The diagnostic usefulness of the biopsy cylinder and biopsy-induced complications were investigated for 458 percutaneous renal biopsies (315 native, 143 transplant kidneys) on 339 patients (average age, 44.6 +/- 18.5 years) under two different biopsy regimes (regime I, manual biopsy technique with Tru-Cut needle, 14 gauge; regime Ii, automated biopsy technique using a Biopty instrument and adapted biopsy needle, 18 gauge). In 435 (95%) of the biopsies, kidney tissue with 9.09 +/- 5.28 glomeruli was obtained (regime I, 93.5%, 9.5 +/- 4.9 glomeruli; regime II, 96.5%, 8.7 +/- 5.6 glomeruli; P > 0.05). Neither with native nor with transplant kidneys was there any evident advantage for a particular regime in terms of the diagnostic usefulness of the cylinder. Bleeding complications (perirenal haematomas, bleeding into the renal pelvis, blood clots in the urinary bladder) were observed in 69 (15.1%) patients (regime I, 15.6%; regime II, 14.6%; P > 0.05). Clinically relevant bleeding complications were significantly rarer under regime II (9.1% versus 3.5%; P < 0.05). Complications were less frequent with transplant than with native kidneys (12.6% versus 16.2%). Doppler sonography of the biopsied transplant kidneys revealed arteriovenous fistulae in nine cases (6.3%; regime I, 8.0%; regime II, 5.4%). In general, use of an automated biopsy instrument and a thinner biopsy needle reduced the number of significant complications following percutaneous renal biopsy, while achieving comparably diagnostic efficacy.