The following article complements my previous one concerning the COVID-19 pandemic and chloroquine. I suggest you read it first, as it will give you an idea of what the scientific literature says about the action of chloroquine on COVID-19 as of April 7th, 2020. However, this article did neither address the question of ethics, nor fully address whether the situation could justify not to follow the usual methodology.
Indeed, this objection is regularly raised: “even though it has not been established that hydroxychloroquine can actually be used as a treatment for COVID-19, ultimately the dosage is well known and there is probably not a great risk in administering it to patients.” Moreover, there is a view which seems to have some popularity, stating that, considering the circumstances, clinicians should not take too much care to comply with methodology, as the emergency dictates to be less rigorous with procedures. These two issues are related. To address them, one must both look at the collective experience and assess what is not known.
A PDF version of this article is available. An audio version (with quite a robotic voice) is also available.
A worrying suspicion
On March 25th, 2020, a study presenting the case of 416 patients with COVID-19 was published1Shaobo Shi, Mu Qin, Bo Shen, Yuli Cain, Tao Liu, Fan Yang, Wei Gong, Xu Liu, Jinjun Liang, Qinyan Zhao, He Huang, Bo Yang, and Congxin Huang, 2020. ‘Association of cardiac injury with mortality in hospitalized patients with COVID-19 in Wuhan, China’, JAMA Cardiology. Doi: 10.1001/jamacardio.2020.0950. Among these patients, 19.7 % had cardiac problems. It is not surprising that patients in respiratory distress, one of the complications of COVID-19, have heart problems. However, the study reports a suspicion that these cardiac problems are not a consequence of this respiratory distress, but directly due to the action of the virus. Another compilation of COVID-19 case analysis also supports this suspicion2Elissa Driggin, Mahesh V. Madhavan, Behnood Bikdeli, Taylor Chuich, Justin Laracy, Giuseppe Bondi-Zoccai, Tyler S. Brown, Caroline Der Nigoghossian, David A. Zidar, Jennifer Haythe, Daniel Brodie, Joshua A. Beckman, Ajay J. Kirtane, Gregg W. Stone, Harlan M. Krumholz and Sahil A. Parikh, 2020. ‘Cardiovascular considerations for patients, health care workers, and health systems during the coronavirus disease 2019 (COVID-19) pandemic’, Journal of th American College of Cardiology. Doi: 10.1016/j.jacc.2020.03.031.
Please note the use of the word “suspicion”: as of April 13th, 2020, there in no certainty COVID-19 causes heart damage. What we do have are elements that point to this. In any case, this question needs to be explored.
The extent of our ignorance
Back to our original questions, since some believe that the combination of hydroxychloroquine and azythromycin can help heal COVID-19, would not it saves time to administer this treatment to all patients without waiting for the end of the clinical trials? This is forgetting the extent of our ignorance.
Let’s remember that hydroxychloroquine can cause heart problems, including a QT prolongation, a disorder in the heart functioning that can lead to cardiac arrest. Considering the suspicion mentioned in the previous section, is it really reasonable to administer such a substance to patients without being fully aware of the possible consequences?
Considering that COVID-19 is suspected of causing cardiac damage and that among the known adverse effects of hydroxychloroquine, as well as azythromycin, there are cardiac complications, it is conceivable (again note that this is not a claim) that these actions combine. As I mentioned in the previous article, the action of hydroxychloroquine in the case of COVID-19, if any, is not elucidated. Therefore, we cannot exclude a possible deleterious effect. In such a case, having administered hydroxychloroquine will not have saved time, but may have led to a deterioration in the health of patients or even to a life-threatening condition.
On the one hand, the undesirable effects of hydroxychloroquine are indeed well known: these are proven risks. There is the suspicion of deleterious effects, these are potential risks. On the other hand, there is no beneficial action observed, the benefit is therefore hypothetical and if there is one, it is very likely to be small. Does such a balance between risk and benefit (I explain this notion in the previous article) really justify rushing the prescription of hydroxychloroquine?
Moreover, it should not be forgotten that hydroxychloroquine is also used against malaria, but also polyarthritis and lupus. These patients need this treatment, and the new enthusiasm for the substance is causing risks of shortages. Difficulties in supply for these patients are also among the risks to be taken into account.
Therefore, is it really reasonable, ethical, to rush the prescribing of this substance in the context of the COVID-19 pandemic?
Of the need for an error methodology
You have to keep in mind COVID-19 is an emerging disease. At the time of this writing, we are still at the beginning of the first COVID-19 pandemic. As a result, there are many things we do not know about it. This means that, necessarily, we are making mistakes. We must be aware of them and arm ourselves to detect them as best we can, in order to minimize the consequences. It is with this objective in mind that the clinical methodology has been set up and is constantly being questioned: it benefits from long experience and each new episode is an opportunity to improve it.
Indeed, methodology has its limits. However, it is much more effective in minimizing risk than relying on one’s own conviction, which is regularly a source of error.
The ethical issue
Since the beginning of this crisis, public debates face a recurring question: is it really ethical to conduct randomized, double-blind, placebo-controlled trials (I explain the principle of such trials in the previous article)? It looks like some patients participating in the test would not receive the best treatment. However, how do we know what the best treatment is?
In my previous article, I presented the basic principles of clinical trials. Of course, I could not give all the details, nor all the variations. In particular, a clinical trial is used to determine how efficient an active substance is. However, a patient clinical care is a much more a complex process than simply administering an active substance. This active substance is only one component of the clinical care. In a clinical trial, regardless of the group to which the patient belongs, he or she always receives this care: a patient participating in the trial is therefore never left without care. In the context of COVID-19, this care makes it possible to significantly improve the patient’s health, even if it is not always sufficient for recovery.
Moreover, if the basic methodology of clinical trials is the placebo-controlled test, when the aim is to highlight the contribution of a new molecule compared to standard care, it is not against a placebo that a double-blind trial is carried out, but against this standard care. This is how the Discovery trials that I presented in the previous article work. Since there is no evidence of efficacy of the molecules being tested, standard care is the best treatment available at the time the trials begin.
It seems medical ethic issues are brand new in public debates. Yet these are issues that health professionals have been facing for a long time. On the ethics of clinical trials, Benjamin Freedman wrote a seminal article3Benjamin Freedman, 1987. ‘Equipoise and the ethics of clinical research’, The New England Journal of Medicine 317 (3), pp. 141 – 145. Doi: 10.1056/NEJM198707163170304. Available on-line..
The first essential point is that medical knowledge must be considered collectively. Indeed, it is not a question of treating just one patient, but a population. Consequently, we cannot stop at the intimate conviction of one physician, the entire medical community must agree on what constitutes the best treatment for a given pathology, for a given population. This is in fact the standard in scientific research: as I regularly point out in this journal, research is a collective work, not a personal adventure.
For most of the diseases we face, there is already a well-established body of knowledge. In these cases, physicians draw on this body of knowledge to determine the best treatment. In contrast, in the case of an emerging disease, such as COVID-19, there is no such knowledge. Because of this absence, the medical community does not have any evidence to determine whether there is an effective treatment for this disease. Eventually, there may be indications that one or another molecule may have an effect, but nothing definitive. For such cases, Benjamin Freedman introduces the concept of “equipoise”. It means that all treatments carry equal importance because the state of general knowledge cannot clearly tell among the various options.
However, even in an equipoise situation, experience with other pathologies generally makes it possible to set up an initial care according to clinical signs. What is the subject of the equipoise is the various molecules that can be envisaged to improve this care. This is the case for COVID-19. One should keep in mind that, very often, one substance alone is not enough, but it is indeed the whole treatment that will help heal.
Of course, it is important, as much as possible, to leave this situation of equipoise, in order to take care of the various patients in the best possible way. Therefore, if a physician has acquired a conviction about a treatment, it is not only the situation of his or her patients that is at stake, but also that of all other patients, present and future. If he or she is right, this information is essential to the medical community. Consequently, ethics dictates him or her to do everything possible to bring to the attention of the medical community as a whole the evidence of the effectiveness of the treatment, over and above his or her simple conviction. In return, ethics commands the community to carefully get informed about the elements transmitted and to methodically search for errors. This work makes it possible to increase knowledge about the pathology and thus to improve its management.
A practical application
As mentioned in my previous article, chloroquine failed to show a real effect against COVID-19 simply through case studies. Therefore, protocols with a higher level of evidence must be used. These protocols are double-blind. The methodology is not a kind of formality: it is what minimizes the risk to patients. To take the current example, COVID-19 is a pathology that could cause cardiac risks. The methodology makes it possible to first clearly elucidate the action of the active substances, including those that may cause heart problems, before administering them on a massive scale.
Conversely, it is unethical to use resources and time to conduct a study with a methodology far too weak for it to have any chance of convincing anyone. In such a case, energy, resources, time, patients’ hopes, possibly patients’ suffering from adverse effects, were used in a study that did not increase knowledge. Moreover, because the pandemic is causing saturation of hospital services, all these resources are limited, even more so than usual. Everyone loses out on mobilizing them in non-significant studies rather than devoting them to following rigorous protocols.
To the reader
Ethical and methodological issues are complex, but have been studied for a long time, and we do have an extensive theoretical material about them. They are too important to pretend to decide them without first seeking information on their ins and outs.
Concerning the COVID-19 pandemic, it must be clear that I overestimate my understanding and you overestimate yours.
Really. It is not just a rhetorical formula. You really need to have an intimate awareness of this fact.
Pandemic management involves many disciplines, including virology, epidemiology, clinical methodology, statistics, logistics, psychology, and many others. One person alone cannot master all aspects of pandemic management. In order to manage such a crisis and to discuss how to deal with it, it is necessary on the one hand to become seriously informed about these issues, and on the other hand to coordinate with each other.
Above all, it is necessary to be able to change our certainties and to know how to rationally change our mind: knowledge about COVID-19 is evolving every day, we necessarily make mistakes. We must therefore be careful and adopt methods to detect our errors and minimize their consequences.
I would like to thank my fellows from Café des sciences for reviewing this article:
- Cyril Vidal, dental surgeon, Dans les dents;
- Marine Brenet, specialist in bioinformatics, GenetiX;
- Jean, specialist in cryosphere, I Need space;
- Erik Gustafsson, psychologist, Science pour parents;
- Sophie, Le Labo de Diti.
|↑1||Shaobo Shi, Mu Qin, Bo Shen, Yuli Cain, Tao Liu, Fan Yang, Wei Gong, Xu Liu, Jinjun Liang, Qinyan Zhao, He Huang, Bo Yang, and Congxin Huang, 2020. ‘Association of cardiac injury with mortality in hospitalized patients with COVID-19 in Wuhan, China’, JAMA Cardiology. Doi: 10.1001/jamacardio.2020.0950|
|↑2||Elissa Driggin, Mahesh V. Madhavan, Behnood Bikdeli, Taylor Chuich, Justin Laracy, Giuseppe Bondi-Zoccai, Tyler S. Brown, Caroline Der Nigoghossian, David A. Zidar, Jennifer Haythe, Daniel Brodie, Joshua A. Beckman, Ajay J. Kirtane, Gregg W. Stone, Harlan M. Krumholz and Sahil A. Parikh, 2020. ‘Cardiovascular considerations for patients, health care workers, and health systems during the coronavirus disease 2019 (COVID-19) pandemic’, Journal of th American College of Cardiology. Doi: 10.1016/j.jacc.2020.03.031|
|↑3||Benjamin Freedman, 1987. ‘Equipoise and the ethics of clinical research’, The New England Journal of Medicine 317 (3), pp. 141 – 145. Doi: 10.1056/NEJM198707163170304. Available on-line.|