- Design
- Delphi consensus definition applied to a single-centre retrospective cohort
- Population
- 220 patients with generalised myasthenia gravis diagnosed 1998 to 2020 at a Spanish referral unit
- Primary outcome
- Proportion adequately controlled at 2-year intervals
- Effect
- 62.0% at 2 years to 81.9% at 8 years; 17.7% never controlled
Spanish neurologists used a Delphi process to define adequately controlled generalised myasthenia gravis. The definition requires a stable post-intervention status of remission or minimal manifestations for a year, an MG-ADL score of 2 or less, no exacerbation needing a change in immunosuppression or rescue therapy in the past year, and no or only mild adverse effects from treatment.
The definition was then applied to 220 patients from one referral unit, diagnosed between 1998 and 2020. The share adequately controlled rose from 62.0% at two years to 81.9% at eight years. Of the 220, 47.3% were controlled at every assessment and 17.7% at none. No clinical feature predicted control. Drug interruptions for adverse effects were more frequent in patients not controlled, and mean annual direct cost per patient was 5,827 to 10,190 euros in that group against 843 to 2,719 euros in those controlled.
The cohort is from one referral centre, which probably over-represents difficult cases. A shared definition of control is useful for deciding when to escalate therapy and for comparing centres.
- Control required stable remission or minimal manifestations, MG-ADL of 2 or less, no recent exacerbations and no more than mild adverse effects.
- The share controlled rose from 62.0% at 2 years to 81.9% at 8 years.
- 17.7% were never adequately controlled at any assessment.
- No baseline feature predicted who would stay uncontrolled.
- Consider using MG-ADL at every visit to judge whether treatment needs escalation.
Why it matters
Without an agreed target, long-term immunosuppression and its adverse effects are hard to justify or compare.
Don't overread it
The definition came from a Spanish expert panel and was applied in one referral unit, so it is not a guideline.
The statistics, in plain English
Control was assessed repeatedly over years in the same patients, so the rise from 62.0% to 81.9% partly reflects who stays in follow-up. The numbers describe one referral cohort and are not population rates.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for neurology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free