The records of 107 patients with displaced subcaptial hip fractures treated by Thompson's femoral head replacements have been reviewed to determine the mortality and certain aspects of systemic morbidity. Comparison was made with a group, matched exactly for age and sex, in which intertrochanteric fractures were treated by nail-plate fixation. The patients treated by Thompson's arthroplasty were further subdivided into two groups: one in which methylacrylic cement was used, and one in which it was not used. We found no difference in the mortality of the patients undergoing replacement arthroplasty and nail-plate fixation. Similarly there was no difference in the incidence of cerebrovascular incidents and myocardial infarction after operation. There was a higher incidence of transient cardiac failure following Thompson's arthroplasty. Hypotension during the operation occurred irrespective of whether cement was used; nor was the use of cement associated with increased mortality. However, it seems that methylacrylic cement does contribute to the cardiac failure that may occur after Thompson's arthroplasty.