Native renal biopsies: complications and glomerular yield between radiologists and nephrologists.

Kidney biopsies have been taken over by radiologists at many institutions. This tendency for radiology referral by nephrologists has two main drawbacks: 1) No follow-up care from the physician performing the procedure and 2) Loss of required procedure training for nephrology fellows. This study assesses for differences in glomerular yield and immediate procedure-related complications between radiology-performed (RP) and nephrology-performed (NP) percutaneous native kidney biopsies at our institution for an 11-month period. All biopsies were done with real time ultrasound guidance. Further analysis was done to look for any correlation between the number of needle insertions ("passes"), needle size and procedure-related complications (graded on a score from 04). A total of 37 native kidney biopsies were performed during the study period. Fourteen biopsies (38%) were performed by radiology while a nephrology fellow performed 23 biopsies (62%). The mean glomerular count for RP biopsies was 15 +/- 10 and for NP 16 +/- 11 (p=ns). The number of passes ranged from 1-5 in the RP group with a total of 33 passes in 14 patients (mean 2.36 passes/patient). Passes ranged from 1-6 in the NP group with 57 passes in 23 patients (mean 2.48 passes/patient). Mean complication scores were similar in both groups. However, severe complications (score 4) were significantly lower in nephrology fellow performed biopsies (p = 0.001). Despite similar pre-biopsy risk assessment and treatment protocols with real-time ultrasound guidance of all biopsies, these results show no statistically significant differences in glomerular yield or overall complication scores but did demonstrate a significantly higher rate of severe complications (score 4) in the RP performed biopsies. Nephrology fellows perform native kidney biopsies at a level equal to or superior to radiologists.