Displaced Intra-articular Calcaneal Fractures: Variables Predicting Late Subtalar Fusion

OBJECTIVEThe goal of the current study was to analyze the prospective clinical outcome of patients who failed closed or open treatment of a displaced intra-articular calcaneal fracture. This cohort of patients required a secondary subtalar fusion by distraction bone-block arthrodesis. DESIGNReview of prospective, randomized trial database. SETTINGFour level I trauma centers. PATIENTSBetween April 1, 1991 and December 31, 1997, 424 patients with 471 displaced intra-articular calcaneal fractures were involved in a large, multicenter, randomized trial. Forty-four patients who required subtalar fusion following initial treatment of a displaced intra-articular calcaneal fracture were compared to the population of patients who did not require subtalar fusion. The variables compared between the two groups included Böhler angles, two computed tomography classification systems, and clinical scores including SF-36, visual analogue score, and oral analogue score. INTERVENTIONSubtalar distraction bone-block arthrodesis with tricortical bone graft was used in all 45 feet. MAIN OUTCOME MEASUREMENTSThe following were examinedx-ray fracture classification, specifically Böhler angles and Essex-Lopresti classification; computed tomography classification, specifically Sanders and Crosby; clinical scores, specifically validated visual analogue score, general health survey scores, oral analogue score, and other factors (i.e., patient demographics including age, sex, profession, smoking history, and Worker's Compensation Board involvement. RESULTSInitial treatment of the 44 patients in our study was nonoperative in 37 (84%) patients and operative (open reduction and internal fixation) in 7 (16%) (1 patient had bilateral heel fractures). Patients requiring fusion differed demographically from those patients not requiring fusion. Mean age was 39 years in both the fusion and nonfusion group. The fusion group had 97% males, whereas the nonfusion group had 89% males. Sixty-four percent of the fusion patients were Worker's Compensation Board claims, whereas 35% of the nonfusion group were Worker's Compensation Board claims. Of those that required fusion, 77% were heavy laborers. On average, the fusion group had a Böhler angle 15° less than the nonfusion group. Forty-six percent of the fusion patients were Sanders-type IV initial fractures. Logistic regression analysis revealed that the primary predictors of requiring fusion were Worker's Compensation Board status (odds ratio = 3.03,