Nutrition in Esophagectomy Care

Evidence

This single-center retrospective cohort study examined an institutional perioperative regimen in patients undergoing two-field esophagectomy for esophageal cancer and tumors of the esophagogastric junction. The investigators retrospectively analyzed patient-, disease-, and surgery-specific information extracted from patient records and collected in a database. The cohort contained 325 consecutive patients: 279 received the modified care regimen and 46 received standard care. The study was reported according to the STROBE statement, and its primary endpoint was the Comprehensive Complication Index, a continuous measure selected by the investigators to assess overall postoperative morbidity. [pmid:42829442]

The modified regimen was a multicomponent institutional pathway rather than a nutrition-only intervention. It incorporated early parenteral nutritional support, in contrast to enteral routes favored by society recommendations, alongside selective oral decontamination, an enforced near-zero fluid balance, hydroxyethyl starch administration, and routine nasogastric decompression. Consequently, the comparison addresses the observed results of two care approaches and does not isolate the contribution of early parenteral nutrition from the other pathway components. [pmid:42829442]

The investigators compared postoperative outcomes before and after propensity-score matching to adjust for baseline differences between the modified-care and standard-care groups. They reported that baseline characteristics were balanced after matching. The postoperative comparisons included complication rates, anastomotic leakage, Comprehensive Complication Index, and textbook outcome rates. Survival data were systematically obtained from the cancer registry of the Comprehensive Cancer Center at the University Hospital of Freiburg, while the clinical cohort data came from institutional records. [pmid:42829442]

For the primary endpoint after matching, the reported Comprehensive Complication Index was 32.01 in modified care versus 35.70 in standard care, with p=0.061. The study defined statistical significance as p<0.05. Thus, the numerical between-group difference in the primary outcome did not meet the study’s stated statistical-significance threshold. The authors also characterized endpoints other than the Comprehensive Complication Index as exploratory and hypothesis-generating rather than confirmatory. [pmid:42829442]

Nutrition-specific reporting was narrow. After propensity-score matching, early parenteral nutrition was reported as similar in the modified-care and standard-care groups. The available report does not provide the corresponding nutritional values or a statistical comparison for that statement. This result therefore documents comparability as reported for early parenteral nutrition after matching, while leaving the magnitude and operational details of the nutritional exposure unspecified in the presented material. [pmid:42829442]

The article also reported that fluid balance on postoperative day 4 was similar after matching, and that acute kidney injury did not occur more frequently after hydroxyethyl starch application. These are pathway-context findings, not direct measures of nutritional delivery. Overall and disease-free survival were reported not to differ between groups, while odds ratios reported for major complications, anastomotic leakage, and intermediate-care-unit readmission favored modified care but were not statistically significant. [pmid:42829442]

Analysis — Observational pathway comparison

The observed pattern is consistent with a retrospective comparison of patients selected into an institutional modified pathway or standard care, followed by propensity-score matching to make recorded baseline characteristics more comparable. The lower numerical Comprehensive Complication Index in modified care was accompanied by a p value above the study’s own threshold, so the primary analysis does not provide a statistically significant between-group result. Similarity in early parenteral nutrition after matching further means that the reported nutrition-specific result is not evidence of a measured nutritional separation between the matched groups. [pmid:42829442]

Interpretation should remain at the level of association because the retrospective design cannot establish causation. The regimen bundled parenteral nutrition with fluid, decompression, decontamination, and other perioperative practices; any observed group pattern cannot be assigned to one component from this design. Propensity-score matching may balance measured baseline characteristics, but it does not convert the comparison into random allocation or rule out differences that were not captured in the matching process. The study’s designation of non-primary outcomes as exploratory also bounds interpretation of those comparisons. [pmid:42829442]

Limitations

Selection bias remains a central concern because care-group assignment occurred in a retrospective single-center cohort rather than through randomization. The comparator was standard care, whereas the modified-care group received a bundled institutional pathway; this limits attribution of observed differences to parenteral nutritional support itself. The retrospective design cannot establish causation. Matching addressed reported baseline differences, but the analysis remains dependent on what was recorded in patient records and incorporated into the matching approach. [pmid:42829442]

The fixed cohort size also constrained inference. The investigators performed a post hoc detectable-effect-size calculation for the primary endpoint rather than a prospective sample-size calculation, and described other endpoints as exploratory. They further stated that the post-matching cohort size and few absolute events limited interpretation for some outcomes, including binary complications. Accordingly, non-significant estimates and numerical trends should not be treated as definitive comparisons. [pmid:42829442]

Generalizability is limited by the absence of total minimally invasive esophagectomy procedures in this cohort, which the authors identified as a constraint in applying the findings to current surgical practice. The available nutrition result is also limited to the statement that early parenteral nutrition was similar after matching; nutritional values and a statistical comparison were not provided in the available report. These limitations define uncertainty in the reported comparison rather than evidence that any unreported measure was absent. [pmid:42829442]