Autoimmune disease clustered in the two years before myasthenia gravis was diagnosed

Design

retrospective population-based matched cohort study using national claims data, Poisson regression

Population

8,355 Korean adults with myasthenia gravis and 83,550 matched controls, mean age 53.7 years

Primary outcome

rate of autoimmune disease in the 10 years before myasthenia gravis diagnosis

Effect

any autoimmune disease RR 2.12 (95% CI 1.97–2.28) over 10 years; 4.27 (3.78–4.83) in the 2 years before

Myasthenia gravis keeps company with other autoimmune disease, but whether those diagnoses precede it, follow it or simply coexist has not been established at population scale. This Korean National Health Insurance study matched 8,355 patients with myasthenia gravis — identified by diagnostic code plus rare disease registration — to 83,550 controls by age, sex and index year, and looked back ten years.

Any autoimmune disease was twice as frequent in the myasthenia group over the full decade (rate ratio 2.12, 95% CI 1.97 to 2.28), and the concentration was close to diagnosis: 4.27 (3.78 to 4.83) in the two years before. The strongest individual associations were systemic lupus erythematosus (4.87, 2.71 to 8.77), Sjogren syndrome (4.75, 3.01 to 7.50), autoimmune thyroid disease (3.66, 3.29 to 4.07) and seropositive rheumatoid arthritis (2.45, 1.89 to 3.19). Psoriasis and type 1 diabetes appeared only within the final two years. Crohn disease and ulcerative colitis showed no association.

The clustering in the last two years is the part that should change behaviour, and it cuts both ways. Some of it is genuine autoimmune co-occurrence; some is ascertainment, because a patient already under a rheumatologist gets investigated faster. Either way, fatigable weakness in someone with known lupus, Sjogren syndrome or thyroid autoimmunity deserves a lower threshold for acetylcholine receptor antibodies than the same symptom in someone without.

• Lower your threshold for myasthenia testing in a patient with lupus, Sjogren syndrome or autoimmune thyroid disease
• Ask specifically about diurnal variation and fatigability rather than accepting fatigue as a connective tissue symptom
• Check thyroid function and antibodies at myasthenia diagnosis if not already done
• Note that inflammatory bowel disease showed no association — do not extend the rule to every autoimmune condition
• Remember that claims data record diagnostic codes, not verified diagnoses

Why it matters

It identifies the two years before diagnosis as the window where myasthenia is most likely to be sitting undetected behind another autoimmune label.

Don't overread it

Part of the clustering close to diagnosis is likely detection bias — patients under specialist follow-up get investigated faster.

The statistics, in plain English

Note how the intervals widen for the rarer conditions: lupus gives a rate ratio of 4.87 but with an interval from 2.71 to 8.77, because few patients in either group had it. Autoimmune thyroid disease, being common, gives 3.66 with a tight 3.29 to 4.07. The rise from 2.12 over ten years to 4.27 in the final two is the pattern that carries the message, and it is also the pattern most vulnerable to the surveillance effect the authors warn about.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.