Predictive value of preoperative systemic inflammation response index for prolonged mechanical ventilation in patients with type A aortic dissection: a propensity-matched analysis

This study evaluated whether the systemic inflammation response index (SIRI) can predict prolonged mechanical ventilation (PMV) in patients with Stanford type A aortic dissection (TAAD). We retrospectively analyzed 489 TAAD patients who underwent total aortic arch replacement at our hospital between January 2021 and December 2025. Patients were divided into PMV and non-PMV groups according to postoperative PMV status. Perioperative clinical data were recorded and compared between the two groups. PMV was defined as mechanical ventilation exceeding 48 h postoperatively or reintubation after extubation with a cumulative ventilation time over 48 h. Propensity score matching (PSM) at a 1:1 ratio balanced baseline differences, after which perioperative outcomes, postoperative complication rates, and 30-day mortality were further analyzed. The optimal cutoff value was determined using receiver operating characteristic (ROC) curve analysis. Univariate and multivariate logistic regression identified risk factors for PMV. Among the 489 TAAD patients, 342 (69.9%) were male and 147 (30.1%) were female, with a median age of 53 years. PMV occurred in 192 patients (39.3%). After PSM, 185 matched pairs were identified with balanced baseline characteristics (standardized mean difference < 0.10). Preoperative white blood cell, neutrophil, monocyte, lymphocyte counts, and SIRI were significantly higher in the PMV group than in the non-PMV group ( P < 0.050). Intraoperatively, cardiopulmonary bypass (CPB) time was longer in the PMV group ( P = 0.004). Regarding early postoperative outcomes, the PMV group showed higher rates of secondary intubation ( P = 0.008), acute respiratory distress syndrome ( P = 0.039), acute kidney injury ( P = 0.023), and 30-day mortality ( P < 0.001), along with longer intensive care unit stays ( P < 0.001) and total hospitalization ( P < 0.001). Spearman correlation analysis revealed a positive relationship between SIRI and PMV ( r = 0.407, P < 0.001). ROC curve analysis identified an optimal SIRI cutoff of 9.28 for predicting PMV, with an area under the curve of 0.830 [95% confidence interval (CI) 0.787–0.873, P < 0.001], sensitivity of 84.2%, and specificity of 80.1%. Multivariate regression confirmed preoperative SIRI ≥ 9.28 as an independent risk factor for PMV [odds ratio (OR) 4.827, 95% CI 1.831–9.766, P < 0.001]. A preoperative SIRI ≥ 9.28 is significantly associated with PMV after total aortic arch replacement in TAAD patients and demonstrates good predictive value for short-term postoperative prognosis. As a convenient and economical inflammatory biomarker, SIRI may assist in preoperative risk stratification and could potentially guide postoperative respiratory management in this population. However, these findings require prospective validation before clinical implementation.