C-reactive protein to high-density lipoprotein cholesterol ratio: an independent risk factor for diabetic retinopathy in type 2 diabetes patients

Abstract

Background and objective: Diabetic retinopathy (DR) is associated with abnormal lipid metabolism and inflammation. However, a single lipid or inflammatory parameter cannot accurately predict the prognosis of DR independently, because it is prone to be affected by various confounding factors. This study aimed to explore the relationship between the inflammation-lipid indicator C-reactive protein (CRP)/high-density lipoprotein cholesterol (HDL-C) and DR occurrence in subjects with type 2 diabetes mellitus (T2DM).

Methods: This hospital-based retrospective study included 784 T2DM patients. Diabetic retinopathy was diagnosed by nonmydriatic fundus photography and/or fundus examination apparatus. T2DM patients were divided into non-DR and DR groups. Demographics variables, clinical history and serum biochemical indicators of the subjects were collected. We also calculated the CRP/HDL-C ratio. The association between the CRP/HDL-C and DR was assessed using multivariate logistic regression analyses.

Results: A total of 784 participants, 612 without DR and 172 with DR, were included in the final sample analysis. Compared with non-DR participants, the DR diagnostic group had significantly higher CRP/HDL-C (4.03 ± 1.67 vs. 2.66 ± 0.97; p < 0.001). Then, the patients were grouped based on the quartiles of CRP/HDL-C, there was a gradual increase in the prevalence of DR was noted in T2DM patients along with the increased quartile of the CRP/HDL-C ratio (Q1: 7.65%; Q2: 15.31%; Q3: 19.90%; Q4: 44.90%; p = 0.028). After adjustment for the impact of various covariates, the odds ratio (OR) of the third and fourth vs. the first quartile of CRP/HDL-C were 2.905 (95% confidence interval [CI]: 1.372 ~ 6.152, p = 0.005) and 9.938 (95% CI: 4.987 ~ 19.804, p < 0.001), respectively. Further, multivariate logistic regression model showed that the CRP/HDL-C ratio (OR 3.176, 95% CI: 1.280 ~ 7.877, p = 0.013) was identified as risk factor for DR. Moreover, the area under the curve (AUC) to evaluate the predictive value of CRP/HDL-C for the risk of DR occurrence was 0.752 (95% CI: 0.711 ~ 0.794).

Conclusion: The ratio of C-reactive protein (CRP) to high-density lipoprotein cholesterol (HDL-C) is associated with DR in patients with T2DM, and CRP/HDL-C may be an effective marker to help identify the risk of DR in patients with T2DM.

Keywords: biomarker; diabetic retinopathy (DR); dyslipidemia; inflammation; the CRP to HDL-C ratio; type 2 diabetes mellitus (T2DM).

Figure 1

Receiver operating characteristic curves (ROC) of DR in T2DM. (A) CRP/HDL-C; (B) CRP/HDL-C combined other 4-indicators (Age, Duration, FPG, and HbA1c).