A 74-year-old Caucasian woman came to the retina clinic as a referral for evaluation of possible hydroxychloroquine toxicity. Her chief complaint was blurry vision in her right eye. She had a history of hydroxychloroquine use for 27 years for Still’s disease, at a current dose of 200 mg every other day. 

 

History of HCQ use

While she never exceeded an HCQ dose of 5 mg/kg/day, she did have a period of simultaneous tamoxifen use while she was undergoing treatment for breast cancer. Her Still’s disease was refractory to multiple other immunomodulatory agents. 

Her ocular history was significant only for recent cataract surgery in the right eye, and she was being monitored as a  glaucoma suspect. On examination, her best-corrected visual acuity was 20/25 OD and 20/20 OS. Intraocular pressures were normal. Pupillary reactions were normal without an afferent pupillary defect, and extraocular movements were full. The slit lamp and dilated fundus examinations didn’t have any significant findings. 

Optical coherence tomography (Figure 1) demonstrated outer retinal loss on either side of the fovea, which was more prominent in the left than the right eye. Given these findings, a multifocal electroretinogram (mfERG) was obtained to evaluate for evidence of photoreceptor dysfunction. The first mfERG was found to have modestly decreased amplitudes, equivocally within the limits of normal, but possibly representing early toxicity (Figure 2A). A repeat mfERG one year later demonstrated more definitive evidence of amplitude reduction, consistent with drug-induced retinal toxicity. 

A second case of HCQ toxicity

A 58-year-old woman of Asian descent was referred to the retina service for blurring in her left eye. She denied any focal scotomas or metamorphopsia. She had an ocular history of dry eye and used Restasis (cyclosporine 0.05%, Allergan/AbbVie). 

Figure 1. Optical coherence tomography of the left (A) and right (B) eyes demonstrate left outer retinal loss in the inner segment/outer segment junction more so than in the right. 

Her medical history included Sjögren’s syndrome, lymphoma and cryoglobulinemia. She had been taking HCQ 200 mg daily, with a short period of 400 mg daily, for 10 years. She denied any allergies to medications. 

Visual acuity was 20/20 in both eyes, with normal intraocular pressures. Pupillary reactions were normal without afferent pupillary defect, and extraocular movements were full. Slit lamp and dilated fundus exams were notable only for a peripheral choroidal nevus in the left eye. 

We ordered fundus autofluorescence to evaluate for possible HCQ toxicity. It demonstrated hyperautofluorescence following the arcades in the left greater than the right eye (Figure 3). OCT demonstrated focal peripheral areas of the inner segment/outer segment (IS/OS) junction loss in a pattern corresponding to the areas of hyperautofluorescence. 

Given these findings, we obtained a mfERG, which demonstrated diffuse slowing and amplitude reductions (Figure 2B) and highlighted dysfunction that wasn’t detected on screening Humphrey visual field (HVF) testing (Figure 4). 

 

Diagnosis and management

Figure 2. Multifocal electroretinogram (mfERG) of patient one (A) and patient two (B) both show reduced amplitudes suggestive of decreased retinal function, more pronounced in patient two. C shows a normal mfERG for comparison. 

Given the multimodal imaging findings, both patients were counseled about the risk of progressive retinal toxicity and irreversible vision loss with ongoing hydroxychloroquine use. After discussing alternative methods of managing their respective systemic diseases with their rheumatologists, both patients ultimately decided to discontinue their HCQ. Both have had stable ocular findings on OCT and fundus exam without evidence of worsening toxicity since.

These cases demonstrate the utility of mfERG in the diagnosis and management of HCQ toxicity. In both patients, mfERG helped demonstrate functional retinal damage at the outset, which served as a helpful tool to both provide objective data in the diagnosis of HCQ toxicity, as well as in counseling them to cease using the medication. Of note, minimal changes were seen on prior screening HVF testing for the second patient leading up to her referral to our clinic, in contrast to the changes seen on mfERG.

 

How to screen for HCQ toxicity

HCQ toxicity can be subtle and usually develops after years of treatment. The risk is low, with incidence previously described at 0.5 to 2 percent of patients with extended use.1-4 However, these estimates are thought likely to underestimate the extent of toxicity.1 

Risk of toxicity has been found to be dose dependent. Patients on doses <5 mg/kg (actual body weight) have a less than 1 percent risk in the first five years and a less than 2 percent risk in 10 years of treatment.5 

Figure 3. Fundus autofluorescence of the right (A) eye shows hyper-autoflourescence in the peripheral macula and around the nerve, and superotemporal hyper- and hypoautoflourescence focally corresponding with a nevus. The left eye (B) shows hyperautoflourescence in the inferior peripheral macula and around the nerve, with speckled hypoautoflourescence in the superonasal macula. 

Increased risk has been noted in >1,000 g cumulative dose, >5 mg/kg/day treatment, more than five years of use, kidney disease, tamoxifen use, as well as history of macular degeneration and other retinal diseases.1,5 Tamoxifen use is especially important to elucidate, given it has been shown to cause a fivefold increased risk in toxicity.1,5 In addition, older age, liver disease and some genetic variants in ABCA4 have been found to increase the risk as well.5

The American Academy of Ophthalmology published updated recommendations for HCQ toxicity screening in 2016, recommending a baseline fundus examination in the first year of use to detect any preexisting retinal abnormalities, then yearly afterward in the absence of any of the previously mentioned risk factors.5 

 

Recommendations for HCQ toxicity

The AAO recommendations recognized that most providers will use spectral-domain OCT and HVF in conjunction with the fundus exam to screen for subtle changes related to HCQ toxicity, but noted this isn’t required at the baseline examination unless abnormalities are apparent on examination. The recommendations encourage use of SD-OCT and HVF at follow-up to help detect toxicity, and strongly recommend these more objective tests if a diagnosis of toxicity is made.5

Ronald B. Melles, MD, and colleagues further described in their study that 50 percent of patients of Asian descent had pericentral changes (more than 8 degrees from the fovea), compared to 2 percent in white patients6—an observation that correlates with the findings in our second patient.6 

Screening recommendations for patients of Asian descent call for 24-2 HVF rather than the 10-2 HVF screening to better determine peripheral changes. However, on both 10-2 and 24-2 HVF nonspecific changes can be difficult to interpret, and the subjective nature of the study limits their use as a screening tool.

Figure 4. On initial screening, Humphrey visual fields of patient two demonstrate minimal changes.


Fundus autofluorescence has also been suggested as a useful tool to augment sensitivity in testing, and can demonstrate early changes prior to retinal pigment epithelium changes from toxicity.5,7,8 

More recently, mfERG has been suggested as an adjuvant tool in suspicious cases to demonstrate functional changes,8 as the findings may be more objective and subject to less variation and confounding factors than HVF testing. Indeed, the cases we present here illustrate the utility of mfERG in management of HCQ toxicity. 

 

Bottom line 

HCQ toxicity is rare, but it can be severe and it is irreversible. The risk of toxicity increases with greater duration of use, higher doses and with retinal disease and systemic toxicity. SD-OCT, FA and HVF augment the sensitivity of the physical exam to pick up on early changes in hydroxychloroquine toxicity. Multifocal electroretinogram may help to correlate functional photoreceptor loss to anatomical changes to better diagnose toxicity and justify stopping HCQ for patients.  RS

 

REFERENCES

1. Melles RB, Marmor MF. The risk of toxic retinopathy in patients on long-term hydroxychloroquine therapy. JAMA Ophthalmol. 2014;132: 1453–1460.

2. Wolfe F, Marmor MF. Rates and predictors of hydroxychloroquine retinal toxicity in patients with rheumatoid arthritis and systemic lupus erythematosus. Arth Care Res (Hoboken). 2010; 62:775-784. 

3. Levy GD, Munz SJ, Paschal J, Cohen HB, Pince KJ, Peterson T. Incidence of hydroxychloroquine retinopathy in 1,207 patients in a large multicenter outpatient practice. Arth Rheum. 1997;40:1482-1486. 

4. Mavrikakis I, Sfikakis PP, Mavrikakis E, et al. The incidence of irreversible retinal toxicity in patients treated with hydroxychloroquine: A reappraisal. Ophthalmology. 2003;110:1321-1326.

5. Marmor MF, Keller U, Lai TYY, Melles RB, Mieler, American Academy of Ophthalmology. Recommendations on screen for chloroquine and hydroxychloroquine retinopathy. Ophthalmology 2016;123(6):1386-1394. 

6. Melles RB, Marmor MF. Pericentral retinopathy and racial differences in hydroxychloroquine toxicity. Ophthalmology. 2015;122:110-116.

7. Marmor MF. Comparison of screening procedures in hydroxy-chloroquine toxicity. Arch Ophthalmol. 2012;130:461-469. 

8. Kellner U, Renner AB, Tillack H. Fundus autofluorescence and mfERG for early detection of retinal alterations in patients using chloroquine/hydroxychloroquine. Invest Ophthalmol Vis Sci. 2006;47:3561–3568.