Fibromyalgia drugs: small benefits for some, no clinically important pooled effect on pain

Design

Systematic review and network meta-analysis of randomised trials, GRADE-assessed

Population

151 trials, 32,592 adults with fibromyalgia

Primary outcome

Meaningful improvement, pain, function, sleep, mental health and adverse events

Effect

Meaningful improvement 105–283 more per 1,000 (low certainty); no drug reached a clinically important pooled effect on pain

This network meta-analysis pooled 151 randomised trials with 32,592 adults with fibromyalgia, comparing drugs and some supplements with placebo, usual care and each other. Certainty was graded with GRADE.

With low certainty, sodium oxybate, selective noradrenaline reuptake inhibitors, SNRIs and gabapentinoids may raise the chance of a patient-reported meaningful improvement, by 105 to 283 more per 1,000. In responder analyses, sodium oxybate, the noradrenaline drugs, SNRIs and cyclobenzaprine gave 111 to 202 more per 1,000 a 30% pain reduction. Tricyclics, sodium oxybate and gabapentinoids may improve sleep. But no drug's average effect on pain, physical health or mental health crossed the threshold for clinical importance. No drug increased serious adverse events or discontinuation in the trials.

For the physician carrying a patient with fibromyalgia and several other conditions, this supports a modest, time-limited drug trial rather than escalating combinations. Most patients will not respond meaningfully; a few will. Agree a target and a review date, and stop what is not working.

• Set a specific, measurable goal before starting any fibromyalgia drug, and review it at a fixed date.
• Consider an SNRI or a low-dose tricyclic where sleep or mood is prominent; expect benefit in a minority.
• Stop a drug that has not met its goal rather than adding a second on top.
• Weigh gabapentinoid sedation, dizziness and misuse risk, especially in older patients or with opioids.
• Sodium oxybate is not available for this use in India; do not count it among options here.

Why it matters

It explains why so many fibromyalgia patients end up on drug combinations that do little: the average benefit is small.

Don't overread it

The responder benefits rest on low-certainty evidence; small averages can hide a few real responders, but also chance.

The statistics, in plain English

'111 to 202 more per 1,000' means that for every 1,000 people treated, roughly 110 to 200 more achieved a 30% pain reduction than on placebo — so about one extra responder for every five to nine treated. 'Low certainty' means further research could change these estimates substantially.

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.