Medicine (Baltimore). 2026 Oct 9;105(41):e51055. doi: 10.1097/MD.0000000000051055.
ABSTRACT
Acute type A aortic dissection (ATAAD) remains a critical cardiovascular emergency with substantial in-hospital mortality. A rapid and interpretable preoperative risk stratification using routinely available biomarkers is clinically desirable. We investigated whether combining blood glucose and the neutrophil-to-lymphocyte ratio (NLR) provides incremental prognostic information beyond either marker alone and developed a bootstrap-validated nomogram for early risk estimation. In this study, 609 patients who underwent Stanford type A acute aortic dissection surgery were enrolled. Logistic regression models were constructed for glucose alone, NLR alone, and combined. Discrimination was assessed using the area under the receiver operating characteristic curve (AUC). The incremental value was evaluated using likelihood ratio tests. A nomogram incorporating age, sex, blood glucose level, NLR, and lactate level was developed. Internal validation was performed to estimate the optimism-corrected performance, and the calibration was assessed using calibration plots and slopes. Admission blood glucose levels and NLR were independently associated with adverse prognosis. The discrimination was good for the glucose model (AUC = 0.80) and modest for the NLR model (AUC = 0.63), whereas the combined model improved discrimination (AUC = 0.82). The model fit improved significantly when NLR was added to the glucose model (P < .001) and when glucose was added to the NLR model (P < .001). In bootstrap validation, the combined adjusted model showed the best performance (apparent C-index = 0.87; optimism-corrected C-index = 0.85), with an optimism-corrected calibration slope of approximately 0.87. The combined assessment of blood glucose levels and NLR provides incremental prognostic value for early risk stratification in patients with ATAAD. This may aid early perioperative risk stratification. The bootstrap-validated nomogram may support rapid and clinically interpretable preoperative risk estimation in high-risk populations.
PMID:42854006 | DOI:10.1097/MD.0000000000051055