Analysis of patient factors associated with selection of corticosteroid injection in the freezing phase of idiopathic adhesive capsulitis.

Abstract

Primary idiopathic adhesive capsulitis (AC) is characterized by shoulder pain and global limitations in range of motion (ROM). The aim of this study was to determine (1) if a spectrum of symptom severity exists during the freezing phase of AC and (2) identify factors associated with patient selection of corticosteroid injection (CSI) for treatment. Patients presenting within 6 months of symptom onset of AC (freezing phase) were enrolled in this single-site retrospective case control study. Visual analog pain scale (VAS) score, shoulder ROM, American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form (ASES) scores, and Patient-Reported Outcomes Measurement Information System (PROMIS function and pain) scores were collected. Each patient was offered oral anti-inflammatory medication, physical therapy, and intra-articular CSI. Patients were divided into 2 cohorts: those electing vs. those deferring CSI. Multivariable logistic regression was performed to identify patient or symptom characteristics predictive of electing CSI. A total of 112 patients (mean age=54.7 ± 8.8 years, female=76 [67.9%], mean symptom duration =13.2 ± 7.9 weeks, elected CSI=74 [66.1%]) were included in our analysis. The overall study population demonstrated a wide spectrum of VAS pain scores (6.0 ± 2.8, range: 0-10) and ROM: forward elevation (99° ± 27°, range: 30°-150°), abduction (82° ± 24°, range: 30°-130°), external rotation (47° ± 13°, range: 0°-90°), internal rotation (38° ± 26°, range: 5°-90°). The CSI group had higher mean VAS pain score (6.6 ± 2.5 vs. 4.9 ± 3.0, P=.005) and greater limitations in ROM for forward elevation (92° ± 27° vs. 113° ± 25°, P=.001) and abduction (77° ± 24° vs. 90° ± 21°, P=.005) compared with the non-CSI cohort. The CSI group demonstrated significantly worse shoulder function based on Constant (P<.05), ASES (P=.001), P-UE (P=.016), P-Intensity (P=.002), and P-Interference (P=.004). Logistic regression demonstrated decreased total shoulder ROM in forward elevation and abduction plane (OR=0.98, 95% CI=0.97-0.99, P=.004). Hispanic ethnicity and increased VAS pain score (OR=1.20, 95% CI=1.01-1.43, P=.04) were associated with increased likelihood of electing CSI. A spectrum of symptom severity exists during the freezing phase of primary AC, despite similar etiology. AC patients with greater pain severity, and greater limitations in ROM at initial evaluation were associated with patient selection of CSI.