Comparative Effectiveness of Robotic-Assisted, Video-Assisted, and Open Thymectomy for Thymoma: A Systematic Review and Meta-Analysis

Abstract

Thymoma is the most common primary mediastinal tumor in adults, typically excised in a single operation. Less invasive approaches, specifically robotic-assisted thymectomy surgery (RATS) and video-assisted thoracoscopic surgery (VATS), have gained popularity subsequent to conventional open thymectomy via median sternotomy. To determine the best surgery for thymoma, this systematic review and meta-analysis compared RATS, VATS, and total thymectomy in terms of efficacy, safety, and prognosis. Literature databases were systematically reviewed for publications, including PubMed, Scopus, and Google Scholar, till February 2025, using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) criteria sectioning the articles based on comparison of RATS to VATS or open thymectomy, as well as quantitative outcomes of reoperation, total survival, recurrence, perioperative complications, length of procedure, and length of stay. The risk of bias was assessed using the Risk Of Bias In Non-randomized Studies of Interventions (ROBINS-I) tool. Eleven articles met the eligibility criteria, and data were analyzed using random-effects model meta-analysis on RevMan (Cochrane, London, UK). RATS had comparable operation times to VATS (pooled mean difference: 4.49 minutes; 95% CI: -39.87 to 48.84; I² = 98%); it was, however, linked with fewer total complications (mean difference: -3.78; 95% CI: -3.89 to -3.70) and less intraoperative blood loss (mean difference: -25.01 mL; 95% CI: -38.03 to -12.00; I² = 0%). Also, RATS showed a reduction in pleural drainage time compared to VATS (mean difference: -0.66 days; 95% CI: -0.97 to -0.35; I² = 0%). RATS led to shorter hospital stays than open thymectomy (mean difference versus VATS: -0.28 days; 95% CI: -1.36 to 0.80; I² = 91%; versus open: -1.38 days; 95% CI: -2.33 to -0.43; I² = 14%) and fewer postoperative complications than open thymectomy but not differences in oncologic outcomes, including mortality and rates of recurrence. The I² values ranged widely from 0% to 98%, indicating variable heterogeneity across outcomes, which limits interpretability in some comparisons. Most studies included were retrospective cohorts, and the risk of bias was inconsistent, with one study deemed to be at low risk and seven deemed to have some concerns. Overall heterogeneity between outcomes ranged from low to moderate (I² = 0%-98%). Issues with costs and surgeon experience variability, however, continue to be major impediments to routine use of RATS. Given the retrospective nature of most included studies and high heterogeneity in key outcomes, conclusions about RATS's comparative advantage should be interpreted cautiously. While RATS demonstrates similar oncologic outcomes with fewer complications and shorter hospital stays, its routine use remains constrained by cost, surgeon training requirements, and a lack of high-quality prospective data. Future multicenter randomized controlled trials and cost-effectiveness studies are necessary to clarify its long-term role in thymoma surgery.

Keywords: open thymectomy; robotic; robotic-assisted thoracoscopic surgery (rats); thymectomy; video-assisted thoracoscopic surgery (vats).

Figure 1. PRISMA flow diagram illustrating the screening and selection process for studies included in the analysis.

PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Figure 2. Weighted bar plot showing the distribution of risk-of-bias judgments across bias domains using the ROBINS-I tool.

Each bar represents the proportion of studies categorized as low, some concerns, or high risk of bias within each domain, providing an overview of methodological quality across the included studies. Image created using Risk-of-bias VISualization (robvis). ROBINS-I: Risk Of Bias In Non-randomized Studies of Interventions.

Figure 3. Traffic light plot displaying domain-level risk-of-bias judgments for each included study using the ROBINS-I tool.

Each row represents an individual study, while each column corresponds to a specific bias domain. Image created using Risk-of-bias VISualization (robvis). ROBINS-I: Risk Of Bias In Non-randomized Studies of Interventions.

Figure 4. Forest plot demonstrating pooled operation time of robotic-assisted and video-assisted thymectomy.

RATS: robotic-assisted thoracoscopic surgery, VATS: video-assisted thoracoscopic surgery.

Figure 5. Forest plot demonstrating pooled hospital stay in days for robotic-assisted and video-assisted thymectomy.

The study by Kamel et al. [19] mislabeled their unit as years, which we changed to days. This mistake was confirmed through their text. RATS: robotic-assisted thoracoscopic surgery, VATS: video-assisted thoracoscopic surgery.

Figure 6. Forest plot demonstrating pooled 90-day mortality for robotic-assisted and video-assisted thymectomy.

RATS: robotic-assisted thoracoscopic surgery, VATS: video-assisted thoracoscopic surgery.

Figure 7. Forest plot demonstrating pooled blood loss (mL) for robotic-assisted and video-assisted thymectomy.

RATS: robotic-assisted thoracoscopic surgery, VATS: video-assisted thoracoscopic surgery.

Figure 8. Forest plot demonstrating pooled pleural drainage (days) for robotic-assisted and video-assisted thymectomy.

RATS: robotic-assisted thoracoscopic surgery, VATS: video-assisted thoracoscopic surgery.

Figure 9. Forest plot of overall complications for robotic-assisted and video-assisted thymectomy.

RATS: robotic-assisted thoracoscopic surgery, VATS: video-assisted thoracoscopic surgery.

Figure 10. Forest plot of pooled hospital days for robotic-assisted thymectomy versus open thymectomy.

RATS: robotic-assisted thoracoscopic surgery, VATS: video-assisted thoracoscopic surgery.