Predictive Ability of Systemic Inflammation Response Index for the Risk of Pneumonia in Patients with Acute Ischemic Stroke

Abstract

Introduction: Several studies have examined the crucial role of inflammatory indexes such as the ratio of monocyte and lymphocyte (MLR), systemic-immune-inflammation-index, and the ratio of neutrophil and lymphocyte (NLR) in stroke-associated pneumonia (SAP). However, the function of the systemic inflammation response index (SIRI) in SAP is not known. This study investigated whether SIRI at admission could predict the incidence of SAP in patients with acute ischemic stroke (AIS). Patients and Methods: 2,802 AIS patients collected from 2013 to 2021 were divided into the SAP and non-SAP groups. The predictive performance of SIRI in SAP was evaluated by the receiver operating characteristic curve. Multivariate regression analysis and the restricted cubic spline (RCS) were performed to explore the relationship between SIRI and SAP risk. Results: The SIRI at admission in SAP patients was significantly higher than that in non-SAP patients (median [IQR]: 3.75 [2.05, 6.99] vs. 1.51 [0.94, 2.62], p < 0.001). SIRI had a predictive ability for predicting the incidence of SAP with area under the curve of 0.757, better than NLR and MLR (both p < 0.05). SIRI ≥2.74 was an independent risk factor for the incidence of SAP (odds ratio: 5.82, 95% confidence interval: 4.54, 7.49, p < 0.001). The RCS model showed an increasing trend of the SAP risk with the increase of SIRI. Conclusion: SIRI showed a good predictive value for SAP. In clinical practice, AIS patients with high SIRI levels (SIRI ≥2.74) should be aware of the risk of SAP.