The Intrathecal Analgesia in Robotic Surgery (IARS) Study: Real-World Associations with Intraoperative Opioid...

Background/Objectives: Subarachnoid analgesia (SA) is increasingly used as an adjunct to general anesthesia in robotic ERAS pathways, but the contribution of the intrathecal local anesthetic, as distinct from that of the intrathecal opioids, has not been isolated. Methods: This study comprised a single-center retrospective cohort of adults undergoing elective robotic surgery (May 2024-July 2025). The exposure was the intrathecal regimen, serving as mutually exclusive categories. The primary outcome was intraoperative opioid exposure (model-derived area under the concentration-time curve per hour, AUC/h); secondary outcomes were standardized intraoperative fluid administration and a stringent pain-free 48 h course (NRS = 0 throughout). Multivariable Gamma and logistic models were supplemented by provider-adjusted mixed-effects analyses, E-values, and a model-free robustness analysis. Results: Of 207 patients, 83.1% received SA; all were given intrathecal morphine and 62.3% levobupivacaine. Compared with no SA, morphine alone was not associated with AUC/h (+14.0%; 95% CI, -0.9 to +31.1), whereas all levobupivacaine-containing regimens were: -36.4% (IM + L5), -24.3% (IM + L5 + S3), and -44.8% (IM + L10 + S3); overall p < 0.001. Algebraically equivalent component reparameterization attributed this to levobupivacaine dose (-44.3% at 5 mg; -59.3% at 10 mg). Remifentanil was used in 98.7%, 88.6%, and 10.0% of patients receiving 0, 5, and 10 mg, respectively. Fluid administration was lower with IM + L5 (-24.9%) and IM + L10 + S3 (-31.0%) after adjustment, but was not with levobupivacaine-graded unadjustment. SA was associated with a pain-free course (adjusted OR, 3.33; 95% CI, 1.06-11.79), an estimate not robust to minimal unmeasured confounding. Conclusions: Levobupivacaine-containing regimens were consistently associated with lower intraoperative opioid exposure, but regimen and opioid strategy were chosen jointly by the same clinician; therefore, these associations between strategies are specific to the delivery context of this study. The fluid and pain-free findings are hypothesis-generating. Trials should vary the local anesthetic dose independently of intrathecal opioids.