To the Editor — The SARS-CoV-2 pandemic has caused more than 1.6 million positive cases and more than 95,000 confirmed deaths as of 10 April 2020 (ref. 1). Although there are no approved drugs to prevent or treat SARS-CoV-2 infection2, a recent report suggested that the combination of hydroxychloroquine and azithromycin (HY/AZ) may have a favorable effect on the clinical outcomes and viral loads of infected patients3; this resulted in massive adoption of the regimen by clinicians worldwide. However, both medications have been independently shown to increase the risk in other populations for QT-interval prolongation, drug-induced torsades de pointes (a form of polymorphic ventricular tachycardia) and drug-induced sudden cardiac death4,5,6. In our center, patients with the respiratory syndrome COVID-19 who are admitted for lower airway disease with features such as non-resolving cough, chest infiltrates on X-ray and persistent fever, with or without blood-oxygen desaturation, are treated with HY/AZ. We reviewed the charts and followed the corrected QT (QTc) interval in a consecutive cohort of 84 patients receiving the regimen. HY and AZ were administered orally for 5 days. HY was given at a dose of 400 mg twice daily on the first day, followed by 200 mg twice daily. AZ was given at a dose of 500 mg per day. The average time of electrocardiograph (ECG) follow-up after HY/AZ exposure was 4.3 ± 1.7 days.

We observed prolongation of the QTc from a baseline average of 435 ± 24 ms (mean ± s.d.) to a maximal average value of 463 ± 32 ms (P < 0.001 (one-sample t-test)), which occurred on day 3.6 ± 1.6 of therapy (Fig. 1). In a subset of nine (11%) of those patients, the QTc was severely prolonged to >500 ms, a known marker of high risk of malignant arrhythmia and sudden cardiac death7. In this high-risk group, the QTc increased from a baseline average of 447 ± 30 ms to 527 ± 17 ms (P < 0.01 (one-sample t-test)). There were no torsades de pointes events recorded for any patients, including those with a severely prolonged QTc. Four patients died from multi-organ failure, without evidence of arrhythmia and without severe QTc prolongation. 64 patients remained admitted and 16 patients were discharged. The clinical and epidemiological characteristics are presented in Supplementary Table 1.

Fig. 1: Changes in QTc on HY/AZ therapy.
figure 1

a, Change in QTc, presented as days after HY/AZ initiation. *P < 0.01, QTc compared with baseline QTc (one-sample t-test to compare each sample against a change in QTc (ΔQTc) of 0 ms (i.e., no change from baseline), with adjustment for multiple testing). Each data point represents a single patient with a single ECG at any given interval (n). STD, standard deviation. b, Frequency of patients with various ranges of QTc prolongation (horizontal axis). Five cardiologists trained and experienced in QT measurement performed all ECG measurements. QT and RR measurements were validated by a senior cardiac electrophysiologist expert in QT measurements. QTc was corrected with the Bazett formula (QTc = QT/RR1/2).

The effectiveness of HZ/AZ in treating SARS-CoV-2 infection has been demonstrated in one small human study so far2. Previously, the combination of HY/AZ resulted in mild QTc prolongation when given to young healthy volunteers8. In our work, we found that in patients with COVID-19 who were treated with HY/AZ, the QTc was significantly prolonged. This discrepancy suggests that QT prolongation may be influenced by patient attributes such as the presence of co-morbidities and the severity of the disease9. Of note, recent guidance suggested ECG screening with QTc assessment for patients with COVID-19 who are candidates for novel therapies, including HY/AZ10. In our cohort, five of nine patients with severe QTc prolongation had a normal QTc at baseline. We therefore suggest that the QTc should be followed repeatedly in patients with COVID-19 who are treated with HY/AZ, particularly in those with co-morbidities and in those who are treated with other QT-prolonging medications.

Ethics declaration

The study was performed according to our Institutional Review Board guidance in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments, with a waiver of informed consent for chart review.