Case Series Evaluating the Operative and Nonoperative Treatment of Scapular Fractures

Scapular fractures occur frequently due to high-energy trauma, with concomitant injuries seen in approximately 90% of cases.^1-4 As a result, treatment is often surrounded by other difficult medical decisions, and factors affecting outcomes can be multifaceted. The gaps in our understanding of long-term outcomes with current treatment modalities have recently come to light, especially when it comes to determining indications for surgery.

Specifically, there is very little literature on radiographic healing and long-term shoulder function in larger samples of scapular fractures; additionally, there is evidence that some patients do not experience full functional recovery.^3,5-7 Studies assessing return of function in patients treated nonoperatively have shown decreased mobility and persistence of pain.⁷ Some of these findings could be due to variability in surgical indications.^2,4 While the majority of fractures are treated nonoperatively, the decision to operate has recently been one of debate. Prior literature has suggested highly variable measurements of angulation and extra-articular displacement at which surgery is recommended.¹ For example, indications for surgery measured by the medial displacement of extra-articular fractures range from >10 mm to >20 mm;^8-11 similarly, the displacement of intra-articular fractures meriting surgery ranges from >2 mm to >5 mm, depending on the author.^12-16

The current debate over surgical indications for less severe scapular fractures, as well as the potential for chronic pain and stiffness calls for a thorough examination of factors affecting functional outcomes. The purpose of this study is to determine which patient factors, fracture patterns, and treatment modalities were associated with differences in healing and return of shoulder function. We hypothesized that certain aspects of the patient’s social history (tobacco, alcohol) as well as concomitant chest wall injuries may be associated with poor outcome scores and lower levels of function. We further hypothesized that glenoid fractures would affect function more than body fractures, and we did not expect to see a significant difference in outcomes between operative and nonoperative treatment.

MATERIALS AND METHODS

This study was approved by the Institutional Review Board. A registry at our level 1 trauma center was queried to identify 663 skeletally mature patients with scapular fractures between 1999 and 2011. Forty-eight patients had died prior to the study, and 21 patients had insufficient radiography and/or clinical follow-up (Figure 1). To be included, patients were required to have at least 1 year of follow-up to assess healing. Data on patient demographics, fracture classification, etiology of injury, concomitant injuries (clavicle fractures, rib fractures, pulmonary injuries), comorbidities, alcohol use, and tobacco use were collected retrospectively for the remaining 594 patients. Patients were then prospectively contacted via telephone and mail, employing 3 Internet search engines as needed, in an attempt to obtain current contact information. Three patients declined to participate, and 438 were not reachable after multiple attempts. Outcome scores for the remaining 153 patients were determined with the Modified American Shoulder and Elbow Surgeons (ASES) Shoulder Form.¹⁷ Scores were measured out of 100, with 0 to 30 representing maximally impaired, 31 to 60 representing moderately impaired, and 61 to 100 representing minimally impaired shoulder function.¹⁸ Due to the retrospective identification of the patients, no pre-injury shoulder function scores were collected. Given that many patients were unreachable, or reachable but not living in close proximity to the hospital, patients did not routinely return for re-evaluation for this study.

Nonoperative management consisted of sling immobilization for comfort for up to 2 weeks, during which time Codman’s exercises and elbow, forearm, wrist, and hand motion were encouraged. Active and passive shoulder mobility without restriction were also recommended progressively as tolerated. Strengthening and unrestricted lifting activities were allowed after approximately 8 to 10 weeks following the injury. Decision for surgery was at the surgeon’s discretion. Surgical indications included articular displacement and severely displaced glenoid neck fractures. Open reduction and internal fixation was performed by 1 of 4 fellowship-trained surgeons. Concomitant surgical procedures were not undertaken in the same setting. Postoperative activity consisted of sling immobilization for comfort for up to 6 weeks, during which time active and passive shoulder mobility without restriction were also recommended progressively as tolerated. Strengthening and unrestricted lifting activities were allowed after approximately 12 weeks following surgery. We considered fractures as healed if either X-rays showed healing progression to complete union or early X-rays showing signs of healing with subsequent follow-up visits indicating clinical healing (absence of pain, absence of shoulder dysfunction).

Continue to: STATISTICAL ANALYSIS...