Session VIII - Reconstruction

*Estimation of the Risk of Nonunion after a Fracture of the Clavicle

Christopher M. Robinson, FRCS; Charles M. Court-Brown, MD; Margaret M. McQueen, MD; Edinburgh Orthopaedic Trauma Unit, Royal Infirmary of Edinburgh, Edinburgh, United Kingdom (a-Scottish Orthopaedic Research Trust into Trauma)

Purpose: Nonunion is usually regarded to be an uncommon complication of clavicular fractures, although recent studies have suggested that the incidence may be higher than previously reported. The aim of this prospective cohort study was to determine the rate of nonunion after a clavicular fracture and to examine in detail the effect of a wide variety of patient- and injury-related factors on the risk of this complication.

Methods: Over a 51-month period, we performed a prospective, observational cohort study of a consecutive series of 868 patients (638 male and 230 female with a median age of 29.5 years and an interquartile range of 19.25 to 46.75 years) with a radiologically confirmed fracture of the clavicle. All patients were treated at a single institution, which functioned as the sole source of trauma care for a well-defined local catchment population. Eight patients were excluded from the study, as they received immediate surgery, for either skin compromise, floating shoulder, or because they had multiple injuries. All the remaining patients were initially treated nonoperatively. All patients were prospectively reviewed both clinically and radiographically at 6, 12, and 24 weeks after injury. We defined fracture union to have occurred when there was evidence of bridging callus on radiographs and there was no evidence of mobility or pain on stressing the site of the fracture. A nonunion was diagnosed if the fracture remained unhealed at 24 weeks after injury. We used a Cox proportional hazards model of survivorship to examine the effect of a wide variety of patient- and injury-related putative factors on the risk of fracture nonunion. All variables that were significantly predictive of nonunion on univariate analysis were included in a multivariate model (with use of forward conditional stepwise methodology) to determine those factors that were independently predictive of nonunion and to generate models to estimate the risk of occurrence of this complication.

Results: There were 581 fractures in the diaphysis, 263 fractures in the outer fifth of the clavicle and 24 fractures in the medial fifth. Fractures of the diaphysis occurred most commonly in younger men (median age of 25 years and interquartile range of 17 to 39.5 years), whereas fractures of the outer fifth were commonest in women with a median age of 40 years and an interquartile range of 26 to 56 years. On survivorship analysis, the overall incidence of nonunion within our study population at 24 weeks after fracture was 6.2% (95% confidence interval [CI], 4.2% to 8.2%). At 24 weeks, 4.4% (CI, 2.4% to 6.4%) of diaphyseal fractures and 11.5% (CI, 5.9% to 17.1%) of lateral fifth fractures remained ununited.

On univariate analysis, after a diaphyseal fracture the risk of nonunion was significantly increased only by advancing age, female sex, displacement of the fracture, and the presence of comminution (P <0.05 in all cases). Further stratification to take into account the degree of fracture displacement did not significantly add to the predictive power for nonunion. On multivariate analysis, all of these factors remained independently predictive of nonunion (P <0.05), and, in the final model, the risk of nonunion was increased by lack of cortical apposition (relative risk [RR] = 0.43; CI, 0.34 to 0.54), female sex (RR = 0.70; CI, 0.55 to 0.89), the presence of comminution (RR = 0.69; CI, 0.52 to 0.91), and age in years (RR = 0.99; CI, 0.98 to 0.99).

On univariate analysis, after a lateral fifth fracture the risk of nonunion was significantly increased only by advancing age and displacement of the fracture (P <0.05 in both cases). As with diaphyseal fractures, further stratification of the extent of displacement did not significantly add to the predictive power for nonunion. On multivariate analysis, both of these factors remained independently predictive of nonunion (P <0.05), and in the final model, the risk of nonunion was increased by lack of cortical apposition (RR = 0.38; CI, 0.25 to 0.57) and age in years (RR = 0.98; CI, 0.97 to 0.99).

Discussion: Our findings confirmed that nonunion at 24 weeks after a clavicular fracture is an uncommon occurrence, although the incidence is higher than previously reported. There are sub-groups of patients who appear to be predisposed to develop this complication, either from intrinsic factors, such as their age or sex, or from the type of injury that they sustain. Multivariate models to estimate the risk of nonunion after diaphyseal and lateral-end fractures were developed.

Conclusions: Our study identified patients at particular risk of nonunion after a fracture of the clavicle. The predictive models that we have developed may be used clinically to counsel patients about their risk of developing this complication immediately after this injury. In view of the increased probability of nonunion in the high-risk sub-groups that we have identified, it is possible that internal fixation of these fractures might be preferable as a primary treatment in these patients. We are about to initiate a prospective randomized controlled trial of primary operative versus nonoperative treatment in stratified high-risk groups to evaluate this conclusion further.

* If noted, the author indicates something of value received. The codes are identified as a-research or institutional support; b-miscellaneous funding; c-royalties; d-stock options and e-consultant or employee.