Background: The impact of preoperative body composition as independent predictor of prognosis for esophageal cancer patients after esophagectomy is still unclear. The aim of the study was to explore such a relationship. Methods: This is a multicenter retrospective study from a prospectively maintained database. We enrolled consecutive patients who underwent Ivor-Lewis esophagectomy in four Italian high-volume centers from May 2014. Body composition parameters including total abdominal muscle area (TAMA), visceral fat area (VFA), and subcutaneous fat area (SFA) were determined based on CT images. Perioperative variables were systematically collected. Results: After exclusions, 223 patients were enrolled and 24.2% had anastomotic leak (AL). Sixty-eight percent of patients were sarcopenic and were found to be more vulnerable in terms of postoperative 90-day mortality (p = 0.028). VFA/TAMA and VFA/SFA ratios demonstrated a linear correlation with the Clavien-Dindo classification (R = 0.311 and 0.239, respectively); patients with anastomotic leak (AL) had significantly higher VFA/TAMA (3.56 ± 1.86 vs. 2.75 ± 1.83, p = 0.003) and VFA/SFA (1.18 ± 0.68 vs. 0.87 ± 0.54, p = 0.002) ratios. No significant correlation was found between preoperative BMI and subsequent AL development (p = 0.159). Charlson comorbidity index correlated significantly with AL (p = 0.008): these patients had a significantly higher index (≥ 5). Conclusion: Analytical morphometric assessment represents a useful non-invasive tool for preoperative risk stratification. The concurrent association of sarcopenia and visceral obesity seems to be the best predictor of AL, far better than simple BMI evaluation, and potentially modifiable if targeted with prehabilitation programs.
