Intraoperative Ketamine and Chronic Opioid Use After Lumbar Spine Surgery in Patients With and Without Opioid-Use Disorder: A Retrospective Database Analysis
Ethan Desverreaux, Victoria Carter, Ryan S. D’Souza, David Hao, Eric Wang, Saba Javed, Nafis B. Eghrari, Nasir Hussain, Jay KarriBACKGROUND:
Persistent opioid use after lumbar spine surgery (LSS) remains a major clinical concern. Ketamine is frequently included in multimodal perioperative analgesic regimens for its attenuating effects on central sensitization and its potential to reduce acute postoperative pain and opioid requirements. However, evidence regarding whether intraoperative ketamine influences longer-term outpatient postoperative opioid use is limited and mixed. We aimed to evaluate whether intraoperative ketamine administration is associated with differences in chronic postoperative opioid use after LSS, stratified by preoperative opioid exposure status.
METHODS:
We performed a propensity-matched retrospective cohort study using the U.S. based TriNetX Global Research Network—a database containing de-identified electronic health records—including adult patients who underwent LSS between January 2010 and December 2024. Patients were grouped based on receipt of an intraoperative ketamine and stratified by preoperative opioid status (opioid-naive or with opioid-use disorder [OUD]). Propensity score matching was used to balance demographic characteristics and relevant comorbidities. The primary outcome was the presence of outpatient opioid prescriptions at 48 hours, 6 weeks, 3, 6, and 12 months after surgery. Secondary outcomes included opioid-related adverse events and new diagnosis of OUD within 12 months of surgery.
RESULTS:
After matching, 53,639 opioid-naive patients and 2031 patients with OUD were analyzed. In opioid-naive patients, intraoperative ketamine exposure was associated with higher rates of outpatient opioid prescribing in the early postoperative period at 48 hours (
CONCLUSIONS:
In this large, real-world cohort, intraoperative ketamine was not associated with lower rates of chronic opioid dispensing after LSS in either opioid-naive or patients with OUD. These database findings reflect the absence of a detectable population-level opioid-sparing signal, and do not exclude dose- or protocol-dependent effects.