Platelet function testing in uraemic patients

Abstract

Introduction: Chronic renal failure (CRF) is associated with excessive bleeding for a variety of reasons. Platelet dysfunction is probably the most consistent and important feature, particularly platelet-platelet and platelet-vessel wall interactions. The skin bleeding time (SBT) is the best established predictor of bleeding in uraemic patients but suffers from poor reproducibility and accuracy. Several newer rapid assays of platelet function are able to provide a means of assessing primary haemostasis, but have not been specifically assessed in uraemic patients. Methods: A single centre, prospective cohort study of patients referred to a tertiary nephrology unit. Patients with both acute and chronic renal impairment were recruited. Laboratory parameters analysed included full blood count, serum creatinine and urea, calculated glomerular filtration rate (GFR) (Cockcroft formulae) APTT PT fibrinogen, SBT, whole blood platelet aggregation (WBPA), platelet function analyzer (PFA-100), thromboelastograph (TEG) and cone platelet analyzer (CPA). Results: This study included 42 patients: nine with CRF (GFR < 30 ml/ min) who were not receiving dialysis; 23 with CRF receiving dialysis; seven who presented in acute renal failure; and three assessed with normal renal function but with nephrotic syndrome and who presented prior to renal biopsy. Twenty-two patients were on low-dose aspirin and four patients were on clopidogrel. There was weak correlation between calculated GFR and SBT (r 2 = 0.1564) with even poorer correlation with serum creatinine and no correlation with urea levels. Overall, PFA-100 was a poor predictor of SBT. Seven of 12 patients with SBT <7 min had abnormal WBPA, 24 of 26 had abnormal WBPA if SBT >7 min. Of those with SBT <7 min, five had abnormal CPA, and >7 min, 15 of 21 had abnormal results. Nineteen patients had abnormal TEG tracings. Six patients underwent renal biopsies with one bleeding complication. Conclusions: In this pilot study we found that prolonged SBT was not predicted by serum creatinine or calculated GFR. Within the limitations of this study, an alternative in vitro test to replicate the SBT has not been identified. © W. S. Maney & Son Ltd 2008.