To the Editor:

Chloroquine (CQ) and hydroxychloroquine (HCQ) are used widely in the treatment of rheumatic disease, and they have the potential to cause serious retinal toxicity when high doses have been taken over a long period of time. What type and what frequency of screening should be performed to recognize toxicity as early as possible? A few authors have proposed that screening is not required, given the relative rarity of toxic retinopathy, but this seems unwise since the result can be blindness. The Physicians' Desk Reference recommends quarterly examinations (Physicians' desk reference. 57th ed. Montvale [NJ]: Thomson PDR; 2003), but this is too frequent relative to the speed of clinical changes and does not seem cost-effective relative to the low incidence. The American Academy of Ophthalmology (AAO) recently assembled a task force to examine the available evidence on these drugs and to develop screening recommendations that would be in accordance with both modern knowledge and the economic realities of practice (Marmor MF, Carr RE, Easterbrook M, Farjo AA, Mieler WF, for the American Academy of Ophthalmology. Recommendations on screening for chloroquine and hydroxychloroquine retinopathy. Ophthalmology 2002;109:1377–82).

The mechanism of CQ and HCQ toxicity is not well understood, and it is unknown whether the primary damage is upon the retinal photoreceptors or the melanotic retinal pigment epithelium (RPE) layer that lies just behind the retina. The typical clinical finding of early retinopathy is paracentral depigmentation surrounding the center of vision, called “bull's-eye maculopathy.” Sometimes a loss of visual sensitivity can be discerned even before fundus changes are visible, by careful testing of paracentral vision. If drug usage continues, toxicity will eventually result in widespread retinal and RPE atrophy. The only known treatment is cessation of the medication, but unfortunately the retinopathy will often progress for some time even after it is stopped. Thus, early recognition is the best defense against serious vision loss.

While the retinopathy is potentially serious and permanent, the incidence of toxicity in clinical practice is very small. Studies of several large series of patients with rheumatic disease have demonstrated little or no toxicity among thousands of subjects. The vast majority of reports of toxicity have occurred in individuals taking >6.5 mg/kg/day of HCQ or 3 mg/kg/day of CQ, for >5 years.

The AAO committee recommended that patients treated with CQ or HCQ take maintenance dosages below these levels, if at all possible. Patients taking the drugs can be divided into 2 groups: low-risk patients, who receive these lower dosages and have used the drug for <5 years, and higher-risk patients, who have used the drug for longer periods, who are taking larger dosages, or who have other risk factors such as high body fat level, concomitant kidney or liver disease, concomitant retinal disease, or older age (>60 years). All individuals starting these drugs should have a complete baseline ophthalmologic examination within the first year of treatment. This should include examination of the retina through a dilated pupil and testing of central visual field sensitivity by either a self-testing grid chart (Amsler grid) or an automated field tester (Humphrey 10-2 testing). If the patient is in the low-risk category and these examination results are normal, the AAO recommendation is that no further special ophthalmologic testing is needed for the next 5 years.

In the US, HCQ is used more than CQ, and it is available only in a 200-mg tablet. Maintenance treatment with 200 mg of HCQ daily will be relatively safe for virtually all but the lowest-body-weight patients, but 400 mg daily puts patients who weigh <135 pounds into the higher-risk category. However, adjustment of HCQ dosage is easy to achieve because the duration of action is so prolonged that daily dosage levels can be calculated on a weekly basis (for example, taking 2 tablets one day, and 1 the next).

For patients in the higher-risk category, which includes anyone using these drugs regularly for >5 years, annual eye examinations are recommended. Although the incidence of toxicity remains relatively low with long-term use, it becomes more difficult to predict who may or may not be at risk, and the consequence of missing the early signs of retinopathy can be permanent vision loss. The screening procedure should include retinal examination and Amsler grid or Humphrey field testing. Should any hint of toxicity appear, more elaborate tests can be performed, such as multifocal electroretinography, which objectively measures the electrical responses from the central retina.

If toxicity is suspected or documented, the drug should ideally be stopped. This is not always an easy decision, however, especially if the impression of toxicity is early or tenuous, or if the treatment has been very effective in controlling a severe or unstable rheumatic condition. The medical physician and the patient should be informed about the ophthalmologic findings and the risks, so a cooperative decision can be made about whether to halt the drug or to cautiously continue it with close monitoring of vision function and with an acceptance that some vision could be lost.

It is hoped that these new recommendations will clarify the rationale for HCQ and CQ screening and will also make it easier during the initial years of treatment, when the risk of retinopathy is low.

Michael F. Marmor MD*, * Stanford University Medical Center, Stanford, CA.