Severe Acute Respiratory Syndrome (SARS) is one of the outcomes related to coronavirus infection, called Sars-CoV-2, and has been configured in a pandemic that has generated social, financial, and psychological implications worldwide.¹ The disease was characterized as a pandemic, with 15,581,009 confirmed cases and 635,173 deaths worldwide as of August 23^rd, 2020.²

The presence of Cardiovascular Disease (CVD) in individuals infected with COVID-19 may result in a worse prognosis, in addition to being associated with a higher lethality rate.³ Data suggest that acute cardiac injury, cardiogenic shock, and cardiac arrhythmia were present, respectively, in 7.2, 8.7, and 16.7% of patients after infection by COVID-19, and highlight that the stay in the Intensive Care Unit (ICU) can increase this prevalence.⁴

CVD stand out as an important public health problem in low- and middle-income countries, in view of their increased burden, whether in relation to their related comorbidities, or due to the public spending involved.⁵ According to data from the National Health Survey ( Pesquisa Nacional de Saúde – PNS), the prevalence of CVD among Brazilian adults aged 18 years old and older was 4.2%, presenting an increasing gradient with the increasing age, highlighting the prevalence of 11.4% among the aged.⁶

Studies involving national data at a time of pandemic help to understand and direct more effective actions and long-term planning. The social isolation actions adopted to face COVID-19 are different among Brazilian macro-regions. The concerns about pre-existing morbidities are highlighted, as is the case of CVD; patients are instructed to follow activity restrictions that impose limitations which may compromise the control of complications resulting from living with CVD, in addition to restricted medical follow-up during this period.

Given this context, the objective of this study was to describe the prevalence of SARS by COVID-19 and to analyze the factors associated with this condition in adults and aged people with chronic CVD in Brazil until 30^thEpidemiological Week of 2020.

Notifications of 116,343 patients with CVD were analyzed in this study. Of these, 61.9% were diagnosed with SARS by COVID-19. Regarding the characterization of the sample, the majority were male (52.8%), white (51.3%), from the southeastern macro-region (58.3%), and lived in urban areas (95.7%). In addition, a higher prevalence of individuals aged 60 years old or older was observed (73.6%). Regarding the bivariate analysis, all variables were associated with a higher prevalence of SARS by COVID-19, except for the age group of 80 years old or older and individuals living in peri-urban areas ( Table 1 ).

Table 1

Characteristic	Percentage of the total sample %	Prevalence of SARS by COVID-19 %	p*	Raw PR (95%CI)
Gender			<0.001
Male	52.8	63.8		1.00
Female	47.2	59.8		0.93 (0.92-0.95)
Age range			<0.001
18 to 29	0.7	53.3		1.00
30 to 39	2.8	63.6		1.19 (1.07-1.32)
40 to 49	7.6	67.3		1.26 (1.14-1.39)
50 to 59	15.3	67.4		1.26 (1.14-1.39)
60 to 69	23.4	65.2		1.22 (1.11-1.34)
70 to 79	25.0	61.1		1.14 (1.04-1.26)
80 or more	25.2	54.9		1.02 (0.93-1.13)
Race			<0.001
White	51.3	55.0		1.00
Black	7.7	60.8		1.10 (1.06-1.14)
Yellow	1.6	64.0		1.16 (1.08-1.24)
Brow	39.2	65.1		1.18 (1.16-1.20)
Indigenous	0.2	73.3		1.33 (1.12-1.58)
Macro-region			<0.001
South	10.9	39.9		1.00
Southeast	58.3	61.3		1.53 (1.49-1.58)
Mid-West	5.6	61.8		1.54 (1.48-1.61)
Northeast	18.8	73.1		1.82 (1.77-1.88)
North	6.8	72.0		1.80 (1.73-1.87)
Residence area			<0.001
Urban	95.7	61.7		1.00
Rural	3.9	53.1		0.86 (0.82-0.89)
Peri-urban	0.4	57.8		0.93 (0.81-1.07)

*Pearson’s χ²test; p-value <0,05; 95%CI: 95% confidence interval.

With regard to signs and symptoms, the majority presented fever (69.5%), cough (79.3%), dyspnea (82.8%), and O₂saturation <95% (74.0%). Regarding clinical factors, the majority did not need to be admitted to the ICU (59.9%) and it was not a nosocomial case (96.7%). However, 50% of the individuals required invasive ventilatory support. The signs and symptoms associated with the higher prevalence of SARS by COVID-19 were: fever, cough, and O₂saturation <95% (p<0.05). Regarding clinical factors, the outcome was associated to ICU admission, use of ventilatory support, and nosocomial case ( Table 2 ).

Table 2

Characteristic	Percentage of the total sample %	Prevalence of SARS by COVID-19 %	p*	Raw PR (95%CI)
Fever			<0.001
No	30.5	49.3		1.00
Yes	69.5	67.4		1.36 (1.34-1.39)
Cough			<0.001
No	20.7	52.1		1.00
Yes	79.3	64.2		1.23 (1.20-1.25)
Dyspnea			0.271
No	17.2	61.6		1.00
Yes	82.8	61.2		0.99 (0.97-1.01)
O 2 Saturation<95%			<0.001
No	26.0	57.4		1.00
Yes	74.0	62.3		1.08 (1.06-1.10)
Hospitalized in the ICU			<0.001
No	59.9	58.6		1.00
Yes	40.1	65.7		1.19 (1.10-1.13)
Use of ventilatory support			<0.001
No	25.7	55.5		1.00
Yes, invasive	24.3	66.7		1.20 (1.17-1.22)
Yes, non-invasive	50.0	61.7		1.12 (1.09-1.13)
Nosocomial case			<0.001
No	96.7	60.0		1.00
Yes	3.3	66.4		1.10 (1.05-1.15)

*Pearson’s χ²test; In bold, p-value <0,05; ICU: Intensive Care Unit; 95%CI: 95% confidence interval.

In the final adjusted model, the prevalence of SARS by COVID-19 was 4% lower in women, when compared to men (PR=0.96; 95%CI: 0.94–0.99) and 18% lower in individuals who lived in rural areas (PR=0.82; 95%CI: 0.77–0.87), when compared to individuals who lived in urban areas. On the other hand, it is noteworthy the prevalence 1.27 times higher in the age group of 50 to 59 years old (95%CI: 1.09–1.48), and 1.81 times higher in the northeast region (95%CI: 1.72–1.91). Fever (PR=1.24; 95%CI: 1.20–1.27), cough (PR=1.12; 95%CI: 1.09–1.16), admission to the ICU (PR=1, 08; 95%CI: 1.05–1.11), use of invasive ventilatory support (PR=1.14; 95%CI: 1.09–1.19), use of non-invasive ventilatory support (PR=1.11; 95%CI: 1.07–1.14), and nosocomial case (PR=1.12; 95%CI: 1.05–1.20) were statistically associated with an increased likelihood of SARS by COVID-19 ( Table 3 ).

Table 3

Characteristics	Final model
	Adjusted PR (95%CI)	p-value*
Gender		0.010
Male	1.00
Female	0.96 (0.94-0.99)
Age range
18 to 29	1.00
30 to 39	1.17 (0.99;1.38)	0.056
40 to 49	1.25 (1.07-1.46)	0.004
50 to 59	1.27 (1.09-1.48)	0.002
60 to 69	1.21 (1.04-1.41)	0.010
70 to 79	1.17 (0.96-1.29)	0.148
80 or more	0.99 (0.85-1.16)	0.981
Macro-region
South	1.00
Southeast	1.45 (1.39-1.51)	<0.001
Mid-West	1.35 (1.26-1.45)	<0.001
Northeast	1.81 (1.72-1.91)	<0.001
North	1.71 (1.62-1.82)	<0.001
Residence area
Urban	1.00
Rural	0.82 (0.77-0.87)	<0.001
Peri-urban	0.92 (0.76-1.12)	0.451
Fever
No	1.00
Yes	1.24 (1.20-1.27)	<0.001
Cough
No	1.00
Yes	1.12 (1.09-1.16)	<0.001
Hospitalized in the ICU
No	1.00
Yes	1.08 (1.05-1.11)	<0.001
Use of ventilatory support
No	1.00
Yes, invasive	1.14 (1.09-1.19)	<0.001
Yes, non-invasive	1.11 (1.07-1.14)	<0.001
Nosocomial case
No	1.00	<0.001
Yes	1.12 (1.05-1.20)

In the final model, the variables were adjusted to each other; *In bold, p-value <0,05; ICU: Intensive Care Unit; 95%CI: 95% confidence interval.

Among hospitalized adults and aged people with CVD, 61.9% were diagnosed with SARS by COVID-19. The prevalence of the outcome was 4% lower in women and 18% lower in individuals living in rural areas. On the other hand, a higher prevalence was observed in the age group from 40 to 69 years old and in the northeast region. Fever, cough, ICU admission, use of ventilatory support, and nosocomial case were significantly associated with an increased likelihood of SARS by COVID-19.

Chronic diseases can be considered risk factors for COVID-19 infection due to their susceptibility to greater associated morbidity and mortality.^{11 , 12} Thus, individuals with previous CVD may be more vulnerable to more severe infections,¹³ considering the fragility of each individual’s system, thus providing the potential action of the virus and corroborating with the data found in this research, whose prevalence of confirmed diagnosis for COVID -19 in hospitalized CVD patients was high.

Even before the pandemic, CVD were common comorbidities in diagnoses of SARS, which can elevate the risk of associated mortality by twelve fold.^{14 , 15} Although the deaths among the participants were not evaluated, a study developed by Zhang,¹⁶ in Wuhan, China, showed that mortality from COVID-19 in patients with CVD had a higher prevalence (22.2%) in relation to the general population of the study (9.8%).

Studies indicate that men are at greater risk of evolving to a more severe condition of COVID-19,¹⁷ indicating a possible influence of biological factors intrinsic to gender as well as socio-cultural and behavioral factors. These data seem to be better consolidated in Chinese¹⁸ and European^{19 , 20} populational studies, in which data disaggregated by gender show similar absolute numbers of contamination between men and women, but with a worse evolution in men, especially with CVD. In a recent publication in the journal “Biology of Sex Differences”, epidemiological data from countries such as Italy, China, Spain, France, Germany, and Switzerland were analyzed and reinforced this hypothesis. These grouped data also indicate that this difference in infection rates and a worse prognosis between genders may be more pronounced in middle-aged individuals (50 to 59 years old).¹⁷

One of the possible explanations for the lower prevalence of SARS by COVID-19 in women is the variation between the immune response and the susceptibility to viral infections between the genders, which can lead to differences in the severity and evolution of the disease.¹⁷ In addition, there seems to be significant differences in the regulation and expression of proteins that participate in the pathophysiological process of SARS-CoV-2 between the genders. Data such as the difference between circulating level, activity, and expression of angiotensin-converting enzyme 2^{21 , 22} and transmembrane serine protease type 2²³ corroborate this theory. In addition, a study carried out in Brazil pointed out that women with and without noncommunicable diseases (NCD) use health services more when compared to men.²⁴ This fact can be attributed to a greater perception of the signs and symptoms of diseases, a higher prevalence of exams, and greater health promotion and prevention practices, contributing to better health outcomes and lower infection rates.²⁴

Despite an 18% lower prevalence of SARS by COVID-19 in rural areas, probably due to the low population density, there is also a high incidence and mortality in rural and remote regions, such as Amazonas and Amapá, which can be justified by the difficulty in access to intensive care.^{25 , 26} Corroborating these findings, an epidemiological analysis carried out in the United States identified a higher rate of SARS-CoV-2 infection in the urban population; however, black individuals, aged between 25 and 49 years, smokers, and obese were related to increased prevalence rates of COVID-19 in rural areas.²⁵

The highest prevalences observed in the northern and northeastern macro-regions may present themselves as a potential public health problem, provided the Brazilian regional inequalities.²⁷ This pandemic situation exposes weaknesses in health care and assistance in Brazil and reinforces the issues of inequality in the north and northeast regions, with regard to the contingency of professionals, infrastructure, and capacity for the production and performance of diagnostic tests, whose issues date prior to the pandemic and persisted in the current epidemiological situation.²⁸

A report produced by the Oswaldo Cruz Foundation (FIOCRUZ) (2020)²⁹ sought to classify vulnerability indicators (A-less vulnerable to E-most vulnerable) at the municipal level, in order to create an estimate of the risk of COVID-19 spreading in Brazilian states. For this, factors such as life expectancy at birth, GINI index — which measures inequality and income distribution —, education component of the human development index (IDHedu), % of population living in extreme poverty, % of population living in urban areas, % of people in households with inadequate water supply and sewage, % of households with running water, and % of households without electricity. The data showed that the municipalities in the north and northeast regions were considered more vulnerable, belonging to classes C, D, and E, and that the less populous capitals, such as Teresina, Maceió, Aracajú, Palmas, Rio Branco, and Porto Velho had high potential for dissemination.

In this study, it was observed that the most frequent signs and symptoms associated with the confirmation of SARS by COVID-19 were dyspnea, cough, and fever, with a statistically significant association only for fever and cough. Data analysis of 4,203 Chinese patients identified that the most common symptoms associated with COVID-19 infection were fever, cough, and dyspnea (80.5, 58.3, and 23.8%); with regard to comorbidities, they were hypertension, CVD, and diabetes (16.4, 12.1, and 9.8%).³⁰ In a retrospective study developed by Zhang,¹⁶ with a sample consisting of 380 individuals and confirmation for COVID-19, it was found that cough with sputum production was the most common condition in patients with CVD when compared to the general population.

In a study conducted by Fang¹² , factors associated with a greater severity of the disease in the general population were considered, with a greater chance of a worse prognosis, admission to the ICU (RR: 5.61, 95%CI: 2.68–11.76) and the use of invasive ventilation (RR: 6.53, 95%CI: 2.70^-15.84). The study by Wang,⁴ on the other hand, showed that individuals with comorbidities had the most severe form of the disease, with greater need for ICU admission, in addition to the association found between the use of ventilatory support and ICU admission for patients with CVD and confirmed infection for COVID -19, corroborating the findings of this study.

Despite efforts to control COVID-19 infections acquired in a hospital environment, studies show that nosocomial infection is an aggravating factor in the control of the disease.^{4 , 30} In the present study, nosocomial infection was significantly associated with confirmed cases of SARS by COVID-19. A study carried out in Wuhan, China, the epicenter of the beginning of the pandemic, showed a prevalence of nosocomial infection of 41% of SARS attributed to COVID-19 infection, having a higher prevalence in relation to diagnoses of SARS in general.^{4 , 30} Even more worrying, in the study by Zhou,³¹ the proportion of nosocomial infections among patients confirmed for COVID-19 in the initial outbreaks of the disease was 29.3%, reiterating the importance of adequate protection, especially in a hospital environment.

It is highlighted that some limitations must be considered when interpreting the results of this study. Data from adults and aged people hospitalized with CVD were analyzed and, therefore, the results cannot be generalized to other populations. In this context, the lack of available variables representative of CVD control, such as medication used and lifestyle data, limits the adjustment of the findings for the CVD status/control factor. Still, there is the influence of the quality of information completion in notification forms and their heterogeneity in the Brazilian regions, as well as the underreporting of cases. In addition, 4.1% of cases were not diagnosed in the laboratory. This fact can be attributed to the lack of diagnostic tests and certified laboratories to perform them in some regions of the country.³² However, other forms of diagnosis were recognized by the Ministry of Health.³³

The study presented its strengths, highlighting that the analysis of secondary databases is one of the best ways to assess the epidemiological situation of a given population, especially banks with national coverage.

It was concluded that there is a high prevalence of SARS by COVID-19 in adults and aged people with CVD in Brazil. Factors related to sociodemographic characteristics, clinical characteristics, signs, and symptoms were associated with this condition. Finally, the data presented in this study will contribute to facing this pandemic by presenting findings from national data. They will also be able to highlight important aggravating factors associated with the confirmation of COVID-19, with the possibility of carrying out monitoring actions in the target audience.