Haptic Virtual Reality Training for Combined Spinal-Epidural Anesthesia in Anesthesiology Trainees: Prospective Randomized Controlled Pilot Trial

Abstract Background Combined spinal-epidural (CSE) anesthesia is a complex psychomotor skill requiring precise tactile discrimination. Traditional manikin-based training may be limited by degraded haptic fidelity and inconsistent tissue feedback. Whether skills acquired through haptic virtual reality (VR) training transfer to supervised clinical procedures remains uncertain. Objective This study assessed the feasibility and early clinical application of CSE skills after high-fidelity haptic VR training in junior anesthesiology trainees vs traditional manikin-based training. Methods We conducted a single-center, prospective, parallel-group, randomized, and outcome assessor-blind pilot trial. A total of 20 fifth-year anesthesiology interns with minimal neuraxial experience were randomly assigned (1:1) by an independent investigator using a card draw to a standardized 7-day curriculum with either a traditional manikin (control group [CG], n=10) or a haptic-VR simulator (simulator group [SG], n=10). SG trainees performed repeated punctures on the haptic-VR CSE Teaching Platform, whereas the CG trainees practiced the CSE procedure on a lumbar puncture manikin. Both interventions were delivered face-to-face in a simulation laboratory by experienced instructors. During training, each trainee performed 1 supervised CSE on a patient. Clinical performance was video recorded and assessed by blinded experts. The primary outcome was the 0 to 100 total clinical performance score, rated by 2 assessors. Secondary outcomes included domain-specific scores, procedural time, puncture-angle adjustments, first-attempt success rate, and adverse events within 24 hours. Results All 20 trainees completed training and clinical assessments and were included in the intention-to-treat analysis. Overall scores were higher in the SG than in the CG (mean difference 4.50 points, 95% CI 1.85-7.15; P =.002). The SG had higher scores for tactile-dependent skills, including recognition by bubble compression (mean difference 1.90 points, loss of resistance 95% CI 1.14-2.66; P <.001) and ligamentum flavum penetration (mean difference 1.90 points, 95% CI 1.18-2.62; P <.001), but lower aseptic technique compliance (mean difference −1.40 points, 95% CI −2.00 to −0.80; P <.001). The SG trainees had a shorter time to ligamentum flavum breakthrough, fewer passive angle adjustments, and more active adjustments. The first-attempt success rate was higher but imprecisely estimated. Primary-outcome interrater reliability was excellent (average-measures intraclass correlation coefficient of 0.977). No significant adverse events occurred. Conclusions Compared with traditional manikin training, haptic-VR training yielded higher overall clinical performance and better tactile-dependent CSE skills during the first supervised patient procedure, whereas manikin training better supported aseptic practice. By evaluating posttraining performance in patients rather than only in simulation, this study extends VR research toward early clinical skill transfer and suggests that the 2 modalities may develop complementary components of procedural competence. In practice, these findings support a blended CSE curriculum in which VR is used for repeated tactile and needle-control training and manikin-based simulation for aseptic preparation and procedural workflow; this approach should be evaluated in larger trials.