Background: Previous studies have shown that bleeding times have positive predictive values of only 5% for perioperative bleeding in unselected populations. Nevertheless, performing bleeding times prior to all renal biopsies is common in nephrology practice.
Methods: We report complications of 112 renal biopsies done at Walter Reed Army Medical Center (WRAMC) from 1996-99 performed without preceding bleeding times. Renal biopsies were done only on normotensive (<140/90) patients who had not recently been taking aspirin or non-steroidal anti-inflammatory agents, under real-time ultrasound guidance with automated 16 g (WRAMC) spring-loaded guns. High-risk patients (with serum creatinine > or = 3 mg/dl or creatinine clearance < or =30 cc/min by Cockroft-Gault formula, N=18, 16%) at WRAMC were treated with pre-renal biopsy estrogens or DDAVP. Factors were tested for their association with complications after renal biopsy using Chi Square testing for categorical variables and student's t-test for continuous variables. A stepwise logistic regression model was used to test for independent significance of factors.
Results: There were two cases each of gross hematuria and inadequate tissue (1.8% each). There were no transfusions or deaths. In univariate analysis, male gender and lower serum creatinine level at time of biopsy were significantly associated with increased risk of complications after biopsy. However, these factors were not significant in logistic regression analysis.
Conclusion: This study suggests that the use of bleeding times does not significantly alter the major complication rates associated with percutaneous real-time ultrasound guided renal biopsy.