Association between neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, and diabetic retinopathy among diabetic patients without a related family history

Abstract

Background: Diabetic retinopathy (DR) is a specific neurovascular complication of diabetes mellitus (DM). Clinically, family history is a widely recognized risk factor for DR, assisting diagnosis and risk strata. However, among a great amount of DR patients without hereditary history like hypertension and diabetes, direct and simple risk factors to assist clinical decisions are still required. Herein, we intend to investigate the associated risk factors for these DR patients based on systemic inflammatory response indexes, neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR).

Methods: We consecutively enrolled 1030 patients with a definite diagnosis of type 2 diabetes mellitus (T2DM) from the endocrinology department of the Second hospital of People in Yun Nan. Based on funduscopy and family history checking, we excluded patients with a family history of hypertension and diabetes and finally enrolled 264 patients with DR and 206 patients with non-diabetic retinopathy (NDR). Through correlation analysis, univariate and multivariate regression, we further explore the association between NLR, PLR, and DR. On top of that, we investigate the effect of NLR and PLR on risk reclassification of DR.

Results: Compared with NDR patients, NLR and PLR levels are significantly higher among DR patients (NLR: 2.36 ± 1.16 in DR group versus 1.97 ± 1.06 in NDR group, p < 0.001; PLR: 11.62 ± 4.55 in DR group versus10.56 ± 4.45 in NDR group, p = 0.012). According to univariate analysis, NLR and PLR add risks to DR. After fully adjusting co-founders, NLR, as both continuous and categorical variate, remains an independent risk factor for DR (OR (95%CI): 1.37 (1.06, 1.78) P = 0.018). And though PLR was not independently associated with DR as a continuous variable (OR (95%CI) 1.05 (0.99, 1.11) p = 0.135), the highest quantile of PLR add two-fold increased risk (OR (95%CI) 2.20 (1.05, 4.59) p = 0.037) in the fully adjusted model for DR. In addition, addition of PLR and NLR to the established factor hemoglobin (Hb) improved the discriminability of the model and assisted the reclassification of DR. After combining PLR and NLR the Area under curve (AUC) of Hb based model raised from 0.76 to 0.78, with a category-free net reclassification improvement (NRI) of 0.532 (p < 0.001) and integrated discrimination improvement (IDI) of 0.029 (p < 0.001).

Conclusions: Systemic inflammatory response indexes NLR and PLR were associated with the presence of DR among patients without associated family history and contributed to improvements in reclassification of DR in addition to Hb.

Keywords: Diabetic retinopathy; Family history; Hemoglobin; Neutrophil-to-lymphocyte ratio; Platelet-to-lymphocyte ratio; Type 2 diabetes mellitus.

Fig. 1

Comparison of NLR and PLR in different group a Boxplot showed that DR patients had higher NLR than the NDR patients (p < 0.001**). b Boxplot showed that DR patients had higher PLR than the NDR patients (p = 0.012)

Fig. 2

Receiver operating characteristic (ROC) curves of NLR, PLR and Hb for DR a The ROC curves using Hb, NLR or PLR alone for discriminating DR (Hb, black, NLR, red, PLR, green) b The ROC curves included and not included NLR and PLR for predicting DR, (Model1: black, a combination of NLR, PLR and Hb for predicting DR, Model2: red, using Hb along to predict DR)