Construction and evaluation of risk prediction model for pulmonary infection after minimally invasive interventional surgery for ruptured aneurysm in department of cerebrovascular surgery

OBJECTIVE  To investigate the risk factors of pulmonary infection after minimally invasive interventional surgery for ruptured aneurysm in the department of cerebrovascular surgery, and to construct and evaluate a risk prediction model for pulmonary infection after minimally invasive interventional surgery.

METHODS  A total of 579 patients who underwent minimally invasive interventional surgery for ruptured aneurysm at the Department of Cerebrovascular Surgery, Qilu Hospital of Shandong University Dezhou Hospital from Jan. 1, 2021 to Dec. 31, 2023 were enrolled. They were divided into a pulmonary infection group (77 cases) and a non-pulmonary infection group (502 cases) based on postoperative pulmonary infection status. The distribution of pathogenic bacteria causing postoperative pulmonary infection was analyzed. A multivariate logistic regression analysis was applied to construct a prediction model, and the goodness-of-fit test was used to evaluate the model. The receiver operating characteristic (ROC) curve was employed to assess the predictive value of the model for pulmonary infection after minimally invasive interventional surgery.

RESULTS  Among 579 patients undergoing minimally invasive interventional surgery, the postoperative pulmonary infection rate was 13.30% (77 cases), with 71 strains of gram-negative bacteria, 30 strains of gram-positive bacteria and 5 strains of fungi isolated. History of diabetes (OR＝9.251), hypoproteinemia (OR＝8.874), duration of ventilation use ≥48 h (OR＝7.231), smoking (OR＝7.158), elevated preoperative neutrophils (OR＝2.738) and elevated preoperative white blood cells (OR＝3.019) were identified as risk factors for pulmonary infection after minimally invasive interventional surgery (P＜0.05). The Hosmer-Lemeshow goodness-of-fit test yielded x²＝4.318, df＝8, P＝0.827, indicating good model fit. The area under the ROC curve of the prediction model was 0.953, with a sensitivity of 92.20% and specificity of 88.40%.

CONCLUSIONS  History of diabetes, hypoalbuminemia, duration of ventilator use no less than 48 hours, smoking, elevated preoperative neutrophils and elevated preoperative white blood cells are risk factors for pulmonary infection after minimally invasive interventional surgery. The constructed risk prediction model for pulmonary infection after interventional surgery demonstrates good predictive value.