Retinal Thickness Changes May Not Predict Hydroxychloroquine-related Retinopathy

By Will Boggs MD

March 25, 2016

NEW YORK (Reuters Health) - Changes in the inner and outer retinal layer do not reliably signal the development of retinopathy in patients taking hydroxychloroquine (HCQ), according to a retrospective study.

"We suspected from an earlier paper of ours that the inner retina (ganglion cell layer) was really not involved with this toxicity to any significant degree," Dr. Michael F. Marmor, from Stanford University in California, told Reuters Health by email. "The most surprising finding was that even the outer retina did not show gradual thinning over years of HCQ usage."

"The photoreceptors are clearly the site of damage clinically," he explained, "and this result shows that there is no chronic or progressive damage visible across the retina until one sees the focal damage that characterizes HCQ retinopathy."

Dr. Marmor's team investigated spectral-domain optical coherence tomography (SD-OCT) findings in 12 short-term (<5 years) and 15 long-term (>15 years) HCQ users without toxic effects whose follow-up ranged from 25 to 52 months.

At baseline, there were no significant differences in full retinal thickness or in inner and outer retina thicknesses between short- and long-term users, according to the March 17 JAMA Ophthalmology online report.

Over time, there were no consistent gains or losses in full retinal thickness in either group, and the rates of inner and outer retinal thickness change were minimal.

In the only patient who developed retinopathy, SD-OCT scans were normal at baseline and one and two years later. At year 4 (after missing her 3-year visit), the patient remained asymptomatic, but her visual field tests showed evidence of early retinopathy, and her SD-OCT scan showed focal outer retinal thinning in the left eye.

"Despite some presumed metabolic stress on the photoreceptors over time, we don't see a gradual loss of cells," Dr. Marmor said. "Rather, there seems to be a critical point in a given patient at which local decompensation and damage finally begins."

"Screening should look at the photoreceptor layers particularly for new regional areas of cellular loss (typically just outside the fovea in European patients, but sometimes near the vascular arcades in Asian patients)," Dr. Marmor said. "New recommendations, endorsed by the American Academy of Ophthalmology, on how to screen for HCQ toxicity are in press, and should be available very soon."

"At this time, the totality of the data appears to indicate that a comprehensive approach should be taken for hydroxychloroquine retinopathy screening," wrote Dr. Judy E. Kim, from the Medical College of Wisconsin, Milwaukee, in an accompanying invited commentary.

"The evaluation should start with an assessment of the patient for high-risk characteristics and calculation of the appropriate daily dosage," she explained. "We should remember that a cumulative dose of 1000 g of hydroxychloroquine is reached in seven years with a daily dose of 400 mg. As for screening tests, multifocal electroretinography and the 10-2 visual field test are sensitive tests when reliable, whereas SD-OCT and fundus autofluorescence are more specific. Therefore, the most practical screening tests for hydroxychloroquine toxicity appear to be a combination of the 10-2 visual field test and SD-OCT."

"Screening earlier than five years after initiation of hydroxychloroquine and, possibly, a more-frequent-than annual screening should be considered when there are risk factors, equivocal test results, or a suspicion of early toxicity," Dr. Kim concluded. "As imaging modalities continue to improve, we should strive to find ways to diagnose hydroxychloroquine toxicity with increased sensitivity and specificity at the earliest stage."

Dr. Marmor noted, "The main key to minimizing toxicity from HCQ, and lengthening its years of use, is proper dose. Recent findings show that the old guidelines no longer hold. The new proper dose recommendation is: stay below 5 mg/kg real (absolute) weight."

Dr. Rishi P. Singh from Cleveland Clinic's Cole Eye Institute, Ohio, told Reuters Health by email, "The most interesting result is that no anatomical changes were appreciated prior to developing toxicity, suggesting that toxicity is not due to an incremental accumulation of HCQ."

"We recently performed a study that demonstrates inner and outer retinal thinning in patients with HCQ toxicity as compared to age-matched controls from our group," Dr. Singh said. "Understanding why retinal thinning occurs precipitously may give us insight into how to better detect and prevent HCQ toxicity."

The Retina Research Foundation and the Retina Society partially supported this research. The authors made no disclosures.


JAMA Ophthalmol 2016.