key: cord-0824311-ulkty104
authors: Vouri, Scott M.; Thai, Thuy N.; Winterstein, Almut G.
title: An evaluation of co-use of chloroquine or hydroxychloroquine plus azithromycin on cardiac outcomes: A pharmacoepidemiological study to inform use during the COVID19 pandemic
date: 2020-04-30
journal: Res Social Adm Pharm
DOI: 10.1016/j.sapharm.2020.04.031
sha: 8454ad60dde8174003dde3c9aa2a925a234bd868
doc_id: 824311
cord_uid: ulkty104

Abstract Background Chloroquine or hydroxychloroquine (chloroquine) plus azithromycin is considered as therapy for COVID19. With benefit evaluations underway, safety concerns due to potential additive effects on QTc prolongation should be addressed. Objective We compared risk of cardiac adverse events between combinations of chloroquine and azithromycin and chloroquine and amoxicillin. Methods We conducted a retrospective cohort study using the IBM MarketScan Commercial Claims and Medicare Supplemental Databases, 2005–2018. We included autoimmune disease patients aged ≥18 years initiating azithromycin or amoxicillin for ≥5 days during chloroquine treatment. Patients had continuous insurance coverage ≥6 months before combination use until 5 days thereafter or inpatient death. Two outcomes were sudden cardiac arrest/ventricular arrhythmias (SCA/VA) and cardiac symptoms. We followed patients for up to 5 days to estimate hazard ratios (HR). Covariates were adjusted using stabilized inverse probability treatment weighting. Results We identified two SVC/VA events among >145,000 combination users. The adjusted incidence of cardiac symptoms among azithromycin and amoxicillin users was 276 vs 254 per 10,000 person-years with an adjusted HR of 1.10 (95%CI, 0.62–1.95). Conclusion Combination use of chloroquine and azithromycin at routine doses did not show pronounced increases in arrhythmias in this real-world population, though small sample size and outcome rates limit conclusions.

using stabilized inverse probability treatment weighting. We conducted two sensitivity analyses 75 that restricted to 1) patients with rheumatoid arthritis or lupus and 2) only the first 76 concomitant use episode. 77 78

There were 69,743 and 72,163 episodes with chloroquine or hydroxychloroquine and 80 azithromycin combination and chloroquine or hydroxychloroquine and amoxicillin combination, 81 respectively (see Table 1 for detailed baseline characteristics). We identified one SVC/VA event 82 per exposure group. There were 29 azithromycin users and 23 amoxicillin combination users 83 with cardiac symptoms. After adjustment, the incidence (per 10,000 person-years) of cardiac 84 symptoms among azithromycin and amoxicillin combination users was 276 (95%CI, 185-410) 85 versus 254 (95%CI, 168-383) with an adjusted HR of 1.10 (95%CI, 0.62-1.95) ( Table 2 ). The 86 sensitivity analyses showed consistent results (Table 2 ) as did stratified analyses of the two data 87 sources, though confidence intervals remained wide due to small event rates (Commercial 88 plans versus Medicare supplemental insurance, Table 3) . 89 90

The risk of SCA/VA was rare in our analysis and yielded inconclusive results. Although the 92 incidence of cardiac symptoms among patients who used chloroquine or hydroxychloroquine in 93 combination with azithromycin was slightly higher than among our active comparison group 94 who used combinations with amoxicillin, results were not statistically significant in our primary 95 or sensitivity analyses. 96

While the low event rates appear encouraging, use of choloroquine and azithromycin 98 combination should still be cautious, especially given other emerging evidence, for a variety of 99 reasons. One recent retrospective cohort study among COVID-19 patients in U.S. Veterans 100

Health Adminstration medical centers found there was no benefits of chlorquine in the 101 reduction of need for mechanical ventilation; however, there was a significantly increased risk 102 of death among chloroquine users as compared to no chloroquine users. 12 Several trials were 103 stopped early due to QTc prolongation related fatalities associated with the use of 104 chloroquine. 4,13 105 106 While controlled evidence on the risk of the combination use is lacking, a recent analysis of the 107 U.S. Food and Drug Administration's Adverse Event Reporting System (FAERS) data found a 108 signal for QTc prolongation or Torsades de Pointes associated with azithromycin. The analysis 109 found no signal associated with chloroquine or hydroxychloroquine when used alone and a 110 weaker association that did not reach thresholds for a safety signal for the combination with 111 azithromycin. 14 Interestingly, although underreporting of events as well as biases in reporting 112 may obscure causal associations in adverse event reporting data, concerns regarding 113 azithromycin's role in severe arrhythmias are corroborated by previous observational studies. A 114 study in Tennessee Medicaid beneficiaries found that azithromycin users had a 2.5 times higher 115 risk of cardiovascular death (95%CI, 1.4-4.5) and a 2.0 times higher risk of death from any cause 116 compared to amoxicillin users (95%CI, 1.2-3.3). 15 In another case-control study using 117 electrocardiogram results and electronic health records, azithromycin users had 43% increased odds of severe QTc prolongation compared to amoxicillin users (95%CI, 1.13-1.82). 16 In 119 contrast, evidence for adverse cardiac adverse events of chloroquine and hydroxychloroquine 120 consists mostly of case series, and the evidence based on more rigorous observational studies is 121

It should be noted that our findings may not be directly applicable to how chloroquine or 124 hydroxychloroquine are being used in combination with azithromycin. Chronic administration 125 among patients with autoimmune disease cannot replicate physiological responses when both 126 drugs are used acutely in a hospitalized setting, as would be expected for COVID-19 patients. 127

Moreover, doses used in routine care (e.g. for lupus erythematosus, adult dosage is only 125 to 128 250 mg chloroquine daily), appear to be lower compared to COVID-19 treatment (1000 mg 129 chloroquine phosphate for day 1 and then 500 mg daily for four to seven days of total 130 treatment), 6 which may suggest an attenuated risk for cardiac adverse outcomes. 18 131 Additionally, our study was under the guidance of a prescribing clinician and cannot replicate 132 scenarios where patients self-medicate. The desire to benefit from potential prophylactic 133 effects has recently claimed one death when a patient used chloroquine available to clean fish 134

There are a several limitations to mention. First, while we addressed confounding via restriction 137 and statistical adjustment, baseline characteristics suggest that chloroquine users with cardiac 138 history were channeled away from azithromycin, and residual confounding may have masked 139 subtle effects. For example, patients with higher risk for cardiac adverse events may be intentionally prescribed other antibiotics when azithromycin would be on option, thus 141 mitigating the effect. Second, we may have underestimated the actual risk of cardiac adverse 142 events among new users of the chloroquine and azithromycin combination, because our 143 population included patients with long-term chloroquine use who may have been tolerating the 144 drug well. Ideally, patients who initiate both drugs simultaneously, as proposed for COVID-19 145 treatment, would be studied, but restriction our analysis to such a population was prohibitive in 146 terms of sample size. Third, restriction to patients with autoimmune disorders aimed to 147 exclude chloroquine use for malaria prophylaxis with unknown exposure period and incomplete 148 capture of outcomes during travel, but indications can only be inferred from claims data. 149

Fourth, in an attempt to maximize sample size, we expanded a validated ICD-9-CM code set for 150 SVD/VA 11 to ICD-10-CM codes, which may have missed or mis-specified events after 2015. We 151 tested our crosswalk in our source dataset and found consistent incidence rates of all endpoints 152 across the ICD transition period. While we have found our code set for cardiac symptoms to be 153 sensitive to capture effects of QTc prolongation in previous studies, 10 We conclude that combination use of chloroquine and azithromycin did not show pronounced 158 increases in arrhythmias in this real-world population. We caution however, against 159 encouraging use of this combination for COVID-19, and in particular self-medication, especially 160 among patients with history or risk of QTc prolongation, until appropriate risk-benefit has been 161 established. 162 

Safety signals for QT prolongation or Torsades de 217 Pointes associated with azithromycin with or without chloroquine or hydroxychloroquine 218

Azithromycin and the risk of 221 cardiovascular death

Risk Evaluation of Azithromycin-Induced QT 223 Prolongation in Real-World Practice

Cardiac Complications Attributed to 226 Chloroquine and Hydroxychloroquine: A Systematic Review of the Literature

FDA Letter to Stakeholders: Do Not Use Chloroquine 231 Phosphate Intended for Fish as Treatment for COVID-19 in Humans

Funding source: None 170

• Chloroquine/hydroxychloroquine plus azithromycin is a potential therapy for COVID19. 3• Each of these drugs is known with an increased risk of QTc prolongation. 4• This combination may have potential additive effects on QTc prolongation. 5• This combination did not show pronounced increases in arrhythmias in our study. 6• We caution against using this combination for COVID-19 until further safety evidence. 7