A randomised, single-blind study was conducted on 802 parturient women who required epidural analgesia, to compare open-end (single hole) with closed-end (three lateral holes) epidural catheters. The complication rate after catheter insertion was not statistically different between the two groups, but the number of unsatisfactory blocks was significantly higher in the open-end group (p less than 0.001). The closed-end catheters were easier and less painful to place, but gave a higher incidence of bloody taps. The open-end catheters caused sensory blockade to be more frequently unilateral and more frequently missed sensory segments. This resulted in a significantly higher number of open-end catheters that required replacement (p less than 0.001). Open-end catheters despite their theoretical advantages in the detection of intravenous and subarachnoid placement caused an unacceptably high incidence of unsatisfactory sensory blockade.