Hydroxychloroquine for COVID-19: Panacea or Wishful Thinking?

Hydroxychloroquine belongs to the quinolone family of medicines. Approved FDA uses include treating certain forms of malaria, lupus, and rheumatoid arthritis. Off-label uses include dermatomyositis, porphyria cutanea tarda, Sjogren, sarcoidosis, and Q fever.

“The evidence you’re talking about … is anecdotal evidence” – Dr. Anthony Fauci, coronavirus task force”

Cortegiani et al published (3/5/2020) a systematic review examining the role of chloroquine in treating COVID-19. This review found six articles from France, China, Italy, and the Netherlands, however, these articles were only “expert consensus” and lacked any clinical data. Furthermore, they identified a need for more coordinated research effort after finding 23 trials from China all simultaneously investigating the role of chloroquine for COVID-19.

Gautret et al published (3/17/2020) a non-randomized study of 26 patients PCR+ patients in France who were treated with hydroxychloroquine. The primary endpoint, the rate of virological clearance at day 6, was statistically lower in the hydroxychloroquine group compared to 16 controls. Six of the patients in the hydroxychloroquine group also received azithromycin and all were found to be PCR negative by day six.

This study has significant limitations. Six of the patients in the hydroxychloroquine group were excluded including three who were transferred to the ICU and one patient who died. Thus, failure to perform an intention-to-treat analysis and exclusion of the sicker patients likely biased the results in favor of the hydroxychloroquine group. Additionally, while the authors should be commended for their attempts to complete a study during this pandemic, the study was underpowered and fell short of their target population of 48 patients. Therefore, in addition to the limited utility of non-randomized and non-blinded studies, there is a significant risk of type 1 error (rejecting the null hypothesis when it is true, or a false positive) due to the very limited sample size. For example, it is very difficult to draw meaningful conclusions from a sample size of five patients (combined azithromycin and hydroxychloroquine) when just one additional patient may have significantly altered the results.

While decreasing the spread of the virus may be helpful in an outpatient setting, this is less applicable in an inpatient setting where patients (ideally) are isolated. Patient-centered outcomes including mortality, length of stay, ventilator days as well as the rate of side effects (QT prolongation, etc) are arguably more important outcomes. These were not reported in this study and, therefore, the implications of this paper are limited at best. 

In conclusion, this study may suggest a signal of benefit, but clinical implications are significantly limited due to the substantial risk of bias and limited design of the study. It is therefore difficult and potentially dangerous to accept the authors’ conclusion: “[they] recommend that COVID-19 patients be treated with hydroxychloroquine and azithromycin to cure their infection.” Larger studies, focused on patient-centered outcomes, must be performed to further investigate this possible intervention.

Gautret et al then published (3/29/2020) a follow up which included 80 patients. All of the patients received hydroxychloroquine and 16/80 received combination therapy with azithromycin. Unfortunately, there is no control group, so no benefit can be scientifically proven. Of note, the majority of the patients enrolled in the study appear to be a healthy population as 15% required oxygen, four were asymptomatic, and 43% of the population had no chronic conditions (such as diabetes, hypertension, chronic respiratory disease). (In our hospital in New York City, we have severe bed shortages and we are generally only admitting the sickest population, who typically require oxygen therapy.) Thus, this study may have limited external validity, and we should be cautious with applying results from a single-center to our own population.

An argument for proponents of the experimental use of hydroxychloroquine is that it is safe. For the most part, this is true. Chloroquine has been in use for over 70 years and hydroxychloroquine is included on the World Health Organization (WHO) list of essential medicines.

In light of the recent two trillion dollar bailout, it also is worth noting that hydroxychloroquine is cheap, costing roughly $25 for a month’s supply.

Despite generally having a good safety profile, hydroxychloroquine has known side effects. Most notably, the prolongation of QTc can cause life-threatening arrhythmias, especially when used in combination with other drugs suspected to prolong QTc (azithromycin, ondansetron, etc). The need for closer monitoring when using experimental therapies may be jeopardized in times of pandemic, when safe staffing and appropriate monitoring may be of limited supply.

Furthermore, if adopted, the widespread use of the medications for COVID-19 may result in drug shortages, especially for those with diseases for which there are proven benefits. For example, patients with lupus may have difficulty obtaining their prescribed hydroxychloroquine.

Lastly, here is a reminder not to ingest chemicals used to clean aquariums.

The entire world is watching us (well, hopefully they’re on their couches watching Tiger King). Is this the time to abandon science and critical thinking? I hope not. Instead, let’s do what we’ve always done. Follow the evidence. For now, the evidence is weak. Hopefully, we will have better data soon. However, if you do feel inclined to offer it to a patient, use shared decision-making and inform them of the risks.