Predictive modeling of aneurysmal subarachnoid hemorrhage with acute lung injury complicating delayed cerebral ischaemia

Objective: Delayed cerebral ischemia (DCI) is a frequent consequence of aneurysmal subarachnoid hemorrhage (aSAH), and severe aSAH is typically accompanied with Acute Lung Injury (ALI). This research examined the risk variables for delayed cerebral ischaemia in aneurysmal subarachnoid hemorrhage patients complicated with ALI, and developed a columnar graph prediction model. Methods: Clinical data from 234 patients with aSAH complicated with ALI, admitted to Lishui People's Hospital between January 2018 and June 2024, were analyzed. The patients were randomly divided into a training group (164 cases) and a validation group (70 cases). Risk factors for the occurrence of delayed cerebral ischaemia (DCI) were identified and incorporated into a model, the differentiation and reliability of the line graph model were validated via the use of ROC curves and calibration curves. Results: Multifactorial logistic regression identified three significant independent risk variables for DCI: elevated positive end-expiratory pressure (PEEP), interleukin-6, and D-dimer (p < 0.05). The column-line plots demonstrated superior discriminatory performance in both the training set (AUC = 0.882, 95% CI: 0.820-0.940) and the validation set (AUC = 0.874, 95% CI: 0.778-0.996), while the calibration curves indicated strong concordance between the training and validation sets. Conclusion: High positive end-expiratory pressure, interleukin-6, and d-dimer are independent risk factors for DCI in patients with aSHA combined with ALI, and the resulting columnar line graphs show significant predictive value and help to better identify patients at high risk of DCI.