Should azithromycin be used to treat COVID-19?
29 May 2020
COVID-19 DPhil Infection, Respiratory and Acute Care
COVID-19 is caused by a virus - so why would we consider treating it with an antibiotic like azithromycin? GP and DPhil Student, Kome Gbinigie, and Postdoctoral Researcher, Kerstin Frie, review the evidence.
To date, there are no known drugs that are effective for treating COVID-19. It is likely to take a long time to develop novel treatments. As the pandemic continues to spread, there have been efforts to re-purpose existing medications as treatments for this disease. One potential treatment is an antibiotic called azithromycin. Azithromycin is widely used to treat respiratory and skin infections, and is generally considered safe for its current approved indications. An online survey found that azithromycin is the second most commonly prescribed treatment for COVID-19 by doctors. US President Donald Trump has recently stated that he has taken a dose of azithromycin as a preventative measure against COVID-19, along with other medicines. Why is it that azithromycin, an antibiotic, is being considered as a treatment for a viral infection?
Feeling confused? Read on!
We don’t know for sure how azithromycin might work against SARS-CoV-2 (the virus that causes COVID-19), but different theories exist. In vitro studies (that is, studies conducted in a petri dish) suggest that azithromycin increases the pH within parts of human cells. The altered pH level might interfere with the ability of SARS-CoV-2 virus to bind to cells, as well as to replicate and spread within cells. Azithromycin may additionally reduce the levels of small proteins called cytokines; some of which promote inflammation that can in turn damage human cells.
We decided to do some digging to find out whether there is any evidence for using azithromycin to treat COVID-19.
Searching feverishly through the depths of the web
We searched a number of electronic databases to find studies assessing the use of azithromycin as a treatment for COVID-19. Due to the unprecedented rate at which new research is being published on COVID-19, we searched repositories that contain manuscripts that have not yet been formally peer-reviewed, such as MedRxiv. We included in vivo studies (studies conducted in humans) and in vitro studies (studies conducted outside of humans). For inclusion in our review, in vivo studies needed to compare outcomes of patients with COVID-19 who did, and did not, receive azithromycin.
Locking down the findings
After screening some hundreds of studies, we found three that were relevant to our research question. Two of the studies were in vitro and one was in vivo. All three studies were published as pre-prints.
In vitro studies
One of the studies found that azithromycin effectively inhibited the activity of SARS-CoV-2 virus. The other study found that azithromycin alone was not able to inhibit replication of SARS-CoV-2 virus. The authors of this study did find, however, that at certain concentrations of azithromycin in combination with hydroxychloroquine there was an inhibitory effect against SARS-CoV-2. Of note, the ratio of virus particles to host cells was much higher in the latter study, meaning that azithromycin was put to a harder test. This may explain some of the difference in the findings.
In vivo study
We identified one study that was suitable for inclusion. This was a small hospital-based trial conducted in France, in which 36 patients with COVID-19 received either hydroxychloroquine alone (14 patients), hydroxychloroquine and azithromycin combined (6 patients) or usual care (16 patients). The authors found that on day six of the study, 100 per cent of the patients receiving azithromycin and hydroxychloroquine combined tested negative for the virus, compared with 57.1 per cent of those receiving hydroxychloroquine alone, and just 12.5 per cent of the control group. This result was statistically significant.
Problems on the horizon?
Whilst these results may seem promising, there are a number of problems with these studies that must be considered. All of the studies have been published as pre-prints, meaning that they have not yet been formally peer-reviewed. The findings of the in vitro studies were conflicting; one found that azithromycin alone inhibited SARS-CoV-2 activity, whilst the other did not. Furthermore, in vitro studies can never fully replicate the conditions in the human body, so we cannot be sure that any effects seen in the laboratory will be reproduced in humans.
The in vivo study identified also has limitations. It was a very small trial with only 36 participants. This means that any differences in outcomes between groups may have arisen by chance. The researchers did not randomly assign patients to the different study groups. When the process of allocating treatments in a study is not random, it can introduce bias. The authors did not report any medium to long-term outcomes, which means that we don’t have information about the effectiveness of the treatment in the long-term. Furthermore, the authors did not report adverse events, which means that we cannot tell how safe azithromycin is for the treatment of COVID-19 from the results of this study.
We didn't identify any data from clinical trials assessing azithromycin alone, which means that any potential effect of azithromycin may be dependent on it being taken together with hydroxychloroquine.
Where do we go from here?
We didn't find any published studies assessing azithromycin as a standalone treatment for COVID-19, so we cannot say whether it is a safe or effective treatment in the context of the current pandemic.
From the extremely limited evidence identified, and from studies with methodological problems, there is possible evidence of a synergy between azithromycin and hydroxychloroquine. What we really need are well-conducted trials of azithromycin with adequate numbers of participants with COVID-19 to confirm or refute these preliminary findings. If proven to be effective, azithromycin would be an affordable and widely available treatment option.
It is also very important that these studies report safety data. Azithromycin can uncommonly prolong the QT interval, which is the time it takes for electrical impulses to spread across the main pumping chambers of the heart (called the ventricles) causing them to contract, and then relax. This is unlikely to cause any problems for most people taking azithromycin.
However, there is an increased risk if azithromycin is prescribed alongside other medications that have the same effect on the QT interval, such as hydroxychloroquine. A prolonged QT interval can lead to serious heart rhythm disturbances in some people. Clinical trials involving azithromycin should therefore carefully consider the criteria that they use to determine patient eligibility, ensuring that included participants are not at a high risk of such a complication.
So, until there is more evidence to suggest otherwise, we would not recommend that clinicians prescribe azithromycin to treat COVID-19 outside of clinical trials. However, we recognise that in line with local/national treatment protocols, some clinicians may wish to prescribe azithromycin if there is evidence of a bacterial super-infection.
Should azithromycin be used to tread COVID-19? A rapid review
Kome Gbinigie, Kerstin Frie
BJGP Open. DOI: 10.3399/bjgpopen20X101094
Kome’s time is funded by the Wellcome Trust (grant reference code: 203921/Z/16/Z).
Kerstin’s time is funded by the Wellcome Trust, Our Planet Our Health (Livestock, Environment and People - LEAP), award number 205212/Z/16/Z.
The views expressed are those of the authors and not necessarily those of the Wellcome Trust.