Abstract
Dear Editor, We read with great interest the publication by Álvarez de Mon-Montoliú et al., “Meta-Analysis of Learning Curve in Endoscopic Spinal Surgery: Impact on Surgical Outcomes.” 1 The article aimed “to evaluate the learning curve in endoscopic spinal surgery, including the time to mastery and challenges faced by novice surgeons, to improve learning and surgical outcomes.” To this end, the authors compared outcomes between surgeons in the “early” and “late” periods of the learning curve. To complement their analysis of this important topic for current and future generations of spine surgeons, we offer the following observations:
First, to properly evaluate the study objective and improve reproducibility and applicability, threshold values for mastery-level performance should be clearly specified for each outcome. 2 Therefore, we respectfully invite the authors to provide operational definitions for “competence,” “proficiency,” and “mastery”, terms used repeatedly throughout the article but not explicitly defined in either the registered PROSPERO protocol or the published manuscript.
Second, as reported in Table 1, there is considerable heterogeneity in cut-offs used to separate surgeons into “early” and “late” groups, ranging from 15 to 214 cases, as well as differences in etiologies, endoscopic techniques, surgeon prior experience, and case mix. 1 Given these factors, we invite the authors to explain why they considered these studies sufficiently homogeneous to combine and analyze. 2
Third, operation time has been consistently characterized as a poor proxy for actual surgical performance and an unreliable measure of learning. Therefore, exploring better outcome measures specific for skill acquisition in endoscopic surgery is critical. 3 One promising direction would be developing and implementing procedure-specific adaptations of the Objective Structured Assessment of Technical Skill (OSATS).4,5 These adapted instruments could be used to establish expert performance-derived thresholds against which novice surgeons could benchmark their progress. 5 Additionally, generating learning curves for measures of surgical process as well as patient-reported outcomes could provide a comprehensive assessment of surgeon performance and better reflect the learning process across multiple domains.3,6
Fourth, learning curves can be analyzed using visual inspection, split-group method, cumulative sum method, and regression.3,7 Because each method has distinct strengths and limitations, future studies should employ multiple complementary approaches. 2 This would allow researchers to assess whether learning occurred, estimate learning rates, detect learning plateaus or regression, and determine whether desired performance levels have been reached.6,7
Fifth, the learning curve of a procedure can be influenced by numerous confounders, including case mix, case cadence, and the surgeon’s prior experience.2,3,6 To address these factors, future studies should report the experience level of each assessed surgeon, standardize equipment and operating room teams, and strive to homogenize the type and complexity of cases.2,3 Additionally, analyses could employ multi-level hierarchical models to account for individual surgeon characteristics and better characterize learning curves for specific endoscopic procedures. 8
Finally, we commend Álvarez de Mon-Montoliú and colleagues for the considerable effort involved in this report. Their work highlights important opportunities for advancing skill acquisition research in endoscopic spinal surgery, and we hope these observations help guide future studies to strengthen the evidence base in this field.
