Segment-specific patterns of regional lymph-node metastasis in primary small-cell carcinoma of the esophagus: A systematic review and meta-analysis

Table 3 Main limitations and safeguards applied in the review

Issue	Risk to inference	Safeguard used here
Retrospective surgical cohorts	Selection bias and variable treatment sequence	Pooled only compatible pathologic surgical denominators; interpreted estimates as surgical-series probabilities
Heterogeneous station maps	Invalid formal station-by-segment pooling	Summarized station evidence as an evidence map
Preprint evidence	Non-peer-reviewed data and possible cohort-overlap uncertainty	Included CHiSCEC for completeness, flagged its preprint status and possible overlap with Xu et al[8], and performed an overlap sensitivity analysis[8,13]
Node-level clustering	Individual nodes are not independent observations	Reported node-level metastatic ratios descriptively only
Unavailable station-positive survival endpoints	Station-specific survival benefit and efficacy index cannot be estimated	Avoided claims that treating any specific station improves survival