Viral persistence is among the leading hypotheses for long COVID, and it generates an obvious test: give a potent antiviral for longer than the acute course and see whether symptoms improve. RECOVER-VITAL did exactly that. At 69 US sites, 959 adults with symptoms persisting at least 12 weeks after SARS-CoV-2 infection were randomised in three arms — 15 days of nirmatrelvir-ritonavir then placebo, 25 days of nirmatrelvir-ritonavir, or 25 days of placebo with ritonavir. Participants were enrolled into one of three symptom phenotypes: cognitive, autonomic or exercise. Median age was 49 and two-thirds were women.
No phenotype benefited. For the cognitive phenotype, adjusted differences against placebo were 3.2% (95% CI -10.4 to 16.8) for 25 days and -2.2% (-15.5 to 11.1) for 15 days. For autonomic, -6.4% (-18.5 to 5.7) and -0.1% (-12.5 to 12.3). For exercise, -7.8% (-19.5 to 3.8) and 0.9% (-11.4 to 13.2). Secondary performance outcomes showed no difference either. Safety was unremarkable: no deaths, and 52 serious adverse events in 42 of 963 participants.
This is a well-designed trial with a ritonavir-containing placebo, which matters because ritonavir's taste and side effects would otherwise unblind everyone. Three phenotypes, two durations, six comparisons, and not one of them positive in any direction. That is about as clean a negative as this field produces.
The honest interpretation is narrower than the headline. It does not disprove viral persistence as a mechanism — it shows that up to 25 days of this particular antiviral, given a median of well over a year after infection, does not reverse established symptoms. Persistence may still be real and simply no longer treatable at that stage, or the reservoir may be in tissue this drug reaches poorly. What it does settle is the clinical question in front of you: there is no basis for prescribing a prolonged nirmatrelvir course to a patient with long COVID, and there is a basis for saying so clearly when asked.
- Do not prescribe extended nirmatrelvir-ritonavir courses for long COVID; there is no benefit at 15 or 25 days.
- The trial used a ritonavir-containing placebo, so the negative result is not an artefact of failed blinding.
- Redirect the consultation to symptom-specific rehabilitation, pacing and management of orthostatic intolerance.
- A negative antiviral trial does not disprove viral persistence as a mechanism in earlier disease.
- Be prepared for patients who have read about this hypothesis and are seeking the drug privately.
The statistics, in plain English
All six confidence intervals straddle zero, and several extend further in the direction of harm than benefit. Consistency across independent comparisons is what makes a negative result convincing: a single null could be underpowering, but six nulls across three symptom groups and two durations point the same way. The intervals are still wide — up to about 17 percentage points — so a small effect has not been excluded, only an effect large enough to be worth the drug, the interactions and the cost.
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 infectious diseases, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free