RESUMO
OBJECTIVES: Our aim in this study was to assess the relationship between serum lipoprotein(a) [Lp(a)] and apolipoproteins and the risk of developing diabetic retinopathy (DR). METHODS: One thousand fifty-seven patients with type 2 diabetes were divided into 2 main groups and followed for 5 years: 637 patients without DR and 420 patients with DR. A group of patients with DR were then divided into 2 subgroups: 162 patients with nonproliferative DR (NPDR) and 163 patients with proliferative DR (PDR). The association between serum Lp(a) and apolipoproteins with NPDR and PDR was assessed using univariate and multivariate regression analyses. Receiver-operating characteristic curve analysis was performed based on the new cutoff values. RESULTS: There was a positive relationship between Lp(a) and the presence of DR as well as a negative correlation between ApoA and DR (p<0.001 and p=0.03, respectively). We also found a positive association between ApoB and the severity of DR (p=0.008). ApoA1 had an area under the curve of 55.0% for the prediction of DR. The calculated cutoff values of ApoB/ApoA1 ratio (0.58 g/L) and ApoB (77.5 g/L) in detection of DR were lower than their standard cutoff values of 0.8 and 90 g/L, respectively. Also, the sensitivity of new cutoff values for ApoB and ApoB/ApoA1 ratio was higher than the standard value, but the specificity of the standard cutoff values for both was higher than our new cutoff value. CONCLUSIONS: Serum Lp(a) and ApoA1 levels were independently associated with DR, and serum ApoB correlated with severity of DR. These measurements can be used for assessment and early treatment of this vision-threatening complication of diabetes.