ERAS‐Based Surgical Bundles and Anastomotic Leakage: Which Components Drive the Effect?

We read with interest the study by Tamura et al. [1] reporting a marked reduction in symptomatic anastomotic leakage (AL) after implementation of an ERAS-based, surgery-focused care bundle in minimally invasive low anterior resection for rectal cancer. The magnitude of the reduction—14.7%–2.5% overall, and 18.0%–1.3% after propensity score matching—stands out and will naturally attract attention from colorectal units seeking reproducible strategies to improve anastomotic safety. A key interpretive issue, however, is whether the observed benefit reflects an independent “bundle effect” or is primarily driven by the near-simultaneous adoption of high-impact technical measures embedded within the bundle. In the late-phase cohort, robot-assisted surgery increased from 6.7% to 88.9%, indocyanine-green perfusion assessment from 4.4% to 98.8%, and oral antibiotic preparation from 0.4% to 86.4%; transanal drainage was also substantially higher, and reinforcement sutures—essentially absent in the early phase—were introduced and used frequently in ultra-LAR. Importantly, the manuscript notes that the start time of individual AL-prevention measures varied, despite a predetermined “official” bundle implementation date. Such simultaneity of multiple effective elements is a well-recognized challenge in evaluating complex surgical bundles, where disentangling component effects from package effects requires dedicated modeling strategies. [2]. In this context, it would be valuable to clarify whether bundle implementation remained an independent predictor of AL after adjusting specifically for these individual components. Could the authors comment on whether any component-level or stepwise multivariable modeling was attempted to disentangle the relative contribution of robot-assisted surgery, routine ICG assessment, reinforcement suturing, transanal drainage, and oral antibiotic preparation from the bundle designation itself? Each of these elements has independently been associated with leakage reduction in colorectal surgery, including randomized evidence for ICG perfusion assessment and meta-analytic support for oral antibiotic preparation. [3] Absent such analyses, a large apparent “bundle effect” may partly reflect secular trends and the cumulative impact of adopting specific technologies and techniques over time. A related point concerns generalizability. The exceptionally low late-phase AL rate is likely achievable in specialized centers, yet many institutions may not have routine access to robotics and universal fluorescence imaging, nor the same volume-related expertise. Additional clarification regarding case-mix evolution (e.g., anastomotic level distribution, proportion of ultra-LAR/ISR, surgeon factors) and how these were handled beyond propensity matching would help readers interpret transferability of the results to lower-volume or resource-constrained settings. We commend the authors for addressing a major clinical problem. Further clarification on the relative contribution and portability of individual bundle elements would strengthen the manuscript's implications for broader adoption. Carlos M. Ardila: conceptualization, investigation, writing – original draft, methodology, validation, visualization, writing – review and editing, formal analysis, supervision. Daniel Gonzalez-Arroyave: conceptualization, investigation, writing - original draft, methodology, validation, visualization, writing – review and editing, formal analysis, supervision. The authors have nothing to report. The authors have nothing to report. The authors have nothing to report. The authors declare no conflicts of interest. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.