Correspondence on ‘Detection and Correlation of Changes on Perimetry and Optical Coherence Tomography in Patients on Chronic Usage of Hydroxychloroquine: A Cross-sectional Study’

To the Editor,

We read with interest the article by Priya et al., published in the recent edition of the Indian Journal of Rheumatology. ¹ This is a valuable addition to the limited data available regarding hydroxychloroquine (HCQ) retinopathy in the Indian population. The study identified macular thinning on Optical Coherence Tomography (OCT) in 52% of the eyes screened and visual field changes in 47%. A few points regarding this study are worthy of discussion.

First, the relatively high incidence of macular thinning identified by the study is surprising. The reported incidence of retinopathy varies widely in literature and depends on the screening modalities used, daily dose and duration of HCQ use, with large series showing an approximate incidence of 7.5%. ² Indian series have shown an incidence varying from 6.36% to 13.5%.^{3, 4} The incidence also depends on the cut-off utilised to define normal versus abnormal. An OCT thickness plot generates a colour-coded chart comparing the patient’s retinal thickness with normative data. In this chart, red indicates a retinal thickness <1% of normal distribution and yellow indicates a thickness <5% of normal distribution. The study has used <5% of normative data as suggestive of macular thinning and this might have led to the observed high incidence. However, there is no clear consensus on whether to use <1% or <5% as the abnormal cut-off. A recent study showed that <1% of normative data has the highest diagnostic yield. ⁵ In general, lowering the cut-off for any test would make it more sensitive but less specific, and this would be true for OCT as well. ⁶ Focal disruption of the IS/OS junction line on OCT is an early sign of HCQ retinopathy. ⁷ It would further add to the information if the authors could point out this finding in the study.

Second, the study reported 32 and 63% prevalence of macular thinning in the <2 year and 2–5 years exposure groups respectively. The retinopathy is seen with longer duration and larger cumulative dose and would be unusual to manifest in first 5 years of usage. ⁷

Third, the study showed a correlation between OCT retinal thickness and perimetry changes. It would be interesting, however, to know how many patients had changes simultaneously on both tests. The study identified paracentral and ring scotoma in approximately 50% of the eyes, which is quite high despite the relatively lower cumulative dose and duration of HCQ use. Visual fields are sensitive but subjective and have variable reliability, hence repeating the visual fields after a few months to confirm the findings might help. ⁷

Early detection of retinopathy is important; however, it is also equally important to avoid unnecessary discontinuation of the drug. Ideally, abnormal results on a test should be verified by other tests or by repeat testing.