Hydroxychloroquine retinal toxicity- Can we do better with SD-OCT?

Abstract

Hydroxychloroquine used in the management of connective tissue diseases also used to treat a wide spectrum of diseases. This disease-modifying anti-rheumatic drug can cause irreversible retinal damage and there are preferred practice guidelines for early screening to aid ophthalmologists and all those involved in the management of these patients. Spectral-Domain Optical Coherence Tomography (SD-OCT) currently an important tool in early detection of possible toxicity. Even with the current guideline-recommended 5mg/kg/day dosage, a baseline SD-OCT of the macula should be taken and repeated at 5 years. Many cross-sectional studies using SD-OCT have shown earlier retina damages, but none had monitored continuously from the beginning of Hydroxychloroquine usage. To evaluate sequential changes of the retina thickness in patients newly treated with hydroxychloroquine. We compared the mean central macular and Retinal Nerve Fiber Layer (RNFL) thickness pre, 3 months and 6 months post hydroxychloroquine using SD-OCT in order to find an association between them over the three time-points.33 previously healthy patients who were newly confirmed clinically and serologically with connective tissue disease planned for hydroxychloroquine who fulfilled the inclusion and exclusion criteria were recruited over 2 years. In consideration of selection criteria and drop-out rate, finally, 26 patients were included in the study.26 eyes of 26 patients (20 females and 6 males) were included in the study. The mean age was 33.92. The macula was significantly thinner (P<0.05) at superior (P<0.032), temporal (P<0.025), and inferior (P<0.016) parafoveal areas. RNFL showed significantly thinning at the temporal quadrant (P<0.041). Further analysis showed significant occurrences of parafoveal inferior thinning at 3 (P<0.035) and 6 (P<0.032) months. Correlation analysis between parafoveal thinning and the mean optic disc RNFL showed a significant result (p =0.004, r=0.55). The mean optic disc RNFL thickness significantly affected lesser (p =0.004, r =0.302) than the perifoveal area (p =0.001, r =0.784) when compared to the typical parafoveal pattern of thinning in this toxicity.There is a need for an earlier repetition of SD-OCT imaging before 5 years to detect potential toxicity. Involvement of RNFL requires further investigation and monitoring.