Case Series Evaluating the Operative and Nonoperative Treatment of Scapular Fractures

STATISTICAL ANALYSIS

Statistical analysis was undertaken with GraphPad software. Associations were tested between positive predictive variables and functional outcomes. Variables included gender, mechanism, fracture classification, patient comorbidities, social factors, associated injuries, and type of treatment. A Mann-Whitney rank test was used to test for associations between nonparametric variables, including patient age. In all cases, P < .05 was considered significant.

RESULTS

Complete clinical and radiographic data were available for 594 patients. This included 462 men and 132 women, with a mean age of 42.8 years (range, 15-92 years). Twenty-four patients (4.0%) sustained bilateral fractures, and 31 fractures (5.0%) were open. All fractures healed primarily. A total of 153 patients completed the ASES questionnaire at a mean of 62 months after injury (Table 1). This group was similar to the entire population with respect to age, gender, and type of treatment. In all, 135 patients had been injured by a high-energy mechanism (88%), and the fracture pattern as per the Arbeitsgemeinschaft für Osteosynthesefragen/Orthopaedic Trauma Association (AO/OTA) classification consisted of 14A (no glenoid involvement) (n = 139; 91%) and 14B/C (glenoid involvement) (n = 14; 9.2%).¹⁹ The mean ASES score for our entire sample was 79.3 (minimally functionally impaired). In all, 117 patients (76%) reported minimal functional deficit (ASES, 61-100), 29 (19%) reported moderate functional deficit (ASES, 31-60), and only 7 (4.6%) reported maximum functional deficit (ASES, 0-30). Gender and age were not associated with functional outcome scores.

Table 1. Patient Demographics and Etiology of Scapula Fractures.

Fifteen patients (9.8%) were treated surgically. They had a higher mean ASES score vs non-surgically treated patients (92.1 vs 77.9; P = .03) (Table 2). However, when patients were divided into 14A and 14B/C fracture patterns, there was only a significant advantage in outcome scores for operative vs nonoperative care in the 14B/C classification (96.0 vs 75.7; P < .05); meanwhile, surgery for scapular body fractures (14A) was not associated with better outcome scores (90.2 vs 78.3; P = .14). Unfortunately, assessment of these comparisons within classification groups resulted in underpowered analyses for these small groups.

Table 2. Number of ASES Surveys Completed and Mean ASES Score for Each Treatment Type and Fracture Classification

^aP < 0.05.

Abbreviation: ASES, American Shoulder and Elbow Surgeons.

Table 3 shows the ASES scores for patients with various types of associated chest and shoulder injuries. Only 7 patients (4.6%) had injuries isolated to the scapula. Thirty-three patients (22%) had associated clavicle fractures, and 102 patients (67%) sustained concomitant chest wall injuries, including rib fractures (n = 88) and pulmonary injuries (n = 71). Patients with associated chest wall injuries did not have worse mean ASES scores than those without chest wall injuries (80.9 vs 78.2; P = .49). Additionally, patients who had concomitant clavicle fractures did not report worse scores than those who did not (83.2 vs 78.6; P = .46).

Table 3. Concomitant Injuries and Mean American Shoulder and Elbow Surgeons (ASES) Scores

The majority of patients were self-reported smokers (54%) and alcohol drinkers (64%) (Table 4). Aspects of social history were associated with differences in functional outcome scores. Non-smokers had a higher mean ASES score than both current smokers (84.5 vs 72.8; P = .02) and patients with any lifetime history of smoking (84.5 vs 73.3; P = .01) (Figure 2). There was no significant difference in shoulder function scores between patients identified as non-drinkers and those who reported consuming alcohol at moderate levels (83.9 vs 78.9; P = .26); however, patients who had a documented history of alcohol abuse had lower mean ASES scores than those who reported being non-drinkers (70.3 vs 83.9; P < .05).

Table 4. Substance Use and Functional Outcome Scores

^aP < 0.05.

Continue to: DISCUSSION...