Predictors of bleeding complications in percutaneous ultrasound-guided renal biopsy

Abstract

The risks associated with performing a percutaneous renal biopsy have substantially decreased in the past two decades because of technical advances in the method. However, bleeding complications still occur, resulting in increased hospital stay and treatment costs. We investigated the predictive value of demographics (age, gender), clinical data (blood pressure), baseline chemistry (hemoglobin/hematocrit, prothrombin time, partial thromboplastin time, bleeding time, serum creatinine, daily proteinuria), and needle size for the risk of major (need for blood transfusion, nephrectomy, or angiography) or minor (no need for any intervention) postrenal biopsy bleeding complications. This was a prospective cohort study of 471 patients who underwent ultrasound-guided biopsy of native kidney by automated needle in a single center; all biopsies were performed by two experienced nephrologists. Patients with transplant kidneys were excluded from the study. Predictors of postbiopsy bleeding were assessed by multiple linear and multivariate logistic regression analysis. Data are presented as unadjusted (OR) and adjusted odds ratios (AOR) with 95% confidence intervals (CI). The study cohort consisted of 471 (277 males, 194 females) patients. Of these, 161 (34.1%) experienced postbiopsy bleeding [157 (33.3%) hematomas, 2 (0.4%) gross hematuria, 2 (0.4%) arteriovenous fistula]. Major complications were seen in 6 (1.2%) patients (blood transfusion, N= 2; angiography, N= 3; nephrectomy, N= 1), but no deaths occurred. The risk of postbiopsy bleeding was higher in women (39.7% women, 30.3% men, AOR 2.05, 95% CI 1.26 to 3.31, P= 0.004), younger subjects (35.0 +/- 14.5 years vs. 40.3 + 15.4, AOR 0.80, CI 0.68 to 0.94, P= 0.006), and patients with higher baseline partial thromboplastin time (102.7 + 11.8% vs. 100.1 + 10.0%, AOR 1.26, CI 1.02 to 1.54, P= 0.032). These findings were independent of size of hematoma. Although the methods for performing a percutaneous renal biopsy have improved in the past two decades, renal biopsy is still not a risk-free procedure. Of the data routinely collected for potential predictors of postbiopsy bleeding complications, only gender, age, and baseline partial thromboplastin time show a significant predictive value. The other variables investigated do not have any predictive value.