Forced-air warming is commonly used to warm patients intraoperatively, but may not achieve normothermia during a short procedure. Comparative trials of a new radiant warming device in general anaesthesia (Suntouch, Fisher and Paykel, Auckland, New Zealand) have had conflicting results. We conducted a randomized controlled trial to compare the efficacy and thermal comfort of the Suntouch radiant warmer and forced-air warming in patients at high risk of hypothermia during neuraxial blockade. With ethics committee approval, 60 patients having transurethral resection of the prostate under spinal were randomized to either radiant warming or forced-air warming. All intravenous and irrigation fluids were warmed but pre-warming was not used. The final intraoperative rectal temperatures for the radiant warming and forced-air warming groups were 36.1 degrees C and 36.4 degrees C respectively (P= 0.03). A large proportion of patients in both groups (46% and 33% respectively, P=0.3) were hypothermic (<36 degrees C) on arrival in the post-anaesthesia care unit. No other patient variables were significantly different. Neither warming device reliably prevented hypothermia, although forced-air warming was slightly superior.