STRENGTHS AND LIMITATIONS OF THIS STUDY
-
In-hospital cardiovascular events and mortality among COVID-19 patients with large sample sizes were studied in Bangladesh, South Asia.
-
Patient cardiovascular risk factors and comorbidities were examined accompanied by clinical information and treatment on admission.
-
Our study provides robust evidence from a secondary analysis of a prospective study.
-
Data on cardiovascular events and mortality among prevaccine COVID-19 patients were presented.
-
Laboratory parameters were not included in this secondary analysis.
Introduction
The global population has been greatly affected by COVID-19, which was declared as a global pandemic by WHO in 2020.1 The first case of COVID-19 in Bangladesh was detected on 8 March 2020 and up to 12 August 2023, >2 million cases were confirmed, resulting in 29 446 deaths.2 Although most of the COVID-19 patients in Bangladesh had only mild symptoms, there were still significant hospitalisations and adverse outcomes.3 A systematic review, assessing the mortality of COVID-19 patients in the publications searched up to April 2020, primarily from China, indicated an increased likelihood of in-hospital mortality among COVID-19 patients after adjusting for age; however, no significant associations were observed between gender or smoking status and in-hospital mortality.4 Another systematic review up to June 2020 also mainly from China concluded that age, gender, smoking and pre-existing conditions such as diabetes, hypertension, cardiovascular diseases, kidney diseases and respiratory diseases contributed to a higher risk of mortality among COVID-19 patients.5 These two systematic reviews shed light on the inconsistencies in the reported magnitude of in-hospital mortality concerning comorbidities and cardiovascular risk factors. It is important to note that most of the studies included in these systematic reviews were conducted in China, where population characteristics and the impact of comorbidities might differ from those in South Asian countries like Bangladesh.
COVID-19 is caused by the SARS-CoV-2 virus which consists of four structural proteins, the spike (S) protein, the nucleocapsid (N) protein, the membrane (M) protein and the envelope (E) protein, and 16 non-structural proteins. The S protein has S1 and S2 subunits which can interact directly with ACE 2, predominantly expressed in the nasal mucous membranes, lung, heart, liver and brain resulting in cell invasion and direct organ damage.6 Elevations of cardiac biomarkers indicating cardiac injury or cardiovascular disease can be affected by COVID-19.7 Clinical outcomes of SARS-CoV-2 infection in terms of hospitalisation, morbidity and mortality vary depending on the variant of concern, risk factors, treatment and vaccination status.6 7 Previous studies have identified associations between adverse outcomes, specifically cardiovascular diseases (CVDs), in COVID-19 patients and cardiovascular risk factors such as older age, male sex, overweight, smoking, hypertension and diabetes.8–14 CVDs are highly prevalent in high-income Asian and Western countries due to population growth and ageing.15
In Bangladesh, after December 2020, Wuhan-like variants were consecutively replaced by the Alpha, Gamma, Eta, Beta, Delta and Omicron variants, respectively.16 17 The infection rates and case fatality rates varied along with the dynamics of the variants throughout the period. The Delta and Beta variants were among the COVID-19 variants associated with higher infection rates in 2021.18 A nationwide immunisation campaign against COVID‐19 was launched among anyone aged 40 years or older in February 2021 and those aged 18 years or older in 10 August 202119; however, the coverage was still low, particularly among the elderly who were the most vulnerable.20 In recent years, Bangladesh has been facing disease transition due to rapid urbanisation and socio-economic and lifestyle changes21 22 and CVD risk factors are now the highest among South Asian countries.23 Furthermore, there is a lack of research investigating the relationship between common clinical presentations of COVID-19 patients and the impact of in-hospital management on patient outcomes.
The protocol of a prospective multicountry study, the ‘World Health Federation (WHF) COVID-19 and Cardiovascular Disease Survey’ was published in April 202124 and the main results of the WHF global study on the description of cardiovascular risk factors and clinical outcomes among hospitalised COVID-19 patients across diverse populations was published in 2022.25 This study involved 23 countries including Bangladesh, which collected data on COVID-19 patients from October 2020 to July 2021 exclusively from patients who had not received a COVID-19 vaccine in Bangladesh, a period during which the Alpha, Beta and Delta variants were reported in the country. The secondary analysis of this study aimed to identify in-hospital cardiovascular events and mortality among COVID-19 patients in Bangladesh and to assess the effects of pre-existing comorbidities and cardiovascular risk factors on in-hospital cardiovascular events and mortality among COVID-19 patients. The information from this analysis can be useful as guidance concerning the clinical benefits and public health implications in the context of various COVID-19 variants, where vaccinations were not yet available.
Methods
Design and study participants
This secondary analysis used the data of adult patients aged ≥18 years with PCR-positive COVID-19 admitted between 10 October 2020 and 31 July 2021 in three COVID-19-designated hospitals in Bangladesh, as per the original study. The original study collected the variables as follows: demographic characteristics, pre-existing medical conditions, clinical condition and treatment during hospital admission and clinical outcomes such as death, major adverse cardiovascular events, renal failure, neurological events or pulmonary outcomes at the time of hospital discharge or at the 30-day follow-up after initial admission if already discharged.24
Variables
The main outcome measures of this study were in-hospital cardiovascular events and mortality. In-hospital cardiovascular events included shock, myocarditis, myocardial infarction, pericarditis, acute heart failure, endocarditis, atrial fibrillation, cardiac arrest, heart blocks and ischaemic stroke, which were included in the composite outcome of major adverse cardiac events used in previous studies26 27 and the study protocol of a main survey.24 Death was recorded for up to 30 days after the initial admission of the COVID-19 patients. Patients who were discharged before 30 days were called to check their status by phone. The recorded causes of death were sudden cardiac death or death due to myocardial infarction, heart failure, stroke, pulmonary embolus, respiratory failure or other causes.
The exposure variables in this study encompassed CVD risk factors and comorbidities. Specifically, the CVD risk factors included age, gender, education level, body mass index (BMI) and smoking status. Education, BMI and smoking were selected as CVD risk factors based on a multicountry community-based prospective cohort study in South Asia.28 Overweight and obesity were defined as a BMI over 25 or 30, respectively. BMI is an anthropometric index of weight and height calculated by dividing a person’s weight by the square of their height (kilograms/metre squared). Comorbidities included both pre-existing cardiovascular and non-cardiovascular conditions prior to admission as reviewed from hospital records. Cardiovascular comorbidities were coronary artery disease, stroke, peripheral vascular disease, atrial fibrillation, heart failure, cardiomyopathies, rheumatic heart disease, Chagas disease, congenital heart disease, valvular disease and hypertension. Non-cardiovascular comorbidities were diabetes, chronic pulmonary disease, chronic kidney disease, asthma, previous organ transplant, tuberculosis, cancer on chemotherapy, HIV, chronic immunosuppression and renal replacement therapy. Comorbidities were then classified into four categories: none, one, two and more than two pre-existing diseases.
The clinical information recorded at admission included temperature, heart rate, SpO2 level, systolic blood pressure and diastolic blood pressure. The treatments on admission were intravenous fluids, antiviral treatment, corticosteroids, antibiotics, ACE inhibitors, angiotensin II receptor blockers and antithrombotic/anticoagulant agents.
Statistical analysis
We used descriptive statistics for categorical variables in counts and percentages, while continuous variables were expressed as means (SD) or medians with IQR. Differences in categorical outcomes between patients who had and those who did not have in-hospital cardiovascular events were tested using the χ2 test. To assess the impact of pre-existing comorbidities—particularly multicomorbidities compared with no comorbidity—and cardiovascular risk factors on in-hospital cardiovascular events among the COVID-19 patients, we used a stepwise binary logistic model to calculate adjusted ORs (adj. ORs) and 95% CIs. Initially, we examined the exposure variables, specifically cardiovascular risk factors, and subsequently considered the clinical information and treatment received on admission. The interactions between comorbidities and other covariates in the models were tested for independence between covariates. The best fitting dataset in the individual model of clinical information and treatment received on admission with comorbidities was selected based on the lowest Akaike information criterion values.
We generated Kaplan-Meier cumulative incidence plots to visualise the survival data comparing individuals with the comorbidities, as well as those who experienced cardiovascular events versus those who did not. The significance of these differences was assessed using the log-rank test. To analyse mortality rates, we estimated HRs using a Cox proportional hazards model. The time-to-event was measured in days from hospital admission to either in-hospital death or death within 30 days of follow-up if the patient had already been discharged. Two Cox proportional hazards models were employed in the analysis. Model 1 examined the effects of comorbidities and CVD risk factors and model 2 included the variables from model 1 as well as clinical information and treatment administered on admission. The assumptions of proportionality of hazard were verified for the application of the Cox regression model using individual and global Schoenfeld tests. For any predictor that violated the proportional hazards assumption, we stratified the model by that predictor, meaning it was no longer directly tested or presented in the final model. We then repeated the Schoenfeld tests to ensure that the assumption was satisfied within the stratified model. Statistical significance was defined as p<0.05. The statistical analyses were performed using R V.4.2.3 (The R Foundation for Statistical Computing Platform, 2022).
Patient and public involvement
The study patients or the public were not involved in the design, conduct, reporting or dissemination plans of our research.
Results
During the study period, 897 adult patients with COVID-19 were admitted to the study hospitals. A description of the cardiovascular risk factors for the study’s COVID-19 patients is provided in table 1. Almost half of the patients were aged between 46 and 65 years and 23.5% of patients were elderly aged >65 years. Male patients slightly outnumbered females. Two-thirds had an education of at least secondary school and had never smoked and almost half were overweight or obese.
Description of cardiovascular risk factors for the study’s COVID-19 patients
Table 2 describes the existing cardiovascular and non-cardiovascular comorbidities for the study’s COVID-19 patients at admission and any cardiovascular events occurring at 30-day of follow-up. Among 897 patients, 276 (30.8%) had no comorbidities and 614 (68.4%) had at least one comorbidity. Of those with comorbidities, 248 (40.4%) had 1, 250 (40.7%) had 2 and 116 (18.9%) had more than 2 comorbidities. Cardiovascular comorbidities were found in 490 patients (54.6%). Among these, hypertension was the most common (n=481, 98.2%), followed by coronary artery disease (n=51, 10.4%), congenital heart disease (n=25, 5.1%) and stroke (n=16, 3.3%). Non-cardiovascular comorbidities were found in 455 patients (50.7%). Among these, diabetes was the most common (n=394, 86.6%), followed by asthma (n=70, 15.4%) and chronic kidney disease (n=62, 13.6%). Cardiovascular events were diagnosed in 167 patients (18.7%). Myocarditis, pericarditis and endocarditis were all found in <l% of the patients (online supplemental table 1). Of all 897 patients, 101 died during admission (11.3%) and an additional 12 died during the 30-day follow-up period (1.5%), resulting in a total of 113 deaths (12.6%). Death was significantly more common in those with cardiovascular events (53/167, 31.7%) than those without events (59/726, 8.1%) (p<0.001) (online supplemental table 2).
Supplemental material
Description of existing cardiovascular and non-cardiovascular comorbidities for the study’s COVID-19 patients at admission and any cardiovascular events occurring at 30-day follow-up
Table 3 presents the associations between pre-existing comorbidities, cardiovascular risk factors and in-hospital cardiovascular events, adjusted for clinical information and management, in the study’s COVID-19 patients. In univariate analysis, patients with two or more comorbidities had higher odds of experiencing in-hospital cardiovascular events compared with COVID-19 patients with no comorbidities. Those who had more than two diseases in the model 1 after adjusting for cardiovascular risk factors and those had two diseases in the model 2 after adjusting for clinical information and treatments on admission showed significantly higher odds of having cardiovascular events. In the model 1, those having education lower than college/university had increased odds of in-hospital cardiovascular events, while former or unknown smoker, or overweight had lower odds. After adjusting for clinical information and treatments on admission as seen in model 2, CVD risk factors were not significantly associated with cardiovascular events. Patients with a higher heart rate at admission (adj. OR 1.03, 95% CI 1.01 to 1.04) and those who were receiving intravenous fluids (adj. OR 2.13, 95% CI 1.23 to 3.70) or antibiotics (adj. OR 4.54, 95% CI 1.47 to 14.01) had significantly higher odds of cardiovascular events. The odds of cardiovascular events were lower in those receiving antiviral medications (adj. OR 0.31, 95% CI 0.18 to 0.53). No interactions between comorbidities and other covariates in the model 1 and model 2 were found.
Effects of pre-existing comorbidities and cardiovascular risk factors, adjusted for clinical information and treatment on admission, on in-hospital cardiovascular events
The factors associated with 30-day mortality in the Cox regression model are shown in table 4. Based on the assumption of proportional hazard, the comorbidities and cardiovascular risk factors were not significantly associated with the mortality (model 1). In model 2, those who received antiviral treatment had a lower hazard of mortality compared with those who did not receive antiviral treatment (adj. HR 0.53, 95% CI 0.34 to 0.81). Almost triple HRs were found in those treated with corticosteroids (adj. HR 2.82, 95% CI 1.48 to 5.38). Mortality increased in patients with cardiovascular events (HR 4.10, 95% CI 2.82 to 5.96); however, this variable was not included in the multivariate Cox regression model due to a violation of assumptions. The probabilities of death in COVID-19 patients within 30-day follow-up stratified by comorbidities and the occurrence of cardiovascular events are presented by the Kaplan-Meier survival curves (figure 1). Four curves of the comorbidities were not statistically significantly different as shown in figure 1A (p=0.450). The probability of death was higher in COVID-19 patients with cardiovascular events compared with those without cardiovascular events (p<0.001) as shown in figure 1B.
Kaplan-Meier survival curves in COVID-19 patients without and different number of comorbidities, n=879 (A) and those with or without cardiovascular events, n=889 (B).
Factors associated with 30-day mortality in a Cox regression model
Discussion
Two-thirds of the COVID-19 patients admitted to the study hospitals during the study period had pre-existing comorbidities, with approximately half having either hypertension or diabetes. Approximately 1 out of 4 patients were aged over 65 years and almost half were overweight or obese. Approximately 19% were diagnosed with cardiovascular events during admission and 11% died during admission and an additional 1.5% died after discharge within 30 days of follow-up. Pre-existing comorbidities and CVD risk factors were associated with in-hospital cardiovascular events. However, after adjusting for clinical information and medications at admission, only the presence of at least two comorbidities remained significant. A significantly higher death rate was found in the COVID-19 patients with cardiovascular events than those without such events. The probability of death within 30 days in the COVID-19 patients aged over 65 years was higher than those aged <26 years in univariate analysis, but the difference was not significant after adjusting for clinical information and medications on admission; there was no evidence of an association with comorbidities.
A Cochrane review found that cardiovascular comorbidities and complications were common in hospitalised COVID-19 patients; however, almost half of their reviewed studies originated from China and the majority of studies were retrospective.29 In early 2020, COVID-19 patients with cardiovascular or non-cardiovascular comorbidities had higher chances of severe outcomes.30 The incidence of cardiovascular events in our study (19%) falls within the range reported in another systematic review, which reported 8.5% in survivors and 21.6% in non-survivors.31 Mortality in hospitalised patients from the systematic review and individual studies varied from 12.0% to 31.8%.4 32–34 The higher mortality rates reported in those previous studies compared with our study may be due to the fact that the majority of hospitalised patients in those studies were older and had more underlying diseases.
For cardiovascular risk factors, our study identified age as a major independent contributor to cardiovascular events, with patients aged over 46 years having a threefold higher risk compared with those aged 25 years or younger. This finding aligns with previous research, indicating a consistent association between advancing age and increased risk of major cardiovascular events and mortality.8 35 We also found a borderline significant difference in in-hospital cardiovascular events between females and males after adjusting for cardiovascular risk factors. These findings may reflect the complexity of other clinical and biological factors in males and females in our study. The previous studies showed that clinical features play a role in determining the severity of a disease and influencing death rates.36 37 The effects of gender on cardiovascular events and death remain uncertain, as previous studies conducted in the USA, the Philippines and China have reported varying findings. These range from no significant differences in death rates by gender to higher mortality rates in males, juxtaposed with higher incidences of CVDs in females.38–40
The significance of education, smoking status or BMI group in our study disappeared after adjusting for clinical information and treatment on admission in the models predicting cardiovascular events and mortality. This may be explained by the finding that the clinical features and medical treatment on admission were more strongly associated with cardiovascular events and mortality than the cardiovascular risk factors. This is consistent with a prior study demonstrating a prediction model of major adverse cardiovascular events in COVID-19 patients.8 A systematic review and meta‐analysis also found an association between CVDs and laboratory results with severe outcomes of COVID-19 patients; however, the certainty of the evidence in this review was limited due to the small number of prospective studies included.5 We did not find any effects of cardiovascular-related medications on death, similar to a previous study which examined the associations of renin-angiotensin-aldosterone system inhibitors, calcium channel blockers, diuretics and beta-blockers on the patients’ prognoses.32
Our study provides robust evidence from a secondary analysis of a prospective study on COVID-19 in-hospital cardiovascular events and mortality, which can guide future healthcare preparedness for care of major adverse cardiovascular events in future pandemic-like COVID-19 situations, particularly in low- and middle-income countries, like Bangladesh. The cardiovascular events examined in our study were similar to the cardiovascular outcomes recorded in the American Heart Association COVID-19 Cardiovascular Disease Registry.41 A recent systematic review published in 2023 supported our findings in univariate analysis that pre-existing CVDs were associated with acute cardiovascular complications.42 Although comorbidities and other cardiovascular risk factors were shown to be associated with cardiovascular events and mortality in COVID-19 patients in previous studies, their effects varied depending on various demographic characteristics of the patients, study settings and designs, analytical methods, treatments on admission and other clinical factors. In our study, the significance of having at least two pre-existing comorbidities on cardiovascular events, but not on mortality, was consistent with the findings from a population-based study in Italy. That study also found that a higher number of comorbidities increased the risks of severe COVID-19-related hospitalisation.43 Cardiovascular events and mortality in our study’s multivariate Cox regression models was similar to the findings from a study in Greece conducted during the dominance of the Delta and Omicron variants.44
Clinical presentation indicating cardiovascular responses such as heart rate, blood pressure and oxygen saturation, and treatment on admission with intravenous fluids, antiviral and antibiotic treatments, and any cardiovascular events found in our study are important for providing proper monitoring and medical care. A high heart rate at admission increased the odds of detection of cardiovascular events after adjusting for comorbidities, cardiovascular risk factors and medications on admission. This finding is consistent with a study conducted in Brazil, which reported higher heart rates in the COVID-19 patients with pre-existing CVDs who had endothelial dysfunction.45 The use of antiviral medications in our COVID-19 patients significantly reduced the rate of cardiovascular events. This finding is supported by a multicentre study in Taiwan, which found antiviral agents were associated with fewer adverse cardiovascular outcomes.46 We could not find any studies which directly supported the mechanism of intravenous fluid increasing the odds of the occurrence of cardiovascular events after admission but a lower risk of mortality in COVID-19 patients; however, it may be related to fluid volume and pressure overloads resulting in cardiac stress and then cardiovascular events.47 Likewise, the use of antibiotics may be coincidently associated with secondary bacterial infections, which could result in a higher incidence of cardiovascular events.
Our study found that managing COVID-19 patients with corticosteroids was associated with higher mortality, which was different from the findings of a study from Italy which collected data until May 2020 and found that corticosteroids had no effect on hospital mortality in patients with COVID-19.48 In contrast, a multicentre study conducted in 36 hospitals in France during January–May 2020 found that dexamethasone had significant association with reducing mortality in patients with respiratory support, defined as the use of oxygen or mechanical ventilation, but not in those without respiratory support.49 Likewise, the trial in the UK included COVID-19 patients admitted from March to June 2020, with 60% of patients receiving oxygen therapy and 16% receiving mechanical ventilation. The trial showed that the use of dexamethasone for up to 10 days in patients receiving mechanical ventilation or oxygen reduced 28-day mortality.50 In addition, the use of corticosteroids for treatment of myocarditis associated with COVID-19 was shown to improve outcomes in a systematic review.48 The different findings of our study compared with previous studies may be explained through several factors: different patient characteristics, with our study having less severe cases as only four patients required mechanical intubation; different study periods, which were influenced by various variants of SARS-CoV-2, the availability of vaccines, the incidence of myocarditis; and different clinical managements, including the use of corticosteroids or respiratory support.
Some findings of our study were different from previous studies due to the rapidly changing nature of the virus, demographical variations of the affected populations, diverse health systems or healthcare management of the COVID-19 patients and a lack of quality research methodology due to the urgent need for COVID-19 research to combat the pandemic and save lives. The sample size in this study was larger than another study conducted in different hospitals in Bangladesh.51 Even though the degrees of SARS-CoV-2 virulence and its clinical outcomes changed over the years of the pandemic, the lessons learnt in terms of optimal treatments of potential CVDs, clinicians’ safety, healthcare providers’ capacity and prompt adoption of relevant health policies should be taken into account,52 considering each patient’s cardiovascular risk factors and comorbidities.
The study had some limitations. First, the data analysed were limited from October 2020 to July 2021 when Bangladesh had been going through the second and third waves of COVID-19 infections and imposed stricter measures to control the devastating situation. Second, some data of the sociodemographic characteristics of the patients were not reported due to restrictions in the COVID-19 dedicated hospitals and the characteristics were collected from the patient history sheets. Third, at the time of data collection for this study, the COVID-19 vaccine was not yet available. Among the patients included in our study, cardiovascular events and death were reported in 18.7% and 12.6% of cases, respectively. Subsequent studies on COVID-19 patients have since shown a decrease in such events among vaccinated individuals compared with those who were not vaccinated.53 54 Fourth, the majority of pre-existing comorbidities in COVID-19 patients in this study were hypertension and diabetes mellitus, which predominated the results of two studied outcomes. Fifth, the laboratory parameters were not included in this secondary analysis as those were analysed and published in the main findings of original study. Finally, the severity of COVID-19 infection related to mortality depended on variants present during the study periods, which may not be generalisable to the current situation.
Conclusions
Our findings support the hypothesis that comorbidities in COVID-19 patients have an adverse impact on the occurrence of cardiovascular events and an association with increased mortality. Close attention to the clinical information of patients diagnosed with COVID-19 on hospital admission is crucial and proper management with respect to the administration of intravenous fluids and antivirals, antibiotics and corticosteroids as well as monitoring for the presence of cardiovascular events should be standard procedure for such patients. Even though the serious health consequences of COVID-19 nowadays may be ameliorated by changing viral virulence and available treatments, the results of this analysis can be a lesson learnt concerning COVID-19 regarding the effects of comorbidities on cardiovascular events and mortality to assist with well-prepared health system responses and clinical care services for future pandemic-like COVID-19 situations.
Data availability statement
Data are available on reasonable request. The datasets used and/or analysed during the current study are available from the organisation of the first author on reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study of secondary analysis was approved by the Human Research Ethics Committee, Faculty of Medicine, Prince of Songkla University (REC.66-055-18-1) and the original study was approved the Bangladesh Medical Research Association (BMRC). Participants gave informed consent to participate in the study before taking part in original study.
Acknowledgments
We would like to express our appreciation to the World Health Federation (WHF) Global Study COVID-19 and Cardiovascular Disease Survey funded by the WHF, and an unconditional research grant by Sanofi and Pfizer for the approval to use the Bangladesh data for this secondary analysis. We also thank Dave Patterson, International Affairs Office, Faculty of Medicine, Prince of Songkla University, for language editing.
This post was originally published on https://bmjopen.bmj.com