The Association between Rotator Cuff Tear Severity on Magnetic Resonance Imaging and Pain Improvement after Referral to Conventional Physical Therapy: A Retrospective Cohort Analysis

Research Objectives: To investigate the association between rotator cuff tear (RCT) pathology on magnetic resonance imaging (MRI) and pain improvement after conventional physical therapy (PT) to optimize appropriate physician referral.

Design: Respective cohort study

Setting: Multiple outpatient PT clinics

Participants: Patients (n=124, 66.9% right shoulders) had a mean age of 58.7 ± 12.3 years old (median: 60 years old) and received a mean of 8.0 ± 5.1 PT visits (median: 6 PT visits). Patients received PT for shoulder pain, had MRI within two years of PT evaluation, and had no history of shoulder surgery.

Interventions: All patients received “conventional PT” as interventions were not controlled for this study and given at the discretion of the treating physical therapist.

Main Outcome Measures: The main outcome measure was magnitude of pain improvement as measured by the Visual Analog Scale (VAS). Patients were grouped into either the “No Tear” group (n=33), the “Partial Tear” group (n=56), and the “Complete Tear” group (n=35) based on MRI RCT findings.

Results: Based on RCT presence or severity, the “No Tear” group had a mean pain improvement of 0.7 ± 2.4 points (median: 0; range: -5 to 9), the “Partial Tear” group had a mean pain improvement of 2.3 points (median: 2; range: -7 to 9), and the “Complete Tear” group had a mean pain improvement of 1.4 points (median: 0; range: -4 to 9). There was a significant association between magnitude of pain improvement and RCT pathology on MRI (p=0.023) with the “Partial Tear” group experiencing greater improvement than the “No Tear” group (p=0.008). There was no significant difference in magnitude of pain improvement between the “Partial Tear” and the “Complete Tear” groups (p=0.106).

Conclusions: The magnitude of pain improvement is highly variable after conventional PT. There was no significant difference in pain improvement between patients with a partial RCT and patients with complete RCT. These results should be taken with caution as shoulder pathologies other than RCT could be causing symptoms.

Author(s) Disclosures: John Martin Leland receives compensation from Mitek.