Risk factors for postoperative intensive care unit admission after minimally invasive esophagectomy following neoadjuvant chemoimmunotherapy in patients with esophageal squamous cell carcinoma: a retrospective cohort study

Full abstract

Read the full abstract

Introduction Patients with esophageal squamous cell carcinoma (ESCC) who undergo minimally invasive esophagectomy (MIE) after neoadjuvant chemoimmunotherapy (nCIT) and subsequently require postoperative intensive care unit (ICU) admission may experience a greater economic burden and poorer clinical outcomes. This study aimed to identify perioperative factors independently associated with postoperative ICU admission in patients with ESCC treated with nCIT followed by MIE. Methods We conducted a retrospective cohort study of patients with ESCC who underwent MIE after nCIT. Baseline demographic characteristics, perioperative variables, postoperative complications, and details of postoperative ICU admission, including incidence, timing, and indications, were systematically collected. Univariable and multivariable logistic regression analyses were performed to identify factors associated with postoperative ICU admission. A nomogram was constructed to provide individualized risk estimates, together with a model-based visualization of the weighted PNI–ASA score and corresponding predicted probability. Results A total of 115 eligible patients were included in the final analysis, of whom 25 (21.7%) required postoperative ICU admission. Compared with patients who did not require ICU admission, those admitted to the ICU had significantly higher rates of pulmonary infection (60.0% vs. 11.1%, p < 0.001) and respiratory failure (80.0% vs. 41.1%, p < 0.001), a longer postoperative hospital stay (26.1 ± 17.1 vs. 18.5 ± 9.1 days, p = 0.003), and higher total hospital costs (¥95,519 ± 40,754 vs. ¥63,426 ± 16,895, p < 0.001). In the multivariable logistic regression analysis, each unit increase in preoperative prognostic nutritional index (PNI) was associated with lower odds of postoperative ICU admission (OR = 0.84; 95% CI, 0.74–0.94; p = 0.002), whereas American Society of Anesthesiologists (ASA) physical status class III was associated with higher odds than class I/II (OR = 7.57; 95% CI, 2.10–27.20; p = 0.002). Operative duration was associated with ICU admission in univariable analysis ( p = 0.024) but not in multivariable analysis ( p = 0.102). The PNI–ASA nomogram showed acceptable discrimination (apparent AUC = 0.810; optimism-corrected AUC = 0.781). Conclusion Lower preoperative PNI and ASA physical status class III were independently associated with postoperative ICU admission in patients with ESCC undergoing nCIT followed by MIE. The PNI–ASA nomogram may support preoperative risk stratification. Although operative duration was not independently associated with ICU admission, it may serve as an intraoperative warning indicator. The dichotomized PNI cutoff derived from the subgroup analyses requires external validation and should not be used as a standalone clinical threshold.