Abstract
BACKGROUND: Learning curves describe the rate of performance improvements corresponding to the surgeon's caseload, followed by a plateau where limited further improvements are observed. This study aimed to determine the learning curve for biportal full-endoscopic posterior cervical foraminotomy (BE-PCF) for the unilateral cervical foraminal disc. METHODS: The learning curve was evaluated using a learning curve cumulative summation test (LC-CUSUM). The goal for the operation time was set to 78 min, which is the mean operation time (mOT) of percutaneous full-endoscopic posterior cervical foraminotomy (PE-PCF) performed by a senior surgeon. Moreover, clinical outcomes and post-operative complications were compared between the early and late learning periods 1 year post-operatively. RESULTS: This study enrolled the first 50 patients who underwent single-level BE-PCF, performed by a single surgeon. The LC-CUSUM signalled competency for surgery at the 20th operation, indicating that sufficient evidence was obtained to prove that the surgeon was competent. The mOT was 71.29 ± 11.69 min in BE-PCF, 71.84 ± 12.61 min in the early learning period, and 67.83 ± 10.31 min in the late learning period (p = 0.254). There was no statistical difference in clinical outcomes, visual analogue scale scores, and neck disability index between both periods (p > 0.05). Four complications were recorded throughout the whole period, with three in the early period and one in the late period (p = 0.285). CONCLUSION: Our study shows that BE-PCF has a learning curve of 20 caseloads to achieve 90% proficiency, and it significantly reduces the operation time based on the performance of a senior surgeon proficient in PE-PCF.