Abstract

Background: Transforaminal percutaneous endoscopic lumbar discectomy (TPELD) has become an alternative and minimally invasive surgical technique for soft lumbar disc herniation. However, the learning curve is relatively long and difficult. This study aimed to summarize the characteristics of the learning curve of TPELD, including the cutoff number of cases to achieve technical proficiency and to discuss the strategies to improve the learning curve.

Methods: The PubMed, Embase, Cochrane Library, and KoreaMed databases were searched for articles describing the learning curves in TPELD. Clinical studies involving human patients and evaluating the learning curve of TPELD with quantitative data were included. Strict quality assessment was completed, and descriptive statistics were calculated.

Results: Of the 6884 screened titles and abstracts, 10 full-text articles, including 958 cases, were considered in the analysis; all were cohort studies grouped into early and late groups based on surgeons' experiences of TPELD. The most commonly used cutoff number to differentiate between these groups was 20, with a mean value of 24.70 ± 18.99 cases (range, 10-72 cases). The most widely used measure was operative time. Although most articles showed better results in operative time or pain scores in the late group, only one article proposed a bona fide learning curve.

Conclusions: We found insufficient evidence to support the cutoff point of 20 or other numbers of cases where the learning curve reaches a plateau. Therefore, these numbers should be interpreted with great care; hence, high-quality prospective studies evaluating the real curve are required.

Keywords: Discectomy; Endoscopic; Learning curve; Lumbar; Outcome; Percutaneous; Transforaminal.

Copyright © 2020. Published by Elsevier Inc.