The Hydroxychloroquine Dose That Works, and the Weight-Based Ceiling That Says It Shouldn't
The dose that has kept her out of a flare for three years is technically above the retinopathy-safety ceiling for her real body weight. The disagreement is whether that ceiling was ever measuring the right thing for a patient built like her.
J.B., a 46-year-old woman who manages warehouse operations, has been on hydroxychloroquine 400mg daily for the past three years, the dose that finally controlled recurrent arthritis and pleuritic flares after two lower doses failed over the preceding two years. She weighs 168 pounds (76kg) at 5'2" — a real body weight that puts her at 5.3mg/kg on her current dose, just over the 5mg/kg real-body-weight ceiling that the American Academy of Ophthalmology's 2016 revised guideline identifies as the single biggest modifiable risk factor for hydroxychloroquine retinopathy. Her annual optical coherence tomography and visual field testing today are both normal. Counting the two years at lower doses that preceded her current one, today is her fifth year on the drug — the cumulative-duration threshold at which the guideline stops treating annual screening as optional, reached at this visit rather than at some point still ahead of her. She has no diabetes, no chronic kidney disease, and takes no tamoxifen, so dose and duration are the only two risk factors she carries at all.
The 5mg/kg threshold comes directly from Melles and Marmor's 2014 dose-response analysis, which found retinopathy risk climbing sharply above that cutoff and comparatively low below it — population-level data, not an arbitrary number. The metric matters as much as the number: the 2016 revision moved the ceiling from ideal body weight to real body weight precisely because real weight predicted risk better, and because thin patients dosed on ideal weight had been systematically underprotected. That closes off the workaround her chart seems to invite, since any adjusted or lean weight is a smaller denominator and would put her milligrams per kilogram higher rather than lower. What the arithmetic does yield is the size of the breach: 5mg/kg of her 76kg is 380mg, so she sits twenty milligrams — one-twentieth of her dose — above her own ceiling, a far narrower gap than the two dose reductions that already failed her.
Annual co-management visit, dose review
Two lower doses failed her over two years before 400 milligrams finally controlled her disease — that's not a preference, that's a documented pattern. Her OCT and visual fields are normal today. I'd rather monitor her more closely than trade a real, proven level of disease control for a population ceiling that may not describe her specifically.
The 5mg/kg ceiling isn't arbitrary — Melles and Marmor's dose-response analysis found retinopathy risk climbing sharply above it. Her current dose puts her at 5.3 milligrams per kilogram, over that line, and counting the two years at lower doses she is at five years today, not approaching it. A good result today on OCT doesn't mean the cumulative risk curve doesn't apply to her.
I take the treatment-failure history seriously — I'm not suggesting we ignore it. I'm saying the ceiling itself was built from real dose-response data, and being above it isn't a technicality.
Before either of you reaches for an adjusted or lean body weight to make this go away — that route isn't open to us. The 2016 revision moved off ideal body weight deliberately, because real weight predicted risk better, and any adjusted figure is a smaller number on the bottom of the fraction, which raises her milligrams per kilogram rather than lowering them.
What I'd rather do is size the breach honestly. Five milligrams per kilogram of 76 kilograms is 380 milligrams. She is twenty milligrams over — not in a different category of patient. Four hundred alternating with three hundred averages 350, which clears her ceiling on tablets she already has, and is a far smaller step down than either of the two that failed her. If even that costs her control, the answer isn't a recalculation, it's methotrexate alongside it so the hydroxychloroquine has room to stay down.
Agreed: alternate 400mg and 300mg for an average of 350mg daily, clearing her own 380mg ceiling by the smallest reduction available rather than by the larger step-downs that failed her twice, and continue annual OCT and visual field testing, which her fifth year on the drug now makes non-optional rather than routine. Not agreed: whether a twenty-milligram breach warranted changing anything at all — the position that a documented, twice-repeated treatment failure outweighs a marginal excursion above a population ceiling was not withdrawn, only outvoted on the grounds that a reduction this small is reversible if her disease answers back. If arthritis or pleuritic pain returns at 350mg, adding methotrexate so the hydroxychloroquine can stay down, rather than returning to 400mg, is the named next step.