Familial Mediterranean Fever · animated companion
FMF Companion
Familial Mediterranean fever is the most common inherited autoinflammatory disease. Here it is in moving pictures: watch a single protein misfire, follow an attack hour by hour, see why colchicine takes months to show its full effect, and look honestly at what a cure would take.
Every animation can be driven from the keyboard: Tab to its timeline, then arrow keys to scrub and Enter to play or pause.
Schematic. Each spike is an attack. Colchicine starts at month 12.
For students
Every module has an Exam hook box and a citation trail. Finish with flashcards, mnemonics and a quiz.
For patients and families
Look for the In plain words boxes. Bring questions to your own doctor: the app reports what guidelines say, not what you should do.
How sure is each claim?
Research claims carry a label: Established Early evidence Hypothesis. Small numbers link to the Sources list.
Part 1 · The disease
What FMF is
Familial Mediterranean fever is the most common inherited autoinflammatory disease 263: short, self-limited attacks of fever with inflammation of the lining of the belly, chest or joints. Pick a population and press play to see how common the gene variants are, and why far fewer people get sick than the variants alone would predict.
Each figure is 1 person in 100. Carrier rates come from different studies that tested different variant panels by different methods, so they are not one comparable series. The predicted figure is a simple estimate: it assumes random partnering, counts only the variants that study tested, and ignores reduced penetrance, so it is not a disease rate.
Why are carriers so common?
Disease variants usually stay rare because they are a burden. FMF variants reached carrier rates of about 1 in 5 in several populations. One explanation is that carrying one copy once helped people survive an infection. Drag through two thousand years.
Schematic model, not a reconstruction: constant selection at the coefficients estimated by Park et al. 2020 (s ≈ 0.077 per generation for M694V and 0.068 for V726A, 25-year generations), each variant starting as one rare copy at its HMM age estimate. The same paper's Gamma method puts the most recent common ancestor further back, about 3,541 years for M694V and 3,372 for V726A, so the start of each curve is the younger of two published estimates. The starting frequency is illustrative. Real selection was probably uneven.
An attack at a glance
How often each site is involved, in the Turkish national registry of 2,838 patients. 264 Tap a bar to see where it is and what it feels like.
Fever and peritonitis together are the classic attack. For what happens hour by hour, with CRP and SAA curves, see An attack, hour by hour.
Key numbers
Part 1 · The disease
The gene & inheritance
FMF comes from variants in one gene, MEFV, which makes the protein pyrin. Press play to zoom from chromosome 16 down to single amino acids, then tap a variant pin. Below, watch alleles pass from parents to children, and practise reading a genetic report.
Chromosome, band and exon sizes are schematic. Protein domains are drawn at UniProt O15553 coordinates; the hot-spot inset is magnified. Pin colour = current Infevers class.
Animated Punnett square
Each of the four boxes is an equally likely outcome for every pregnancy. "Two variants" means the genotype, not certain illness: penetrance is incomplete.
Pseudodominance. Because carriers are so common, an affected parent often has a carrier partner. Then half the children inherit two variants, and FMF appears in every generation, as if it were dominant. Cazeneuve and colleagues documented this pseudodominant transmission in Armenian families. 33
Read a genetic result
Pick the two alleles a report might list. The tool shows each variant's current Infevers class and how the pair counts in the Eurofever/PRINTO 2019 classification criteria. It is a study aid, not an interpretation of anyone's test. 75,108
Beyond the textbook pattern
Part 1 · The disease
Pyrin, animated
Pyrin is a guard protein inside neutrophils and monocytes. It does not look for microbes directly. It watches a cellular switch called RhoA, and fires when that switch goes quiet. FMF variants make the guard jumpy. Pick a genotype and a stimulus, then press play.
Schematic, not to scale.
Outcome grid
| Small everyday dip in RhoA | Pathogen switches RhoA off |
|---|
Part 1 · The disease
An attack, hour by hour
A typical FMF attack lasts 12 to 72 hours and then stops by itself. Choose where it strikes, then scrub through the attack. The tissue scene and the four curves move together.
Illustrative. No published study gives an hour-by-hour fever or pain curve. Shapes follow published descriptions (onset within hours, resolution in 12–72 h, attack and between-attack SAA/CRP medians). Individual attacks vary.
Attack patterns over the years
Part 2 · Colchicine
Colchicine inside the cell
Colchicine comes from the autumn crocus and has been the backbone of FMF treatment since the 1970s. Follow one dose from the mouth to the inside of a neutrophil, where it takes apart the cell's microtubule scaffolding. Four scenes, one timeline.
Schematic, not to scale.
How sure are we about the mechanism?
Part 2 · Colchicine
Colchicine over months
Colchicine is not a painkiller and does not stop an attack that has already started. Its benefit builds as the drug settles into white cells and as weeks pass without the inflammation that would have happened. Drag through two years, then zoom out to decades.
Illustrative model. No study has published a week-by-week curve of SAA or attack rate after starting colchicine. The shapes follow what is published: leukocyte pharmacokinetics, response judged over 3–6 months, SAA falling after dose increases, and long-term cohort outcomes. It is not a prediction for any person.
Zoom out: decades and amyloidosis
Part 2 · Colchicine
Dosing, safety, interactions
Colchicine works at low doses but has a narrow safety margin, and a few common drugs can push its blood level up several-fold. Pick an interacting drug and press play to watch ten days of daily dosing, then explore the published dose ranges, the organs that are monitored, and what the guidelines say about kidneys, pregnancy and lifelong use.
Dosing for any person is set by their own clinician. This module reports what EULAR 2016, the 2024 EULAR/PReS update and the US label (COLCRYS) publish. It is a study aid, not a dosing guide. 63,199,203
1 · The exposure window
Schematic. The curve heights come from the label's single-dose (0.6 mg) AUC studies in healthy volunteers, scaled to once-daily dosing with a half-life of about 29 hours; inhibition is modelled as a mix of higher absorption and slower clearance. That 29 hours sits inside the label's repeated-dose range of 26.6–31.2 hours, but published half-lives run from about 7 hours to about 58 hours depending on study design, so the curve would be flatter or steeper with other values. 16,63,66 The zone boundaries are illustrative: there is no sharp line between safe and toxic levels. 1.8 mg/day is simply a value inside the label's adult FMF range, chosen for the picture.
| Added drug or food | Label group | Mean colchicine AUC change (single 0.6 mg dose) | Label FMF cap if there is no alternative |
|---|---|---|---|
| Clarithromycin | Strong CYP3A4 inhibitor | +282% | 0.6 mg/day |
| Ketoconazole (itraconazole listed with it) | Strong CYP3A4 inhibitor | +212% (ketoconazole) | 0.6 mg/day |
| Ritonavir (and most protease inhibitors) | Strong CYP3A4 inhibitor | +296% | 0.6 mg/day |
| Cyclosporine | P-gp inhibitor | +259% | 0.6 mg/day |
| Verapamil | Moderate CYP3A4 inhibitor | +103% | 1.2 mg/day |
| Diltiazem | Moderate CYP3A4 inhibitor | +93% | 1.2 mg/day |
| Azithromycin | Not in the cap lists we cite | +57% | None listed in our sources |
| Grapefruit juice | Moderate CYP3A4 inhibitor | No meaningful mean change in the label's own study (Cmax and AUC each moved by under 3%) | 1.2 mg/day; label tells patients to avoid it |
| Statins, fibrates | Added muscle toxicity | Not quantified in our sources | Warning: myopathy and rhabdomyolysis |
Sources: US label interaction studies and dose-adjustment lists; the caps also apply within 14 days of taking the interacting drug. 63,257
The clarithromycin study
10.2% vs 3.6%In a retrospective study, 9 of 88 patients (10.2%) given clarithromycin and colchicine at the same time died, versus 1 of 28 (3.6%) given them one after the other. Baseline kidney impairment, a longer overlap and pancytopenia predicted death. 103
Azithromycin raised colchicine exposure much less and is suggested in reviews as the safer macrolide. 16,257
Statins and fibrates
The concern here is muscle: colchicine and these drug classes can each damage it. The label reports myopathy and rhabdomyolysis when one is added to a stable regimen of the other, and notes that CK monitoring will not necessarily prevent severe myopathy. 63
One review suggests pravastatin or rosuvastatin may be preferred over CYP3A4-metabolized statins; myopathy has still been reported with pravastatin. 16,243
Grapefruit: a label paradox
Both sides of this sit in the same document. The label lists grapefruit juice in its table of moderate CYP3A4 inhibitors, applies the 1.2 mg/day FMF cap to it and tells patients to avoid it. Yet the label's own interaction study (240 mL twice daily for 4 days) found mean Cmax and AUC each moved by under 3%, which is no meaningful change. Neither source we cite resolves the contradiction; the effect may vary with the product and the person. 63,147
2 · Published doses by age
Bars report published ranges, not a dose for anyone; a dose for a person is set by their clinician. Dark lilac: EULAR 2016 starting dose. Light lilac: room for stepwise increases up to the published maximum (red line). The guidelines give the maxima for "children" and "adults" without naming an age cut-off, so age 18 is this app's assumption, not a guideline figure.
What EULAR publishes
- Start colchicine as soon as a clinical diagnosis is made. 199,200
- Starting doses: ≤0.5 mg/day under 5 years, 0.5–1.0 mg/day at 5–10 years, 1.0–1.5 mg/day over 10 years and in adults. 199,200
- Maximum 2 mg/day in children and 3 mg/day in adults; the 2024 update keeps both ceilings. 43,200,203
- Higher starting doses can be used with high disease activity, M694V homozygosity, or complications such as amyloidosis. 199,200
- The maxima are published for "children" and "adults"; no source we cite names the age at which one becomes the other. 199,200,203
Single or divided?
Both EULAR 2016 and the 2024 update allow one daily dose or divided doses, chosen by tolerance and adherence. Dividing can ease gut side effects; once daily can help adherence. 199,200,203
In a 24-week randomized trial (79 children completed it), once-daily dosing was non-inferior to twice-daily for attacks, CRP, ESR and SAA. Diarrhoea was slightly more frequent with once-daily dosing. 213
Steps up and the label
If attacks or subclinical inflammation persist in an adherent patient, the dose is raised stepwise: 0.5 mg/day steps in EULAR 2016, 0.25 or 0.5 mg steps in a German consensus, 0.3 mg/day increments in the US label. CRP or SAA is checked at least every 3 months while escalating. 59,63,199,200,222
US label FMF ranges: 0.3–1.8 mg/day at 4–6 years, 0.9–1.8 at 6–12 years, 1.2–2.4 over 12 years and in adults. A 2007 paediatric consensus put the ceiling at 2.5 mg/day. 59,63,117
A steady-state study across ages 2 to under 16 found broadly similar colchicine exposure in children and adults on standard starting doses, despite the higher mg/kg doses in younger children; exposure ran about 25% higher in the 4 to under 6 year group. 22
3 · Toxicity and monitoring
Click an organ or a button. Schematic body map.
Narrow margin, serious overdose. Colchicine has a narrow therapeutic index, with no clear line between non-toxic, toxic and lethal doses. Overdose can be fatal: poisoning moves over days from gut symptoms and fluid loss to multi-organ failure. The 2024 update asks that any intentional or accidental overdose be assessed carefully, and the label advises keeping tablets away from children. 43,63,67,147
What is checked, and how often. EULAR 2016: review response, toxicity and adherence every 6 months (more often in the first year or when inflammatory markers are raised, yearly once stable), with SAA and/or CRP, blood count, liver enzymes, kidney function, CPK and urinalysis. The 2024 update asks for regular monitoring, more often at diagnosis and when disease is uncontrolled, without a fixed interval. 199,203,222
4 · Kidneys and liver
Creatinine clearance (CrCl) in mL/min. Categories follow the US label's FMF section.
Liver impairment
Clearance may be much lower and the half-life longer in chronic liver disease, with wide variation between patients. The label (FMF) advises close monitoring in mild to moderate impairment and considering a dose reduction in severe disease; there are no pharmacokinetic data for Child-Pugh C. In one report the half-life was 2–3 times longer in severe kidney failure and about 10 times longer with kidney failure plus cirrhosis. 16,63
5 · Pregnancy, breastfeeding, fertility
Pregnancy Established
EULAR 2016: colchicine should not be stopped during conception, pregnancy or breastfeeding, and amniocentesis is not justified because of it. The 2024 update keeps both points. 43,199,200
- 179 colchicine-exposed FMF pregnancies vs 197 untreated FMF and 312 healthy: no difference in early or late miscarriage or malformations. 15
- 238 exposed pregnancies (87% FMF): major anomalies 4.5% vs 3.9% in controls (P = 0.648), no cytogenetic anomalies. 53
- 2018 meta-analysis: no increase in miscarriage or major malformations. 107
Breastfeeding
Fetal plasma levels are reported at about 15% of maternal levels; breast-milk levels are similar to maternal serum. 63
In 4 nursing mothers with FMF, milk levels (1.9–8.6 ng/mL) tracked serum levels; the authors judged infant intake small and nursing safe. The label estimates infant exposure at under 10% of the maternal weight-adjusted dose and urges caution. 9,63
Male fertility
EULAR 2016: men generally need not stop colchicine before conception; only rare, proven colchicine-related azoospermia may call for a temporary dose reduction. 199,200
In some FMF men with azoospermia, testicular biopsy showed amyloidosis, so the disease, not the drug, may be the cause. The label calls drug-related male infertility rare and reversible. 10,63,91
6 · Resistance, adherence, lifelong use
Resistance and intolerance
5–10%One review reports non-response at 5–10%; another puts it at up to 5%, plus 2–5% who cannot tolerate the drug; a third gives 5–15% and notes the top figure may be inflated by missed doses. 13,16,200
EULAR 2016: ≥1 attack a month after ≥6 months at the maximum tolerated dose may indicate resistance. A 2021 consensus: ≥1 attack a month over 3 months, or persistent lab inflammation, with adherence checked. The 2024 update: add-on therapy, IL-1 blockers best supported; colchicine usually continues. 59,199,202,203
Adherence
13%of 38 patients in an Israeli clinic filled every prescription in a year; 8% filled none. 13
- 324 Turkish children (MASIF scale): 72.8% poorly adherent. 60
- 104 Turkish adults (MMAS-8): low 28.8%, moderate 45.2%, high 26.0%. 24
Guidelines rule out missed doses before labelling anyone resistant. 59,200
Lifelong, and the 5-year rule
FMF treatment is lifelong, with adherence as the cornerstone of the 2024 principles. 43,200,203
EULAR 2016 allowed a dose reduction to be considered only after more than 5 years without attacks and with normal acute-phase reactants, after expert consultation and with continued monitoring. A 2025 editorial warns that older age or carrying one variant should not automatically lead to dose reduction. 89,199,200
Part 3 · Complications & diagnosis
AA amyloidosis
The most serious complication of uncontrolled FMF builds silently over years. The liver keeps making serum amyloid A (SAA), a fragment of it misfolds into fibrils, and the fibrils clog the kidney's filters. Press play to watch two decades, then switch to "SAA brought below 10 mg/L" to see what controlling inflammation changes.
Schematic. The time axis follows the reported median of about 17 years from onset of inflammation to diagnosis, which varies greatly between people. Deposits are drawn in Congo-red colour. The controlled mode reflects Lachmann 2007 (deposits regressed in 60% of patients with median SAA below 10 mg/L), not a prediction for anyone.
Organs fill in with the timeline above (schematic shading). Frequencies from EULAR 2016 and a 2024 review. 177,199
Under the microscope: Congo red
Schematic colours. Real slides vary with section thickness, stain quality and fibril orientation.
Who is at risk
Log scale. Each bar is an odds ratio (or risk ratio, for M694V in Kasifoglu 2014) from a different study, population and statistical model. They are not comparable with each other and cannot be multiplied into a personal risk. The SAA1 and male-sex figures come from one mainly Armenian cohort; a separate study of 124 M694V homozygotes put male sex lower, at OR 2.37, alongside arthritic attacks at OR 3.27. 34,78,121,262,273
Factors without a single number
These are listed as risk factors in reviews and cohorts without an effect size attached. 89,121,262,273
Delayed diagnosis raised amyloidosis frequency from 2.6% to 10% in one cohort, and M694V homozygotes develop amyloidosis earlier (about age 27) than heterozygotes (about age 41). 89
Why country matters most
In 2,482 genetically confirmed patients from 35 centres in 14 countries, country of recruitment was the strongest risk factor, ahead of M694V homozygosity. Country risk tracked infant mortality, which points to an environmental factor. Amyloidosis was reported in about 24% of Armenians in Armenia but was essentially absent among Armenian Americans in California. 231,262
How common, then and now
What else to know
Why colchicine prevents it even if attacks persist
Amyloid is built from SAA, so risk follows how much SAA is made over years, not just how many attacks occur. SAA can stay raised between attacks. In 26 children with no attacks in the previous year but high SAA, raising the colchicine dose lowered median SAA from 244 to 35.5 mg/L by the 2-month recheck. In Zemer's 1,070-patient cohort, cumulative proteinuria was 1.7% at 11 years in adherent patients versus 48.9% at 9 years in those who said they were not taking it regularly. 58,141,286
Treating established amyloidosis
EULAR 2016 advises intensifying to the maximum tolerated colchicine dose plus biologics as needed; the 2024 update frames add-on therapy the same way, with IL-1 blockers the best supported. In 68 patients with FMF amyloidosis followed at least 5 years, proteinuria regressed in 15, and worsening was linked to starting creatinine ≥1.5 mg/dL and to doses ≤1.5 mg/day. A 2025 systematic review found the evidence for treating established disease observational but judged it reassuring; retrospective data suggest tocilizumab may help selected patients. 158,172,199,203,229
Amyloid storm
A rapid deterioration, often set off by infection: creatinine and proteinuria at least double and CRP rises more than tenfold within 2 weeks. In a 20-patient series, 16 reached end-stage kidney disease or death within a year (3 of 20 controls); infection triggered 17 of the 20 storms. 89,139
Eprodisate Mixed evidence
A drug that interferes with fibril formation. In the 183-patient EFAST trial, worsening kidney function or death occurred in 27% vs 40% on placebo (P = 0.06; HR 0.58), with slower clearance decline but no significant effect on dialysis or death. The 261-patient confirmatory phase 3 trial missed its primary endpoint in 2016. 48,126
Kidney transplantation
- Egypt, living donors (23 amyloid recipients, 16 FMF): graft survival like controls; 10-year patient survival 68% vs 87%; recurrence in 1. 237
- Turkey, 17 FMF recipients: about 89–90% graft and patient survival at 5 years; recurrence in 2. 124
- Israel: survival 73% vs 84% at 5 years and 39% vs 63% at 15 years. 88
Cardiovascular disease drives post-transplant deaths; recurrence in the graft has become rare. 177
Phenotype II
Amyloidosis before, or instead of, typical attacks. It is uncommon: among 461 relatives of FMF-amyloidosis patients, only 2 had significant proteinuria without symptoms, and their rectal biopsies showed no amyloid. Estimates differ: at most 0.3% of the 2,838-patient Turkish registry, and about 0.6% in a later review, whose authors also doubt it is a separate entity. 89,171,264
Monitoring
- EULAR 2016: urinalysis at every visit to exclude proteinuria. 199
- In amyloidosis, frequent SAA with a target below 10 mg/L, plus proteinuria and GFR; CRP is an imperfect substitute. 199
- The 2024 update asks for complete, sustained control of biochemical inflammation. 203,222
- Amyloid kidneys are sensitive to low blood flow, high blood pressure, nephrotoxic drugs and surgery. 199
Why it matters
Leading cause of deathIn a Turkish cohort of 587 patients, amyloidosis was the leading cause of death and the only significant predictor of mortality. In an Israeli cohort of 1.25 million adolescents (1,225 with FMF), mortality was higher with FMF, and kidney amyloidosis caused 35% of deaths in men and 60% in women with FMF. 2,266
Part 3 · Complications & diagnosis
Diagnosis & criteria
There is no single test that proves FMF. The diagnosis rests on the attack pattern, blood markers in and between attacks, the gene test and the response to colchicine. Tick features in the calculators below, then press play on the lab timeline and the diagnostic pathway.
Eurofever/PRINTO 2019 calculator
These are classification criteria, built so that research studies enrol comparable patients. They are not a diagnostic test for one person, and a doctor can diagnose FMF in someone who does not meet them. 75
Sensitivity and specificity from Gattorno et al. 2019 (derivation set, then validation set). In Turkish children the criteria showed sensitivity 96% and specificity 73.1%; a Turkish referral cohort classified 98.2% of homozygotes and 80.2% of heterozygotes, and no controls. 75,228,253
The older criteria sets
Blood markers across an attack
Curve shapes are schematic. The attack and between-attack values for SAA and CRP are the medians from Lachmann et al. 2006; the other rows show direction of change only.
The diagnostic pathway
A simplified teaching sequence drawn from published reviews and guidelines. Real work-ups overlap and loop back.
Part 3 · Complications & diagnosis
The differential
Recurrent fever with pain has many causes. The rhythm is often the first clue: how long each episode lasts, how regularly it comes back and whether it ends by itself. Press play to sweep through 60 days, then click a disease to see how it is told apart.
Across the hereditary recurrent fevers, how long an episode lasts is the single most discriminating variable: FCAS under a day, FMF 1–3 days, PFAPA 3–6 days, MKD 3–7 days, TRAPS 7 days or more. The FMF, PFAPA, MKD and TRAPS lengths are cited in the table below; the FCAS figure is standard teaching, not tied to a fetched source. 62,75,144,258
Schematic 60-day window. Episode lengths follow the sources in the table; the spacing is illustrative. The right-hand tag says whether a row's rhythm is cited ("cited") or comes from standard teaching not tied to a fetched source ("teaching").
Fever + abdominal pain: decision points
Pick a clue from the story. The matching rows light up in the animation above and the box below names the test that separates them.
The full comparison
Click any row to load it into the note panel above.
| Disease | Gene / inheritance | Typical attack length | Key distinguishing features | Distinguishing test |
|---|
Part 4 · Treatment & the future
The treatment ladder
Published guidance climbs one rung at a time: colchicine from the moment of clinical diagnosis, then the maximum tolerated dose, then IL-1 blockade added on top if attacks and inflammation persist. Press play to light the rungs in order, or click a rung to read what the guidelines say about it. This describes guidelines and trials, not advice for any person.
Click a rung to read it. Schematic, not to scale.
What the trials measured
Each group is a separate trial with its own design and endpoint, so bars are not comparable across groups. The attack-rate bars are the figures each trial published; the Cochrane re-analysis printed under them is more cautious, and found no clear difference for anakinra or rilonacept in the number of patients having attacks. 283
Sources, left to right: CLUSTER phase 3 randomised trial of canakinumab, 63 patients in the FMF cohort 44,283 · anakinra randomised trial, 25 colchicine-resistant patients 20 · rilonacept randomised alternating-treatment trial, 14 patients 93 · tocilizumab phase II and phase III trials 98,132.
Part 4 · Treatment & the future
Living with FMF
FMF is managed across years, not visits. Press play to walk a single year around the ring: routine checks arrive on schedule, and the toggles drop real-life situations into the same year, each with what the published literature says about it. Everything here is what studies and guidelines report, not advice.
Schematic year. Intervals follow published guidance, not any one clinic.
What the trigger evidence actually shows
Questionnaire percentages are what patients report, which is not the same as what causes attacks. The case-crossover design compares each patient with themselves, so it is the stronger evidence. 118,209,279
Diet is the weakest link in this literature: a review found only 7 eligible studies, with conflicting results on fatty and salty food and a possible role for wheat, and no robust trial showing that any diet prevents attacks. 167
School, work and mood
Group findings from separate studies, each with its own population. 151,236,251
Prognosis, monitoring and people
The monitoring checklist in the 2016 guideline
- Response, toxicity and adherence reviewed every 6 months, more often in the first year or when acute-phase markers are raised, and yearly once stable.
- Tests listed: SAA and/or CRP, blood count, liver enzymes, kidney function, CPK, and urinalysis for proteinuria.
- Urinalysis at every clinic visit, to catch proteinuria early.
- Colchicine is reduced if liver enzymes go above twice the upper limit of normal.
- High SAA values are shared with the patient, deliberately, to support adherence.
- The 2024 update sets no fixed interval, and asks for monitoring more often at diagnosis and when disease is uncontrolled, against a minimum outcome set. 199,203,222
Questions people bring to a clinic visit
Plain-words versions of what the guidelines themselves say should be discussed. They are prompts for a conversation with a clinician, not a script and not advice.
- What is my SAA or CRP between attacks, and is it in the normal range?
- Was there any protein in my urine today?
- Am I on the dose the guidance would call maximum tolerated for me, and is there room to go up?
- How many attacks have I had in the last 3 months, counted rather than remembered?
- If attacks continue at this dose, what would the next step be?
- What should happen to my treatment if I plan a pregnancy, or need an operation, or need vaccinations?
- How often should I be seen, and which tests each time?
Part 4 · Treatment & the future
Nursing care
What the published guidance describes for the four things a nurse does with someone in an attack: look, give the medicines safely, teach, and write it down. Every figure here is quoted from the literature, not from local practice.
1 · Assessment during an attack
Illustrative. The observation curves follow published descriptions of the attack time course; no study publishes an hour-by-hour chart. The findings and thresholds are cited.
Measuring how active the disease is
2 · Giving the medicines
Injectable IL-1 blockers
Sites shown are the ones the labels name for subcutaneous injection. Rotation is standard practice for repeated subcutaneous injections.
3 · What to teach, and what gets misunderstood
4 · What goes in the notes
Entry built from what you ticked
A worked example of the items the guidelines list. Your service decides the format.
The monitoring year
Part 4 · Treatment & the future
Toward a cure
No approved therapy cures FMF today. Colchicine and IL-1 blockade control it, for life. This module takes the question seriously anyway: where in the pathway could you intervene, what evidence exists for each route, and what stands in the way. Every claim carries an evidence label. Press play to send the signal down the pathway, then pick a route to drop a barrier in front of it.
Schematic pathway, not to scale.
Replacing the marrow: what it would take
Hypothetical sequence drawn from transplant and gene-editing biology in other diseases. There is no FMF-specific human dataset behind these curves.
The routes, side by side
| Route | What it would fix | Best evidence | Level | Main obstacle | Status |
|---|
What a cure would have to achieve. Two things at once, not one. It would have to stop attacks, and it would have to switch off the quiet inflammation between them, because AA amyloidosis is driven by chronically raised SAA rather than by the attacks themselves. The 2024 EULAR/PReS update makes control of subclinical inflammation an explicit treatment target. A cure would also have to reach the right cells: MEFV messenger RNA in blood is restricted to myeloid cells, and the gene is switched on as CD34+ progenitors differentiate into granulocytes, which puts the target inside the bone-marrow stem and progenitor compartment. Pyrin protein has also been reported in dendritic cells and in synovial fibroblasts, so a purely blood-cell fix might leave a reservoir behind. 35,54,203
Thirteen open questions
These are the gaps the literature itself leaves open, not rhetorical questions. Each one is a place where a study could change the picture.
What a next trial might look like
Read this part carefully. The three designs below are illustrations written for this page. They are not registered trials, not proposals from any investigator, and not advice. No trial of gene editing, RNA therapeutics, HSCT or a pyrin-targeted small molecule in FMF is registered as of the October 2026 sweep of ClinicalTrials.gov. 42
The statin caution, and its limits. Statins deplete geranylgeranyl pyrophosphate, which stops RhoA from sitting in the membrane, and in cell experiments that activates the pyrin inflammasome; statin-induced IL-1β release was reversed by adding geranylgeranyl pyrophosphate back or by activating PKN. This is in-vitro mechanism. Its clinical relevance is unknown: no clinical study has looked at whether statins affect FMF, and patients with cardiovascular indications routinely receive them. Nobody should change or stop a medicine on the strength of this, and any question about a statin belongs with the prescriber. It appears here as one of the open questions, not as a recommendation. 207
Part 5 · Context & review
History & people
A disease first sketched in 1908, named in 1958, made treatable in 1972 and explained at the molecular level only after 2014. Press play to run the playhead from 1900 to today, or jump between eras. The scene above the timeline redraws itself as each era arrives.
The scene is a drawing of what changed in each era; every caption in it repeats a milestone below, with that milestone's sources in the note panel. The axis is stretched after 1940 and again after 1990 so crowded recent decades stay readable. The outlook band is qualitative.
Names it has carried
Before 1958 the same illness went by many names, often after the doctors who described it or the people who had it. Click a dated name to jump the timeline. 1
Every milestone
The dots on the timeline are also listed here, so they can be reached with the keyboard and read in order.
What it was called
Part 5 · Context & review
Flashcards & mnemonics
Tap the card (or press Space) to flip it. Arrow keys move through the deck. Filter by part, or shuffle.
Mnemonics
Part 5 · Context & review
Quiz
Pick an answer to see the explanation. The mixed quiz draws 20 questions at random; or choose a part.
Part 5 · Context & review
Glossary
Every specialist term used in the app. Type to filter; the arrow on each entry jumps to the module where it matters most.
Part 5 · Context & review
Sources
Every reference returned by the six research passes behind this app: journal articles, guidelines, drug labels, trial registrations and a few database or search records. The small numbers throughout the app point here.
Research for each area was done separately, so a few works appear under a merged entry. Entries marked check could not be fully verified against the original, so treat them with extra care.
In October 2026 every flagged entry was re-checked against Europe PMC, Crossref, DOAJ and the ClinicalTrials.gov registry. Six turned out to name the wrong authors, the wrong journal or the wrong paper, and two gave a paraphrased title. Those entries now carry a short note saying what was corrected, and several paywalled links were replaced with free copies.