A method for the assessment of the accuracy of reduction in postreduction roentgenograms was developed. The final clinical results, one to 7.5 years after injury, were studied in 117 patients with 118 displaced bimalleolar or trimalleolar fractures at the ankle treated at the Los Angeles County-University of Southern California Medical Center between January 1966 and December 1970. Three variables were found to significantly affect the final clinical results: (1) the amount of talar displacement prior to reduction; (2) the type of fracture; and (3) the presence of rupture of the deltoid ligament.