy and doubling of serum creatinine.

Results: For all three patient outcomes, the area under the curve was higher (non- significantly) for unadjusted PCR and ACR than for the PCR or ACR values that had been adjusted for ECE (regardless of weight measurement used). A ROC curve for all-cause mortality is shown below as an example.

Conclusion: Adjusting PCR and ACR for estimated creatinine excretion does not improve their ability to predict patient outcomes, despite improving prediction of 24h proteinuria. Several explanations for this finding are possible, but practically, it may not be necessary to use different PCR/ACR thresholds according to age and gender to predict patient risk.

No external funding and no conflict of interest

A9. Aetiology of Staphylococcus Aureus Bacteraemias in Patients ReceivingRenal Replacement Therapy

Jennifer S Lees, Wendy Metcalfe, Kristjan Helgason, Jane Goddard

Background: Bacteraemia is a major cause of morbidity amongst dialysis patients, and mortality due to sepsis in this cohort is at least 100 times higher than in the general population. This is of particular concern in haemodialysis patients, in whom the use of tunnelled central venous catheters (CVCs) contributes substantially to the tally of bacteraemias. It is unclear to what extent other sources are accountable for bacteraemia in patients requiring renal replacement therapy (RRT). Our aim was to describe the aetiology of Staphylococcus aureus bacteraemias (SABs) in patients on RRT within the renal unit of the Royal Infirmary of Edinburgh (RIE).

Methods: We included patients who were receiving RRT, who had proven SABs in the period 01 Jan 2006 - 31 Dec 2010 within RIE renal unit. Episodes of SABs were identified from positive blood culture results obtained in RIE bacteriology laboratory. Results were cross-checked with the Scottish Renal Registry records. We included all modalities of RRT (transplant, peritoneal dialysis (PD) and haemodialysis (HD)). We excluded patients with acute kidney injury.For each SAB episode, we collected mode of RRT, vascular access for HD, source of SAB and time interval between SAB and death (where applicable). To calculate this interval in any patient with more than one SAB episode, the date of the final episode was used.

Results: We identified a total of 169 SABs in 108 individuals. One hundred and sixty (94.7%) of SABs were in HD patients, with 3 (1.8%) and 6 (3.6%) in PD and transplant patients respectively.The numbers of SABs in HD patients in each of the five years 2006-2010 were 41, 32,35, 23, 29 respectively. The percentage of patients dialysed via CVC also increased from20% in Jan 2006 to 31% in 2010. During the same time period the number of HD patients in the unit increased from 241 on 01 Jan 2006 to 272 on 31 Dec 2010. In this 5 year cohort at the time of SAB 27.2% had only an AVF (graft in one patient), 53.1% CVC (1.7% non-tunnelled) and 19.8% both.In all HD patients, including those with arterio-venous fistulae (AVF), 65.6% of SABs originated from tunnelled CVCs, 1.2% from non-tunnelled lines, 7% likely metastatic infection from a line, 17.5% were other infections and 11.3% were unknown. In patients with an AVF, 11.4% of SABs were due to an infected CVC, in 36.3% the source of SAB was unknown. The transplant patient and PD patients had low rates of SAB.Fifty six (52%) patients identified are now deceased. For those patients the median time interval between SAB and death was 86 days (IQR 19-287).

Conclusion: The use of central venous catheters was the single greatest contributor to SABs in this patient cohort. In the haemodialysis population, reducing our dependence on tunnelled CVCs would afford us the greatest opportunity to reduce our rate of SABs, along with patient morbidity and mortality.

A10. Renal impairment after high dose flucloxacillin and single dose gentamicin prophylaxis in patients undergoing elective hip and knee replacement.

Sudhakar Rao Challangundla1, David Knox1, Amanda Hawkins1, David Hamilton2, RobertFlynn4, Sue Robertson3, Chris Isles3.Departments of Orthopaedics1, Microbiology2 and Renal Medicine3, Dumfries andGalloway Royal Infirmary; Medicines Monitoring Unit, Ninewells Hospital, Dundee4

Background: We switched our antibiotic prophylaxis for elective hip and knee surgery from cefuroxime to flucloxacillin with single dose gentamicin in order to reduce the incidence of C.diff diarrhoea. More patients subsequently appeared to develop acute kidney injury (AKI).Methods: We examined the incidence of AKI in 193 patients undergoing elective hip or knee surgery. These patients were given the following prophylactic antibiotics: cefuroxime (n = 46); then high dose flucloxacillin (5-8g) with single dose gentamicin (n =50); then low dose flucloxacillin (1-4 g) with single dose gentamicin (n = 45); and finally cefuroxime again (n = 52).

Results: There were no statistically significant differences between the four groups by chi-square tests for age, gender, nature of operation, American Society of Anaesthesia grade, mode of anaesthesia, baseline serum creatinine, pre-operative co-morbidity, pre- operative medication and post operative hypotension. Patients receiving high dose flucloxacillin required more vasopressors during surgery (p = 0.02). The proportion of patients in each antibiotic group with any form of AKI by RIFLE criteria was: first cefuroxime (9%), high dose flucloxacillin (52%), low dose flucloxacillin (22%), second cefuroxime (14%) (p < 0.0001). Odds ratios for AKI derived from a multivariate logistic regression model, adjusted also for sex and ACE inhibitors/ARBs, with the first cefuroxime group as a reference category were: high dose flucloxacillin 14.5 (95% CI,4.2-50.2); low dose flucloxacillin 3.0 (0.8-10.9); second cefuroxime 1.9 (0.5-7.4). Three patients required temporary haemodialysis. Biopsies in two of these showed acute tubulo-interstitial nephritis. All three patients belonged to the high dose flucloxacillin group. None of the patients developed C Diff diarrhoea.

Summary: We have shown a strong association between prophylactic antibiotic regimen and subsequent development of AKI that was principally attributed to the use of high dose flucloxacillin with single gentamicin. We found no evidence to suggest that the strong association was confounded by any of the co-variates we measured.

A11. Two cases of profound metabolic acidosis associated with renal impairment and paracetamol use caused by acquired 5-oxoprolinuria.

Robert Hunter, Cate Preece and John NearyDepartment of Renal Medicine, Royal Infirmary of Edinburgh

Introduction: We present two cases in which a profound metabolic acidosis was associated with renal impairment, malnutrition and paracetamol use. In both cases there was an elevated anion gap with no obvious cause. In both, the suspected diagnosis of 5- oxoprolinuria was confirmed by the presence of large amounts of 5-oxoproline (pyroglutamic acid) in the urine.

Case 1: A 48 year-old female with a history of adult polycystic kidney disease (baseline Cr = 250 microM) presented to hospital following ten days of vomiting, abdominal pain and dyspnoea. She was malnourished (serum albumin = 20 g / L) and admitted to excessive alcohol consumption. She had been using paracetamol intermittently at therapeutic doses for back pain in the preceding two months. She was diagnosed with a lower respiratory tract infection and extracellular fluid overload and treated with ceftriaxone, diuretics and regular paracetamol. Despite an improvement in her clinical condition, she developed a progressive metabolic acidosis (serum bicarbonate = 11 mM, corrected anion gap = 25). Lactate levels were normal and there were no urinary ketones. A urinary organic acid profile detected large quantities of 5-oxoproline. Paracetamol was discontinued and her acidosis resolved completely over the next three days.

Case 2: A 67 year-old male was admitted to the orthopaedic wards with a wound infection, one month after fixation of a fractured ankle. His wound swabs grew Staphylococcus aureus and he was treated with intravenous flucloxacillin and regular paracetamol. Over the following nine days he developed a progressive impairment of kidney function and a metabolic acidosis (Cr = 295 microM, bicarbonate < 5 mM). He had an elevated anion gap (corrected anion gap = 27) with a normal lactate and a trace of ketones in his urine. His urine contained a large amount of 5-oxoproline and small amounts of ketone bodies. Paracetamol and flucloxacillin were discontinued and he received supplemental bicarbonate. His acidosis resolved over the following days.

Discussion: Accumulation of 5-oxoproline (pyroglutamic acid) is a rare but recognised cause of metabolic acidosis with an elevated anion gap. It is caused by defects in the - glutamyl cycle which may inherited or acquired (Wellner et al., 1974). The latter are commonly associated with renal impairment, malnutrition and the use of paracetamol and/ or other drugs including flucloxacillin (Fenves et al., 2006). This cause of a metabolicacidosis – which is almost certainly under-diagnosed – should always be considered in the absence of an alternative explanation for an elevated anion gap as it is easily treated by the cessation of paracetamol.

References:Fenves et al. (2006), Clin J Am Soc Nephrol 1: 441 – 447. Wellner et al. (1974), PNAS 71(6): 2505 – 2509.

Conflict of interest: none.

A12. Acute Renal Impairment in Emergency Surgical Admissions: A ProspectiveAudit of Incidence, Age, Length of Stay and Mortality

Callum Carruthers, FY2 Western Infirmary, Glasgow

Background: Acute kidney injury (AKI) has been associated with increased hospital stay and mortality, and even small increases in serum creatinine have been linked to worse outcomes1,2. This audit aims to quantify the incidence of AKI in patients admitted to a surgical receiving unit at a university teaching hospital and evaluate the relative ages of patients in this group, length of stay and mortality.

Methods: Data was collected on all patients admitted under one surgical receiving team that had a serum creatinine measurement. Data was collected for 9 receiving weeks. Base line creatinine, admission creatinine and highest recorded creatinine during admission were noted. Electronic records were revisited at the end of the data collection period when information on length of stay and mortality were documented. All information and statistics were collected using Microsoft Excel. Acute Kidney Injury was classified byUK Renal Society guidelines, based on RIFLE criteria3.

Results: Data was collected on 515 patients referred to the general surgeons. Incidence of AKI at admission was up to 5.6% and during whole admission up to 7.6%.Patients with renal impairment on admission were significantly older; 70 years vs. 49 years (p=<0.005); as were those with renal impairment across whole admission; 68 years vs. 49 years (p=<0.005).Length of stay was longer in those with AKI on admission; 13.9 days (95%CI 7.6-20.2)vs. 2.3days (95%CI 1.9-2.7); and AKI across whole admission; 13.4 days (95%CI 8.8-18.0) vs. 2.1 days (95%CI 1.8-2.4).All cause mortality was higher in the admission AKI group (17.2% vs. 3.5%) and whole admission AKI group (17.5% vs. 3.2%.)

Conclusion: Results demonstrate that AKI is prevalent in patients referred with a primarily surgical issue. Elderly patients are at higher risk and those with AKI have a significantly longer stay and increased mortality. Further research should evaluate the aetiology, recognition and management of acute kidney injury in surgical patients.

1. Chertow GM et al. Acute Kidney Injury, Mortality, Length of Stay, and Costs inHospitlized Patients. J Am Soc Nephrol 2005; 16: 3365-3370.2. Praught ML, Shilpak MG. Are small changes in serum creatinine an important risk factor? Curr Opin Nephrol Hypertens 2005; 14: 265-270.3. Lewington A, Kanagasundaram. Acute Kidney Injury. UK Renal Society. http://www.renal.org/Libraries/G u idelines/A c ute_Ki d ne y _Injury_- Final_Draft_08_January_2011.sflb.ashx (accessed 21 February 2011)

No funds were received for this work and I declare no conflict of interest.

Perc

ent s

urvi

val

A13. Buttonhole Needling Of Arteriovenous Fistulas Improves Primary Patency

Stewart Lambie, Raigmore Hospital, Inverness

Introduction and Aims: Arteriovenous fistulas (AVF) are the gold standard for vascular access in haemodialysis patients. Current needling techniques can lead to problems including pain, bruising, prolonged bleeding and aneurysm formation. AVFs are also affected by stenosis and thrombosis. Buttonhole or constant site needling has been shown to reduce pain on needling and aneurysm formation. This study aims to assess whether buttonhole needling has any effect on stenosis and thrombosis of AVFs

Methods: Subjects with existing AVFs were recruited from a district general hospital renal unit and two satellite units for a randomised controlled trial of buttonhole versus traditional needling technique. At the end of the trial after one year initial results were available. These indicated a reduction in local anaesthetic use and patient preference for buttonhole needling but as there was still a concern about an increased infection rate the results were held to be equivocal. Subjects and their medical teams were then allowed clinical freedom to choose to use either technique, as was the case for all other patients with AVFs . We now present follow up data from the original trial subjects after a further two years of follow-up.

Results: 59 patients were recruited (mean age 60 30 years; 34 male, 25 female; 15 diabetic), three patients withdrew prior to randomisation (two died, one transplanted). Patients were randomised in blocks of four for logistical reasons, 28 to intervention and28 to the control group. Three patients withdrew from the intervention group (one no reason given; one due to pain; and one had a thrombosed fistula). In the follow-up period two patients reverted to traditional technique while 10 patients converted from traditional to buttonhole. AVF survival without needing surgical or radiological intervention to maintain patency, analysed according to intention to treat, is shown in the figure below demonstrating an advantage in primary patency for buttonhole needling (p<0.05 for the difference between the two curves, hazard ratio 0.42). Over the total time period there were two documented infections in the buttonhole group and one in the traditional group.

Conclusions: Buttonhole technique reduces the need for intervention to maintain patency of AVFs. Increased infection rate has not been demonstrated by our study. Our unit has now changed its policy to buttonhole needling as our standard technique.

Fistula survival without intervention100

Buttonhole80 Traditional

60

p<0.0540

20

00 300 600 900 1200

Days

A14. Clearance of Beta 2 Microglobulin as a Marker of Dialysis Efficiency

Claudine Jennings, Iain Henderson, Alison Severn (Renal Unit, Ninewells Hospital, Dundee), Eleanor Oakes and Bill Bartlett (Dept of Biochemistry, Ninewells Hospital)

Background: Beta 2 microglobulin (B2-m) is recognised as a surrogate marker of middle molecule uraemic toxins. Levels are increased up to 60 fold in patients with end stage renal disease and high levels are associated with dialysis related amyloidosis as well as being an independent predictor of increased mortality.B2-m is not removed by low flux dialysis membranes however removal is very effective using high flux dialysers. The use of HDF rather than conventional HD also improves clearance of middle molecules and there is some evidence of improved long term outcomes with HDF over HD potentially partly attributable to increased middle molecule clearance. The Dundee dialysis units have routinely used HDF with high flux dialysis membranes in all patients for the past two years.

Method: Blood samples were taken for B2-m from all Dundee dialysis patients (including those at Perth and Arbroath satellite units) on two occasions. Pre-dialysis levels were taken in May 2010 and both pre and post dialysis levels were taken in November 2010 at the same as the routine monthly bloods. The samples were sent to a reference lab in Sheffield for analysis. In total 140 dialysis patients had complete result sets.Dialysis information for the day of sample collection including length of dialysis, patient weight and amount of UF was collected from our electronic dialysis records.Kt/V for B2-m was calculated using a formula created by Casino et al[1] as a simple tool to evaluate middle molecule clearance and this method compared favourably with more complex methods previously used. This was compared to Kt/V urea and URR which are routinely calculated from our electronic records.

Results: Baseline B2-m levels were lower in this dialysis population than that seen in previous studies (mean 22.4mg/L). Mean post dialysis level was 5.5mg/L indicating good clearance on high flux HDF. Kt/V B2-m was generally slightly higher than Kt/V urea for the dialysis session analysed. For our patient population dialysis efficiency (from both URR and Kt/V urea) was above recommended targets.

Analysis of the data is ongoing and more detailed results will be presented at SRA.

Discussion: According to current guidelines high flux HDF provides high efficiency clearance of small and middle molecules. Calculation of Kt/V B2-m is not routine and may not provide additional useful information above Kt/V urea for patients receiving HDF however this may not be the case for other forms of dialysis. Further analysis of this data set is required to look at the subset of patients whose clearance of middle molecules is poor and whether dialysis efficiency measured by Kt/V B2-m can be linked to clinical outcomes.

1. Casino et al. A simple approach for assessing equilibrated Kt/V B2-m on a routine basis. Nephrology Dialysis Transplant 2010; 25: 3038-3044.

A15. Dialysis in Octogenarians: A single centre study.

Izhar Khan and Yvonne Law, Renal Unit, Foresterhill Aberdeen AB25 2ZN

Introduction And Aims: The incidence of CKD-5 increases dramatically with age and older patients are being accepted for renal replacement therapy (RRT). Our previous research and that of others have suggested that co-morbidity rather than age alone has a significant effect on patient survival. The Scottish Renal Registry (SRR) reported a total of 546 patients over the age of 75 years out of 4278 patients receiving RRT in Scotland (population 5 million) in 2009. A study published by the SRR in 2005 showed that 213 out of 3944 patients were aged over 80 years at time of commencing RRT between the years 1994-2003. The median survival of these patients was only 328 days. The aim of our study was to determine if there has been an improvement in survival of octogenarians commencing RRT in recent years.

Methods: The Aberdeen renal Unit serves a population of 0.5 million in the North- East of Scotland. Using our renal database (Clinical Vision v4.0, CCL) we identified all patients who commenced RRT and were aged 80 years or older between the years 2001-2009. We recorded renal diagnosis, modality of treatment, presence of permanent dialysis access and date of death if applicable. Survival analysis was carried out using Kaplan-Meier method with the help of MedCalc statistical package.

Results: In the Aberdeen Renal Unit a total of 48 patients were over 80 years when commencing RRT. 37 patients had permanent access at initiation of RRT and all were commenced on haemodialysis. The median survival for all patients was 671 days in this cohort of patients compared with the earlier SRR cohort who had a median survival of 328 days. After exclusion of deaths in the first 90 days, the median survival of the Aberdeen cohort was 697 days compared with 459 days for the SRR cohort.

Conclusions: This study shows a dramatic improved survival in octogenarians commencing dialysis in recent years in our unit compared with an earlier Scottish cohort of octogenarians. The numbers treated were not different from those in the rest of Scotland, thus excluding the likelihood of selection bias. Possible causes of improved patient survival include better pre-dialysis care with planned access, early referral and attention to comorbid conditions before RRT is commenced.

A16. Assessment of cardiac troponin T in haemodialysis patients without acute coronary syndrome using a high sensitivity troponin T assay

David Counter, Gareth Jones, Kim Heathcoat, Rob Flynn, Ken Donaldson, SueRobertson, Alison Almond, Chris IslesRenal Unit and Dept of Biochemistry, Dumfries Infirmary, Dumfries; and MedicinesMonitoring Unit, Ninewells Hospital, Dundee

Background: Elevated cardiac troponin is frequently seen in haemodialysis (HD) patients in the absence of acute coronary syndrome (ACS) and have been associated with increased all-cause and cardiovascular mortality. This has presented difficulties in interpreting troponin levels when HD patients present with symptoms suggestive of ACS. The advent of new highly sensitive troponin (hsTnT) assays capable of accurately detecting even lower troponin concentrations, is likely to increase the prevalence of elevated baseline troponin levels in HD patients and further confound their use as diagnostic tools in suspected ACS.

Methods: We measured hsTnT in our HD patients using an Elecsys 2010 analyser (Roche Diagnostics). The 99th percentile upper reference limit for this assay in healthy subjects is 0.014μg/l. The limit of detection is 0.005μg/l and the assay imprecision, expressed as the coefficient of variation at the 99th percentile is 10%.

Results: All 53 hospital haemodialysis patients (59% male, mean age 66 years) were included in the study. None were experiencing chest pain at the time. Median hsTnT was0.056μg/l pre-dialysis and 0.039μg/l post-dialysis. hsTnT was elevated in 50/53 (94%) and was slightly higher in those with a history of vascular disease (0.060μg/l) than in those without (0.047μg/l) but not significantly so (p=0.40). Intra-individual variation in pre-dialysis hsTnT, comparing two measurements taken one month apart, was small and insignificant (median difference =0; IQR -0.004,0.003; p=0.49).

Conclusions: Nearly all of our HD patients, including those with no clinical evidence of vascular disease, had raised hsTnT. hsTnT is cleared by dialysis but remains elevated in most patients. There was little intra-individual variation in measurements taken a month apart. This suggests that a single hsTnT can be used as a baseline against which to judge further measurements when HD patients present with symptoms suggestive of ACS.

A17. Renal Recovery In Dialysis Patients: Myth Or Reality?

Mark Findlay and Margaret McMillan, Renal Unit, Western Infirmary, Glasgow

Anecdotal reports of recovery of renal function give hope to patients starting dialysis. We reviewed the characteristics and clinical course of patients at the Western Infirmary who stopped chronic dialysis because of renal recovery in 1970-2009.

Patients stopping dialysis were identified from the proton database. Only patients on dialysis for at least 90 days before renal recovery were included; their electronic and paper case notes were reviewed. Data were analysed using Microsoft Excel.

Only 32 patients over the 40-year period met the inclusion criteria. Seventeen were male. Eleven had glomerulonephritis and 5 renovascular disease. Median (range) age was 66 (30-77 years). Median serum creatinine on starting dialysis had been 733.5 (436-1524) micromols/l. At the time of stopping dialysis patients had dialysed for a median of 261.5 (range 95-1869) days. One month after stopping one patient had had to restart; serum creatinine in the others ranged from 197-1158 micromols/l (median376). A year after stopping 8 patients had restarted and another 4 had died. Median follow-up period is 13 years (range 1-33 years). Only 6 patients stayed alive off dialysis for at least 5 years.

Our results confirm that it is rare to stop dialysis after 90 days and rarer still to achieve sustained freedom from dialysis.