Abstract
Objective
To evaluate the role of neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), monocyte-to-lymphocyte ratio (MLR), platelet-to-neutrophil ratio (PNR), platelet-to-monocyte ratio (PMR), and neutrophil-to-monocyte ratio (NMR) as predictors for lupus nephritis (LN) patients without infection or as biomarkers for distinguishing between infection or flare with LN patients.
Methods
LN patients were divided into three groups: LN without infection, LN with infection, and LN with flare. A total of 57 healthy subjects were enrolled as controls. The differentiation was analyzed between LN without infection and control group, and LN with infection and LN with flare. Correlations among variables were assessed in the LN group without infection. Receiver operating characteristic curves were constructed in two comparable groups.
Results
NLR, PLR, and MLR were increased significantly in the LN group without infection as compared with those in healthy controls. NLR (area under the curve (AUC): 0.75) and MLR (AUC: 0.79) were useful for distinguishing between LN patients without infection and healthy subjects. In differentiating LN patients without infection from the controls, optimal cutoffs of NLR and MLR were 3.43 (sensitivity: 45.6%, specificity: 96.5%, and overall accuracy: 68.8%) and 0.24 (sensitivity: 75.0%, specificity: 73.7%, and overall accuracy: 73.6%), respectively. In addition, NLR (r = 0.322, p = 0.011) and PLR (r = 0.283, p = 0.026) were positively correlated with CRP. Importantly, NLR and NMR were increased while PNR was decreased in the LN group with infection in comparison with those in the LN group with flare. NLR (AUC: 0.80), NMR (AUC: 0.78), and PNR (AUC: 0.74) were useful in differentiating LN patients with infection and flare, and their optimal cutoffs were 4.02 (sensitivity: 82.6%, specificity: 69.6%, and overall accuracy: 75.5%), 12.19 (sensitivity: 80.4%, specificity: 73.9%, and overall accuracy: 77.5%), and 28.26 (sensitivity: 65.2%, specificity: 76.8%, and overall accuracy: 71.6%), respectively.
Conclusions
We demonstrated, for the first time, that MLR or NMR had the best accuracy in differentiating LN patients without infection from healthy subjects, or differentiating infection from flare in LN patients, respectively. Our results implied that NLR, MLR, PNR, and NMR may be useful biomarkers in predicting LN.
Keywords
Get full access to this article
View all access options for this article.