Neutrophil to Lymphocyte Ratio and Cardiovascular Disease Incidence in HIV-Infected Patients: A Population-Based Cohort Study

Abstract

Neutrophil to lymphocyte ratio (NLR) has been shown to predict occurrence of cardiovascular events in the general population. The aim of our study was to evaluate the role of NLR to predict major cardiovascular disease (CVD) events in HIV-infected subjects. We performed a retrospective cohort study of HIV-infected patients residing in the Local Health Authority (LHA) of Brescia, northern Italy, from 2000 to 2012. The incidence of CVD events in HIV-positive patients was compared with that expected in the general population living in the same area, computing standardized incidence ratios (SIRs). To evaluate the predictive role of NLR, univariate and multivariate Cox regression models were applied, computing hazard ratios (HRs). A total of 3766 HIV-infected patients (mean age 38.1 years, 71.3% males) were included (person-years 28768.6). A total of 134 CVD events occurred in 119 HIV-infected patients. A 2-fold increased risk (SIR 2.02) of CVD was found in HIV-infected patients compared to the general population. NLR levels measured at baseline and during follow-up were independently associated with CVD incidence, when also adjusting for both traditional CVD risk factors and HIV-related factors (HR 3.05 for NLR≥ 1.2). The area under the receiver operating characteristics (ROC) curve showed a modest, not statistically significant, increase, from 0.81 to 0.83, with addition of NLR to Framingham risk score model covariates. In conclusion an elevated NLR is a predictor of risk CVD in HIV-infected patients, independently from the traditional CVD risk factors.

Fig 1. Cumulative incidence CVD curves in HIV-infected subjects with NLR<1.2 and NLR ≥1.2.

Abbreviation: NLR, neutrophil to lymphocyte ratio.

Fig 2. Risk of CVD event (Hazard Ratio) according to NLR distribution.

NLR was modeled by cubic spline (solid line) with three knots in Cox regression model adjusted for age, sex, intravenous drug use, diabetes and tobacco smoking as fixed covariates, and CD4 cell count, antiretroviral therapy, presence of hypertension, SBP, total cholesterol and HDL as time dependent covariates. The reference value is 1.2. The 95% confidence intervals are shown as dashed lines. Vertical axes have a logarithmic scale. Abbreviations: NLR, neutrophil to lymphocyte ratio; SBP, systolic blood pressure; HDL, high density lipoprotein.