Context History taking and clinical tests are commonly used to diagnose shoulder
pain. Unclear is whether tests and history accurately diagnose instability
or intra-articular pathology (IAP).
Objective To analyze the accuracy of clinical tests and history taking for shoulder
instability or IAP.
Data Sources Relevant studies identified through PubMed, EMBASE, CINAHL, and bibliographies
of known primary and review articles.
Study Selection Studies comparing the performance of history items or physical examination
with a reference standard were included. Studies on fibromyalgia, fractures,
or systemic disorders were excluded. Of 1449 articles, 35 were eligible, and
17 were selected.
Data Extraction Data were extracted on study population, clinical tests, reference tests,
and outcome. The studies’ methodological quality (patient spectrum,
verification, blinding, and replication) was assessed with the Quality Assessment
of Diagnostic Accuracy Studies (QUADAS) checklist.
Data Synthesis Six tests showed positive likelihood ratios (LRs) and confidence intervals
(CIs). Tests favoring the diagnosis for establishing instability included:
relocation (LR, 6.5; 95% CI, 3.0-14.0) and anterior release (LR, 8.3; 95%
CI, 3.6-19). Tests showing promise for establishing labral lesions included:
the biceps load I and II (LR, 29; 95% CI, 7.3-115.0 and LR, 26; 95% CI, 8.6-80.0),
respectively, pain provocation of Mimori (LR, 7.2; 95% CI, 1.6-32.0), and
internal rotation resistance strength (LR, 25; 95% CI, 8.1-76.0). The apprehension,
clunk, release, load and shift, and sulcus sign tests proved less useful.
Results should be cautiously interpreted because studies were completed in
select populations in orthopedic practice, mostly assessed by the test designers,
and evaluated in single studies only. No accuracy studies were found for history
taking or for clinical tests in primary care.
Conclusions Shoulder complaints are frequently recurrent. Instability might cause
some of these complaints. Best evidence supports the value of the relocation
and anterior release tests. Symptoms related to IAP (labral tears) remain
unclear. Most promising for establishing labral tears are currently the biceps
load I and II, pain provocation of Mimori, and the internal rotation resistance