RESUMO
AIM: To derive a simple risk score to predict the individual risk of major complications for patients undergoing a percutaneous renal biopsy procedure of native kidneys. METHODS: The risk score was derived from a cohort of 1205 adult patients subjected to percutaneous renal biopsy and assigned to training and validation datasets. Factors associated with major complications were derived from univariate analysis and then modelled by stepwise multivariate logistic regression. Based on the odds ratio, independent predictors were assigned a weighted integer. The risk score is calculated from the sum of the integers. RESULTS: The overall incidence of major complications was 3.2%. Independent factors associated with MC were lower pre-biopsy haemoglobin, lower platelets, higher blood urea nitrogen, documented chronic kidney disease features in pre-biopsy ultrasound (US) and the presence of haematoma in the post-biopsy US. A score for pre-biopsy evaluation included the first four predictors and stratified patients in three categories with increasing risk at higher scores (low-risk 0.1%, moderate-risk 3.0% and high-risk 26.1%). The score demonstrated good discriminative power (AUC = 0.872). The addition of post-biopsy US findings increased the discriminative power (AUC = 0.938). A higher post-biopsy risk score was also associated with a higher incidence of MC (low-risk 0.2%, moderate-risk 2.7%, high-risk 16.9%). CONCLUSION: The risk of major complications after a percutaneous renal biopsy can be assessed by a simple risk score calculated from readily available information.