This week we have results from the “long awaited” (in COVID-19 time) randomized trials of hydroxychloroquine and dexamethasone for treatment of COVID-19 as well as new guidelines from the CDC for testing. We also have new information on the safety of convalescent plasma in treating COVID-19. Note that the hydroxychloroquine and dexamethasone trials have not yet been peer reviewed but the protocol, data, etc. are available online here.
1) Hydroxychloroquine does not work for hospitalized patients. We already know that hydroxychloroquine does not work to prevent infection. We now have information that it doesn’t work to treat COVID-19. This is a randomized, controlled trial of 11,000 plus hospitalized COVID-19 patients in the United Kingdom. (the “RECOVERY” trial). Within this study, 1542 patients were randomized to hydroxychloroquine and 3132 got usual care without hydroxychloroquine. The endpoint was mortality at 28 days. There was no difference in mortality (25.7% with hydroxychloroquine and 23.5% with standard care (hazard ratio 1.11 [95% confidence interval 0.98-1.26]). There was also no difference in hospital stay or other endpoints. Read more here.
In addition to this negative study, there are concerns that hydroxychloroquine may reduce the antiviral activity of remdesivir (more here). Given the preponderance of data, the FDA in the US has revoked the authorization to use hydroxychloroquine for COVID-19 (more here). There is now good evidence that hydroxychloroquine should not be used for treating COVID-19 in any clinical setting.
- A Randomized Trial of Hydroxychloroquine as Postexposure Prophylaxis for Covid-19 NEJM. June 3, 2020 DOI: 10.1056/NEJMoa2016638
2) Dexamethasone is the first drug to reduce mortality in patients with COVID-19 who require respiratory support. This is an analysis of another arm of the “RECOVERY” trial. In this arm, 2104 patients were randomized to dexamethasone 6mg once a day for 10 days and were compared to 4321 patients randomized to usual care alone. The 8-day mortality was highest in those who required ventilation (41%), lower in those patients who required oxygen only (25%), and lowest among those who did not require any respiratory support (13%).The NNT to prevent one death in patients on a ventilator is 8. The NNT for patients treated with oxygen alone (non-invasively) is 25. Patients who did not require oxygen did not benefit.
To put this in perspective, the numbers have been reported as “a miracle” with a reduction of death of 1/3 in ventilated patients and 1/5 of those on non-invasive oxygen. But those are relative reductions and not absolute reductions in mortality. The NNT is a better measure of the absolute reduction in mortality. Still a positive outcome, but certainly not a “a miracle” cure.
The results of this study are in contrast to the recommendations not to use steroids in COVID-19. It reminds us that “expert opinion” can be wrong and a well conducted study can change practice. The same shift has occurred with hydroxychloroquine: “Expert opinion” suggested using it, well controlled studies have shown there is no benefit.
3) Convalescent plasma is safe when used to treat COVID-19 although we still do not have solid data on efficacy. The first study is a retrospective analysis of 5000 hospitalized patients who got convalescent plasma therapy. Sixty-six percent of the patients were in the ICU. Overall 7-day mortality was 14.9% with a mortality of 0.3% within 4 hours of plasma infusion. This is consistent with the expected mortality based on historical data (10-20%). Adverse events were rare and there was no signal suggesting convalescent plasma is harmful. They report 4 deaths, transfusion associated fluid overload (TACO) in 7 patients, transfusion related lung injury in 11 patients and severe allergic reaction in 4.
A second study of 25 patients treated at Huston Methodist hospitals came to a similar conclusion regarding the safety of convalescent plasma.
- J Clin Invest. June 11, 2020. https://doi.org/10.1172/JCI140200. (https://www.jci.org/articles/view/140200/pdf)
- Am J Pathol 2020,-:1e11;https://doi.org/10.1016/j.ajpath.2020.05.014
4) There is a new, comprehensive summary statement about testing released by the CDC. The full document is here. A brief summary:
- Antibody testing is not to be used for diagnosing an acute infection.
- An authorized PCR or antigen detection assay should be used to test patients with symptoms.
- Asymptomatic patients with known or suspected exposure to COVID-19 should be tested, hopefully as part of a contact tracing program.
- For those asymptomatic individuals without known exposure in special settings such as group homes or nursing homes there are several recommended options. These include(verbatim):
Approaches for early identification of asymptomatic individuals include:
- Initial testing of everyone residing and/or working in the setting,
- Regular (e.g., weekly) testing of everyone residing and/or working in the setting, and
- Testing of new entrants into the setting and/or those re-entering after a prolonged absence (e.g., one or more days)
Settings for which these approaches could be considered include:
- Long-term care facilities
- Correctional and detention facilities
- Homeless shelters
- Other congregate work or living settings including mass care, temporary shelters, assisted living facilities, and group homes for individuals with intellectual disabilities and developmental disabilities
- High-density critical infrastructure workplaces where continuity of operations is a high priority
The bottom line is that this is a comprehensive, complex document that can be found here.