JSES - 2026-03-19 - Journal Article
Cannabis Use Disorder Is Associated With Increased Early Postoperative Opioid Use and Pain but No Long-Term Differences After Arthroscopic Rotator Cuff Repair: A Retrospective Cohort Study Using TriNetX.
Sontam TR, Tummala S, Khela HS, Miller L, Kelly JD
Topics
Key Takeaway
Cannabis use disorder before arthroscopic rotator cuff repair is associated with 17% higher early opioid use rates (42.0% vs. 35.9%) and fewer PT visits within 90 days, but no difference in 2-year complications or revision rates.
Summary Depth
Choose how much analysis to show on this article page.
Summary
This study used the TriNetX database to compare 90-day and 2-year outcomes after arthroscopic RCR in patients with versus without a CUD diagnosis within 3 months preoperatively, after 1:1 propensity-score matching. CUD patients had higher opioid use rates (42.0% vs. 35.9%, p<0.0001), higher acute pain coding (4.2% vs. 2.6%, p=0.0008), and fewer PT visits through day 42 in the early postoperative period. No significant differences were found at 2 years for revision RCR, conversion to arthroplasty, shoulder stiffness, or new opioid-related disorders.
Key Limitation
ICD-coded CUD captures only clinically documented disorder, excluding the far larger population of regular cannabis users without a formal diagnosis, making the true exposure-outcome relationship unquantifiable.
Original Abstract
BACKGROUND
Cannabis use has become increasingly common in the general population, with cannabis use disorder (CUD) defined as clinically significant impairment and/or distress related to cannabis consumption. Prior studies have associated CUD with increased postoperative pain and opioid use, but its impact following arthroscopic rotator cuff repair (RCR) remains unclear. This study aimed to compare short- and long-term surgical outcomes following RCR between patients with and without CUD.
METHODS
Using the TriNetX database, we conducted a retrospective cohort study of patients who underwent arthroscopic RCR for full-thickness rotator cuff tears. Patients diagnosed with CUD within three months prior to surgery were compared to patients without any documented history of CUD. 1:1 propensity-score matching was performed to control for demographics and comorbidities, resulting in two matched cohorts. Outcomes within 90 days included postoperative pain, opioid use, mean number of opioid prescriptions, shoulder stiffness, and physical therapy (PT) utilization. Long-term outcomes assessed within two years included upper limb mononeuropathies, shoulder pain or stiffness, conversion to total shoulder arthroplasty, revision RCR, arthroscopic debridement, and new-onset opioid-related disorders.
RESULTS
After matching, 3,012 patients remained in each cohort with no significant differences in baseline characteristics. In the first 14 days postoperatively, CUD patients had higher rates of opioid use (42.0% vs. 35.9%, p < 0.0001), acute pain (4.2% vs. 2.6%, p = 0.0008), and a greater number of opioid prescriptions (0.677 ± 1.137 vs. 0.539 ± 0.959, p < 0.0001). These differences persisted through postoperative days 15-42, during which CUD patients also attended fewer PT sessions (2.685 ± 1.941 vs. 3.158 ± 2.334, p < 0.0001). From days 43-90, opioid-related outcomes were similar between groups, though CUD patients continued to have fewer PT visits (3.858 ± 3.187 vs. 4.584 ± 3.719, p < 0.0001). No significant differences were observed in long-term complications at two years.
CONCLUSION
For patients undergoing arthroscopic RCR, a coded diagnosis of CUD was associated with increased early postoperative pain, greater opioid requirements, and lower PT participation within 90 days of surgery. More research is required to investigate the nature of this association.