Effect of early treatment with fluvoxamine on risk of emergency care and hospitalisation among patients with COVID-19: the TOGETHER randomised, platform clinical trial
Reis G, Dos Santos Moreira-Silva EA, Silva DCM, et al. Lancet Glob Health. 2021;27;10(1):e42-e51.
SUMMARY:
The theoretical benefit of fluvoxamine stems from the possibility that serotonin dysregulation may play a role in illness, and that fluvoxamine has both anti-inflammatory and antiviral properties.
The TOGETHER trial is a randomized, adaptive platform trial from 11 cities in Brazil investigating the efficacy of repurposed treatments for COVID-19 among high-risk adult outpatients. A total of 8 treatments were tested, but this report is on only fluvoxamine.
The study enrolled adults with confirmed COVID-19 and fewer than 7 days of symptoms with at least 1 high-risk feature (diabetes, hypertension, obesity, immunosuppression, age >50 years, or unvaccinated status).
The study excluded patients who were hospitalized, were vaccinated, had chronic lung issues, were already taking SSRIs, or had an uncontrolled psychiatric disorder.
Of 9,020 screened patients, 1,472 were randomized and enrolled in this trial (739 to fluvoxamine and 733 to placebo). The remainder of the eligible patients (1,826) were randomized to another therapy.
The mean age was 50, and the symptoms were present for a mean of 3.8 days before enrollment.
The primary outcome was a composite of medical admission due to COVID-19 (disease specific) or observation in the ED for >6 hours within 28 days of randomization.
The primary outcome was found in 11% of the fluvoxamine group vs 16% of the placebo group.
No significant differences were observed in any secondary outcomes, including viral clearance, mortality (2% vs 3%), hospitalization for COVID-19, all-cause hospitalization, time to hospitalization, hospital length of stay, days on a ventilator, or time to recovery.
In the infographic for this publication, the authors claim a mortality decrease of up to 91%. How can this be? When the analysis was limited to patients who took >80% of their assigned therapy (per protocol), the mortality was 12 patients vs 1 patient. This approach might have been subject to bias and has become a main area of scrutiny regarding this article.
Using a disease-specific reason for hospitalization biases the findings in favor of the intervention arm, because admissions due to adverse events (eg, serotonin syndrome or suicidality) would not have been counted.
EDITOR’S COMMENTARY: In this large randomized controlled trial from Brazil, the authors report that 100 mg fluvoxamine twice per day for 10 days decreased the need for admission or ED observation for >6 hours among unvaccinated patients with confirmed COVID-19 and at least 1 high-risk comorbidity. Don’t be misled by the data, because there was no difference in hospitalization. It is up to you to decide whether the chance of preventing ED stays >6 hours outweighs the potential harms from the medications or the possibility that the study findings represent a false-positive result among the 8 interventions assessed simultaneously. For me, these data are just not convincing enough to act on in my practice.
EMA 2022 FebruaryFull episode audio for MD edition193:35 min - 190 MB - M4AEMA 2022 February - Abstracts PDF522 KB - PDFEMA 2022 February - MP3s235 MB - ZIPTime to Talk a Little Nerdy: Survivorship Bias224 KB - PDF
To earn CME for this chapter, you need to subscribe.
Sign up today for full access to all episodes and earn CME.
