Bleeding Complications of Native Kidney Biopsy: A Systematic Review and Meta-analysis

Background Kidney biopsy provides important information for nephrologists, but the risk of complications has not been systematically described. Study Design Meta-analysis of randomized controlled trials and prospective or retrospective observational studies. Setting & Population Adults undergoing native kidney biopsy in an inpatient or outpatient setting. Selection Criteria for Studies MEDLINE indexed studies from January 1980 through June 2011; sample size of 50 or more. Intervention Native kidney biopsy with automated biopsy device and real-time ultrasonographic guidance. Outcomes Macroscopic hematuria and erythrocyte transfusion rates and factors associated with these outcomes. Results 34 studies of 9,474 biopsies met inclusion criteria. The rate of macroscopic hematuria was 3.5% (95% CI, 2.2%-5.1%), and erythrocyte transfusion was 0.9% (95% CI, 0.4%-1.5%). Significantly higher rates of transfusion were seen with the following: 14-gauge compared with smaller needles (2.1% vs 0.5%; P = 0.009), studies with mean serum creatinine level ≥2.0 mg/dL (2.1% vs 0.4%; P = 0.02), ≥50% women (1.9% vs 0.6%; P = 0.03), and ≥10% of biopsies for acute kidney injury (1.1% vs 0.04%; P < 0.001). Higher transfusion rates also were observed in studies with a mean age of 40 years or older (1.0% vs 0.2%; P = 0.2) and mean systolic blood pressure ≥130 mm Hg (1.4% vs 0.1%; P = 0.09). Similar relationships were noted for the macroscopic hematuria rate with the same predictors, but none was statistically significant. Limitations Publication bias, few randomized controlled trials, and missing data. Conclusions Native kidney biopsy using automated biopsy devices and real-time ultrasonography is associated with a relatively small risk of macroscopic hematuria and erythrocyte transfusion requirement. Using smaller gauge needles may lower complication rates. Patient selection may affect outcome because studies with higher serum creatinine levels, more women, and higher rates of acute kidney injury had higher complication rates. Future studies should further evaluate risk factors for complications. Kidney biopsy provides important information for nephrologists, but the risk of complications has not been systematically described. Meta-analysis of randomized controlled trials and prospective or retrospective observational studies. Adults undergoing native kidney biopsy in an inpatient or outpatient setting. MEDLINE indexed studies from January 1980 through June 2011; sample size of 50 or more. Native kidney biopsy with automated biopsy device and real-time ultrasonographic guidance. Macroscopic hematuria and erythrocyte transfusion rates and factors associated with these outcomes. 34 studies of 9,474 biopsies met inclusion criteria. The rate of macroscopic hematuria was 3.5% (95% CI, 2.2%-5.1%), and erythrocyte transfusion was 0.9% (95% CI, 0.4%-1.5%). Significantly higher rates of transfusion were seen with the following: 14-gauge compared with smaller needles (2.1% vs 0.5%; P = 0.009), studies with mean serum creatinine level ≥2.0 mg/dL (2.1% vs 0.4%; P = 0.02), ≥50% women (1.9% vs 0.6%; P = 0.03), and ≥10% of biopsies for acute kidney injury (1.1% vs 0.04%; P < 0.001). Higher transfusion rates also were observed in studies with a mean age of 40 years or older (1.0% vs 0.2%; P = 0.2) and mean systolic blood pressure ≥130 mm Hg (1.4% vs 0.1%; P = 0.09). Similar relationships were noted for the macroscopic hematuria rate with the same predictors, but none was statistically significant. Publication bias, few randomized controlled trials, and missing data. Native kidney biopsy using automated biopsy devices and real-time ultrasonography is associated with a relatively small risk of macroscopic hematuria and erythrocyte transfusion requirement. Using smaller gauge needles may lower complication rates. Patient selection may affect outcome because studies with higher serum creatinine levels, more women, and higher rates of acute kidney injury had higher complication rates. Future studies should further evaluate risk factors for complications.