key: cord-1030356-zu5mg2im
authors: Fiolet, Thibault; Guihur, Anthony; Rebeaud, Mathieu; Mulot, Matthieu; Peiffer-Smadja, Nathan; Mahamat-Saleh, Yahya
title: Effect of hydroxychloroquine with or without azithromycin on the mortality of COVID-19 patients: a systematic review and meta-analysis
date: 2020-08-26
journal: Clin Microbiol Infect
DOI: 10.1016/j.cmi.2020.08.022
sha: 5239c87ed5ae4edf712c54eee5a22a3baf16063b
doc_id: 1030356
cord_uid: zu5mg2im

BACKGROUND: Hydroxychloroquine or chloroquine with or without azithromycin have been widely promoted to treat COVID-19 following early in vitro antiviral effects against SARS-CoV-2 OBJECTIVE: The aim of this systematic review and meta-analysis was to assess whether chloroquine or hydroxychloroquine with or without azithromycin decreased COVID-19 mortality compared to the standard of care. DATA SOURCES: Pubmed, Web of Science, Embase Cochrane Library, Google Scholar and MedRxiv were searched until 25 July 2020. STUDY ELIGIBILITY CRITERIA: We included published and unpublished studies comparing the mortality rate between patients treated with chloroquine or hydroxychloroquine with or without azithromycin and patients managed with standard of care. PARTICIPANTS: Patients ≥18 years old with confirmed COVID-19. INTERVENTIONS: Chloroquine or hydroxychloroquine with or without azithromycin. METHODS: Effect sizes were pooled using a random-effects model. Multiple subgroup analyses were conducted to assess the drug safety. RESULTS: The initial search yielded 839 articles, of which 29 articles met our inclusion criteria. All studies except one were conducted on hospitalized patients and evaluated the effects of hydroxychloroquine with or without azithromycin. Among the 29 articles, 3 were randomized controlled trials (RCT), one was a non-randomized trial and 25 were observational studies, including 10 with a critical risk of bias and 15 with a serious or moderate risk of bias. After excluding studies with critical risk of bias, the meta-analysis included 11,932 participants for the hydroxychloroquine group, 8,081 for the hydroxychloroquine with azithromycin group and 12,930 for the control group. Hydroxychloroquine was not significantly associated with mortality: pooled Relative Risk RR=0.83 (95% CI: 0.65-1.06, n=17 studies) for all studies and RR=1.09 (95% CI: 0.97-1.24, n=3 studies) for RCTs. Hydroxychloroquine with azithromycin was associated with an increased mortality: RR=1.27 (95% CI: 1.04-1.54, n=7 studies). We found similar results with a Bayesian meta-analysis. CONCLUSION: Hydroxychloroquine alone was not associated with reduced mortality in hospitalized COVID-19 patients but the combination of hydroxychloroquine and azithromycin significantly increased mortality.

On December 31, 2019, the World Health Organization (WHO) identified an unknown pneumonia 114 caused by a new coronavirus, SARS-CoV-2, in Wuhan, China. By July 30, 2020, WHO confirmed 115 more than 17 million cases and 667,935 deaths [1] . Chloroquine (CQ) and its derivative 116 hydroxychloroquine (HCQ) were rapidly identified as potential drug candidates since CQ had an 117 antiviral activity against Middle East respiratory syndrome (MERS) and severe acute respiratory 118 syndrome (SARS) in vitro [2] . An in vitro antiviral activity of the aminoquinolines HCQ and CQ was 119 confirmed against SARS-CoV-2 and a study reported a synergistic effect of the HCQ with 120 azithromycin (AZI) against SARS-CoV-2 [3]. These drugs appeared as potential low-cost treatments 121

for COVID-19 patients [4] [5] [6] [7] and received wide and speculative coverage by the international press 122 and the United States President [8] . 123 124 Subsequently, HCQ and AZI were tested in a study where macaques were infected by SARS-CoV-2 125 and received either a high dose of HCQ (90 mg/kg on day 1 then 45 mg/kg) or a low HCQ dose (30 126 mg/kg on day 1 then 15 mg/kg) [9] . HCQ with or without AZI did not improve the time to viral 127 clearance regardless of the stage of disease: prophylaxis, early treatment or late treatment. between study variance τ². The Bayesian random-effect model assumes these parameters are random 207 with a probability distribution. Two prior distributions were tested µ~Normal (1,100) with a large 208 variance and τ ~Half-Cauchy (0,0.5) and a second scenario with µ~Normal (1,1) and τ ~Half-Cauchy 209 (0,0.5). The Bayesian analysis was conducted with the R package "brms" [34] . 210 211

Subgroup analyses were further conducted according to the quality assessment to explore the source of 213 heterogeneity among observational studies. We performed stratified analyses by type of article (peer-214 reviewed vs unpublished), use of an adjustment on confounding factors (studies with RR unadjusted vs 215 between the loading dose and the maintenance doses. Additionally, influence analysis was conducted 219 by omitting each study to find potential outliers [34] . Influence analysis is used to detect studies which 220 influence the overall estimate of a meta-analysis the most, omitting one study at a time (leave-one-out 221 method). 222 223 A two-sided p-value <0.05 was considered statistically significant. All analysis were conducted using 224 R version 3.6.1 with meta package and robvis package [35] . 225

A flow chart is presented in Figure 1 . After searching Pubmed, Cochrane Review and Web of Science, 228 839 articles were identified. After screening the title and the abstract, only 21 articles about 229 hydroxychloroquine and COVID-19 were included for further consideration. We excluded 564 articles 230 that did not meet the inclusion criteria. We did not find any non-English articles meeting our inclusion 231 criteria. Two duplicate studies on the same cohort were excluded [12, 36] . Two Chinese randomised 232 controlled trials on hydroxychloroquine reported zero deaths in both treatment and control groups 233

[37,38] and thus their results were not included in our meta-analysis. Ten articles from 234 Medrxiv/Google Scholar were added. Thus, 29 articles were included, of which 25 were observational 235 studies, one was an interventional non-randomized study and three were randomized controlled trials 236 (RCT). These studies included27 articles for HCQ [14] [15] [16] [17] [18] [19] 23 lowered by the lack of information about the assignment of treatment, the time between start of 267 follow-up and start of intervention), some unbalanced co-intervention with other antiviral and 268 antibiotic drugs and imbalance between groups for confounders such as comorbidities and age. (Table 1) . There was a significant subgroup difference between RCT and 276 non-randomized studies (P heterogeneity between = 0.03) with respectively RR RCT =1.09 (95%CI: 0.97-1.24) 277

and RR non-randomized = 0.79 (95%CI: 0.60-1.04) (Figure 2 ). Among observational studies with a moderate 278 risk of bias, we found no association between HCQ and mortality RR moderate bias =1.03 (95%CI: 0.91-279

1.17, I²=0%, n=7 studies) with no subgroup heterogeneity (Table S4, Figure S3 ). Results remained 280 nonsignificant with influence analysis ( Figure S4 ). The Bayesian meta-analysis led to similar results 281 with a pooled RR for mortality of 0.93 (95%CI: 0.72-1.14, n=17 studies) (Table S5, Figure S5 ). In 282 sensitivity analysis, after inclusion of studies with critical risk of bias, the global RR was marginally 283 not significant 0.80 (95%CI: 0.65-1.00) ( There was a significant higher heterogeneity among non-randomised studies as compared to RCT (I² 290 =84%, P heterogeneity within <0.01). In fact, heterogeneity was null for RCT. Egger's test (p= 0.68) and 291

Begg's test (P=0.13) were not significant for asymmetry of the funnel plot indicating that there was no 292 major publication bias for non-randomized studies ( Figure S6) . Hydroxychloroquine with azithromycin and mortality 296

After exclusion of studies with critical bias, the pooled RR for COVID-19 mortality was 1.27 (95%CI: 297 1.04-1.54, n=7) indicating an increased mortality linked to the use of hydroxychloroquine with 298 azithromycin. With a baseline hospital mortality of 26%, we identified a significant absolute risk 299 difference of +7%.

We found an increased risk of mortality in patients treated with 300 hydroxychloroquine and azithromycin compared to standard of care (RR: 1.29 (95%CI: 1.06-1.58, 301 n=6)) among non-randomized studies but this relationship was not found in the single Brazilian RCT, 302 with no heterogeneity observed across the study design (P heterogeneity between = 0.28) (Figure 3) . There was 303 a low heterogeneity across the included studies (I² =38%, p=0.14). Egger's test (p= 0.70) and Begg's 304 test (p=0.65) were not significant but the asymmetry in the funnel plot indicates that a publication bias 305 could be present ( Figure S7 ). However, the number of included studies was small. Subgroup analyses 306 are described in supplementary material (Table S4, Figure S8 ). The Bayesian meta-analysis led to 307 similar results with a pooled RR for mortality of 1.32 (95%CI: 0.97-1.68, n=7 studies) (Table S5 , 308 Figure S9 ). The increase in mortality was also significant with influence analysis ( Figure S10) . Our meta-analysis reported a high heterogeneity for hydroxychloroquine alone, but this heterogeneity 320 was lowered among RCT, studies with moderate risk of bias and for the association of HCQ+AZI. The 321 various quality of studies (not reporting HCQ dose, the lack of adjustment in reported estimates) may 322 explain one part of the heterogeneity observed according to our subgroup analysis (Table S4) . could not conduct subgroup analysis for severity since most of studies reports do not use the same 381 definition of severity and do not report the same biological and clinical outcomes. We also noted a 382 high level of heterogeneity in the administration of HCQ (dosing, timing between hospital 383 administration and intervention, duration…). In some studies, these data were not reported at all. 384

Another limitation comes from the studies which did not report adjusted effect size when mortality 385

was not the primary endpoint, leading to a high risk of confounding bias. As is usually done, this 

World Health Organization (WHO)

Middle East Respiratory 423 Syndrome and Severe Acute Respiratory Syndrome: Current Therapeutic Options and Potential 424 Targets for Novel Therapies

In vitro testing of 427 combined hydroxychloroquine and azithromycin on SARS-CoV-2 shows synergistic effect

Minimum costs to manufacture new treatments for 430 COVID-19

Optimized Dosing Design of Hydroxychloroquine for the Treatment of Severe Acute Respiratory 433

SARS-CoV-2)

Hydroxychloroquine, a less toxic derivative of 435 chloroquine, is effective in inhibiting SARS-CoV-2 infection in vitro

Hydroxychloroquine in the treatment and prophylaxis of SARS-CoV-2 infection in 438 non-human primates 2020

The Risks of Prescribing Hydroxychloroquine for Treatment of COVID-440 19-First, Do No Harm

Hydroxychloroquine use 443 against SARS-CoV-2 infection in non-human primates

446 Review of trials currently testing treatment and prevention of COVID-19

Anticovid, a 449 comprehensive open-access real-time platform of registered clinical studies for COVID-19. J 450 Antimicrob Chemother n

Clinical impact of 452 COVID-19 on patients with cancer (CCC19): a cohort study

Clinical efficacy of 455 hydroxychloroquine in patients with covid-19 pneumonia who require oxygen: observational 456 comparative study using routine care data

Low dose of hydroxychloroquine reduces fatality 458 of critically ill patients with

Observational Study of 461 Hydroxychloroquine in Hospitalized Patients with Covid-19

Hydroxychloroquine Is Associated with an Increase of Survival in COVID-19 Patients: An 465 Observational Study 2020

The association between treatment with heparin and survival in 467 patients with Covid-19

Treatment With Hydroxychloroquine or Azithromycin With In-Hospital Mortality in Patients 471 With COVID-19 in

Hydroxychloroquine in patients with COVID-19: A 473 Systematic Review and meta-analysis

Virological and clinical 476 cure in COVID-19 patients treated with hydroxychloroquine: A systematic review and meta-477 analysis

Does Adding of Hydroxychloroquine to 479 the Standard Care Provide any Benefit in Reducing the Mortality among COVID-19 Patients?: a 480 Systematic Review

Cochrane Handbook for Systematic Reviews of Interventions. Part 3: Special 483 topics. 13.6.2.2 Combining studies (version 6)

Hydroxychloroquine in Hospitalized Patients with COVID-19: a Quasi-Randomized Comparative 487 Study

Hydroxychloroquine and 492 azithromycin as a treatment of COVID-19: results of an open-label non-randomized clinical trial

How to perform a meta-analysis with R: a practical tutorial

ROBINS-I: a 498 tool for assessing risk of bias in non-randomised studies of interventions

RoB 2: a revised tool 501 for assessing risk of bias in randomised trials

Meta-analysis in clinical trials

Features of 20 133 506 UK patients in hospital with covid-19 using the ISARIC WHO Clinical Characterisation Protocol: 507 prospective observational cohort study

Quantifying heterogeneity in a meta-analysis

Operating characteristics of a rank correlation test for publication bias

Targeting endosomal acidification by chloroquine analogs as a promising 513 strategy for the treatment of emerging viral diseases

A re-evaluation of random-effects meta-analysis

Outlier and influence diagnostics for meta-analysis

Risk-of-bias VISualization (robvis): An R package and Shiny web app 520 for visualizing risk-of-bias assessments

Mortality in an US Urban Healthcare System

Hydroxychloroquine in patients with 526 mainly mild to moderate coronavirus disease 2019: open label, randomised controlled trial

A pilot study of hydroxychloroquine in treatment of patients with 529 moderate COVID-19

Hydroxychloroquine in Nonhospitalized Adults With Early COVID-19. Annals of Internal 533 Medicine 2020

Hydroxychloroquine in Hospitalized Patients with COVID-19: Preliminary results from a multi-536 centre, randomized

Outcomes of 539 hydroxychloroquine usage in United States veterans hospitalized with COVID-19

Tocilizumab Therapy in COVID-19 Patients -An Observational Study

Force on the clinical characteristics and short-term outcome of hemodialysis 546 patients with SARS-CoV-2 infection

Report of the COVID-19 550 Registry of the Spanish Society of Nephrology (SEN). Nefrología (English Edition) 2020

RECOVERY trial: the UK covid-19 study resetting expectations for clinical trials

Compassionate use of 555 hydroxychloroquine in clinical practice for patients with mild to severe Covid-19 in a French 556 university hospital

Utilization of COVID-558 19 treatments and clinical outcomes among patients with cancer: A COVID-19 and Cancer 559

Consortium (CCC19) cohort study

Hydroxychloroquine with 562 or without azithromycin and in-hospital mortality or discharge in patients hospitalized for 563 COVID-19 infection: a cohort study of 4,642 in-patients in France

COVID-19 in patients with 566 lung cancer

Treatment with 568 hydroxychloroquine, azithromycin, and combination in patients hospitalized with COVID-19

Hydroxychloroquine with or without Azithromycin in Mild-to-Moderate Covid-19

COVID-19 and Kidney 575 Transplantation: Results from the TANGO International Transplant Consortium

Factors Associated With 578 Death in Critically Ill Patients With Coronavirus Disease 2019 in the US

Comparison of 581 hydroxychloroquine, lopinavir/ritonavir, and standard of care in critically ill patients with SARS-582

CoV-2 pneumonia: an opportunistic retrospective analysis

SARS-CoV-2 infection in dialysis 585 patients in northern Italy: a single-centre experience

Risk Factors for 588 Mortality in Patients with COVID-19 in New York City

Covid-19 transmission, outcome and associated risk factors in cancer patients at the first 592 month of the pandemic in a Spanish hospital in Madrid

Outcomes of Hydroxychloroquine Treatment 595

Among Hospitalized COVID-19 Patients in the United States-Real-World Evidence From a 596

Federated Electronic Medical Record Network

Outcomes of 599 3,737 COVID-19 patients treated with hydroxychloroquine/azithromycin and other regimens in 600

A retrospective analysis

ADL-603 dependency, D-Dimers, LDH and absence of anticoagulation are independently associated with 604 one-month mortality in older inpatients with Covid-19

Effect of High vs Low 607 Doses of Chloroquine Diphosphate as Adjunctive Therapy for Patients Hospitalized With Severe 608

Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Infection: A Randomized Clinical Trial. 609

The Effect of Chloroquine

Hydroxychloroquine and Azithromycin on the Corrected QT Interval in Patients with SARS-CoV-613

Preliminary evidence from a multicenter 615 prospective observational study of the safety and efficacy of chloroquine for the treatment of 616 COVID-19

Clinical 618 characteristics and outcomes of the first 63 adult patients hospitalized with COVID-19: An 619 experience from Oman

COVID-19 pneumonia in a kidney 622 transplant recipient successfully treated with tocilizumab and hydroxychloroquine

Hydroxychloroquine is 625 associated with slower viral clearance in clinical COVID-19 patients with mild to moderate 626 disease: A retrospective study

Hydroxychloroquine or

Chloroquine for Treatment or Prophylaxis of COVID-19: A Living Systematic Review

Safety of 632 hydroxychloroquine, alone and in combination with azithromycin, in light of rapid wide-spread 633 use for COVID-19: a multinational, network cohort and self-controlled case series study. 634 Rheumatology

Rating the quality of evidence-study limitations (risk of bias)

Assessment of QT 639 Intervals in a Case Series of Patients With Coronavirus Disease

Treated With Hydroxychloroquine Alone or in Combination With Azithromycin in an Intensive 641 Care Unit

Assessment of QT 643 Intervals in a Case Series of Patients With Coronavirus Disease

Treated With Hydroxychloroquine Alone or in Combination With Azithromycin in an Intensive 645 Care Unit

QT Interval 647 Prolongation and Torsade De Pointes in Patients with COVID-19 treated with 648

Risk of QT Interval 651 Prolongation Associated With Use of Hydroxychloroquine With or Without Concomitant 652

Azithromycin Among Hospitalized Patients Testing Positive for Coronavirus Disease 2019 653 (COVID-19)

Toxicities Associated With Hydroxychloroquine and Azithromycin: An Analysis of the World 656 Health Organization Pharmacovigilance Database

COVID-19) Update: FDA Revokes Emergency Use 659 Authorization for Chloroquine and Hydroxychloroquine

COVID-19 : 663 l'ANSM souhaite suspendre par précaution les essais cliniques évaluant l'hydroxychloroquine 664 dans la prise en charge des patients -Point d'Information -ANSM : Agence nationale de 665 sécurité du médicament et des produits de santé n

COVID-19: reminder of risk serious side effects with chloroquine and 670 hydroxychloroquine

Hydroxychloroquine prophylaxis for COVID-19 contacts in 674

The Lancet Infectious Diseases

COVID-19 coronavirus 676 research has overall low methodological quality thus far: case in point for 677 chloroquine/hydroxychloroquine

NIH halts clinical trial of hydroxychloroquine

Solidarity" clinical trial for COVID-19 treatments n

Does the inclusion of grey literature influence 686 estimates of intervention effectiveness reported in meta-analyses?

Grey literature in meta-analyses of randomized 689 trials of health care interventions. Cochrane Database of Systematic Reviews

Hydroxychloroquine plus Azithromycin | Coronavirus Disease COVID-19

Discovery stopping inclusions in two treatment groups