Lower peripheral T- and NK-cell counts are associated with systemic inflammation and nutritional status in hospitalized gastrointestinal surgery patients

Background Systemic inflammation and impaired nutritional status may be associated with altered cellular immune profiles in surgical inpatients. However, peripheral T- and NK-cell profiles and their clinical correlates remain insufficiently characterized in hospitalized gastrointestinal surgery patients. Methods This retrospective observational study included hospitalized patients from a gastrointestinal surgery department who underwent peripheral blood lymphocyte subset testing between January 2024 and May 2025. Patients were classified into disease categories according to discharge diagnoses. T lymphocytes, natural killer (NK) cells, B lymphocytes, and NKT cells were analyzed together with inflammatory markers, including interleukin-6 (IL-6), C-reactive protein (CRP), white blood cell count, neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), and nutritional markers including serum albumin and hemoglobin. Group comparisons, correlation analyses, and multivariable regression were performed. Continuous T-cell and NK-cell counts were the primary outcomes; cohort-defined low-count status was a secondary exploratory outcome. Sensitivity analyses included multiple imputation and Firth penalized logistic regression. Results After excluding one patient with missing diagnosis, 887 patients were included, of whom 623 had complete data for multivariable analysis. Higher neutrophil-to-lymphocyte ratio (NLR) was associated with lower continuous T-cell and NK-cell counts and higher odds of low-count status across complete-case and sensitivity analyses. In models pooled across 50 imputed datasets, obstruction/perforation-related acute abdomen, compared with benign anorectal diseases, was associated with lower T-cell counts (adjusted geometric mean ratio [aGMR] = 0.822, 95% CI 0.706–0.958, P = 0.012) and lower NK-cell counts (aGMR = 0.821, 95% CI 0.677–0.995, P = 0.045). These associations did not reach statistical significance in the complete-case models, indicating sensitivity to missing-data handling. Albumin associations varied by outcome and analysis. Conclusion The primary findings concern associations of continuous T-cell and NK-cell counts with inflammatory and nutritional markers, with NLR showing consistent associations across analyses. Cohort-defined low-count status was a secondary, exploratory outcome. These findings do not establish causality or validated clinical thresholds and require prospective validation with standardized sampling and perioperative data.