Missed fractures of the greater tuberosity

Fractures of the greater tuberosity may result from a variety of mechanisms. The most common are avulsion injuries such associated with anterior shoulder dislocation, or direct trauma, as might occur in a fall on the shoulder or with hyperabduction and impaction of the greater tuberosity against the surrounding bone structures [1, 2]. These fractures can be misdiagnosed, as radiographs of the shoulder are often insufficient to confirm the diagnosis [2], especially in the case of undisplaced fractures and if the radiographic series does not include an anteroposterior (AP) view with the arm in external rotation [2, 3]. Patients may complain of persistent rotator cuff symptoms and may be referred for further examination. Missed injury remains a persistent problem, both from a clinical and medico-legal point-of-view. Few studies on this topic are available in the literature.

We present the clinical and radiological findings of a consecutive series of 17 patients who were diagnosed and managed with undisplaced greater tuberosity fractures.

Few months ago we decided to publish a retrospective study of a consecutive series of patients who sustained an occult greater tuberosity fracture and were managed at our institution between 2006 and 2008. All patients gave written consent to participate in the study. The study was submitted and approved by the ethics committee of “Campus Bio Medico” of Rome.

Of the 17 included patients, 16 agreed to participate in the study. 1 patient made a formal complaint against the initial treating physician and he declined to participate in the study. 3 patients were not contactable for final follow-up.

8 of the 17 patients had a sports related injury.

Our average follow up was 16.5 months, with all patients having 6 months or more follow up, except one patient who was fully recovered at 3 months with an OSS of 46. The internal consistency of OSS score was measured by the Cronbach’s alpha, with 0.89 at the preoperative assessment and 0.92 at 6-month follow-up. A coefficient of test-retest reliability of 6.8 was obtained using the Bland and Altman method. A significant correlation was obtained with Constant score, SF36 and Health Assessment Questionnaire Disability Index. The sensitivity to change of the study questionnaire was examined by comparing scores before and 6 months after operation, and it showed that the OSS is sensitive to clinical changes [5, 6]. The OSS in our patients showed good long term clinical results.

An issue which we highlight is that 3 of our patients were not able to attend for the final follow up, and we had to rely on the latest available clinical outpatient follow up result.

After trauma, all patients had pain in elevation and external rotation of the humerus. At diagnosis, all patients had minimal or no displacement of the fracture fragment. All fractures were undisplaced; therefore non-operative management was performed. Of the 16 included patients, OSS data were available for 13 of them. At the final follow up the mean OSS was 38.3 (range 17–46; SD 9.11).

Data on demographics, interval from initial injury to diagnosis, associated injury at MRI, type of primary injury, treatment and complications are reported in Table 1. Figures 1 and 2 depicts the typical radiographic and MRI findings in these patients.

Following initial treatment all fractures united with no secondary displacement. Three patients required glenohumeral joint injections for post traumatic stiffness between 6 and 12 weeks post injury and three patients required arthroscopy and subacromial decompression for post-traumatic impingement between 9 and 20 months post injury.

The main finding of this study is that in the majority of patients with a traumatic undisplaced fracture of the greater tuberosity, non-operative management was effective, allowing a safe and prompt return to activities.

Isolated fractures of the greater tuberosity account for approximately 20% of all proximal humeral fractures and are associated with a glenohumeral dislocation in approximately 10—30% of cases [2]. Kim et al. [8] reported that isolated greater tuberosity fractures occurs frequently in male patients with a mean age of 42.8 years. Moreover patients with isolated greater tuberosity fractures had fewer medical comorbidities than those with surgical neck fractures [8]. Tuberosity avulsion or fracture may occur after a fall onto an outstretched upper extremity due to an eccentric load applied by the attached rotator cuff on the tuberosity, often in the setting of a traumatic glenohumeral dislocation. The majority of greater tuberosity fractures are undisplaced, however the impingement of the tuberosity against the acromion or the impact against the anteroinferior glenoid during glenohumeral dislocation/subluxation could cause the inferior displacement of the tuberosity [8].

Non-displaced or minimally displaced (< 5 mm) fractures of the greater tuberosity are usually treated non operatively. Indications for surgery is a displacement > than 5 mm and take into account factor such as fracture characteristics and patient characteristics (age, comorbidities, extremity dominance, pre-injury shoulder and individual level of function, local bone quality). Surgery may be performed with an open techniques with a standard deltopectoral approach or through a deltoid splitting approach or with an arthroscopically assisted technique.

According to our experience is rare that, if the arm is placed at the patient’s side in a sling without pillow in slight internal rotation, an undisplaced fracture become displaced.

Also, we used validated questionnaire-based outcome measures.

Patients with painful abduction and external rotation after shoulder trauma with no abnormality on plain radiographs should be always considered as potentially to have sustained an undisplaced greater tuberosity fracture. A consistent clinical finding that may differentiate from a rotator cuff injury is tenderness laterally over the greater tuberosity. Poor quality of radiographs, lack of the external rotation view or lack of clinical experience could be causes of missed diagnosis. MR examination subsequently performed due to persisting symptoms, revealed the fracture in all patients. Therefore, MRI should be always performed in patients with persistent pain, bony tenderness and decreased range of motion despite negative plain radiographs. This can avoid missed diagnosis and a potential source of patient dissatisfaction satisfaction and litigation.

Prevalence of rotator cuff tear associated with occult fractures of the greater tuberosity was lower than reported in literature (2 of 17 patients in our series), even though our series is too small to draw definitive conclusions. The involvement of the rotator cuff in previous reports was found in 11 of 24 patients, and 7 of 25 patients. Further studies are needed to better understand the relationship of symptoms caused by trauma and the presence of partial tendon tears, with or without fracture [4, 9‐12].

The majority of patients were asymptomatic at the final follow up. Therefore healing of undisplaced greater tuberosity fracture is reliably achieved with non-operative treatment and surgical intervention should be only considered in case of persisting pain and shoulder function.

Strengths of the present study are that 2 fully trained shoulder surgeons performed all the diagnosis and treatment, and that the follow up evaluations were performed by an independent assessor.

In conclusion, undisplaced greater tuberosity fractures may be managed non-operatively with good results in the majority of cases. Nevertheless, shoulder MRI is warranted to confirm the diagnosis in patients with persistent post-traumatic shoulder pain and limitation of shoulder function with negative radiographs. Prompt identification of these fractures can facilitate patient management and information, particularly in counselling patients regarding the risk of stiffness and post-traumatic impingement. This will avoid a source of patient dissatisfaction and litigation.