The part of M35.4 treatment nobody warns you about is not the diagnosis and not the first prescription. It is the descent. Glucocorticoids work fast — patients often describe range of motion returning within days[1] — and that speed sets an expectation the rest of the treatment cannot meet.
What follows is the reverse: a slow, deliberately unhurried reduction measured in months, with the dose stepping down while the disease is watched for signs of coming back. This article describes how that year is structured in the literature and what happens to the body along the way.
Dosage is determined exclusively by the treating physician individually for each patient, depending on clinical picture, body weight, comorbidities, and other factors. The values cited below come from the scientific literature and do not constitute an indication for self-dosing.
Why the descent is the long part
The reason the taper is slow is that steroids suppress the inflammation without resolving what drives it. While the dose is high, the fascia is quiet. As the dose falls, the question being tested every few weeks is whether the disease has actually burned out or has merely been held down.
Two figures from the published series frame the problem. Roughly 82 % of patients respond to treatment in some measurable way, but only about 56 % respond to corticosteroids alone — the rest need a second drug added. And among those who do respond, about a quarter relapse. The taper is the period in which that quarter declares itself.
Induction at roughly 1 mg/kg held for six to eight weeks, then a stepwise reduction every few weeks down to a maintenance band, and — with stable remission — withdrawal at around a year. Every plan is built individually with a rheumatologist.
The curve has three distinct phases rather than one straight line. The induction window (first 6–8 weeks) holds the starting dose steady while inflammatory markers and skin tightness normalise — nothing is reduced here. The reduction phase steps the dose down by roughly 10–20 % every two to four weeks; this is where most relapses surface. The maintenance band of 5–10 mg/day is the longest and slowest stretch, often occupying several months on its own, because the final milligrams are the ones the adrenal axis has to relearn to produce. Complete withdrawal inside nine months is the exception, not the rule, and a relapse at any step pushes the dose back up and restarts the descent from a higher point.
Lebeaux & Sène 2012 · Mazori 2017
The three phases in plain terms
- Induction — the starting dose, commonly cited in the range of 0.5–1 mg/kg of body weight per day, held for six to eight weeks before the first formal response assessment. Nothing is reduced during this window.
- Reduction — stepwise decrements, typically on the order of 10–20 % every two to four weeks, guided by skin tightness, range of motion, and inflammatory markers rather than by the calendar alone.
- Maintenance — a low band, often described around 5–10 mg/day, held while remission proves itself stable. This phase is frequently the longest of the three.
Why a second drug appears mid-taper
Methotrexate is the drug most often added to a steroid regimen in eosinophilic fasciitis, and the reason is arithmetic rather than pessimism. Its purpose is steroid-sparing — it holds remission at a lower prednisone dose than the patient could otherwise tolerate, which shortens the total steroid exposure even though it lengthens the drug list.
The timing is the part that surprises patients. Methotrexate needs three to six months before a verdict on its effect is reasonable, so it is usually started while the steroid dose is still meaningful, not after the steroid has failed. The two curves overlap deliberately.
From the patient’s side, the hardest stretch is rarely the second drug itself. It is the gap between the fast, almost theatrical response to steroids and the months of waiting to find out whether the slower drug has taken hold.
What the body does while the dose falls
Steroid side effects do not switch off at the moment the dose starts dropping; they fade across the descent, and they fade at different rates. Appetite and sleep tend to recover early. Bone density does not — losses accumulated in the first six months can be permanent, which is why vitamin D and calcium supplementation is generally described as starting with the first prescription rather than at the first symptom.
Proportion of patients on medium- to high-dose prednisone who develop each effect during the first six months. Most ease as the dose is reduced; bone-density loss is the notable exception.
Appetite change and weight gain affect the large majority of patients on medium- to high-dose prednisone — this is pharmacology, not a lapse of discipline, and it is the effect patients most often blame themselves for. Sleep disturbance and irritability sit at 50–70 % and are most pronounced above roughly 20 mg/day. The metabolic effects — hypertension, new glucose intolerance, bone-density loss — affect a quarter to a third of patients and are the reason monitoring runs on a schedule rather than on symptoms. Almost everything on this list eases once the dose drops below 10 mg/day; bone loss is the exception, which is why supplementation is framed as starting on day one rather than being added later.
rheumatology literature · Mazori 2017
What gets monitored, and how often
Monitoring during a taper serves two separate purposes that are easy to conflate: watching the disease for relapse, and watching the drugs for toxicity. They run on different clocks.
Disease monitoring asks 'is it coming back?'. Drug monitoring asks 'is the treatment causing harm?'. Both run through the same taper year, on different schedules.
The two tracks answer different questions and are easy to confuse when both appear on the same lab request form. Disease monitoring is largely clinical — skin tightness, range of motion, and the return of morning stiffness carry more information than any single blood value, with eosinophil count and inflammatory markers used as supporting evidence. Drug monitoring is almost entirely laboratory-driven and runs on a fixed schedule regardless of how well the patient feels, because the effects it looks for (marrow suppression, liver enzyme rise, glucose intolerance, bone loss) are silent until they are advanced. A patient can be doing well on the first track and still need action on the second.
Lebeaux & Sène 2012 · Mazori 2017
On the drug side, the schedule described for methotrexate in the literature is a blood count every four to six weeks, liver enzymes and kidney function alongside it, and folic acid supplementation on a different day of the week from the methotrexate dose itself.
When a flare interrupts the descent
A relapse during a taper is common enough to be planned for rather than treated as failure — roughly a quarter of initial responders experience one. The signals that most often precede it are the same ones that opened the disease:
- Morning stiffness returning in an area that had already cleared
- New swelling or induration appearing in a location that was previously free
- A measurable loss of range of motion in a joint that had regained it
- Fatigue disproportionate to the day’s activity
- A rise in peripheral eosinophil count on a routine follow-up count
A flare during the taper does not mean the treatment has failed. Most respond to returning to a higher dose, and the earlier the change in the disease is reported to the treating clinician, the smaller the step back tends to be.
After zero
Reaching the end of the taper is a smaller event than most patients expect. The adrenal axis, suppressed for months, needs time to resume normal cortisol output, and fatigue in the weeks after the last dose is frequently described. The second drug often continues past the point where the steroid stops.
What the long-term series show is broadly reassuring on function: most patients regain the weight, sleep, and mood they had before treatment, and the contracture risk — the thing that actually determines long-term disability in M35.4 — is driven far more by how early treatment started than by how the taper ended[5].
TIP: We recommend that you consult your doctor. Thank you!
References
- Lebeaux D, Sène D. Eosinophilic fasciitis (Shulman disease). Best Pract Res Clin Rheumatol · 26(4):449-458. 2012. PubMed · 23040360
- Lakhanpal S, Ginsburg WW, Michet CJ, Doyle JA, Moore SB. Eosinophilic fasciitis: clinical spectrum and therapeutic response in 52 cases. Semin Arthritis Rheum · 17(4):221-231. 1988. PubMed · 3232080
- Wright NA, Mazori DR, Patel M, Merola JF, Femia AN, Vleugels RA. Epidemiology and treatment of eosinophilic fasciitis: an analysis of 63 patients from 3 tertiary care centers. JAMA Dermatol · 152(1):97-99. 2016. DOI · 10.1001/jamadermatol.2015.3648
- Mazori DR, Femia AN, Vleugels RA. Eosinophilic fasciitis: an updated review on diagnosis and treatment. Curr Rheumatol Rep · 19(12):74. 2017. DOI · 10.1007/s11926-017-0700-6
- Mango RL, Bugdayli K, Crowson CS, et al.. Baseline characteristics and long-term outcomes of eosinophilic fasciitis in 89 patients seen at a single center over 20 years. Int J Rheum Dis · 23(2):233-239. 2020. DOI · 10.1111/1756-185X.13770