Fluvoxamine modestly reduced long COVID fatigue; metformin was inconclusive

Design

Randomised, placebo-controlled, adaptive trial, Brazil

Population

399 adults with fatigue 90 days to 24 months after SARS-CoV-2 infection

Primary outcome

Change in Fatigue Severity Scale score

Effect

Fluvoxamine vs placebo: mean difference -0.43 at day 60 and -0.58 at day 90 (posterior probability of superiority 99%)

This adaptive, placebo-controlled trial at outpatient sites in Brazil enrolled 399 adults with fatigue 90 days to 24 months after confirmed SARS-CoV-2 infection. Participants received fluvoxamine 100 mg twice daily, extended-release metformin or placebo for 60 days. The outcome was change in the Fatigue Severity Scale.

Fluvoxamine lowered the score against placebo by a mean 0.43 points at day 60 and 0.58 at day 90, each with a 99% posterior probability of superiority, and improved quality of life. Metformin results were inconclusive: the dose was cut from 750 mg to 500 mg twice daily after adverse effects caused withdrawals, and that arm stopped for futility. Adverse events were less frequent with fluvoxamine (24.0%) than with placebo (35.5%) or metformin (about 43% to 45%).

Baseline scores were about 5.6 to 5.9, so the gain is modest and benefit beyond day 90 is unknown. Fluvoxamine interacts with several common drugs through CYP1A2 and CYP2C19, which matters in multimorbid patients.

• Fluvoxamine 100 mg twice daily lowered fatigue scores by 0.43 points at 60 days and 0.58 at 90 days against placebo.
• Metformin results were inconclusive because of a dose change and early stopping.
• Adverse events were less common with fluvoxamine than with placebo (24.0% vs 35.5%).
• Check interacting drugs before considering fluvoxamine, including theophylline, warfarin and other CYP-metabolised drugs.
• Benefit past 90 days and effects on other long COVID symptoms are not known.

Why it matters

It is one of few randomised signals for a treatment of long COVID fatigue that a general physician can prescribe.

Don't overread it

A single adaptive trial with modest gain, a foundation funder and follow-up of only 90 days has not changed any guideline.

The statistics, in plain English

A posterior probability of superiority of 99% comes from a Bayesian analysis and means that, given the data, the drug is very likely better than placebo. It does not say the effect is large: the average difference was under 0.6 points on a 7-point scale.

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.