The Utility of CHA(2)DS(2)-VASc Scores as a Risk Assessment Tool in Low-Risk In-Hospital Patients With Coronavirus Disease 2019 Infection.

Coronavirus disease 2019 (COVID-19) infections can have serious consequences such as cardiac manifestations, severe coagulopathy, and thromboembolism.1, 2, 3, 4 We read with interest the study by Ruocco et al published in the American Journal of Cardiology. This study, performed on an Italian cohort, showed that the CHA(2)DS(2)-VASc score could aid prognostication of mortality and a composite end-point of inpatient death or invasive ventilation across CHA(2)DS(2)-VASc tertiles in COVID-19 patients.

In Singapore, there has been a demographic shift in COVID-19 cases initially involving at-risk elderly population in the community, and transitioning to cluster outbreaks in densely-populated foreign worker dormitories. Active case finding of dormitory residents resulted in a surge of swab-positive COVID-19 patients who were either asymptomatic or mildly symptomatic. They were admitted to hospital for risk assessment before transfer to a nonhospital isolation facility. This policy provided an opportunistic insight into the low-risk COVID-19 cohort.

Our hospital registry of 554 in-hospital COVID-19 patients, recruited between 23rd January to 30th April 2020, were admitted to a tertiary healthcare institution for risk assessment before down-triaging to a nonhospital isolation facility. They were allocated into 3 groups based on the CHA(2)DS(2)-VASc scores (CV ≤1; CV2-3; CV ≥4). Primary outcomes were intensive care unit (ICU) admissions, mechanical ventilation, all-cause mortality, and a composite end-point. The composite end-point was defined as the presence of one of the following: ICU admissions, mortality, mechanical ventilation, or any COVID-related organ involvement (pneumonia, acute respiratory distress syndrome, acute kidney injury, pulmonary embolism, coagulopathy, acute myocardial infarction, heart failure, and stroke).

The baseline characteristics of our study population differed from the Ruocco et al study. Our cohort had lower median age of 34 (28 to 43) years compared with the previous study (65 [53 to 76] years). Our median CHA(2)DS(2)-VASc score of 0 was also lower (2 [1 to 3] in the cited study). Seventy-six study participants (13.7%) remained asymptomatic. There was lower prevalence of co-morbidities in our cohort compared with the cited study (hypertension 12.3% vs 48.6%, diabetes 5.1% vs 15.7%, ischemic heart disease 1.2% vs 11.2%, stroke 0.5% vs 7.6%, and heart failure 0.7% vs 6.1%, respectively). Chronic medication use was less in our study cohort. In terms of study outcomes, there were 2 deaths in our low-risk group, both of whom were in the CV ≤1 group. There were 19 (3.4%) patients requiring ICU admission, 16 (2.9%) requiring mechanical ventilation, and 59 (10.6%) with the composite end-point. We observed significant increases in ICU admissions (CV ≤1: 2.8%, CV2-3: 11.1%, CV ≥4: 50.0%, p <0.001), mechanical ventilation (CV ≤1: 2.3%, CV2-3: 11.1%, CV ≥4: 50.0%, p <0.001), and composite end-point (CV ≤1: 9.8%, CV2-3: 16.7%, CV ≥4: 100.0%, p <0.001) across the groups (Table 1 ). Univariable logistic regression analysis demonstrated significantly increased risk of mechanical ventilation in the CV2-3 (odds ratio [OR] 5.778, 95% confidence interval [CI] 1.187 to 28.121, p = 0.030) and CV ≥4 (OR 28.889, 95% CI 4.415 to 189.038, p <0.001) groups compared with CV ≤1 group (reference group). There was significantly increased risk of ICU admission in the CV ≥4 (OR 22.844, 95% CI 3.551 to 146.951, p = 0.001) group compared with CV ≤1 group (reference), with a trend toward increased risk of ICU admission in the CV2-3 group (OR 4.569, 95% CI 0.958 to21.790, p = 0.057). Significant increased risk of the adverse composite end-point was observed in the CV ≥4 group (OR 36.923, 95% CI 4.051 to 338.550, p = 0.001) compared with the CV ≤1 group (reference). There was no statistical difference in composite end-point between CV2-3 (OR 1.978, 95% CI 0.550 to 7.110, p = 0.296) and CV ≤1 groups.

Table 1

Clinical characteristics and outcomes of patients hospitalised for COVID-19 infection categorised according to CHA(2)DS(2)-VASc score

Variable	All patients (N = 554)	CHA(2)DS(2)-VASc scores			p Value
		≤1 (N = 532)	2-3 (N = 18)	≥4 (N = 4)
Sex (male)	482 (87.0)	473 (88.9)	9 (50.0)	0	<0.001
Age (years)	36 (11)	36 (10)	54 (13)	67 (19)	<0.001
Smoking status					0.330
Current smoker	30 (5.6)	27 (5.3)	3 (16.7)	0
Ex-smoker	3 (0.6)	3 (0.6)	0	0
Hypertension	53 (12.3)	34 (8.3)	16 (88.9)	3 (75.0)	<0.001
Diabetes mellitus	21 (5.1)	8 (2.0)	10 (62.5)	3 (75.0)	<0.001
Hyperlipidemia	34 (8.1)	17 (4.3)	13 (72.2)	4 (100.0)	<0.001
Atrial fibrillation	0	0	0	0
Ischemic heart disease	5 (1.2)	1 (0.3)	3 (16.7)	1 (25.0)	<0.001
Congestive heart disease	3 (0.7)	0	1 (6.3)	2 (50.0)	<0.001
Stroke	2 (0.5)	0	0	2 (50.0)	<0.001
Chronic kidney disease	3 (0.7)	1 (0.3)	0	2 (50.0)	<0.001
Medications
Statin	26 (6.2)	11 (2.8)	11 (61.1)	4 (100)	<0.001
Beta-blocker	8 (1.9)	4 (1.0)	3 (16.7)	1 (25.0)	<0.001
Calcium channel blocker	29 (6.9)	17 (4.3)	10 (55.6)	2 (50.0)	<0.001
ACE-I	6 (1.5)	2 (0.5)	3 (16.7)	1 (25.0)	<0.001
ARB	12 (2.9)	8 (2.0)	3 (16.7)	1 (25.0)	<0.001
Diuretics	5 (1.2)	3 (0.8)	0	2 (50.0)	<0.001
Metformin	18 (4.3)	8 (2.0)	10 (62.5)	0	<0.001
Insulin	4 (1.0)	2 (0.5)	1 (6.3)	1 (25.0)	<0.001
Study outcomes
Mortality	2 (0.5)	2 (0.5)	0	0	0.952
Intensive care unit admission	19 (3.4)	15 (2.8)	2 (11.1)	2 (50.0)	<0.001
Mechanical ventilation	16 (2.9)	12 (2.3)	2 (11.1)	2 (50.0)	<0.001
Composite end-point	59 (10.6)	52 (9.8)	3 (16.7)	4 (100.0)	<0.001

ACE-I = angiotensin converting enzyme inhibitor; ARB = angiotensin receptor blocker; COVID-19 = coronavirus disease 2019.

Similar to the previous study, our receiver operating characteristics (ROC) analysis (Figure 1 ) confirmed the prognostic ability of CHA(2)DS(2)-VASc score in the low-risk COVID-19 cohort for ICU admissions, mechanical ventilation requirement, and study composite end-point.

Figure 1

Receiver operating characteristic curves for mechanical ventilation, intensive care unit admission, and study composite end-point for the predictor of CHA(2)DS(2)-VASc score. Composite end-point was defined as the presence of one of the following: ICU admissions, mortality, requiring mechanical ventilation, or one of the COVID-related organ involvements (pneumonia, acute respiratory distress syndrome, acute kidney injury, pulmonary embolism, coagulopathy, acute myocardial infarction, ventricular tachycardia, myocardial injury or myocarditis, heart failure, and stroke).

To date, a clinically simple risk stratification score for COVID-19 patients is lacking. Ruocco et al have called for an urgent need to characterize these patients to identify the at-risk patients of acute respiratory distress syndrome. Our findings reinforce CHA(2)DS(2)-VASc score as a potential tool to identify at-risk COVID-19 individuals in a generally young, low-risk, asymptomatic, or mildly symptomatic cohort. The study demonstrated that CV2-3 and CV ≥4 groups displayed higher rates of mechanical ventilation, ICU admissions, and composite end-point, compared with the CV ≤1 group. However, we did not see a trend for all-cause mortality due to the low-risk nature of the cohort with its overall mortality rate of 0.5%.

The CHA(2)DS(2)-VASc score is indeed a simple and widely-available stratification tool that can be used in the outpatient setting or upon admission, as it is not restricted by laboratory measurements that is required in other proposed risk scores. This is important as it allows clinicians to identify those who are at higher risk in the community, and may benefit from closer in-hospital monitoring.

Disclosure

The authors have no conflicts of interest to disclose.

Financial disclosure: None.