The effects of oligomeric proanthocyanidins on hydroxychloroquine-induced retinopathy as revealed by nationwide medical claim data for South Korea

Although hydroxychloroquine (HCQ) is considered to be safe drug with anti-inflammatory, immunomodulatory, and antithrombotic effects,[1,2] rare cases of permanent visual impairment.[2,3] The prevalence of HCQ retinopathy is about 7.5% for more than five years, increasing after 20 years.[4] Oligomeric proanthocyanidins (OPCs) are antioxidants and rich in a grape seed extract.[5-7] The extract is effective in some patients with nonproliferative diabetic retinopathy.[8,9] The extract may protect retinal cells by inhibiting the oxidative stress-mediated apoptosis mediated by activation of the nuclear factor erythroid 2-related factor 2 (Nrf2) pathway in patients with diabetic retinopathy.[9] No study has explored whether the extract may protect against HCQ retinopathy.[10]

We used nationwide medical claims data of the Korean Health Insurance Review and Assessment Service (HIRA) to find the effect of co-administration of OPC, as single product name Entelon®, on the incidence of HCQ retinopathy was investigated in patients with rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) who were receiving HCQ from 2007 to 2019 (Figure 1). Statistical analysis was performed using the SAS software (version 9.4; SAS Institute Inc., Seoul, Korea) and R statistical software (version 4.0.3; R Foundation for Statistical Computing, Vienna, Austria).

We included 31,140 patients on HCQ in a no-OPC group and 7,785 in an OPC group after propensity score matching (Table 1a). The mean duration of HCQ use was 17.6±32.5 months and the mean cumulative dose 32.3±124.0 g (Table 1b). It was divided into three quantiles according to the HCQ durations and cumulative doses: low, medium, and high. By the duration of HCQ use, the cumulative incidence of retinopathy was 0 in the low and medium groups, but 3.92 (95% confidence interval [CI]: 3.52-4.35) in the high group (p<0.001). The cumulative incidence of HCQ retinopathy increased with HCQ duration and cumulative dose, as in previous studies (Figure 2). The cumulative incidence rate of retinopathy was somewhat higher in the OPC than in the no-OPC group, but it did not reach statistical significance (0.64 vs. 0.56 per 1,000 PY, p=0.681) (Figure 3a). When the incidence of HCQ retinopathy by OPC use was analyzed separately in RA and SLE patients, OPC made no significant difference to the incidence of retinopathy (Figure 3b, c). In the SLE subgroup, the cumulative incidence of retinopathy was somewhat lower in the OPC group without statistical significance. Multivariate analysis showed that such retinopathy was not affected by the duration of OPC use or cumulative dose of OPC. Longer duration of HCQ use (hazard ratio [HR]: 1.035, 95% CI: 1.032-1.037, p<0.001) were risk factors for HCQ retinopathy (Table 2).

In conclusion, the HIRA aids research on rare conditions, adverse events, and preventative effects of drugs in socially marginalized groups such as older and disabled adults who are seldon enrolled in RCTs. Although co-use of OPC did not significantly reduce this, it may be necessary to repeat the analysis, when SLE patients accumulate longer OPC exposure. Efforts to avoid permanent visual loss caused by HCQ retinopathy must be redoubled.