Session III - Upper Extremity
Fri., 10/8/04 Upper Extremity, Paper #23, 4:01 pm
Sequelae after Clavicular Fractures: A Three-Year Prospective Study of the Natural History and Predictive Risk Factors Associated with Nonoperative Treatment
Purpose: Clavicular fractures are frequent injuries; they constitute approximately 4% of all fractures in adults and about 35% of all fractures that occur in the shoulder region. The aim of this prospective study was to describe the natural history of clavicular fractures after nonoperative treatment and to identify predictive risk factors. In particular, the following symptoms were investigated: pain at rest, pain during activity, cosmetic defects, reduction in strength, paraesthesia, and nonunion until 24 weeks after injury.
Methods: During a 3-year period, 245 consecutive patients 15 years of age with a radiographically verified fracture of the clavicle were included in a prospective study. All patients received the same nonoperative treatment, including wearing a sling for a week and taking analgesics. At the time of initial treatment, the patients answered a specific questionnaire and were interviewed by one of the authors to clarify different aspects and details related to the fracture. Additional standardized clinical examinations were carried out at 1, 4, 8, 12 and 24 weeks by the same examiner for all patients. Radiographic follow-up examinations were carried out until 24 weeks and included anteroposterior 0° and 45° tilted views. The baseline data were age, sex, smoking, fractured side and dominant side, fracture location, fracture type, and shortening, displacement, or bony contact between the fragments. The endpoint sequelae were dichotomous, whereas other endpoints (minimum pain/pain at rest, maximum pain/pain during activity, strength reduction, and cosmetic defects) were based on answers recorded on a visual analog scale (0 to 10). Paresthesia was recorded as yes if the patient had paresthesia in the arm or hand or both. Nonunion was recorded as yes if the fracture had not healed after 24 weeks.
Results: A total of 222 of 245 patients attended the final follow-up at 24 weeks. There was no difference in demographic and baseline data between the 245 included and the 222 who attended the final follow-up. The median time to complete recovery without sequelae was 16 weeks with a minimum of 3 weeks. Ninety-three of the 222 patients (42%) who were examined at 24 weeks still had sequelae. Displacement of more than one bone width was the strongest radiographic risk factor for symptoms and sequelae. A comminuted fracture and older age were significantly associated with an increased risk for symptoms remaining at 24 weeks. Shortening did not predict the functional outcome, nor did the site of the fracture in the clavicle. Nonunion occurred in 15 of 222 patients (7%) and was significantly more common in women (13%) than in men (3%). Smokers did not have a higher occurrence of nonunions.
Conclusion/Significance: The proportion of symptomatic patients was larger than expected. This was especially true for patients with risk factors such as comminuted or displaced fractures and for elderly patients. Patients with cosmetic defects and pain at rest at 3 months continued to have symptoms at 6 months. Why nonunion was more common in women could not be explained. Primary displacement was the strongest radiographic risk factor for sequelae at 24 weeks. Both radiographic projections used (0° and 45° tilted view) provided important information because transversally placed fragments and ventrodorsal displacement cannot be visualized in the 0° view. Whether a more active treatment protocol including surgery would have reduced the proportion of patients with sequelae can obviously not be answered by the present study. However, it seems reasonable for future studies to evaluate the use of surgical treatment for patients with the defined risk factors shown in the present study.