Background: We aimed to study the relationship between the number of fractured scapular regions, and the severity and distribution of associated injuries in blunt trauma patients.

Methods: 107 consecutive patients with fractured scapulae (100 males) with a mean age of 35 (8–65) years were prospectively studied between January 2003 and December 2005. Mechanism of injury, associated injuries, injury severity scale (ISS) and the number of anatomical scapular regions involved in each fracture were studied. Patients were divided into single-region, two-region fracture, and more than two-region fracture groups. Computer tomography was used for fracture classification in 99 patients and plain X-rays in the remaining 8.

Results: Road traffic collisions were the most common cause of scapular fracture. 95 patients (89%) sustained associated injuries. The most frequent was chest injury (68 (64%)). The median ISS was 9 (4–57) for the single-region fracture group (n 55), 20 (4–59) for the two-region fracture group (n 30), and 22.5 (4–54) for more than two-region fracture group (n 22) (p=0.02, Kruskal Wallis test). The median values of abbreviated injury scale (AIS) for chest injuries for the three groups were 1 (0–4), 3 (0–5) and 3 (0–5), respectively (p=0.001, Kruskal Wallis test). The single-region fracture group had significantly less posterior structure injury (9/55) compared with the multiple-region fracture group (46/52) (p=0, Fisher’s exact test).

Conclusion: Associated injuries are common in patients having scapular fractures. ISS and AIS for chest injuries are higher and posterior structure injuries more frequent in patients with fractures involving multiple scapular regions.

Introduction: Fractured scapula is rare because the scapula is mobile and well protected. We report the first prospective study in the literature of scapular fractures caused by blunt trauma.

Patients: 67 consecutive patients (64 males and 3 females, with a mean age of 33 (8–60) years) were included between January 2003 and September 2004. Data regarding the mechanism of injury, associated injuries, the Injury Severity Score (ISS), the location of the scapular fracture, whether it was isolated or involved more than one part of the scapula, and the accuracy of radiographic modality used were analyzed.

Results: 50 (75%) fractures were caused by road traffic accidents, 11 by afall from height, 4 from a fall from the same level and 2 by heavy object. 56 patients (84%) sustained associated injuries of which chest injury occurred in 48 (72 %) and was the most frequent. Many patients had more than one substantial chest injury. 30 (45%) patients had lung contusion, 27 rib fracture, 23 haemothorax and 22 pneumothorax. The abbreviated injury score for chest injury for the isolated fractures (n=41) was 2.1 (0–5) and for the combined fracture group (n=26) was 2.4 (NS). Other associated injuries were upper limb fractures in 29 patients, head and facial injuries in 25, 17 had pelvic fractures, lower limb fractures occurred in 16 patients, 15 had spinal fractures and 10 abdominal injuries. No brachial plexus or subclavian artery injury occurred. The mean ISS was 20 for both fracture groups. 41 (61%) of fractures were isolated. Of these, 31 (75%) involved the body of the scapula, 4 acromion, 2 glenoid, 2 coracoid, 1 neck and 1 spine. For the 26 combined fractures, the body was involved in 24 (92%), 13 the neck, 12 spine, 4 glenoid, 4 acromion and 3 coracoid. Plain chest X-ray was done in 63 patients and the scapular fracture was shown in 40 (63%). 42 patients had scapular X-rays and the fracture was seen in 35 (83%). The fracture was shown in all 42 patients that had computer tomography (CT) of the chest. The same was true for the 19 patients who had CT of the scapula. In total, CT was done in 61 (91%) of the 67 patients.

Discussion; Scapular fracture has a high rate of associated injuries, mainly to the chest. Lung contusion, haemo- and pneumothorax were very frequent in our series compared with other reports. The liberal use of trauma CT protocols and the prospective nature of the study may explain this finding.