Abstract
Introduction : Hydroxychloroquine (HCQ) alone or in combination
with azithromycin (AZ) is one of the many therapies being explored for
the treatment of Coronavirus 2019 (COVID-19). We performed a systematic
review regarding the effects of HCQ versus HCQ+AZ on corrected QT
interval (QTc) and cardiovascular outcomes.
Methods : We performed a systematic search, using PubMed,
EMBASE, SCOPUS, and Google Scholar from inception to May
3rd, 2020, with studies fulfilling the following
inclusion criteria: (1) compared HCQ versus HCQ+AZ in COVID-19; (2)
reported change in QTc interval and/or cardiovascular outcomes. The
primary outcome was change in QTc (maximum QTc–baseline QTc) and
incidence of TdP in COVID-19 patients on HCQ vs. HCQ+AZ.
Results: A statistically significant change in QTc interval was
observed with HCQ+AZ compared to HCQ alone (WMD 9.13 ms, 95%CI
3.74-14.01, p=0.01, I2 =29.04%). However, no
significant difference in the risk of development of
QTc>500 ms was observed between two groups (10.6% in HCQ
vs. 14.7% in HCQ+AZ, RR 0.71, 95% CI 0.32-1.59, p =0.40,I2 = 35.8%). Also, no significant difference
in risk of TdP was observed between the two groups (0% vs. 0.5%, risk
difference -0.002,95% CI-0.02 to 0.02, p=0.83,I2 =0%, respectively). However, one patient
experienced TdP, three days after discontinuation of HCQ+AZ for
prolonged QTc (499 ms).
Conclusion: The risk/benefit for HCQ and AZ should be carefully
contemplated in view of risk of QTc prolongation. Until further safety
data is available, we recommend close monitoring of QTc interval and
electrolytes, avoiding drug-drug interactions in these high-risk patient
populations.
Keywords : Hydroxychloroquine, Azithromycin, Torsades de
pointes, COVID-19
Novel Coronavirus 2019 (2019-nCoV, COVID-19), now declared a pandemic,
has infected 3.5 million people and killed 243,401 people worldwide1. Several pharmacologic approaches have been
recommended for the management of COVD-19 infected patients. One such
therapy is the use of hydroxychloroquine (HCQ) alone or in combination
with azithromycin (AZ). Hydroxychloroquine alone or in combination with
azithromycin has a potential for QT prolongation, torsades de pointes
(TdP), and sudden cardiac death (SCD). Therefore, to assess the
arrhythmic safety, we performed a systematic review regarding the
effects of hydroxychloroquine alone versus hydroxychloroquine plus
azithromycin on corrected QT interval (QTc) and cardiovascular outcomes
in COVID-19 patients.
We performed a systematic search, using PubMed, EMBASE, SCOPUS, and
Google Scholar from inception to May 3rd, 2020, with
studies fulfilling the following inclusion criteria: (1) compared HCQ
versus HCQ plus AZ in COVID-19; (2) reported change in QTc interval
and/or cardiovascular outcomes; (3) included human subjects; and (4)
studies in the English language. The data from included studies were
extracted using a standardized protocol and a data extraction form. Any
discrepancies were resolved with a consultation with the senior
investigator (JG). Two independent reviewers (SS and KS) extracted the
following data from the eligible studies: author name, study design,
publication year, follow-up duration, number of patients, age, gender,
diabetes mellitus (DM), hypertension (HTN), coronary artery disease
(CAD), sudden cardiac death, arrhythmias, QT interval, and outcomes. The
primary outcome was change in QTc (maximum QTc – baseline QTc) and
incidence of TdP in COVID-19 patients on HCQ vs. HCQ + AZ.
Mantel-Haenszel risk ratio (RR) random-effects model (DerSimonian and
Laird method) was used to summarize data between the
groups2. We used the Wan method to estimate the mean
and standard deviations 3. We then calculated the
pooled weighted mean difference (WMD) to evaluate the change in QTc
interval between the groups. A value of I2 of
0–25% represented insignificant heterogeneity, 26–50% represented
low heterogeneity, 51–75% represented moderate heterogeneity, and more
than 75% represented high heterogeneity4. A
two-tailed p < 0.05 was considered statistically significant.
Statistical analysis was performed using Comprehensive Meta-Analysis
version 3.0 (Biostat Solutions, Inc. [BSSI], Frederick, Maryland).
This systematic review of 4 studies [two retrospective5,6 and two prospective 7,8;
excluding case reports/series], incorporated a total of 448 patients.
The mean age was 60±13 years, and 58.7% were males. Hypertension was
the most common comorbidity (56.5%) followed by diabetes (31.3%),
atrial fibrillation (10.1%), coronary artery disease (CAD) (9.6%), and
heart failure (HF) (6.1%) (Table 1 ). A statistically
significant change in QTc interval was observed with HCQ + AZ compared
to HCQ alone (weighted mean difference 9.13 ms, 95% CI 3.74 - 14.01, p
= 0.01, I2 = 29.04%) (Figure 1A ).
However, no significant difference in the risk of development of QTc
>500 ms was observed between the two groups (10.6% in HCQ
vs. 14.7% in HCQ + AZ, RR 0.71, 95% CI 0.32-1.59, p = 0.40,I2 = 35.8%) (Figure 1B ). Also, no
significant difference in risk of TdP was observed in COVID 19 patients
treated alone or with combination drug therapy (0% vs 0.5%, risk
difference -0.002, 95% CI -0.02 to 0.02, p = 0.83,I2 = 0%, respectively) (Figure 1C).
Drug-induced QT prolongation exhibits 2-3 fold increased risk of TdP (a
potentially fatal ventricular arrhythmia); hence are typically avoided
if QTc >500 msec. The result of our pooled analysis
demonstrated no significant difference in the risk of QTc prolongation
>500 ms or TdP between the two groups for the treatment of
COVID-19 infection. Although QTc >500 ms was observed in
13% patients overall, TdP occurred in only one patient, three days
after discontinuation of HCQ+AZ for prolonged QTc (499 ms). This was
likely triggered by additional risk factors, including bradycardia,
propofol infusion, cardiomyopathy, and hypothermia6.
Also, non-sustained ventricular tachycardia (VT) was observed in 2.5%
of patients, while sustained VT occurred in one patient (in the setting
of viral myocarditis treated with HCQ)8.
Azithromycin has known proarrhythmic effect (regardless of QT
prolongation), which is typically dose-dependent, likely related to
increased intracellular sodium current, with resultant increased risk of
SCD. However, only 23% of patients completed the five-day course for
the treatment of COVID-19 infection 8. Moreover, the
prevalence of cardiovascular diseases (such as CAD and HF, which can
potentially increase the risk) was low in our pooled analysis. In
addition, the risk of drug-induced TdP is higher in critically ill
patients (likely due to increased prevalence of other comorbid
conditions – advanced age, underlying cardiac disease, electrolyte
imbalance, drug-drug interaction, and genetics (as evident in one
patient who had TdP despite drug discontinuation). Furthermore, patients
have an increased risk of ventricular arrhythmias from COVID-19 disease
itself (likely secondary to cardiomyopathy and myocarditis). These
factors alone or in combination, suggest the possibility of
underestimating the risk of TdP and SCD in our pooled analysis (also
evident by the fact that only 15.6% of patients in our study were
treated in an intensive care unit setting).
The risk of TdP may be higher in such vulnerable populations (critically
ill patients at risk for cardiomyopathy and myocarditis, electrolyte
derangements, drug-drug interactions, and those with history of
cardiovascular diseases), where extreme caution is warranted. Therefore,
risk/benefit for HCQ and AZ should be carefully contemplated. Until
further safety data is available, we recommend close monitoring of QTc
interval and electrolytes, avoiding drug-drug interactions in these
high-risk patient populations. There is a critical need for larger
randomized controlled studies to assess the risk/benefit profile of
these drugs in COVID-19 therapy.