Evaluation of Acromiohumeral Interval on Chest Radiographs in Predicting Rotator Cuff Pathology: A Cross-Sectional Study

Background The acromiohumeral interval (AHI) on shoulder radiographs is a useful measure of the subacromial space. Prior studies have shown that a narrowed AHI (typically <6-7 mm) is often associated with large, full-thickness rotator cuff tears. However, it remains uncertain whether AHI can reliably detect partial-thickness tears or smaller tears. This knowledge gap limits the clinical utility of using routine radiographs to predict rotator cuff pathology. Purpose This study aimed to evaluate the AHI measured on standard chest radiographs as a diagnostic marker for rotator cuff tears, and correlate the radiographic AHI with tear severity as confirmed on shoulder MRI. We sought to determine whether an AHI narrowing on a chest X-ray could predict the presence and extent (partial vs. full-thickness) of rotator cuff tears. Methods We conducted a cross-sectional study over three months at a tertiary hospital. Adult patients with clinical suspicion of rotator cuff injury undergoing routine chest radiographs (posteroanterior view including the shoulders) were enrolled. Those with a history of shoulder surgery, recent trauma, or metal implants were excluded. The AHI was measured on the radiograph for each shoulder; an AHI <7 mm was predefined as positive for rotator cuff tear. All patients subsequently underwent shoulder MRI, which served as the reference standard for diagnosing and grading rotator cuff tears (intact tendon vs. partial-thickness vs. full-thickness tear). The required sample size was calculated using a two-mean comparison formula, yielding a minimum of 85 subjects; we enrolled 90 to account for dropouts. Radiographic measurements were performed by two independent observers, and MRI findings were interpreted by a radiologist blind to the X-ray results. Diagnostic performance of AHI <7 mm for rotator cuff tears was assessed. Results A total of 90 patients (mean age ~60 years) were evaluated. MRI confirmed rotator cuff tears in 63 shoulders (38 full-thickness, 25 partial-thickness); 27 had intact cuffs. Mean AHI was significantly lower in full-thickness tears (~6 mm) compared to partial tears or intact tendons (~8-9 mm; p < 0.001). An AHI <7 mm on chest radiograph showed high specificity (~95%) and a positive predictive value (~90%) for full-thickness tears but limited sensitivity (~50%), as most partial tears had AHI ≥7 mm. In full-thickness tears, AHI <7 mm identified ~80% of cases. AHI negatively correlated with tear severity on MRI (r ≈ -0.6, p < 0.001). Conclusion The AHI measured on routine chest radiographs provides a simple, highly specific indicator of rotator cuff tears. A markedly narrowed AHI (<7 mm) on a radiograph strongly predicts the presence of a full-thickness rotator cuff tear and correlates with more advanced tear severity on MRI. However, a normal AHI does not exclude rotator cuff pathology, i.e., partial-thickness tears and smaller tears often show a preserved AHI, limiting sensitivity. These findings suggest that while AHI assessment on plain radiographs can aid in identifying patients likely to have major rotator cuff tears (and thus expedite definitive imaging or surgical referral), it should be used in conjunction with clinical evaluation. Radiographs remain a valuable initial tool in shoulder pain, but an AHI ≥7 mm cannot rule out a tear.