A total of 112 women was entered into a randomised study to investigate the effects of suction drainage on haematoma formation in breast biopsy wounds. Of 107 evaluable patients, 52 received drainage and 55 had no drain. The wounds were examined 1 week after operation when the volume of wound haematoma was measured using ultrasonography. Haematomas were present in 48 (87%) patients with undrained wounds compared with 34 (65%) patients with drained wounds (P = 0.014, chi 2 test). Median haematoma volume was 20 ml (range 0-172.5 ml) in the undrained group compared with 2 ml (range 0-100 ml) in the drained group (P = 0.001, Mann-Whitney U test). There was one wound infection in each group. There was no difference in median pain score or duration of pain between the groups. While suction drainage significantly reduces the incidence and volume of wound haematomas, this does not appear to influence outcome.