Current best evidence for clinical care (more info)
Hydroxychloroquine has been promoted for its use in treatment of COVID-19 patients based on in-vitro evidences. We searched the databases to include randomized and observational studies evaluating the effect of Hydroxychloroquine on mortality in COVID-19 patients. The outcome was summarized as odds ratios (OR) with a 95% confidence interval (CI).We used the inverse-variance method with a random effect model and assessed the heterogeneity using I2 test. We used ROBINS-I tool to assess methodological quality of the included studies. We performed the meta-analysis using 'Review manager software version 5.3'. We identified 6 observationalstudies satisfying the selection criteria. In all studies, Hydroxychloroquine was given as add on to the standard care and effect was compared with the standard care alone. A pooled analysis observed 251 deaths in 1331 participants of the Hydroxychloroquine arm and 363 deaths in 1577 participants of the control arm. There was no difference in odds of mortality events amongst Hydroxychloroquine and supportive care arm [1.25 (95% CI: 0.65, 2.38); I2 = 80%]. A similar trend was observed with moderate risk of bias studies [0.95 (95% CI: 0.44, 2.06); I2 = 85%]. The odds of mortality were significantly higher in patients treated with Hydroxychloroquine + Azithromycin than supportive care alone [2.34 (95% CI: 1.63, 3.34); I2 = 0%]. A pooled analysis of recently published studies suggests no additional benefit for reducing mortality in COVID-19 patients when Hydroxychloroquine is given as add-on to the standard care. Graphical Abstract.
| Discipline / Specialty Area | Score |
|---|---|
| Infectious Disease | |
| Family Medicine (FM)/General Practice (GP) | |
| General Internal Medicine-Primary Care(US) | |
| Public Health | |
| Emergency Medicine | |
| Intensivist/Critical Care | |
This is a fatally flawed meta-analysis that pools retrospective cohort studies with controlled clinical trials. It is meaningless. Moreover, new trials are emerging weekly that are methodologically stronger and actionable (many HQ trials have been paused or terminated).
Given recent anecdotal data that has driven the adoption of chloroquine compounds for COVID19 management, this meta-analysis is timely and informative. We really need to stop adopting unproven therapies in the absence of solid data. At a minimum, if we are going to use therapies as "salvage" then RCTs should be performed in parallel.
As a clinician, I believe this meta-analysis shows that hydroxychloroquine has no real value, not by itself, or by adding it to Azythromycin.