Synergistic impact of nutritional risk, glycemic control, and systemic inflammation on Abdominal Compartment Syndrome in diabetic patients following complex ventral hernia repair: a development and validation study.

内容摘要

Abdominal Compartment Syndrome (ACS) is a catastrophic complication following complex hernia repair, particularly in patients with Type 2 Diabetes Mellitus (T2DM) who exhibit a phenotype of "metabolic vulnerability." The interplay between preoperative nutritional depletion, chronic hyperglycemia, and systemic inflammation remains poorly understood in this context. Current risk assessment tools rely heavily on anatomical metrics and often fail to capture the synergistic impact of immuno-metabolic fragility. We aimed to develop and prospectively temporally validate a dynamic nomogram that integrates immuno-nutritional markers with surgical variables to predict ACS in diabetic patients. We conducted a two-stage, prospective temporal validation study involving 555 diabetic patients undergoing elective complex hernia repair at a tertiary referral center. Phase I (January 2015 to December 2021) comprised a retrospective derivation cohort (N = 461) that was randomly split into a training set (n = 323) and an internal testing set (n = 138) to identify predictors and construct the model. Phase II (January 2022 to December 2024) established a prospective temporal validation cohort (N = 94) to verify model performance in a real-world clinical setting at the same institution. We utilized determining factors including the Hernia Sac Volume to Abdominal Cavity Volume (HSV/ACV) ratio, Nutritional Risk Screening 2002 (NRS-2002), Systemic Inflammatory Response Index (SIRI), and Glycated Hemoglobin (HbA1c). The primary endpoint was the development of ACS within 7 days postoperatively. Pre-specified sensitivity analyses included LASSO penalized regression with bootstrap optimism correction (1,000 resamples), modeling of all continuous predictors as restricted cubic splines, and comparison of nested models (anatomical-only vs. anatomical + metabolic vs. full nomogram) by decision curve analysis, net reclassification improvement (NRI), and integrated discrimination improvement (IDI). The study population exhibited a high baseline metabolic burden, with a mean body-mass index of 31.2 kg/m2 and a mean HbA1c of 7.9%. Multivariable logistic regression identified seven independent predictors: HSV/ACV ratio &#x2265; 0.25 (Odds Ratio [OR], 2.75; 95% Confidence Interval [CI], 1.60-4.85), use of tension reduction procedures (OR, 2.45), operative time > 200 min (OR, 2.12), BMI &#x2265; 30 kg/m2 (OR, 1.88), NRS-2002 score &#x2265; 3 (OR, 2.18), SIRI &#x2265; 1.6 (OR, 1.98), and HbA1c &#x2265; 6.0% (OR, 1.65). A positive correlation between SIRI and nutritional risk (Spearman &#x3c1; = 0.42, P < 0.001) was observed, consistent with an immuno-nutritional axis. The nomogram demonstrated good and stable discrimination, with an Area Under the Curve (AUC) of 0.89 (95% CI, 0.85-0.93) in the training cohort, 0.86 in the internal testing cohort, and 0.84 (95% CI, 0.77-0.91) in the prospective temporal validation cohort. Comprehensive calibration assessment in the prospective cohort showed a calibration slope of 0.91 (95% CI 0.74-1.08), calibration intercept of -0.09 (-0.31 to 0.13), Brier score 0.112, and a non-significant Hosmer-Lemeshow test (P = 0.45). Bootstrap-based internal validation (1,000 resamples) yielded an optimism-corrected C-index of 0.87 and a calibration slope of 0.93. Risk stratification categorized patients into low, intermediate, and high-risk groups, yielding ACS incidence rates of 1.0%, 10.1%, and 72.7%, respectively (P < 0.001). Compared with the anatomical-only baseline model, the full nomogram achieved an NRI of 0.31 (95% CI 0.18-0.44) and an IDI of 0.094 (95% CI 0.061-0.127), both P < 0.001. The high-risk phenotype was associated with significantly greater postoperative fluid sequestration, prolonged intensive care unit stays, and increased 30-days mortality. The integration of immuno-nutritional markers (SIRI, NRS-2002) and glycemic control with anatomical parameters provides a temporally validated tool with good and stable discrimination for ACS in diabetic patients. Its principal clinical utility lies in reliable risk exclusion (negative predictive value 95.5% in the prospective temporal validation cohort), supporting safe rule-out of low-risk patients, while the more modest positive predictive value (46.4%) indicates that a high score should prompt enhanced surveillance and individualized decision-making rather than constitute a deterministic indication for pre-emptive open-abdomen management.