RightRep EN · PT Open the app
BLOG

Can You Train With Rheumatoid Arthritis? The Inflammatory-Joint Audit: Where EULAR Writes Exercise Into Standard Care (4 Principles + 10 Recommendations), the Two-Year Strength Trial Gained 19–59% Strength With Bone Density Rising While Controls' Fell, and RAPIT's 309-Patient High-Intensity File Slowed Hip-Bone Loss From −1.6% to −0.5% a Year While X-Ray Damage Ran 3.5 vs 5.7 Larsen Points (p=0.045) — Loading Did Not Damage the Joints; Where the Disease's Signature Lesion Is Muscle (−13% Body Cell Mass on a +12% Metabolic Rate, Sarcopenia in 24–30%) and 24 Weeks of Resistance Training Restored Lean Mass, Function and Muscle IGF; Where the Deadliest Door Is Cardiac (CVD Mortality 1.5×) and Six Months of Individualised Training Cut 10-Year Risk and Disease Activity Together; and the One Fence No Other File in This Library Carries: the Neck

By RightRep AI Assistant · · 21 min read

Rheumatoid arthritis is the file where the gym's two default scripts — shut it down, or push through — fail in an unusual way: the disease punishes both, and the treatment file has already solved the question they fight over. The shut-it-down script prescribes stillness to a disease whose signature lesion is muscle loss on a sped-up metabolism — body cell mass 13% below matched controls while resting energy expenditure runs 12% higher — which makes rest not caution but the complication [10]. The push-through script ignores that untreated or flaring synovitis is a medical file, not a programming variable, and that RA carries a door most lifters have never heard of: the neck. So the fences come first, and this file's fences are sharper than most. First: morning stiffness that lasts more than 30–60 minutes, small joints swollen on both sides of the body, symptoms that ease as the day warms up — that pattern is inflammatory arthritis until a rheumatologist says otherwise, and it is a this-month door, not a maybe door, because the same trials that follow show the training file assumes the disease is being treated to target [3][6]. Second: one joint hot, red, swollen and severely painful with fever — the infection door, emergency department today. Third, the one no other file in this library carries: long-standing RA and the neck — atlantoaxial subluxation shows on 16.4–95.7% of radiographs in symptomatic series, severe involvement runs 2.1% in a 717-patient real-life file, and about half of screened asymptomatic high-risk patients carried subclinical involvement on MRI — so loaded neck work, neck manipulation and any new neck neurology (pain radiating down the arms, electric shocks on flexion, new numbness, gait or bladder change) belong to a clinician's clearance, urgently, before any bar touches the cervical spine [15][16]. Fourth: the medication ladder — DMARDs, biologics, steroids — belongs to the prescriber; this file covers the training-load question, and its pt-BR sibling question (the drug file) already has its own audit in the prednisone guide [3]. And fifth: a flare re-doses the program, it does not cancel it — that is the guideline's own architecture [3]. With the fences standing, the evidence underneath is the calmest inversion in the whole inflammatory family: the question "can I train?" was answered by rheumatology itself, with two-year randomized trials, X-rays, bone scans and a recommendation set that writes physical activity into standard care "throughout the course of these diseases" [3][4][6][7][8].

The short answer: yes — the training is the co-prescription, not the alternative

Strip the file to its spine. The guideline floor is unambiguous: EULAR's task force from 16 countries built four overarching principles and ten recommendations for physical activity in inflammatory arthritis, and its summary sentence is that physical activity "is advocated as integral part of standard care throughout the course of these diseases" [3]. The safety question was X-rayed, not assumed: the Cochrane review of dynamic exercise programs in RA pooled eight randomized trials and found moderate evidence for improved aerobic capacity (pooled effect size 0.99, 95% CI 0.29–1.68) and muscle strength (0.47, 0.01–0.93) — with the safety line stated flatly: "no deleterious effects were found in any of the included studies" [4]. High-intensity loading was tried for two years and acquitted on film: the RAPIT trial's 309 patients who weight-trained at high intensity for two years developed less radiographic damage than the usual-care group — Larsen score up 3.5 points versus 5.7, p=0.045 — with the effect most pronounced in the feet [8]. Strength training changes the disease's signature lesion: 24 weeks of twice-weekly progressive resistance training raised lean mass, cut trunk fat by 2.5 kg, improved training-specific strength 119%, and restored the muscle IGF system the disease had suppressed [9]. The deadliest door is cardiac, and training is its countermeasure: cardiovascular mortality runs 1.5 times the general population's [13], and six months of individualised high-intensity aerobic and resistance work improved VO₂max, blood pressure, lipids, 10-year cardiovascular risk — and disease activity — together [14]. And the fence with no sibling in this library: the neck gets cleared once, clinically, and then stays out of direct loading [15][16]. That is the audit in one paragraph: the reader standing in this file needs a rheumatologist, a barbell, and a very precise rule about the neck.

What the condition is: a systemic disease wearing a joint costume

The mechanics in a paragraph: rheumatoid arthritis is immune-driven inflammation aimed at the synovium — the lining of joints, starting characteristically and symmetrically in the small joints of the hands and feet — but the inflammation is systemic, and the damage bill runs to four organs: the joints (erosions visible on X-ray within years of untreated disease), the muscle (cachexia, below), the bone (glucocorticoid and inflammation-driven osteoporosis — the osteoporosis audit covers that mechanism), and the heart and vessels (the 1.5× mortality file) [13]. The scale: the Global Burden of Disease file puts age-standardised prevalence at 246.6 per 100,000 in 2017 — roughly one in 400 adults, up 7.4% since 1990 — higher in women throughout, with prevalence peaking in the 70s but incidence arriving in exactly the decades that fill gyms [1]. The Brazil anchor: in the COPCORD census of Montes Claros, Minas Gerais — 3,038 adults surveyed door-to-door — rheumatic disease ran at 7.2% of the city and RA at 0.46%, with women predominant and a mean age of 53.4 at diagnosis [2]. That is the pt-BR lane's audience: one in every two hundred Brazilian adults, disproportionately women, told for decades to "protect the joints" — and holding, in the sarcopenia file below, the exact opposite prescription [2][10][11].

The guideline floor: what EULAR actually recommends

The folklore this post audits — rest the inflamed joint, avoid loading, certainly avoid intensity — is not what any guideline says. EULAR's 2018 recommendation set for physical activity in inflammatory arthritis and osteoarthritis was built by a task force from 16 countries through thirteen research questions and a systematic literature review, and it landed on four overarching principles and ten recommendations, with task-force agreement scoring 8.8–9.8 out of 10 [3]. The load-bearing sentence: physical activity "is advocated as integral part of standard care throughout the course of these diseases" [3]. Two structural facts matter for a lifter. First: exercise is not the alternative to methotrexate or biologics — it is the co-prescription, written in the same document, assumed to continue underneath the drug ladder [3]. Second: "throughout the course" is doing real work in that sentence — it includes flares (at reduced dose), remission, and the decades after; the recommendation set is a permission architecture with dose rules, not a wellness footnote. This is the same architecture this library found in the ankylosing spondylitis file's ASAS-EULAR guideline and the arthritis umbrella audit — the inflammatory family's guidelines all point the same direction, and RA's is the oldest and calmest of them [3][4].

The permission RCTs: two years, X-rayed

The trial a skeptical lifter should read first is Häkkinen's. Seventy patients with early RA — disease under two years, all on medication aimed at remission — randomized to two years of progressive dynamic strength training (every major muscle group, 50–70% of one-rep max, twice weekly) or range-of-motion exercise without resistance [6]. The result: maximum strength rose 19–59% across every measured group (knee extensors, trunk flexors and extensors, grip), with disease activity, function (HAQ) and walking speed all improving more in the training group — and bone mineral density at the femoral neck and spine rising (+0.51% and +1.17%) while the control group's fell (−0.70% and −0.91%) [6]. Then the question every RA patient actually asks: but what about the joints themselves? That question has a direct, randomized, radiographic answer. The RAPIT trial put 309 RA patients on two years of high-intensity weight-bearing exercise or usual physical therapy, and its radiographic substudy scored the hands and feet of the 281 completers: after two years, the Larsen damage score rose 3.5 points in the exercise group versus 5.7 in usual care (p=0.045) — progression was slower in the loaded joints, most visibly in the feet, and the rate of damage tracked disease activity and glucocorticoid use, not exercise [8]. The same trial's bone file closed the osteoporosis loop: hip bone loss slowed progressively in the training arm (−1.6% in year one, −0.5% in year two), and the benefit tracked with strength gains (OR 1.75) and aerobic fitness (OR 1.79) — not with attendance [7]. Adaptation, not presence, pays. And the pooled file agrees: the Cochrane review's eight trials found no deleterious effects anywhere [4], and the 14-trial, 1,040-patient aerobic meta-analysis found quality of life (SMD 0.39, P<0.0001), function (HAQ 0.24, P=0.0009) and pain (0.31, P=0.02) all improved with DAS28 disease activity and joint counts identical to controls [5]. Read together, these four files retire the joint-protection folklore the way the spondylolysis file retired rest: not with rhetoric, with films [4][5][6][7][8].

The cachexia door: the disease's signature lesion is muscle

Here is the part of the file most gym culture has never heard, and it inverts the whole "protect yourself" script. RA's best-described lesion outside the joint is rheumatoid cachexia: in the study that named the mechanism, 23 RA patients against 23 matched controls carried 13% less body cell mass (P<0.00001) while burning 12% more energy at rest (P<0.008) — a body simultaneously losing muscle and running hot, driven by TNF-α and IL-1β, with food intake adequate the whole time [10]. The disease dismantles muscle from the inside; rest prescribes exactly the environment it dismantles it in. The modern sequel file quantifies the prevalence: sarcopenia runs in 24–30% of RA patients depending on criteria, tracks higher disease activity (+0.39 DAS28 per unit in pooled analysis), correlates with glucocorticoid exposure, and arrives with its own bill — lower bone density, more falls, more fractures [11]. And the countermeasure is randomized: Lemmey's group took 28 patients with established, medically controlled RA and randomized them to 24 weeks of twice-weekly high-intensity progressive resistance training or range-of-motion home exercise — lean mass and appendicular lean mass rose (P<0.01), trunk fat fell 2.5 kg, training-specific strength rose 119%, chair stands 30%, knee-extensor strength 25%, arm curls 23%, walk time improved 17% — and muscle IGF-1 and IGFBP-3, suppressed by the disease, were restored by the training [9]. The authors' own conclusion is the sentence this section was built for: "PRT should feature in disease management" [9]. For a lifter that translates cleanly: in RA, progressive resistance is not a risk factor to manage — it is the missing half of the prescription.

The heart door: the 1.5× file and the training that answers it

The mortality file first, because it reframes every other section: across 24 observational studies totalling 111,758 RA patients and 22,927 cardiovascular events, cardiovascular death ran at 1.50 times the general population's rate (95% CI 1.39–1.61) — ischaemic heart disease 1.59, cerebrovascular disease 1.52 [13]. The systemic inflammation bills the heart, the blood-pressure-lowering and zone-2 files explain the mechanism, and RA's own inactivity census explains the opportunity: 110 RA patients against 440 matched controls spent 2,392 versus 2,494 kcal/day (P=0.003) — and the missing kilocalories sat specifically in moderate-intensity activity, the exact zone the heart file needs, with low disease activity and lower fatigue independently predicting a bigger daily budget [12]. The interventional file closes the loop: 40 matched RA patients randomized to six months of individualised high-intensity aerobic and resistance training, three times a week, versus lifestyle advice — and VO₂max (P=0.001), systolic and diastolic blood pressure (P<0.001 / P=0.003), triglycerides (P=0.030), HDL (P=0.042), BMI (P=0.001), body fat (P=0.026), 10-year cardiovascular event probability (P=0.012), CRP (P=0.042), DAS28 (P=0.008) and HAQ (P=0.003) all improved, with the VO₂max gain the strongest predictor of every other improvement [14]. One program, both engines, and the disease-activity needle moved with the fitness needle — the direction the push-through/shut-it-down debate never considers, because neither script was ever individualized or dosed [14].

The neck fence: the rule no other file in this library carries

Every chronic-condition file in this library has its own distinctive fence — the concussion file has the CT rule, the spondylolysis file has the cord signs. RA's is the cervical spine, and it deserves its own section because almost nobody outside rheumatology knows it exists. The systematic review of 1,878 RA patients imaged for cervical involvement found atlantoaxial subluxation on 16.4–95.7% of plain radiographs in symptomatic series, subaxial subluxation in 10–43.6%, and spinal cord compression in 0–13% [15]. The severity file is the one to carry: among 717 real-life RA patients, severe atlantoaxial involvement — the kind that matters — ran at 2.1% (15 of 717), an incidence of 5.2 per 1,000 person-years, concentrated in men, ACPA-positive patients, long disease duration and extra-articular disease; and when the same investigators screened 30 asymptomatic patients with that risk profile, about half showed subclinical atlantoaxial involvement on MRI [16]. The gym translation writes itself: in long-standing RA, the neck is cleared clinically before it is loaded directly, direct neck training and any cervical manipulation are off the program without that clearance, and any new neck symptom with radiating pain, shocks down the spine on flexion, new numbness or weakness, or gait and bladder change is an urgent clinical door — before the next session, not after [15][16]. The rest of the spine and body trains normally; this fence is specific, literal, and load-bearing [15][16].

The program in gym terms: what trains, what re-doses, what stops

Trains always (disease controlled, no flare): everything. Two years of progressive barbell-range loading at 50–70% 1RM and above was the intervention arm of the best trials in this file — squats, presses, pulls, carries — with intensity and progression held like any trainee's, plus the aerobic engine the heart file demands [6][7][14]. The warm-up earns its keep doubly here: morning stiffness means the first minutes of a session are the stiffest minutes of the day [3].

Re-doses during a flare: volume and intensity down, pain-free arcs only, keep moving — the guideline's "throughout the course" architecture in gym terms [3]. The joint that is hot and swollen rests from direct loading; everything it doesn't hurt to train, trains. A flare is also a rheumatology conversation about medication, not a programming puzzle to solve alone [3].

Doses the hands specifically: grip is often the first thing RA taxes, and the morning-stiffness minutes are the calibration window — a grip pattern that fights the stiff morning hand gets re-angled (straps, neutral handles, thicker or thinner bars) rather than abandoned, the same way the tennis elbow and wrist files re-route load around an irritated tissue. Machine stations (machines vs free weights) are legitimate mainline tools here, not a downgrade — grip-independent loading is a feature during hand flares [6][9].

Graduates: after a flare settles, intensity climbs back over days-to-weeks on the same 24-hour symptom budget this library uses everywhere — the morning after each step decides the next [3][4].

Stops and refers: the top fences — new-onset inflammatory pattern (this month's rheumatology door), a hot single joint with fever (today's ED door), any neck neurology or the intention to load the neck directly in long-standing disease without clearance (urgent clinical door), and a flare that won't settle (medication conversation, not persistence) [3][15][16].

The Brazil anchors: 0.46%, and the audience this file's twin exists for

The COPCORD census of Montes Claros gives the pt-BR lane its scale: 3,038 adults door-to-door, RA at 0.46% of the population over 16 — behind only osteoarthritis (4.14%) and fibromyalgia (2.5%) among rheumatic diseases — with women predominant and 35.2% of the rheumatic-disease group unemployed, a number that makes the function file (grip, chair stands, walking speed — the exact measures Lemmey's resistance trial improved 17–30%) an economic variable, not a vanity one [2][9]. The pt-BR twin of this file exists because that population's default advice has historically been the shut-it-down script — and because the file that replaces it was randomized, X-rayed, and written into a guideline years ago [3][4][8].

So what do you actually do? The practical file

Six rules, every one sourced from the files above:

  1. Sort the doors first. New symmetric small-joint swelling with long morning stiffness = rheumatology, this month — the training file assumes the disease is treated to target [3][6]. One hot joint with fever = emergency today. Any neck neurology, or long-standing disease and a plan to load the neck = clinical clearance before the bar [15][16].
  2. Train like the trials did. Progressive whole-body strength at real intensity, twice a week or more, plus individualised aerobic work — that combination raised strength 19–59%, bone density, VO₂max and function, and lowered disease markers, across two-year randomized files [6][7][14].
  3. Retire the joint-protection folklore with the films. The two-year X-ray file is the answer to "won't loading damage me?": 3.5 vs 5.7 Larsen points, p=0.045 — the loaded joints did better [8]. The honest caveat lives alongside: damage tracked disease activity and steroids, so the medication file matters more than the loading file ever will [3][8].
  4. Treat muscle as the disease's second front. Cachexia runs on a −13%-cell-mass, +12%-metabolism engine and sarcopenia sits at 24–30% — and resistance training is the one intervention shown to reverse both directions at once (+119% strength, lean mass restored, muscle IGF back) [9][10][11].
  5. Run the heart program as prevention, not cardio vanity. 1.5× cardiovascular mortality is the file's deadliest number, and the moderate-intensity zone is exactly where the disease deletes your daily budget (2,392 vs 2,494 kcal/day) — individualised aerobic plus strength work moved 10-year risk, CRP and DAS28 in the same direction in six months [12][13][14].
  6. Keep the flare ledger. A dated log of morning-stiffness minutes, which joints are hot, what trained and what the morning after looked like — that's the dose instrument for every rule above and the document both your rheumatologist and your programming decisions run on; the tracking habit is load-bearing medicine here [3][4].

How RightRep fits into this

The RA file is, at its core, a dose-and-symptom ledger question — and it is unanswerable from memory:

  • The flare dose made visible. Sessions logged with load, volume and a next-morning note turn "re-dose, don't cancel" from a slogan into data — which session size your flares tolerate is a measurable number, not a guess [3][4].
  • The two-engine program. The trials that moved every marker ran strength and aerobic work together — a logged program is how both engines actually get their weekly dose [6][14].
  • The morning-stiffness window. Stiffness minutes against training time is the RA-specific trend line — the graph that tells you whether the program is dosing with the disease or against it [3][12].
  • Form under fatigue. Camera-scored reps keep intensity honest on the days the disease makes everything feel heavier — calibration, not grinding, is what the two-year trials actually prescribed [6][9].

The bottom line

  • The guideline floor: EULAR's 4 overarching principles + 10 recommendations — physical activity "integral part of standard care throughout the course of these diseases" [3].
  • The safety spine: Cochrane, 8 RCTs of dynamic exercise — moderate evidence for aerobic capacity (ES 0.99) and strength (0.47), "no deleterious effects were found in any of the included studies" [4]. The aerobic meta (14 RCTs, 1,040 patients): QoL +0.39, HAQ +0.24, pain +0.31 — DAS28 and joint counts unchanged [5].
  • The permission RCTs: Häkkinen's two-year early-RA strength trial — +19–59% strength, disease activity/function/walking better than controls, femoral BMD +0.51% and spine +1.17% while controls fell [6]. RAPIT's bone file: hip loss slowed −1.6%→−0.5%/year, tracking strength (OR 1.75) and fitness (OR 1.79), not attendance [7].
  • The X-ray acquittal: 281 completers, two years of high-intensity weight-bearing — Larsen damage +3.5 vs +5.7 (p=0.045); loading did not damage the joints, and the feet were protected [8].
  • The cachexia file: body cell mass −13% on resting expenditure +12%, TNF-driven, on adequate food [10]; sarcopenia 24–30%, tracking disease activity and steroids, billing falls and fractures [11]; reversed direction by 24 weeks of resistance training — +119% strength, lean mass up, trunk fat −2.5 kg, muscle IGF restored [9].
  • The heart door: CVD mortality 1.50× (1.39–1.61) across 111,758 patients [13]; the daily-activity deficit sits in the moderate zone (2,392 vs 2,494 kcal/day) [12]; six months of individualised aerobic+resistance 3×/week improved VO₂max, 10-year CVD probability (P=0.012), CRP, DAS28 (P=0.008) and HAQ together [14].
  • The neck fence: atlantoaxial subluxation 16.4–95.7% of symptomatic radiographs [15]; severe involvement 2.1% of 717 real-life patients, ~50% of high-risk asymptomatic patients subclinical on MRI — cleared before loaded, urgent if neurology appears [16].
  • The Brazil anchor: RA 0.46% of 3,038 adults in the Montes Claros COPCORD census, women predominant — the audience the pt-BR twin serves [2].

Clear the fences, treat the disease to target, load the body like the trials did — and let the camera count the calm reps at rightrep.app.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. Safiri S, Kolahi AA, Hoy D, et al. Global, regional and national burden of rheumatoid arthritis 1990–2017: a systematic analysis of the Global Burden of Disease study 2017. Annals of the Rheumatic Diseases, 2019;78(11):1463–1471. pubmed.ncbi.nlm.nih.gov
  2. Senna ER, De Barros AL, Silva EO, et al. Prevalence of rheumatic diseases in Brazil: a study using the COPCORD approach. The Journal of Rheumatology, 2004;31(3):594–597. pubmed.ncbi.nlm.nih.gov
  3. Rausch Osthoff AK, Niedermann K, Braun J, et al. 2018 EULAR recommendations for physical activity in people with inflammatory arthritis and osteoarthritis. Annals of the Rheumatic Diseases, 2018;77(9):1251–1260. pubmed.ncbi.nlm.nih.gov
  4. Hurkmans E, van der Giesen FJ, Vliet Vlieland TP, Schoones J, Van den Ende EC. Dynamic exercise programs (aerobic capacity and/or muscle strength training) in patients with rheumatoid arthritis. Cochrane Database of Systematic Reviews, 2009;(4):CD006853. pubmed.ncbi.nlm.nih.gov
  5. Baillet A, Zeboulon N, Gossec L, et al. Efficacy of cardiorespiratory aerobic exercise in rheumatoid arthritis: meta-analysis of randomized controlled trials. Arthritis Care & Research, 2010;62(7):984–992. pubmed.ncbi.nlm.nih.gov
  6. Häkkinen A, Sokka T, Kotaniemi A, Hannonen P. A randomized two-year study of the effects of dynamic strength training on muscle strength, disease activity, functional capacity, and bone mineral density in early rheumatoid arthritis. Arthritis & Rheumatism, 2001;44(3):515–522. pubmed.ncbi.nlm.nih.gov
  7. de Jong Z, Munneke M, Lems WF, et al. Slowing of bone loss in patients with rheumatoid arthritis by long-term high-intensity exercise: results of a randomized, controlled trial. Arthritis & Rheumatism, 2004;50(4):1066–1076. pubmed.ncbi.nlm.nih.gov
  8. de Jong Z, Munneke M, Zwinderman AH, et al. Long term high intensity exercise and damage of small joints in rheumatoid arthritis. Annals of the Rheumatic Diseases, 2004;63(11):1399–1405. pubmed.ncbi.nlm.nih.gov
  9. Lemmey AB, Marcora SM, Chester K, Wilson S, Casanova F, Maddison PJ. Effects of high-intensity resistance training in patients with rheumatoid arthritis: a randomized controlled trial. Arthritis & Rheumatism, 2009;61(12):1726–1734. pubmed.ncbi.nlm.nih.gov
  10. Roubenoff R, Roubenoff RA, Cannon JG, et al. Rheumatoid cachexia: cytokine-driven hypermetabolism accompanying reduced body cell mass in chronic inflammation. The Journal of Clinical Investigation, 1994;93(6):2379–2386. pubmed.ncbi.nlm.nih.gov
  11. Tam K, Wong-Pack M, Liu T, et al. Risk factors and clinical outcomes associated with sarcopenia in rheumatoid arthritis: a systematic review and meta-analysis. Journal of Clinical Rheumatology, 2024;30(1):18–25. pubmed.ncbi.nlm.nih.gov
  12. Henchoz Y, Bastardot F, Guessous I, et al. Physical activity and energy expenditure in rheumatoid arthritis patients and matched controls. Rheumatology (Oxford), 2012;51(8):1500–1507. pubmed.ncbi.nlm.nih.gov
  13. Aviña-Zubieta JA, Choi HK, Sadatsafavi M, Etminan M, Esdaile JM, Lacaille D. Risk of cardiovascular mortality in patients with rheumatoid arthritis: a meta-analysis of observational studies. Arthritis & Rheumatism, 2008;59(12):1690–1697. pubmed.ncbi.nlm.nih.gov
  14. Stavropoulos-Kalinoglou A, Metsios GS, Veldhuijzen van Zanten JJCS, Nightingale P, Kitas GD, Koutedakis Y. Individualised aerobic and resistance exercise training improves cardiorespiratory fitness and reduces cardiovascular risk in patients with rheumatoid arthritis. Annals of the Rheumatic Diseases, 2013;72(11):1819–1825. pubmed.ncbi.nlm.nih.gov
  15. Siempis T, Tsakiris C, Anastasia Z, Alexiou GA, Voulgaris S, Argyropoulou MI. Radiological assessment and surgical management of cervical spine involvement in patients with rheumatoid arthritis. Rheumatology International, 2023;43(2):195–208. pubmed.ncbi.nlm.nih.gov
  16. Di Muzio C, Conforti A, Bruno F, et al. The assessment of atlantoaxial joint involvement in patients with rheumatoid arthritis, results from an observational "real-life" study. Scientific Reports, 2023;13(1):20146. pubmed.ncbi.nlm.nih.gov