RESUMO
PURPOSE: Renal biopsy (RB) is occasionally performed in critically ill patients. The safety and impact of RB in this setting have not been reported. METHODS: A 10-year (2000-2009) retrospective multicentre study was conducted in ten French intensive care units (ICU) on patients who underwent RB during their management. Medical files were retrieved for data analysis. RESULTS: Seventy-seven patients underwent an RB of which 68 (88 %) were on a native kidney and 9 (12 %) on a transplanted kidney. Percutaneous ultrasound-guided RB was used in most cases (87 %). Fifty-seven per cent of the patients were on mechanical ventilation at the time of RB. RB-related complications occurred in 17 (22 %) patients, two were graded as severe (requirement for kidney embolization, eventually successful). In 35 (51 %) non-transplanted patients, RB established a specific diagnosis other than acute tubular necrosis (ATN), which was diagnosed in only 18 % of patients. In the remaining patients, only non-specific lesions were observed. Therapeutic modifications followed RB in 14 (21 %) non-transplanted patients. Presence of signs of systemic disease involving the renal tract, occurrence of renal failure before hospital admission, and absence of any factor usually associated with ATN significantly predicted the presence of a specific diagnosis at RB other than ATN. CONCLUSIONS: In this cohort, the contribution of RB to diagnosis and treatment was undeniable, but at the expense of frequent adverse events although most of them were not considered severe.