I thank Lyons for his comments and for alluding to Marmor’s editorial, providing an opportunity to respond to both. His first concern is that the use of hydroxychloroquine not be halted without good cause. One can understand this concern based on the statistics from his practice, in which 89 (14.8%) of 600 screened patients had retinal toxicity confirmed by the combination of visual fields and multifocal electroretinography. If all these patients were asked to stop hydroxychloroquine, the rate would be 13-fold that in our sample (2 of 183 cases screened, or 1.1%) and 10-fold that of Elder and Rahman (4 of 262 cases screened, or 1.5%). One wonders if we define hydroxychloroquine toxicity the same way. Moreover, an ophthalmologist’s potential response to concern about toxicity is more varied than simply recommending cessation of hydroxychloroquine, as Lyons’ comment might be taken to imply. Table 2 in my paper showed that ophthalmologists recommended dosage reductions approximately 6 times as often as they recommended cessation of drug when toxic dosing without manifest toxicity was discovered.

His second concern is that cumulative dosing of more than 870 g be recognized as more strongly associated with toxicity than adjusted daily dose. My article makes no claim about the relative importance of the association of cumulative dose and risk of toxicity compared with adjusted daily dose. I agree that it is important, but the issue is irrelevant to clinical care. The 2002 American Academy of Ophthalmology guidelines already advocated yearly screening after 5 years of drug use (ie, at a cumulative dose of approximately 730 g), so recognizing the importance of cumulative dose at his suggested threshold leads to no different action. However, recognizing the importance of toxic daily dosing does lead the clinician to an action that may prevent, not just detect, retinopathy. The ophthalmologist can advise the internist to reduce a toxic dose. This is the reason for emphasizing it, and for understanding the nuances of adjusted daily dose.

These nuances bear elaboration and reference to Marmor’s editorial. Lyons does not say whether his statistic of less than 50% of toxic patients taking less than 6.5 mg/kg refers to actual or ideal body weight (IBW). If he means actual body weight, then we need to know the corresponding statistic using IBW. In an overweight society, the latter percentage generally will be higher. We also should challenge Marmor’s implication that his formula for IBW is preferred. There are many investigators who use other formulas. It matters which algorithm one uses. For example, if one considers a woman who is 63 inches tall and weighs 140 pounds, the IBWs using different algorithms in published use by ophthalmologists include 98 pounds, 115 pounds (Marmor’s algorithm), 111 to 124 pounds (Metropolitan Life Insurance Small Frame Table), 135 pounds, and 131 to 147 pounds (Metropolitan Life Insurance Large Frame Table).

No study indicates that Marmor’s choice is the best one, and there are reasons to believe that it is not. For example, applying Marmor’s algorithm to a sample of 235 women screened in our clinic for whom height was recorded (median, 64 inches; interquartile range, 62 to 66 inches) implies that 82% of women taking 400 mg/d of hydroxychloroquine will be deemed as taking more than 6.5 mg/kg daily of IBW compared with 28% using the National Heart, Lung and Blood Institute table. It is impractical to suggest that 82% of screened women taking the most common daily dose will need a call to the internist to reduce the dose.

Multifocal electroretinography and spectral-domain optical coherence tomography are welcome ancillary tests in selected clinical situations, but evidence has not been published that their universal application in practice, as advised in the revised guidelines, reduces the incidence of hydroxychloroquine retinopathy, nor that this is a prudent use of health care dollars. However, detecting and reducing toxic dosing, rather than accepting it as an unalterable risk factor such as age or cumulative dose, adds no cost in a screening environment and can prevent a fraction of cases that would occur if toxic dosing were allowed to continue.