Safe dosing for HCQ dependent on body type

Physicians should consider maximum HCQ  dose on Ideal weight for patients who are short and obese and real weight for short, thin individuals.

Recent years have brought major changes to the recommendations on screening patients for chloroquine and hydroxychloroquine (HCQ) retinopathy.

Daily dosing is the only modifiable risk factor. Previous guidance from 2011 said this was unimportant, except in short obese patients and recommended <6.5 mg/kg of ideal body weight, said David J. Browning, MD, PhD.

In 2016, the AAO revised its recommendations, stating that patients taking doses >5.0 mg/kg of real body weight per day were at greater risk of developing toxicity.1

"Real weight correlates better with risk compared with ideal weight," the authors stated.

The real weight guidance is based on data from a retrospective chart review of 2,361 patients who had taken HCQ for >5 years; authors Melles and Marmor used logistic regression and receiver operating curves to analyze the data,2 Dr. Browning noted.

The risk of retinopathy as per body habitus indicated by body mass index (BMI) based on ideal weight showed the risk was greater for aesthenic somatotypes.

"This was not the case, however, when they looked at real body weight-instead there was an invariant risk over BMI (Figures 1 and 2)," he said.

When the two graphs are plotted on the same axis, the data reveal that for obese patients with BMIs >30 kg/m2, the ideal body weight method yields lower risk (Figure 3).

"This presents a paradoxical situation in which the guidelines encourage the real weight method using a new threshold of 5 mg/kg, but the data show that dose increases the risk of toxicity in patients who are obese," Dr. Browning said.

Dr. Browning's group tried to replicate the results of Melles and Marmor by analyzing an independent database including 567 patients taking HCQ of whom 41 had retinopathy.3

"We found that for the ideal body weight method, the results mirrored those of the authors-there was higher risk at lower BMIs-but we could not replicate the results using the real body weight method (Figures 4 and 5). Instead of invariance, there was a higher risk for the real body weight method at lower BMIs," he said.

Melles and Marmor calculated receiver operating characteristic curves to assess the sensitivity and specificity of real versus ideal body weight in predicting toxicity, noting that a higher area under the curve (AUC) implies a better ability to predict retinopathy.

"They reported AUCs of 0.78 for real and 0.75 for ideal weight (Figure 6)," Dr. Browning said. "Although this difference in the two methods is statistically significant, it is clinically unimportant and achieves statistical significance by having a massive sample size (n = 2,361). Furthermore, the analysis was univariate, and it ignored the effects of covariates such as gender, age, cumulative dose, and BMI."

When Dr. Browning's group put this information into a model and analyzed the complete data, they found that the AUCs were identical for the two methods (Figure 7).

"The results on which the guideline change was based have not been replicated, and in fact, are in dispute," Dr. Browning contended.

Instead, it makes more sense to use the ideal body weight method for short, obese patients and the real body weight for short, asthenic patient, he said.

"Our country is the midst of an obesity epidemic," Dr. Browning emphasized. "During the past 50 years, the typical patient who takes HCQ has had an increase in weight of 27.5 lbs which correlates to a 63 mg increase in the ceiling dose if one uses the AAO guidelines. HCQ is concentrated in lean tissues and stored less in fat, making this a dangerous situation."

Dosing HCQ by ideal body weight is unsafe for short asthenic somatotypes, and dosing by real body weight is unsafe for short obese somatotypes. Using the lower dose between ideal and real body weight is safer for all patients.


This article was adapted from Dr. Browning's presentation during the 2017 meeting of the American Academy of Ophthalmology. He had no disclosures that were relevant to this talk.


1. Marmor MF, Kellner U, Lai TYY, et al. AAO Statement. Recommendations on screening for chloroquine and hydroxychloroquine retinopathy 2016 revision. Ophthalmology. 2016;123:1386-1394.
2. Melles RB, Marmor MF. The risk of toxic retinopathy in patients on long-term hydroxychloroquine therapy. JAMA Ophthalmol. 2014;132:1453-1460.
3. Browning DJ, Lee C. Somatotype, the risk of hydroxychloroquine retinopathy, and safe daily dosing guidelines. Clinical Ophthalmology. 2018;12:811-818.