乌鸦传媒

As a proposed treatment for COVID-19 pneumonia, hydroxychloroquine (HCQ, Plaquenil) on its own may not be expected to lead to poor cardiovascular outcomes, but serious adverse events emerged when rheumatoid arthritis (RA) patients took azithromycin at the same time, according to an observational study.

Short-term use of HCQ among adult RA patients was associated with a similar 30-day safety profile as that for sulfasalazine (SSZ), according to Patrick Ryan, PhD, of Janssen Research and Development in Titusville, New Jersey, and colleagues of the OHDSI-COVID-19 consortium. The consortium pooled together claims databases and electronic health records dating back to 2000 from Germany, Japan, the Netherlands, Spain, the U.K., and the U.S.

But patients who took both HCQ and azithromycin for short periods fared significantly worse at 30 days than peers who were on HCQ-amoxicillin. Study authors showed that the HCQ-azithromycin combination was associated with increased risk of:

• Cardiovascular mortality: calibrated HR 2.19 (95% CI 1.22-3.94)
• Chest pain/angina: calibrated HR 1.15 (95% CI 1.05-1.26)
• Heart failure: calibrated HR 1.22 (95% CI 1.02-1.45)

A caveat on the study: the paper was on medRxiv, so it has not been peer reviewed.

"Short-term HCQ treatment is safe, but addition of [azithromycin] may induce heart failure and cardiovascular mortality, potentially due to synergistic effects on QT length. We call for caution if such combination is to be used in the management of COVID-19," Ryan's group concluded.

HCQ is an antimalarial drug and a mainstay of therapy for systemic lupus erythematosus. It recently received Emergency Use Authorization by the FDA for use to treat COVID-19 patients. President Trump and others have touted the potential benefits of HCQ-azithromycin therapy since March.

For weeks, however, critics have warned of the potential heart risks of HCQ taken alone or in conjunction with azithromycin, typically citing the QT segment prolongation that could lead to arrhythmia and cardiovascular death.

"Regarding safety, we know from the last 70 years of using HCQ in rheumatology, that the drug has a good safety profile. The findings of the study that addition of azithromycin adds toxicity is important and suggests that caution is warranted," agreed Jinoos Yazdany, MD, MPH, of the University of California San Francisco and Zuckerberg San Francisco General Hospital.

"There is still no evidence that HCQ works for the treatment or prevention of COVID-19," Yazdany told 乌鸦传媒. "It is important for the medical community to maintain equipoise and realize that our inferences are extremely limited from currently available uncontrolled or poorly controlled studies."

Some have resorted to hoarding HCQ in response to the coronavirus crisis, leaving patients who use the medication for other indications scrambling to maintain their own supplies.

"As a rheumatologist, I am distraught by the inability for my patients to access HCQ for conditions where they are absolutely known to be effective; for my patients skipping HCQ may mean a major disease flare in the midst of a pandemic," said Katherine Liao, MD, of Brigham and Women's Hospital in Boston.

Ryan and colleagues noted that they had chosen SSZ and amoxicillin as active comparators in the study because they have similar indications as HCQ monotherapy and combination therapy.

People included in the study were new users of HCQ (n=956,374), SSZ (n=310,350), HCQ-azithromycin (n=323,122), and HCQ-amoxicillin (n=351,956).

The investigators achieved balanced comparison groups after propensity score stratification.

Nevertheless, this kind of observational study is inherently subject to confounding, such as confounding by indication, according to Liao.

"In this case, neither HCQ or SSZ are first line therapies for RA," she explained. "They are frequently used in combination with other disease-modifying anti-rheumatic drugs. When they are used as monotherapy, there is usually a reason and typically different reasons for using HCQ versus SSZ. For example, either they have mild RA or they have contraindications to other therapies."

"It is also important to remember that extrapolation of the findings of these claims-based analyses to the COVID-19 pandemic is limited. The use of antimalarials at higher doses and in critically ill patients may worsen their toxicity profile, and that is a question that this study did not address," according to Yazdany.

Assessing HCQ's efficacy in COVID-19 will require randomized trials, but large real-world registries, such as the present one, are still valuable because they have the advantage of looking at low-frequency signals such as sudden cardiac death, commented C. Michael Gibson, MD, of Beth Israel Medical Center in Boston.

"If anyone attempted to do such a project even 10 years ago, it would have been difficult to have this number of participants in a study in this amount of time. I laud this multi-national team for pooling their resources and performing this distributed analysis," said Liao.

Comment