QTc Interval Prolongation and Life-Threatening Arrhythmias During Hospitalization in Patients With Coronavirus Disease 2019 (COVID-19): Results From a Multicenter Prospective Registry

Abstract

Background

Prolonged QTc intervals and life-threatening arrhythmias (LTA) are potential drug-induced complications previously reported with antimalarials, antivirals, and antibiotics. Our objective was to evaluate the prevalence and predictors of QTc interval prolongation and incidences of LTA during hospitalization for coronavirus disease 2019 (COVID-19) among patients with normal admission QTc.

Methods

We enrolled 110 consecutive patients in a multicenter international registry. A 12-lead electrocardiograph was performed at admission, after 7, and at 14 days; QTc values were analyzed.

Results

After 7 days, 15 (14%) patients developed a prolonged QTc (pQTc; mean QTc increase 66 ± 20 msec; +16%; P < .001); these patients were older and had higher basal heart rates, higher rates of paroxysmal atrial fibrillation, and lower platelet counts. The QTc increase was inversely proportional to the baseline QTc level and leukocyte count and directly proportional to the basal heart rate (P < .01).

We conducted a multivariate stepwise analysis including age, male gender, paroxysmal atrial fibrillation, basal QTc values, basal heart rate, and dual antiviral therapy; age (odds ratio [OR], 1.06; 95% confidence interval [CI], 1.00–1.13; P < .05), basal heart rate (OR, 1.07; 95% CI, 1.02–1.13; P < .01), and dual antiviral therapy (OR, 12.46; 95% CI, 2.09–74.20; P < .1) were independent predictors of QT prolongation.

The incidence rate of LTA during hospitalization was 3.6%. There was 1 patient who experienced cardiac arrest and 3 with nonsustained ventricular tachycardia. LTAs were recorded after a median of 9 days from hospitalization and were associated with 50% of the mortality rate.

Conclusions

After 7 days of hospitalization, 14% of patients with COVID-19 developed pQTc; age, basal heart rate, and dual antiviral therapy were found to be independent predictors of pQTc. Life-threatening arrhythmias have an incidence rate of 3.6%, and were associated with a poor outcome.

arrhythmia, COVID-19, ECG, QTc prolongation, risk prediction

Coronavirus disease 2019 (COVID-19) is an infectious disease caused by a newly discovered coronavirus, presenting mainly as a severe acute respiratory syndrome [1]. First reported in China in December 2019, it has quickly spread all over the world, becoming a pandemic in a few months. There is no standard therapy and no clear consensus from scientific societies in the absence of solid clinical data. The first observational data suggest that a combined approach with an antimalarial drug (hydroxychloroquine) and an antibiotic, macrolide, may have some effect [2]. Other suggested approaches consist of antivirals, such as remdesivir or lopinavir/ritonavir [3], or anti–interleukin-6 in cases of an increased inflammatory response [4]. However, no therapies have been shown as effective to date [5].

Some of the drugs currently used for this disease may have interactions with myocardial cells, especially during the repolarization phase, and may result in QTc interval prolongation and torsade de pointes [6]. In cases of QTc prolongation >500 msec, drugs should be withdrawn or continuous electrocardiograph (ECG) monitoring should be started. The arrhythmic risk in COVID-19 patients seems to be increased in relation to several factors, such as increased sympathetic activity and direct myocardial injury [7].

The aim of the study was therefore to evaluate potential predictors of QTc interval prolongation and incidences of life-threating arrhythmias in COVID-19 patients admitted with a normal QTc interval who started therapy during hospitalization.

METHODS

Study Population

We prospectively enrolled 154 consecutive patients with a diagnosis of COVID-19 from 25 February to 30 March 2020, admitted into 4 hospitals: the infectious diseases unit and intensive care unit of Hospital-University Polyclinic of Bari, Italy; the Department of Infectious disease, Vittorio Emanuele II Hospital, Bisceglie, Italy; the Department of Infectious Disease, San Carlo Hospital, Potenza, Italy; and the First Department of Medicine, Faculty of Medicine, University Medical Centre Mannheim, Germany. The study was approved by the institutional review board of each hospital involved. All patient information was deidentified. All procedures were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Inclusion Criteria

The diagnosis of COVID-19 was based on the definition of the World Health Organization [8].

Exclusion Criteria

Patients already in treatment with antiarrhythmic drugs or drugs prolonging the QTc interval were excluded from the study. Patients already in treatment with anti–COVID-19 drugs before hospitalization were excluded. Patients with an ECG recorded >12 hours after admission (n = 14) and patients with an admission QT > 450 msec in men and 470 msec in women (n = 30) were excluded from the analysis of 1 clinical end point: prolonged QT evaluation.

Clinical Data

All patients underwent a clinical examination; age, gender, medical history, and previous therapy were recorded.

Blood Sample Collection

Circulating levels of C-reactive protein, high-sensitivity troponin, D-dimer, lactate dehydrogenase (LDH), ferritin, creatinine, electrolytes (sodium, potassium, calcium, magnesium), and blood count with formula were obtained by venipuncture at the admission. Normal values were <1 mg⁄L for C-reactive protein, <19.8 pg/ml for high-sensitivity troponin, 313–618 U/L for LDH, 0.61–1.24 for creatinine, 150–400 × 10⁹/L for the platelet count, 4.3–10.0 × 10⁹/L for the total white blood cell count, 2.0–7.0 × 10⁹/L for neutrophils, and 0.95–4.5 × 10⁹/L for lymphocytes.

Treatment Approach

Treatment was based on operator choice, mainly according to Italian guidelines of the Infectious Disease Society [9].

Electrocardiogram Analysis

Standard 12-lead ECGs were serially recorded within 12 hours after admission and were then repeated after 7 and 14 days. The QT and RR intervals were measured in 3 consecutive beats in the sinus rhythm and 5 consecutive beats in atrial fibrillation, and were then averaged [10]. The Framingham formula was used in all cases in order to get the best assessment of QT intervals, also in a bundle branch block [11].

Definition of Outcome

Clinical end points were QTc interval prolongation and life-threatening arrhythmias during hospitalization.

QTc Prolongation Group

The patients included presented with normal QTc interval values at admission and, after 7 days, developed QTc interval values greater than 450 ms for men and 470 ms for women [12]. Patients presenting at admission with QTc intervals greater than 450 ms for men and 470 ms for women were excluded from the study.

Life-Threatening Arrhythmias

Life-threatening arrhythmias included ventricular tachycardia (VT), ventricular fibrillation, torsade de pointes, asystole, or complete atrioventricular block. The presence of rhythm disorders was assessed by ECG and independently reviewed by 2 experienced cardiologists during the hospital stay. Nonsustained VT was defined as a ventricular rhythm faster than 100 bpm lasting less than 30 seconds.

Statistical Analysis

Continuous variables were reported as means ± standard deviations or medians with interquartile ranges. Variables were compared using a Student t test for either paired or unpaired groups, the Mann-Whitney test (for unpaired data), or the Wilcoxon rank test (for paired data) as required, with dichotomic variables as percentages that were compared with the χ ² test or Fisher test, as required. Repeated measures were analyzed with an analysis of variance test. A multiple regression analysis was used to identify predictors for prolonged QTc after 7 days and for correcting bias of principal confounders; odds ratios (ORs) and 95% confidence intervals (CIs) were also calculated. A P value <.05 was considered as statistically significant.

RESULTS

Patients’ Characteristics

We enrolled 110 consecutive patients in the study. All baseline features are reported in Table 1; 66% were male, and the mean age was 58 ± 14 years. The mean QTc interval at admission was 409 ± 26 msec. At admission, 5% of patients presented with atrial fibrillation. The admission heart rate was 73 ± 15 bpm. Overall, the death rate during hospitalization was 9% (10 out 110 patients).

Table 1.

Baseline Clinical Features During Hospitalization in Patients with Prolonged QT Interval at 7 Days After Admission

	General population	Prolonged QTc	No.	P value
Number of patients	110	15	95
Age, y	58 ± 15	66 ± 12	56 ± 14	.01
Male sex	80%	80%	64%	.23
Clinical baseline profile
Hypertension	39%	20%	43%	.09
Diabetes	13%	0%	18%	.80
Obesity, BMI >30	15%	13%	16%	.91
Renal insufficiency, ClCr <30 ml/min	8%	0%	10%	.21
History of lung disease	16%	20%	17%	.78
History of heart disease	21%	13%	26%	.44
History of cancer	8%	13%	7%	.43
Clinical features at admission
Fever	75%	80%	74%	.88
Hypo/anosmia	6%	6%	6%	.86
Diarrhea	6%	0%	6%	.38
Myalgia/artralgia	18%	6%	19%	.44
Laboratory data
Admission LDH levels, U/L	328 ± 286a	240 ± 89a	380 ± 403a	<.001
Admission D-dimer levels, ng/mL	753 ± 976a	592 ± 407a	773 ± 2493a	.06
CRP peak during hospitalization, mg/dl	75 ± 132a	70 ± 94a	78 ± 146a	.78
Admission creatinine levels, mg/dl, .2–8.5	.9 ± .3a	1 ± .3a	.9 ± .3a	<.05
Admission leucocytes count, x109/L	6325 ± 4150a	4600 ± 2550a	6534 ± 4385a	.02
Admission lymphocyte, x109/L	866 ± 527a	840 ± 570a	876 ± 516a	.75
Admission platelets count, x109/L	187 ± 101a	161 ± 48a	197 ± 101a	.02
Complications
ICU stay	6%	13%	4%	.15
Death	9%	0%	11%	.19

Data are provided as mean ± SD unless otherwise noted, and counts are as shown as percentages. Abbreviations: BMI, body mass index; ClCr, clearance; CRP, C-reactive protein; ICU, intensive care unit; IQR, interquartile range; LDH, lactate dehydrogenase; QTc, corrected QT; SD, standard deviation.

^aMedian and IQR.

QTc Prolongation During Hospitalization

The mean QTc interval after 7 days was 429 ± 30 msec, and after 14 days the mean was 435 ± 26 msec (Figure 1). After 7 days, 15 (14%) patients developed a prolonged QTc interval (mean QTc at admission 414 ± 16 vs 481 ± 21 msec after 7 days; mean QTc interval increase 66 ± 20 msec [+16%]; P < .001; Figure 2). These patients who developed a prolonged QTc were older (66 ± 12 vs 56 ± 14 years, respectively; P < .05), had higher basal heart rates (87 ± 26 vs 71 ± 11 bpm, respectively; P < .001), had higher rates of paroxysmal atrial fibrillation (20% vs 2%, respectively; P < .01), and had lower platelet counts (169 ± 41 vs 231 ± 112 *1000/mm³, respectively; P < .05; Tables 1–2) than patients who did not develop a prolonged QTc.

Table 2.

Electrocardiographic Features During Hospitalization in Patients with Prolonged QTc Intervals 7 Days After Admission

ECG features	General population	Prolonged QTc	No	P value
Number of patients	110	15	95
Admission
Negative T waves	9%	0%	12%	.16
ST elevation	0%	0%	0%	.99
ST depression	1%	0%	1%	.69
Mean QTc interval, msec	409 ± 26	415 ± 15	408 ± 27	.36
7 days
Negative T waves	9%	0%	10%	.35
ST elevation	0%	0%	0%	.99
ST depression	0%	0%	0%	.99
Mean QTc interval, msec	429 ± 30	481 ± 21	421 ± 22	.01
14 days
Negative T waves	5%	0%	6%	.42
ST elevation	0%	0%	0%	.99
ST depression	1%	0%	2%	.69
Mean QTc interval, msec	435 ± 26	464 ± 28	430 ± 23	.01

Data are provided as means ± SDs, and Counts are shown percentages. Abbreviations: ECG, electrocardiograph; QTc, corrected QT; SD, standard deviation.

Figure 1.

QTc intervals over time in general population. Data are shown as mean ± 95% confidence interval; P < .001 for all repeated measures. Abbreviation: QTc, corrected QT.

Figure 2.

QTc intervals over time among patients that did or did not develop prolonged (above cut-off levels) QTc at the seventh day of hospitalization. Data are shown as mean ± 95% confidence interval; P < .001 for all repeated measures. Abbreviation: QTc, corrected QT.

At 7 days, no relevant differences in electrolyte concentrations were found. When comparing patients with and without QT prolongation, no differences were found in terms of hydroxychloroquine use (93% vs 91%, respectively; not significant); the use of other drugs (93% vs 32%, respectively, used lopinavir/ritonavir [P = .01]; 20% vs 76%, respectively, used azithromycin [P < .001]; 60% vs 26%, respectively, used cephalosporin [P = .01]; 0% vs 27%, respectively, used tocilizumab [P < .05]). However, the use of single (100% vs 74%, respectively; P = .03) or dual antiviral therapy (40% vs 9%, respectively; P < .01) was significantly different between patients with and without QTc prolongation (Table 3).

Table 3.

Drug Therapy During Hospitalization Among Patients With and Without Prolonged QT Interval During the Seventh Day of Hospitalization

Therapy during hospital stay	General population	Prolonged QT group	Non prolonged QT group	P value
Use of cloroquine or similar	93% (100)	93% (14)	91% (86)	.98
Use of corticoids	31% (34)	0%	35% (34)	.01
Use of antiviral drugs	77% (85)	100% (15)	74% (70)	.03
Lopinavir/ritonavir	40% (44)	93% (14)	32% (30)	.01
Darunavir/cobicistat	27% (30)	13% (2)	29% (28)	.23
Oseltamivir	13% (14)	26% (4)	11% (10)	.09
Tenofovir	6% (7)	7% (1)	6% (6)	.99
Use of tocilizumab	24% (26)	0%	27% (26)	.01
Use of antibiotics	95% (98)	87% (13)	90% (85)	.76
Macrolid, azitromicin	71% (75)	20% (3)	76% (72)	.01
Beta-lactames	10% (13)	13% (2)	12% (11)	.88
Cephalosporin	30% (34)	60% (9)	26% (25)	.01
Anticoagulation	30% (29)	13% (2)	28% (27)	.01
Antiplatelets	7,% (12)	7% (1)	12% (11)	.80
ACE-I	4% (3)	0%	3% (3)	.25
ARB	9% (8)	13% (2)	6% (6)	.29

Data are reported as percentage and number of patients.

Abbreviations: ACE-1, Angiotensin-converting enzyme-1; ARB, Angiotensin II receptor blocker.

Both in absolute and in relative terms, the increase in QTc values was inversely proportional to baseline QTc levels (r, -0.52 and -0.57, respectively; P < .001) and leucocyte count (r, -0.20 and r -0.19, respectively; P < .05), and directly proportional to basal heart rates (r, 0.33 and 0.28, respectively; P < .01; Figure 3).

Figure 3.

Correlations between QTc increase, baseline QTc levels and basal heart rate. P values < .01. Abbreviations: QTc, corrected QT.

We conducted a multivariate forward stepwise logistic regression analysis including age, male gender, presence of paroxysmal atrial fibrillation, basal QTc values, basal heart rate, and dual antiviral therapy; age (OR, 1.06; 95% CI, 1.00–1.13; P < .05), basal heart rate (OR, 1.07; 95% CI, 1.02–1.13; P < .01), and dual antiviral therapy (OR, 12.46; 95% CI, 2.09–74.20; P < .1) were independent predictors of QT prolongation (OR, 1.05; 95% CI, 1.00–1.09; P = .03), with model accuracy (receiver operating characteristic area under the curve) of 0.85 (95% CI, .76–.91).

Life-Threatening Arrhythmias

There was 1 patient who experienced cardiac arrest due to asystole and was resuscitated with advanced circulatory support, and 3 patients had nonsustained ventricular tachycardia managed with an intravenous beta-blocker (metoprolol; Table 4). Patients experienced life-threatening arrhythmias after a median of 9 days (6–21 days) from hospitalization. Life-threatening arrhythmias were associated with a poor outcome: 2 out of 4 patients died 24 ± 12 hours after the index event.

Table 4.

Clinical Features, In-Hospital Therapy, and Adverse Events in Patients With Life-Threatening Arrhythmias During Hospitalization

	Patient 1	Patient 2	Patient 3	Patient 4
Age, y	79	74	59	53
Sex	F	F	F	M
Clinical baseline profile
Hypertension	Y	N	Y	N
Diabetes	N	Y	N	N
Obesity, BMI >30	N	Y	Y	N
Renal insufficiency, ClCr <30 ml/min	N	N	N	N
History of lung disease	N	N	N	N
History of heart disease	N	N	N	N
History of cerebrovascular disease	N	Y	N	N
History of cancer	N	N	N	N
ECG features
Admission QTc interval, msec	425	380	410	449
7 days QTc interval, msec	409	470	410	460
14 days QTc interval, msec	470	580	451	407
Use of vasoactive drugs	Y	Y	Y	N
Hydroxycloquine administration	Y	N	Y	Y
Antiviral therapy
Lopinavir/ritonavir	N	Y	N	N
Darunavir/cobicistat	Y	Y	N	N
Antibiotic therapy, azythromicin	Y	Y	Y	Y
Arrhythmias during hospital stay
Cardiac arrest	Y	N	N	N
VT	N	N	N	N
NSVT	N	Y	Y	Y
Days from admission	21 days	6 days	12 days	6 days
Death	Y	Y	N	N

Abbreviations: BMI, body mass index; ClCr, clearance; ECG, electrocardiograph; F, female; QTc, corrected QT; M, male; N, no; NSVT, non sustained Ventricular Tachycardia; VT, ventricular tachycardia; Y, yes.

Interestingly, 1 patient showed a transient “Brugada type I” ECG pattern during fever that disappeared after paracetamol infusion; the patient was continuously ECG monitored and did not experience arrythmia during hospitalization. He had no history of syncope and was treated during hospitalization with hydroxychloroquine and azithromycin.

Drug Therapy During Hospitalization

In the overall population, 93% of patients received therapy with hydroxychloroquine, 95% received antibiotic therapy (71% macrolide [azithromycin], 30% cephalosporin, 10% beta-lactam), and 77% received antivirals (40% lopinavir/ritonavir, 27% darunarvir/cobicistat, 13% oseltamivir, 6% tenofovir). Among patients receiving dual antiviral therapy (13%), 66% received lopinavir/ritonavir and oseltamivir, 18% received lopinavir/ritonavir and darunarvir cobicistat, and 16% received oseltamivir and darunarvir cobicistat.

DISCUSSION

We report one of the first multicenter registries on COVID-19 patients that aims to evaluate electrocardiographic changes during hospitalization and life-threatening arrhythmias. We found that:

• After 7 days of hospitalization, 14% of patients developed prolongation of the QTc interval; dual antiviral therapy, age, and basal heart rate were independent predictors of a prolonged QT interval.

• Life-threatening arrhythmias have an incidence rate of 3.6% and may present after a median of 9 days of hospitalization with cardiac arrest and nonsustained VT.

• Life-threatening arrhythmias had a poor outcome, with 50% mortality.

QTc prolongation is quite common among patients treated with antimalarials, antivirals, and antibiotics. Therefore, the aim of the study was to evaluate changes in QT intervals during hospitalization, and their potential correlation with life-threatening arrythmias among patients with COVID-19 and normal admission QTc intervals.

Although no adequately sized randomized trials have been published on treatment of COVID-19, a combination of hydroxychloroquine and azithromycin is commonly used to treat this infection. Moreover, some centers have also started using a combination of antiviral drugs, such as lopinavir/ritonavir or remdesevir. These drugs, especially in combination, may increase the risk of QT interval prolongation and, subsequently, ventricular arrhythmias [12]. Hydroxychloroquine blocks the KCNH2-encoded hERG/Kv11.1 potassium channel and can potentially the prolong QTc interval. Some case reports showed life-threatening arrhythmia due to chronic use of hydroxychloroquine [13, 14]. However, in a registry of 28 patients with systemic lupus erythematosus receiving chronic treatment (7 months) with chloroquine, no conduction disturbances were reported [15]. Antiviral drugs, such as lopinavir/ritonavir, may also induce QT prolongation [16]. An arrhythmogenic effect of azithromycin has been reported in single cases [17], but conflicting data have been published on this risk [18, 19].

In a randomized trial with COVID-19 patients, Borba et al [20] evaluated the effect of high-dosage hydroxychloroquine (ie, 600 mg twice daily for 10 days) versus low-dosage (ie, 450 mg twice daily on Day 1 and once daily for 4 days). The high-dose group had a higher prevalence of QTc intervals greater than 500 milliseconds (18.9 vs 11% in the low-dose group) [20]. Chorin et al [21] found a mean baseline QTc prolongation from 435 to 463 ms after 3.6 days of therapy in COVID-19 patients treated with hydroxychloroquine (400 mg twice daily on the first day, followed by 200 mg twice daily) and azithromycin (500 mg daily). The QTc was severely prolonged, to >500 ms, in 9 (11%) patients. There were no torsade de pointes events recorded for any patients, including those with a severely prolonged QTc [21].

In the present study, we found that a combination of antivirals may predict a prolonged QTc interval; however, due to the heterogenous antiviral combinations, no conclusion can be provided. There were 15 patients who received 2 antiviral drugs; among these, 10 patients were treated with lopinavir/ritonavir (200/50 mg daily) and oseltamivir (75 mg daily).

Ritonavir and oseltamivir were both associated with QT prolongation in previous studies [22, 23]; therefore, their combination could increase this risk.

In the present study, age and basal heart rate were also found to be predictors associated with a prolonged QT interval. These findings may reflect patient’s comorbidities and the higher fragility of this subset of patients. Indeed, there was a trend, although not statistically significant, of a higher rate of intensive care stay for patients that developed a prolonged QT interval after 7 days (13 vs 4% in patients without a prolonged QT interval).

Life-threatening arrhythmias in patients with viral infection have been previously described in the context of myocardial inflammation. Several mechanisms are involved in this process: increased oxidative stress and inflammation can increase the release of inflammatory cytokines, such as tumor necrosis factor–alpha and interleukin-6, leading to Ca2+/calmodulin protein kinase II activation [24]. Viruses can also alter the function or expression of ion channels or induce structural remodeling of the myocardium. Coxsackie virus can increase the calcium current (ICa), leading to action potential duration prolongation [25].

In the present study, most of the arrhythmias were cardiac arrest and nonsustained ventricular tachycardia. Arrhythmias occurred mainly during the second week of hospitalization, and 2 patients died after 24 ± 12 hours from the index event due to multiorgan failure. Management was conservative, and betablockers were administered. As shown also in previous studies [24], no patients experienced ventricular fibrillation following torsade de pointes.

Moreover, 3 out of 4 patients that experienced life-threatening arrhythmias were in the intensive care unit; therefore, arrhythmias seem to be related more to severe progression of the disease. During intensive care unit stays, sedatives—hypnotics, benzodiazepines, or alpha-2 adrenergic agonists—were also administered to obtain light sedation, which was useful for tolerating mechanical ventilation. However, the use of dexmedetomidine, an alpha-2 adrenergic agonist, was associated with bradycardia that could increase the risk of QTc prolongation [26].

This study shows that young patients without significant comorbidities may be candidates for domiciliary treatment with a relative low risk of arrhythmic complication. However, baseline ECGs, with exclusion of patients with inherited long QT syndromes and conduction disturbances (such as a bundle branch block), should be warranted in all patients before treatment.

Limitations

Some limitations have to be considered for the present investigation. The study evaluated a relatively low number of patients. Continuous ECG monitoring was performed only in cases of intensive care unit hospitalization. The population enrolled consisted of symptomatic patients that required hospitalization.

CONCLUSIONS

During hospitalization, at 7 days after admission, 14% of patients developed QTc prolongation; dual antiviral therapy, age, and basal heart rate were the only independent predictors of QT prolongation at 7 days. Life-threatening arrhythmias have an incidence rate of 3.6%, and were associated with a poor outcome.

Note

Potential conflicts of interest. The authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest.

Author notes

F. S., F. Monitillo, P. R., L. D. B., and N. D. B. contributed equally and should be considered as senior authors.