Skip to main content
Device Global Header

Subscribe
or Renew

Advanced Search

Editor’s Note: These letters were published on July 15, 2020, at NEJM.org.

Correspondence

Hydroxychloroquine as Postexposure Prophylaxis for Covid-19

To the Editor

The window for postexposure prophylaxis against Covid-19 is narrow.1-3 Therapy that is initiated up to 4 days after exposure to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is early treatment, not postexposure prophylaxis. The trial described in the article by Boulware et al. (published online on June 3 at NEJM.org)4 was therefore largely about the prevention of symptoms in persons who may already have been infected. The trial was designed to detect a 50% relative reduction in new cases of symptomatic Covid-19; this estimate was overly optimistic. The trial was not powered to detect an important, but lesser reduction. Regardless, the authors found a nonsignificant (P=0.35) absolute difference of −2.4 percentage points (a 17% relative reduction) in the incidence of new symptomatic illness compatible with Covid-19 between the percentage of participants who received hydroxychloroquine within 4 days after exposure and those who received placebo. The upper boundary of the 95% confidence interval was an absolute reduction of approximately 7 percentage points (a relative reduction of approximately 50%), which was the investigators’ prespecified target effect size.

We can draw three conclusions. First, hydroxychloroquine might be effective in early treatment, since the absence of evidence is not evidence of absence. Second, a larger trial involving participants with a virologic diagnosis should be conducted to detect a meaningful early treatment effect (e.g., a trial involving 8000 participants could detect a reduction in the incidence of symptomatic Covid-19 from 15.0% to 12.5%). Third, other trials examining preexposure prophylaxis and early postexposure prophylaxis should be considered.

Michael S. Avidan, M.B., B.Ch.
Washington University School of Medicine, St. Louis, MO

Hakim-Moulay Dehbi, Ph.D.
University College London, London, United Kingdom

Sinead Delany-Moretlwe, M.B., B.Ch., Ph.D.
University of the Witwatersrand, Johannesburg, South Africa

No potential conflict of interest relevant to this letter was reported.

This letter was published on July 15, 2020, at NEJM.org.

  1. 1. Linton NM, Kobayashi T, Yang Y, et al. Incubation period and other epidemiological characteristics of 2019 novel coronavirus infections with right truncation: a statistical analysis of publicly available case data. J Clin Med 2020;9:538-538.

  2. 2. Ferretti L, Wymant C, Kendall M, et al. Quantifying SARS-CoV-2 transmission suggests epidemic control with digital contact tracing. Science 2020;368:eabb6936-eabb6936.

  3. 3. He X, Lau EHY, Wu P, et al. Temporal dynamics in viral shedding and transmissibility of COVID-19. Nat Med 2020;26:672-675.

  4. 4. Boulware DR, Pullen MF, Bangdiwala AS, et al. A randomized trial of hydroxychloroquine as postexposure prophylaxis for Covid-19. N Engl J Med 2020;383:517-525.

To the Editor

The primary results of the trial conducted by Boulware et al. were nonsignificant, and the trial was interpreted as negative. A few points warrant careful consideration. First, the reverse fragility index was only 5 (i.e., 5 events would have to change in order for the primary end point to move from nonsignificant to significant). The reverse fragility quotient (the reverse fragility index divided by the sample size) was 0.006, which indicates that the nonsignificance of the results was contingent on only 0.6 events per 100 participants. Second, the absolute risk reduction was 2.4 percentage points. This magnitude of absolute risk reduction is similar to that in other positive trials.1,2 The number needed to treat was 42, which may be acceptable considering the current escalating pandemic. Third, the trial was powered to detect a 50% relative reduction in new symptomatic infections, which is an extraordinary robust estimate. Therefore, in accordance with the guidelines of the Journal and the American Statistical Association, it is critical to not view results in a dichotomized manner on the basis of P values, especially when the results are fragile, the trial is underpowered, and other statistical measures suggest a possible benefit with no signal for harm.3,4

Muhammad-Shahzeb Khan, M.D.
Cook County Hospital, Chicago, IL

Javed Butler, M.D., M.P.H.
University of Mississippi, Jackson, MS

Dr. Butler reports receiving consulting fees from Abbott, Adrenomed, Amgen, Array BioPharma, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, CVRx, G3 Pharmaceuticals, Impulse Dynamics, Innolife, Janssen, LivaNova, Luitpold Pharmaceuticals, Medtronic, Merck, Novartis, Novo Nordisk, Relypsa, Roche, V-Wave, and Vifor Pharma. No other potential conflict of interest relevant to this letter was reported.

This letter was published on July 15, 2020, at NEJM.org.

  1. 1. Tavazzi L, Maggioni AP, Marchioli R, et al. Effect of n-3 polyunsaturated fatty acids in patients with chronic heart failure (the GISSI-HF trial): a randomised, double-blind, placebo-controlled trial. Lancet 2008;372:1223-1230.

  2. 2. Pitt B, Remme W, Zannad F, et al. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med 2003;348:1309-1321.

  3. 3. Wasserstein RL, Schirm AL, Lazar NA. Moving to a world beyond <0.05. Am Stat 2019;73:Suppl:1-19.

  4. 4. Harrington D, D’Agostino RB Sr, Gatsonis C, et al. New guidelines for statistical reporting in the Journal. N Engl J Med 2019;381:285-286.

To the Editor

The trial by Boulware et al. that showed ineffectiveness of hydroxychloroquine for prevention of Covid-19 nevertheless reestablished the safety of hydroxychloroquine in otherwise healthy persons. The primary conclusions of the trial regarding the ineffectiveness of hydroxychloroquine to protect against Covid-19 rely heavily on the incidence of new illness compatible with Covid-19 (reported in 11.8% of the participants in the hydroxychloroquine group and 14.3% of those in the placebo group) rather than on laboratory-confirmed diagnoses (in only 2.7% and 2.2% of the participants, respectively). Did the participants who had symptoms compatible with Covid-19 also undergo laboratory testing for SARS-CoV-2 infection? It would be important for all 821 asymptomatic participants with high-risk exposure who were enrolled in the trial to undergo testing for SARS-CoV-2 infection before and after the administration of hydroxychloroquine.1

Asymptomatic persons with Covid-19 pose a considerable challenge with respect to the hazard of exposure and community spread of SARS-CoV-2 infection.2 The classification and identification of Covid-19–positive persons on the basis of symptoms may be misleading because of the variable and overlapping symptoms of the disease.3 The results of the trial conducted by Boulware et al. with respect to the moderate risk of Covid-19 and the low incidence of hospitalizations in a population with high-risk exposure to SARS-CoV-2 are important for disease management.4

Babu L. Tekwani, Ph.D.
Southern Research, Birmingham, AL

Dr. Tekwani reports receiving support to his laboratory from the collaborative Antiviral Drug Discovery and Development Center, which is led by the University of Alabama at Birmingham School of Medicine and funded by a Centers of Excellence for Translational Research grant (U19 AI142759) from the NIAID. No other potential conflict of interest relevant to this letter was reported.

This letter was published on July 15, 2020, at NEJM.org.

The opinions expressed in this letter are the responsibility of author and do not represent the official views of Southern Research, the University of Alabama at Birmingham, or the National Institute of Allergy and Infectious Diseases (NIAID).

  1. 1. La Scola B, Le Bideau M, Andreani J, et al. Viral RNA load as determined by cell culture as a management tool for discharge of SARS-CoV-2 patients from infectious disease wards. Eur J Clin Microbiol Infect Dis 2020;39:1059-1061.

  2. 2. Kong W, Wang Y, Hu J, Chughtai A, Pu H, Clinical Research Collaborative Group of Sichuan Provincial People’s Hospital. Comparison of clinical and epidemiological characteristics of asymptomatic and symptomatic SARS-CoV-2 infection: a multi-center study in Sichuan Province, China. Travel Med Infect Dis 2020 May 31 (Epub ahead of print).

  3. 3. Centers for Disease Control and Prevention. Coronavirus disease 2019: symptom-based strategy to discontinue isolation for persons with COVID-19. 2020 (https://www.cdc.gov/coronavirus/2019-ncov/community/strategy-discontinue-isolation.html).

  4. 4. Hafner CM. The spread of the Covid-19 pandemic in time and space. Int J Environ Res Public Health 2020;17:3827-3827.

Response

The authors reply: In response to Avidan and colleagues: we agree that the postexposure prophylaxis window for SARS-CoV-2 is narrow; for this reason, we used a pragmatic trial design, and trial medication was shipped overnight to participants (68% of the participants began to receive hydroxychloroquine or placebo within 1 to 3 days). Although Avidan et al. suggest that a trial involving 8000 participants is indicated in order to detect an absolute reduction in symptoms of 2.4 percentage points, our trial was designed to detect a 50% relative reduction in the incidence of symptomatic Covid-19 and was completed in 7 weeks in the setting of a pandemic in order to provide meaningful timely results. It would have been impractical to undertake an unfunded trial that was 10 times as large as our trial. For a therapy to be widely used as postexposure prophylaxis, it should provide a significant reduction in the incidence of infection. For example, preexposure or postexposure prophylaxes against human immunodeficiency virus infection or bacterial meningitis are more than 90% effective.1,2 We agree that trials evaluating the benefit of hydroxychloroquine as preexposure prophylaxis remain valuable.

In response to Khan and Butler: the reverse fragility index for our trial was 10, which implies that if there was a 20% reduction in the percentage of participants with new Covid-19–compatible illnesses in the intervention group (from 49 to 39 participants), the results would be different (39 of 414 participants in the hydroxychloroquine group and 58 of 407 participants in the placebo group, P=0.04). However, this is not what we found. In this context, we think that widespread use of hydroxychloroquine as postexposure prophylaxis is not warranted.

In response to Tekwani: we agree that classification of participants on the basis of illness compatible with Covid-19 is a limitation, and we acknowledge this limitation in our article. In most locales in March and April 2020, SARS-CoV-2 testing was unavailable to most persons in the United States who were not hospitalized, including 60% of the health care workers in our trial. Thus, we relied on the identification of symptoms consistent with Covid-19 put forth by the Council of State and Territorial Epidemiologists.3 In a perfect world, polymerase-chain-reaction (PCR) testing would have been preferable; however, even PCR is not perfect for detecting early infections. On the first day of symptoms, false negative PCR results are estimated to occur in 38% of patients (range, 18 to 65), and these false negative results decrease over time.4 An effective prophylaxis would reduce the incidence of symptomatic disease as well as that of laboratory-confirmed disease. We hope that the scientific community can build on our results with additional, well-designed randomized clinical trials.

Elizabeth C. Okafor, B.Sc.
Katelyn A. Pastick, B.Sc.
Radha Rajasingham, M.D.
University of Minnesota, Minneapolis, MN

Since publication of their article, the authors report no further potential conflict of interest.

This letter was published on July 15, 2020, at NEJM.org.

  1. 1. Dworzack DL, Sanders CC, Horowitz EA, et al. Evaluation of single-dose ciprofloxacin in the eradication of Neisseria meningitidis from nasopharyngeal carriers. Antimicrob Agents Chemother 1988;32:1740-1741.

  2. 2. Grant RM, Lama JR, Anderson PL, et al. Preexposure chemoprophylaxis for HIV prevention in men who have sex with men. N Engl J Med 2010;363:2587-2599.

  3. 3. Standardized surveillance case definition and national notification for 2019 novel coronavirus disease (COVID-19). Atlanta: Council of State and Territorial Epidemiologists, 2020 (https://cdn.ymaws.com/www.cste.org/resource/resmgr/2020ps/interim-20-id-01_covid-19.pdf).

  4. 4. Kucirka LM, Lauer SA, Laeyendecker O, Boon D, Lessler J. Variation in false-negative rate of reverse transcriptase polymerase chain reaction-based SARS-CoV-2 tests by time since exposure. Ann Intern Med 2020 May 13 (Epub ahead of print).

Citing Articles (4)