In eight healthy patients, the influence of the train-of-four (TOF) response of prolonged neuromuscular monitoring and of different peripheral temperatures was studied during normal core temperature. Anaesthesia was induced and maintained with midazolam-fentanyl and a 70/30% mixture of nitrous oxide and oxygen. The mechanical TOF response of the adductor pollicis muscle (twitch tension and TOF ratio), was recorded simultaneously in both hands using supramaximal TOF stimulation of the ulnar nerve at the wrist. One arm was kept normothermic. The other arm was cooled using cold infusions and cold packings. Skin, muscle and core temperatures were continuously measured. In the normothermic arm (skin temperature >32.0d̀C), the twitch tension and TOF ratio were unchanged following 130–230 min of continuous nerve stimulation. In the hypothermic arm the twitch tension and TOF ratio showed only minor variations above a skin temperature of 32.0d̀C (corresponding to a mean muscle temperature of 34.50.3d̀C). Below a skin temperature of 32.0d̀C a progressive decrease in TOF response was recorded. A linear relationship was found between skin temperature and TOF response as well as between muscle temperature and TOF response. At a skin temperature of 27.0d̀C (corresponding to a mean muscle temperature of 30.80.4d̀C), an approximate 20d̀ reduction in twitch tension and a 10% decrease in TOF ratio were recorded with a considerable interindividual variation. We conclude that prolonged TOF nerve stimulation does not change the mechanical twitch response in patients with a normal central and peripheral temperature. A peripheral skin temperature below 32.0d̀C with sustained and normal body temperature is, however, associated with changes in both twitch tension and TOF ratio that may be a source of error when evaluating neuromuscular function.