Rheumatoid Arthritis in Apparent Remission: Widespread Pain and the Limits of the DAS28
A single patient whose composite disease-activity score says one thing and whose exam and imaging say another. The disagreement isn’t whether her pain is real — it’s whether a number built on subjective inputs should drive an escalation her objective findings directly contradict.
Beatriz N., a 58-year-old woman, spends most weekends at an easel set up in her converted garage, working through a stack of landscape commissions she says she never quite catches up on, though for the past six months she has had to prop her forearms on a cushion just to hold a brush steady through a full session. She was diagnosed with seropositive rheumatoid arthritis nine years ago, and the last six of those, on adalimumab and methotrexate together, have been genuinely quiet by every objective measure her rheumatologist tracks — stable hand and foot films, a C-reactive protein that has stayed under 3 for more than two years, and, until recently, joints she describes as simply not bothering her day to day.
What has changed is pain that no longer maps to any one joint: it runs through her shoulders, hips, low back, and both forearms, worse on days she hasn’t slept well, which lately is most of them, and comes with a fogginess she describes as losing her train of thought mid-sentence while sketching, something that never happened before this year. At her last visit her DAS28-ESR came out at 5.3 — high disease activity, since that band begins above 5.1 — but the number is almost entirely a product of her tender joint count (14 of 28) and her own global assessment (72 out of 100). Her swollen joint count is zero, her CRP drawn the same day is 2, and the sedimentation rate carrying the remainder of the score is 24, which sits inside the normal range for a woman of 58 and is doing arithmetic work that the inflammation it is supposed to be measuring is not. A musculoskeletal ultrasound with power Doppler, ordered specifically to settle the question the numbers alone couldn’t, found no synovitis anywhere examined. Her widespread pain, taken on its own criteria — pain in multiple body regions, more than three months’ duration, unrefreshing sleep, and cognitive symptoms — meets the threshold for fibromyalgia in its own right under Wolfe et al.’s 2016 revisions to the American College of Rheumatology’s 2010/2011 criteria. Strip the two subjective terms out of that 5.3 and what is left — a swollen joint count of zero, a CRP of 2, an age-normal sedimentation rate — is the profile of a patient in remission. The instrument has not detected her arthritis flaring. It has detected that she hurts, which is a different finding with a different treatment.
What the DAS28 is actually measuring today
Her DAS28-ESR is 5.3 — above the 5.1 line, so high disease activity by the formula’s own threshold, and treat-to-target strategy says a score like that should trigger escalation, whether that means increasing or switching her biologic or adding a second conventional DMARD. The entire point of a validated composite score is that it doesn’t require re-litigating each individual patient’s subjective inputs case by case.
You’re right that treat-to-target strategies work precisely because they don’t require re-litigating every case — but this exact scenario has already been studied, and it isn’t a case-by-case override, it’s a described failure mode of the instrument itself.
Ranzolin et al., in a 2009 study of 270 rheumatoid arthritis outpatients in Arthritis Care & Research, found that comorbid fibromyalgia specifically inflates DAS28 in rheumatoid arthritis patients through the subjective components — tender joint count and patient global assessment — without corresponding synovitis, which is exactly the pattern in front of us: zero swollen joints, a normal CRP, and a negative power Doppler ultrasound. Escalating an immunosuppressive regimen here would mean treating the population Ranzolin’s own paper describes, aimed at a mechanism her imaging already ruled out.
I agree escalation isn’t supported by what we actually have — but neither escalating nor simply holding her RA regimen treats the pain generator most consistent with her presentation. Central sensitization has its own real, effective first-line pharmacotherapy that hasn’t been offered yet.
Start duloxetine, an SNRI with genuine efficacy specifically in fibromyalgia’s central pain amplification — a distinct mechanism from anything in her RA regimen — and address her sleep directly. Reassess in eight to twelve weeks — but not by swapping in another composite. The CDAI is the obvious substitution and it’s the wrong one: it drops the acute-phase reactant altogether and keeps both the tender joint count and her global assessment, so on today’s numbers it lands in the low twenties and the two terms already inflating her DAS28 would carry even more of it. Track the objective limb by itself instead — swollen joint count, CRP, and a repeat power Doppler if her exam changes — so the RA question isn’t permanently deferred, just asked with the parts of the assessment her fibromyalgia cannot move. I’ll concede the fibromyalgia diagnosis itself rests on criteria, not a single confirmatory test — a real, if minor, source of uncertainty here.
Agreed: hold the current RA regimen unchanged, start duloxetine, address sleep directly, and reassess in eight to twelve weeks on the objective measures alone — swollen joint count and CRP — rather than on any composite score containing her tender joint count or global assessment.
Not agreed, and left as a real open branch rather than smoothed over:
No need to repeat the power Doppler ultrasound at follow-up if her swollen joint count stays at zero and her own report improves.
Repeat it regardless, to avoid missing a genuine early flare hiding behind the same subjective symptoms the fibromyalgia treatment is meant to address.