This study examines the relationship between pneumatic tourniquet cuff size, occlusion pressure and the resulting pain. Two tourniquet cuff widths were used, a wide (14 cm) and a narrow cuff (7 cm). Twenty volunteers were divided into two groups for tourniquet application: a pressure group in which the tourniquet was inflated to a pressure equal to the systolic pressure + 100 mmHg, and a saturation group in which the tourniquet was inflated to 10 mmHg above the loss of arterial pulse, as indicated by cessation of pulse waveform on an oximeter. According to a randomised cross-over protocol, subjects were studied using wide and narrow cuffs simultaneously and/or successively on both arms. Pain was assessed by subjects by means of a visual analogue score (0-10 cm). Occlusion pressures were similar for all volunteers in the pressure group and significantly higher than those in the saturation group with both the wide and narrow tourniquets. The wide cuff data turned out to be significantly lower than the narrow cuff results. Subjects in the pressure group could tolerate pain with the narrow cuff for significantly longer than with the wide cuff. However, in the saturation group, volunteers tolerated the wide cuff for longer. Pain intensity increased more rapidly in those in the pressure group with the wide cuff than with the narrow cuff. In contrast, volunteers in the saturation group found the narrow cuff to be more painful than the wide cuff. In conclusion, this study has shown that a wide tourniquet cuff is less painful than a narrow cuff if inflated at lower pressures and at these lower pressures it is still effective at occluding blood flow.