Comparison of intrathecal morphine versus thoracic paravertebral block for video-assisted thoracoscopic surgery: a randomized non-inferiority trial

BACKGROUND: Although widely performed, video-assisted thoracoscopic surgery (VATS) is associated with significant postoperative pain. Thoracic paravertebral block (TPVB) is a guideline-favored regional technique, while intrathecal morphine (ITM) is a technically less demanding and more widely applicable neuraxial alternative. We hypothesized that ITM would provide analgesic efficacy non-inferior to TPVB in terms of 24-hour intravenous morphine milligram equivalents (IV-MME). METHODS: This was a prospective, randomized, controlled, observer-blinded non-inferiority trial. Adults undergoing elective VATS were randomized 1:1 to TPVB (0.25% bupivacaine, 0.4 mL/kg+epinephrine 1:400,000) or ITM (5 mcg/kg based on ideal body weight). The primary endpoint was 24-hour IV-MME (intraoperative opioids excluded). Secondary endpoints included 12-hour IV-MME, pain (Numeric Rating Scale at rest/activity), rescue analgesia, time to first opioid demand, 15-item Quality of Recovery, safety (postoperative nausea/vomiting, pruritus, respiratory depression), and 30-day complications. The prespecified non-inferiority margin was 5.05 mg. Analyses followed intention-to-treat with per-protocol sensitivity. RESULTS: Of 82 screened, 72 were randomized (ITM n=36; TPVB n=36); 2 TPVB conversions to thoracotomy were excluded from the per-protocol set (n=70). For the primary endpoint, the median difference (ITM-TPVB) in 24-hour IV-MME was 6.5 mg (95% CI 1 to 17; p=0.009) in intention-to-treat and 9.0 mg (95% CI 2 to 18; p=0.002) in per-protocol; in both analyses, the upper CI bound exceeded 5.05 mg, and non-inferiority was not met. Individual 24-hour and 12-hour IV-MME favored TPVB; rescue analgesia was more frequent with ITM (21/36 vs 12/36; p=0.033). Pain outcomes, quality of recovery scores, and time to first opioid demand were similar, with no respiratory depression. Although mild 30-day Clavien-Dindo grade I events differed between groups, no apparent clinically meaningful safety signal was observed. CONCLUSIONS: ITM did not demonstrate non-inferiority to TPVB for 24-hour IV-MME; TPVB was associated with lower systemic opioid use, while pain outcomes were similar. TRIAL REGISTRATION NUMBER: NCT07126483.