The utility of hydroxychloroquine in the management of Covid-19 often seems to be more of a political than a medical debate. Many politicians, journalists, hacks, and a whole range of social media experts have an amazing ability to simultaneously micro-analyse and also ignore the scientific evidence that’s out there. Many authors of papers that do not support HCQ use have been challenged and abused online in a way that I find deeply disturbing. It’s not science and it’s not good for managing the pandemic.
Here at St Emlyn’s we’ve taken the approach that we will carefully review and consider evidence as it is published. We keep an open mind and try to use our powers of critical appraisal to come to an informed decision on all topics, and especially the controversial ones such as HCQ. In a previous post we reviewed the initial findings of the RECOVERY trial on HCQ that demonstrated no benefit to patients admitted to hospital with COVID19 (2,3) and as the evidence currently stands that is still our view. However, one of the criticisms of that study was that the HCQ was given too late. As an antiviral it may well be that benefits would be found earlier in the disease process and before the apparently cytokine mediated phase of the disease (which is typically when hospital admission is required). In that regard our minds are still open and the question of early use is still one that might be worth pursuing.
Perhaps the earliest use of a drug in a disease such as Covid-19 would be in the prophylactic use of HCQ in preventing the disease developing in those exposed to it. Arguably that would be the most logical time to give an antiviral drug to potential patients, and this week we have a paper in the NEJM that tests this approach. (1) The abstract is below, but as always we strongly recommend that you read the full paper.
What kind of paper is this?
This is a placebo controlled randomised trial which, as this is an interventional trial of a medication, is the ideal study design to assess the effectiveness of HCQ in preventing COVID19 disease.
What was the intervention.
Individuals who were deemed to be at risk of COVID19 were randomised to placebo or to hydroxychloroquine was 800 mg (4 tablets) once, then 600 mg (3 tablets) 6 to 8 hours later, then 600 mg (3 tablets) daily for 4 more days for a total course of 5 days (19 tablets total).
Tell me about the participants.
This is where this paper gets interesting and where I have some concerns about the design. Patients were recruited through social media and other media platforms across the US and Canada. Such recruitment strategies inevitably affect the type of participants. Factors such as age, language, access to technology are factors here that may not be associated with disease severity (and thus importance). This is evidenced by the median age of participants (40) and the proportion of health care workers (66.4%). This is therefore quite a unusual sample of participants which will affect the generalisability of the results. 821 participants were recruited.
To get into the trial the participants had to have had an exposure to COVID19 defined as household or occupational exposure to a person with confirmed Covid-19 at a distance of less than 6 ft for more than 10 minutes while wearing neither a face mask nor an eye shield (high-risk exposure) or while wearing a face mask but no eye shield (moderate-risk exposure). Interestingly I’ve had moderate risk exposure on many, many shifts over the pandemic (and I’m still antibody negative) as we don’t routinely wear eye shields when seeing COVID19 patients in moderate risk areas (no AGPs).
What about the outcomes?
Again this is an area of concern in this paper. We really want to know whether or not participants developed COVID19, and to me that means laboratory confirmation of infection. This is especially important in a COVID19 study as we know that asymptomatic infection is common, and especially common in the young (and this was a young cohort). However, in this study they used a combination of outcomes to determine a positive test. PCR was used in a proportion but for the majority ‘symptomatically compatible’ symptoms were also used as a positive outcome. As the symptoms for COVID19 are relatively non-specific and asymptomatic disease is common can we really be sure that this is a robust outcome? PCR was used to diagnose 20 positive patients. Symptoms alone were used to diagnose the other 107 positives. So most outcomes in this study were based on self reported clinical symptoms and that’s not as robust as we would like to see.
What were the main results?
The headline figures are that there was no difference in the primary outcome between those on the HCQ and placebo groups. The incidence of new illness compatible with Covid-19 did not differ significantly between those receiving hydroxychloroquine (49 of 414 [11.8%]) and those receiving placebo (58 of 407 [14.3%]) (P = 0.35). The absolute difference was −2.4 percentage points (95% confidence interval, −7.0 to 2.2). interestingly the 95% confidence intervals cross the point of a 50% reduction in COVID19 disease which was part of the original sample size calculation.
Patients taking HCQ were more likely to suffer mild side effects and were a little less compliant with the medication.
What does this paper really tell us?
At first glance it’s great to see an RCT on an important topic such as this. It would be great if had an antiviral that prevents COVID19 infection but this paper suggests not. However, I have concerns about the design, particularly in the participant selection and especially in the outcome measure. It is possible that there is an effect here, but that the lack of a robust gold standard for the diagnosis and the rather bespoke participant selection means that I’m still unsure.
The authors and the accompanying commentary are honest and open about the limitations of the study, and whilst inconclusive as to whether there is benefit this is further evidence that HCQ can only be justifiably prescribed as part of well designed clinical trials. Should any of those show benefit then we will of course change our practice in the spirit of evidence based agility.
- A Randomized Trial of Hydroxychloroquine as Postexposure Prophylaxis for Covid-19 https://www.nejm.org/doi/full/10.1056/NEJMoa2016638
- Dexamethasone, COVID-19 and the RECOVERY trial. St Emlyn’s https://www.stemlynsblog.org/dexamethasone-covid-19-and-the-recovery-trial-st-emlyns/
- Horby P, Lim WS, Emberson J, et al. Effect of Dexamethasone in Hospitalized Patients with COVID-19: Preliminary Report. 2020; published online June 22. DOI:10.1101/2020.06.22.20137273.
- Covid19 resources on St Emlyn’s https://www.stemlynsblog.org/?s=covid19