<- Back to digest

JSES - 2026-07-03 - Journal Article

Assistant Experience Enhances the Learning Curve in Reverse Total Shoulder Arthroplasty: A Comparative CUSUM Analysis of Three Surgeons.

Taniguchi S, Hirakawa Y, Manaka T, Nakazawa K, Ito Y, Terai H

retrospective cohortLOE IIIn = 180 (60 consecutive rTSA cases per surgeon, 3 surgeons)N/A (operative and perioperative outcomes only)

Topics

shoulder elbowsports
PMID: 42398667DOI: 10.1016/j.jse.2026.06.020View on PubMed ->

Key Takeaway

Surgeon C, who assisted under a mature standardized rTSA workflow, reached operative efficiency plateau at 15–20 cases versus 30–40 cases for Surgeon A with no prior assistant exposure.

Summary Depth

Choose how much analysis to show on this article page.

Summary

This study used CUSUM analysis of operative time across the first 60 consecutive rTSA cases for three surgeons with identical subspecialty training but differing assistant experience pathways to quantify learning curve differences. Surgeon C (assistant under mature, standardized workflow) stabilized at 15–20 cases; Surgeon B (assistant during early technique evolution) at 25–35 cases; Surgeon A (no prior assistant exposure) at 30–40 cases. Complications and blood loss variability were concentrated in the early ascending CUSUM phase for all three surgeons, with Surgeon C demonstrating stable blood loss from the outset.

Key Limitation

With only three surgeons, each representing a single data point per assistant-experience category, no statistical inference about the effect of assistant experience can be drawn; the observed differences may reflect individual aptitude rather than training pathway.

Original Abstract

BACKGROUND

Reverse total shoulder arthroplasty (rTSA) is an established treatment for cuff tear arthropathy, irreparable rotator cuff tears, and advanced glenohumeral osteoarthritis. However, despite advances in implant design and preoperative planning, rTSA remains technically demanding during early experience, and the influence of assistant experience on learning curves is unclear.

METHODS

This retrospective study evaluated the early learning curves of three shoulder surgeons (Surgeons A-C), each assessed over their first 60 consecutive rTSA cases. All surgeons shared similar subspecialty training in shoulder surgery but differed in their assistant experience pathways before independent practice: no prior assistant exposure, assistant experience during early technique evolution, and assistant experience under mature and standardized surgical workflows. Learning trends were analyzed using cumulative summation (CUSUM) analysis of operative time, with each surgeon's individual mean used as the reference, and breakpoints were identified by changes in the CUSUM slope. Intraoperative blood loss was assessed using box-and-whisker plots, with cases categorized into three sequential groups of 20 cases. Intraoperative and postoperative complications were recorded and overlaid on the CUSUM curves to evaluate their temporal distribution during the learning process.

RESULTS

CUSUM analysis demonstrated distinct learning patterns among the surgeons. Surgeon A showed a pronounced initial upward CUSUM trend, with a breakpoint at approximately 30-40 cases, followed by stabilization. Surgeon B exhibited a similar biphasic pattern with a more gradual initial slope and an earlier breakpoint at approximately 25-35 cases. In contrast, Surgeon C demonstrated minimal early CUSUM elevation and reached a plateau at approximately 15-20 cases, indicating more rapid stabilization of operative efficiency. Intraoperative blood loss decreased progressively with experience for all surgeons; however, reductions in blood loss lagged behind the stabilization of operative time. Surgeons A and B showed wide variability in blood loss during the first 20 cases, with gradual narrowing in later phases, whereas Surgeon C demonstrated relatively stable blood loss from the early phase. Both intraoperative and postoperative complications were predominantly observed during the early ascending phase of the CUSUM curves, with few events occurring after stabilization.

CONCLUSION

Proficiency in rTSA develops in a stepwise manner and is strongly influenced by prior assistant experience. Assistant training under proficient surgeons facilitates earlier stabilization of operative time, blood loss, and perioperative safety, suggesting that high-quality assistant experience is essential for optimizing the rTSA learning curve.