JOA - 2026-10-07 - Journal Article
Cluster B Personality Disorder Is Independently Associated With Opioid Abuse at Two and Five Years After Primary Total Knee Arthroplasty.
Chebi P, Mzeihem M, Alvarez J, Macgregor K, Rechenmacher A, Mejia A, Gonzalez M, Farid Y
Topics
Key Takeaway
Cluster B personality disorder independently increases odds of opioid abuse at 2 years (OR 1.39) and 5 years (OR 1.43) after primary TKA, despite equivalent 90-day perioperative opioid exposure.
Summary Depth
Choose how much analysis to show on this article page.
Summary
This study used a national all-payer claims database (2010–2022) to evaluate whether personality disorder (PD) independently predicts complications and opioid abuse after primary TKA, using 1:1 propensity matching on demographics, comorbidities, and preoperative opioid abuse. After matching and multivariable adjustment, any PD was associated with opioid abuse at 2 years (OR 1.26) and 5 years (OR 1.36), while paradoxically showing lower odds of wound and infectious complications (OR 0.80). Cluster B PD—75.7% of the PD cohort, predominantly borderline PD (81.1%)—drove the opioid signal exclusively, with no difference in 90-day morphine milligram equivalent exposure between groups.
Key Limitation
Claims-based PD identification captures only coded diagnoses, producing a severely underrepresented and likely more severely affected PD cohort that limits generalizability to the broader population of patients with undiagnosed or subclinical personality pathology.
Original Abstract
BACKGROUND
Psychiatric comorbidities are associated with worse results after total knee arthroplasty (TKA), but outcomes in patients who have personality disorders (PD) remain poorly defined. This study evaluated complications and opioid-related outcomes after primary TKA in patients who have PD.
METHODS
Patients who underwent primary TKA from 2010 to 2022 were identified in a national all-payer claims database. Patients who had PD were identified by diagnosis codes and classified by the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition cluster. Because they carried a greater baseline comorbidity burden, patients who had PD were matched 1:1 to patients who did not have PD on demographics, comorbidities, psychiatric conditions, and preoperative opioid abuse. Multivariable regressions evaluated medical, wound and infectious, orthopaedic, and opioid-related outcomes.
RESULTS
Of 698,040 patients, 2,961 (0.4%) had PD; 75.7% had Cluster B PD, of which 81.1% were borderline. After matching (n = 2,922 per group) and adjustment, PD was independently associated with opioid abuse at two years (odds ratio [OR] 1.26; 95% confidence interval [CI], 1.04 to 1.52) and five years (OR 1.36; CI, 1.15 to 1.60), and with lower odds of wound and infectious complications (OR 0.80; CI, 0.67 to 0.95); all P < 0.05. Total 90-day opioid exposure in morphine milligram equivalents did not differ. Only Cluster B PD remained associated with opioid abuse at two years (OR 1.39; CI, 1.19 to 1.63) and five years (OR 1.43; CI, 1.25 to 1.64; both P < 0.001).
CONCLUSIONS
The greater baseline comorbidity burden of patients who have PD accounts for most of the excess perioperative risk. However, Cluster B PD remains independently associated with opioid abuse at two and five years after TKA despite similar opioid exposure, supporting preoperative psychiatric screening and targeted opioid stewardship.