The ratio of monocytes to lymphocytes multiplying platelet predicts incidence of pulmonary infection-related acute kidney injury

Abstract

Abstract Background Inflammation is a crucial factor in the pathogenesis and development of acute kidney injury (AKI). Macrophages, as an important innate immune cell, regulate immune response and play a pathophysiological role in AKI. This study aimed to evaluate the predictive capacity of peripheral blood monocytes for the incidence of pulmonary infection-related AKI. Methods We recruited 1038 hospitalized patients with pulmonary infections from January 1 to December 31, 2019, in Zhongshan Hospital, Fudan University. Patients were divided into derivation and validation cohorts. Data on demographic characteristics, disease history, and biochemical indexes were retrieved from the electronic medical system. The composite inflammatory indexes were calculated as monocyte/(lymphocyte × platelet ratio) (MLPR). We applied dose–response relationship analyses to delineate the nonlinear odds ratio (OR) in different MLPR levels and integrated it into a logistic model to predict the risk of AKI. Results The incidence of hospital-acquired AKI was 18.8% in the derivation cohort. Compared to non-AKI, the MLPR levels were significantly higher in AKI patients. Dose–response curve revealed that the increase of AKI risk was faster in the first half of MLPR and then tended to flatten. After classifying the MLPR levels into six groups, the AKI incidence increased from 4.5% to 55.3% with a peaking OR of 24.38. The AUC values of the AKI model only including MLPR were 0.740, and after gradually integrating other covariates, the area under the receiver operating characteristic (AUC) value reached 0.866, which was significantly higher than the AUC of full models without MLPR (0.822). Moreover, the better prediction ability of AKI was observed in the external validation, with an AUC of 0.899. Conclusion MLPR has good predictive efficiency in AKI, which can be used as a simple and easy clinical composite index to effectively predict early pulmonary infection-related AKI.