Diagnosis of shoulder pain in primary care
Shoulder pain is a common and disabling complaint that is associated with high morbidity and significant associated health care costs. Shoulder pain is a common reason for primary care medical consultation, however the clinical diagnosis of shoulder pain is complicated by the similar presentations of different shoulder conditions and a lack of validated clinical examination tests and diagnostic criteria in primary care populations. Radiological imaging is also widely available and is increasingly being utilized to aid in the diagnostic process however the relevance of imaging to symptoms of shoulder pain remains unclear. The difficulties associated with the diagnosis of shoulder pain frequently result in delayed diagnosis and delays in the implementation of appropriate management. An improvement in the ability to accurately diagnose painful shoulder conditions would assist in optimising patient outcomes in primary health care.
The aim of this thesis was to evaluate the diagnostic accuracy of a clinical examination for identifying a predominant subacromial, acromioclavicular joint (ACJ) and glenohumeral joint (GHJ) pain source, and to assess the added value of diagnostic imaging findings for identifying symptomatic pathology affecting these structures. The diagnostic accuracy of clinical examination findings for detecting the presence of rotator cuff tears that may require early referral for specialist evaluation was also assessed. A review of the literature highlighted the poor specificity of many commonly used clinical tests, a lack of information regarding the relationship between imaged pathology and symptoms with little information to guide decisions regarding the use of diagnostic imaging investigations for shoulder pain.
This project consisted of a reliability study in which the reliability of clinical examination tests was first evaluated, followed by a diagnostic accuracy study in which consecutive patients with shoulder pain were recruited from primary health care physiotherapy and medical practices. All participants received a clinical examination and a series of diagnostic imaging investigations (x-ray and diagnostic ultrasound scan) (index tests) followed by a diagnostic injection of local anaesthetic (diagnostic block) into the subacromial bursa (SAB) and ACJ (reference standard tests). Those not reporting at least 80% reduction in pain (positive anaesthetic response (PAR)) following the SAB or ACJ diagnostic block also received a GHJ diagnostic block performed as part of a magnetic resonance arthrogram (MRA) investigation. Results of the clinical examination and diagnostic imaging investigations (index tests) were compared with results of the reference standard tests to estimate the ability of these clinical examination and imaging findings to accurately identify a predominant subacromial, ACJ or GHJ pain source and to detect the presence of rotator cuff tears.
Combinations of clinical features were identified with the ability to accurately rule-in a PAR following SAB and ACJ diagnostic block. When only a small number of these clinical features were present, confirmation of supraspinatus or ACJ pathology on ultrasound improved the ability to rule-in a PAR following SAB and ACJ diagnostic block respectively. Overall the added diagnostic value of imaging findings for predicting an 80% PAR was limited due to the low prevalence of specific imaging findings, resulting in identification of only a small additional number of cases in whom a PAR could be predicted. Additional diagnostic investigations such as clinically-administered diagnostic injections of local anaesthetic may provide more information regarding the likelihood of a predominant subacromial or ACJ pain source in a larger proportion of patients. Analysis of diagnostic accuracy of clinical examination and imaging findings for predicting a PAR following GHJ diagnostic block was beyond the scope of this thesis but will be the subject of ongoing analysis. Clinical examination predictors of a large or multi-tendon rotator cuff were also identified that were able to accurately identify the presence of a large or multi-tendon rotator cuff tear that may require specialist evaluation.
In conclusion, the ability to accurately diagnose painful subacromial and ACJ disorders in primary care begins with information gathered from the clinical examination however, for many patients the accurate diagnosis of these disorders may also require additional diagnostic investigations including diagnostic imaging or diagnostic injections. Combinations of clinical examination findings alone are likely to be sufficient to identify a large or multi-tendon rotator cuff tear that may require specialist evaluation. Results of this research may provide a framework that can be used by primary care practitioners to guide diagnostic processes for painful shoulder disorders, enabling more accurate and efficient identification of these conditions. This has the potential to reduce health care costs, reduce the burden on secondary care services, enable more timely application of appropriate treatment interventions and improve outcomes for patients suffering from shoulder pain.