One in Five Patients Admitted With COVID-19 Die in the Hospital
Prognosis is poor for those with CV risk factors/comorbidities, but complications like PE and DVT are less frequent than expected.
More than 20% of patients hospitalized with COVID-19 in the United States die before discharge or are referred for hospice care, with respiratory complications being the dominant cause of death, according to a snapshot of nearly 15,000 people included in the American Heart Association’s (AHA) COVID-19 CVD Registry.
That grim finding is backed up by a second analysis, this one based on the electronic health records (EHRs) of more than 28,000 patients across 52 health systems, which also showed that 20.7% of patients hospitalized with COVID-19 die in the hospital.
If any good news is to be found, it’s that these studies, both presented during the virtual AHA 2020 Scientific Sessions, suggest the risk of in-hospital cardiac complications might be lower than some others have suggested. For example, 3% to 5% of patients had an MI in the hospital, stroke rates were less than 2%, and deep vein thrombosis or pulmonary embolism (PE) occurred in fewer than 4% of hospitalized COVID-19 patients, researchers report.
James de Lemos, MD (UT Southwestern Medical Center, Dallas, TX), who presented results from the AHA COVID-19 CVD Registry, said patients who develop cardiovascular and thrombotic complications represent an important, high-risk subgroup, but “the in-hospital complications occurred less frequently than we feared when we initially set up the registry and were evaluating early single-center experiences.”
Although these complications occurred less often than expected in both studies, Ann Marie Navar, MD, PhD (UT Southwestern Medical Center), who led the EHR analysis, pointed out that patients with preexisting cardiovascular risk factors and comorbidities represent a particularly vulnerable population. For example, roughly one-third of patients with chronic kidney disease died in the hospital, as did nearly 30% of those with preexisting CAD. More than 20% of patients with diabetes or hypertension also died while in the hospital.
To put the numbers in context, Navar said that 10% of the US population has diabetes and half have high blood pressure.
“This represents about 16 million US adults over age 80 years,” she said. “Given the hospitalization rates that we’re seeing, that translates into a remarkably high risk of death for a large proportion of the population. It seems almost impossible then to be advocating for a strategy of focused protection or herd immunity, as some in the administration are calling for, given the absolutely alarming rates of these different comorbidities and risk factors in the population.”
The AHA registry was launched in early April and includes 109 participating centers. For his analysis, de Lemos focused on 14,889 patients hospitalized with COVID-19 as of September 30, 2020.
Of those included, nearly 60% had hypertension, 40% had diabetes, and 45% were obese (body mass index > 30 kg/m2). Prevalent cardiovascular disease was also common, with 5% of patients having a prior MI, 3% to 4% having previously undergone coronary revascularization, and prior stroke and prior heart failure each noted in more than 10% patients.
In terms of complications, approximately 3% of all hospitalized COVID-19 patients had an MI, fewer than 2% had a stroke, and less than 4% developed deep vein thrombosis or PE, reported de Lemos. Overall, a composite cardiovascular endpoint that included cardiovascular death, MI, stroke, heart failure, or cardiogenic shock occurred in 8.8% of hospitalized patients. In-hospital myocarditis was also rare, occurring in just 0.3% of COVID-19 patients.
Navar and colleagues focused their analysis on 19,584 patients hospitalized with COVID-19 (median age 52 years; 47% female) who were discharged home or died in the hospital as of July 1, 2020. As in the registry data, they found cardiovascular comorbidities and risk factors were common: 31.1% had diabetes, 50.4% had hypertension, 14.3% had prior heart failure, 18.0% had CAD, and 5.6% had end-stage renal disease.
Overall, approximately 5.0% of patients had an in-hospital MI, 2.0% developed PE, and 1.5% had a stroke. However, mortality among those who developed a complication was high. For those who had an MI, PE, or stroke, the in-hospital mortality rate was 55.5%, 26.5%, and 56.0%, respectively. Nearly three-quarters of the patients who required mechanical ventilation, which was necessary in 32.6%, did not survive to discharge. Older patients were most at risk.
“What’s most striking is that starting at about age 55, the risk of death amongst patients who are hospitalized increases almost linearly,” said Navar.
The researchers also saw differences in mortality based on race, with the highest risk of death observed in American Indian/Alaskan Natives (24.1%) followed by Black/African-American (22.7%) patients. Race was not statistically associated with mortality in the study, but Hispanic ethnicity appeared to be associated with a lower risk of death. In a separate analysis from the AHA COVID-19 Registry, which was presented by Fatima Rodriguez, MD (Stanford University, CA), race and ethnicity were also not associated with outcomes, but Black or Hispanic patients “bore a greater burden of morbidity and mortality” because they disproportionately accounted for positive cases.
Navar said understanding which COVID-19 patients are most at risk is critical when a vaccine eventually comes online, though she stressed it’s also important for people to know their own risk of complications so that they can make appropriately informed choices. Additionally, understanding risk is important for healthcare teams so they can give prognostic information to patients once they’re hospitalized, she said.
De Lemos JA, on behalf of the COVID-19 CVD Registry. Design, implementation, and initial results. Presented at: AHA 2020. November 17, 2020.
Navar AM. Impact of cardiovascular disease on outcomes among hospitalized COVID-19 patients: results from > 28,000 patients across the United States. Presented at: AHA 2020. November 17, 2020.
- De Lemos reports research grants and/or honoraria from Roche Diagnostics, Ortho Clinical Diagnostics, Amgen, Regeneron, Novo Nordisk, Siemens Diagnostic, Janssen, Quidel Cardiovascular, and Eli Lilly.
- Navar reports serving as an unpaid advisor to Cerner Corporation.