Cardiac Injury in COVID-19 Sharply Increases Risk of Death

New data out of Wuhan, China, confirm that cardiac injury is common in patients hospitalized with COVID-19 and, unsurprisingly, is tightly linked to higher mortality.

One in five patients had cardiac injury; among those who did, half died, Shaobo Shi, MD (Renmin Hospital of Wuhan University, China), report in JAMA Cardiology.

Although “mounting evidence substantiates the presence of cardiac injury in patients with COVID-19,” the researchers say, “the association between COVID-19–associated cardiac injury and risk of mortality remains unclear.” They sought to fill that gap by detailing their own experience in the city from which the SARS-CoV-2 virus first emerged mere months ago.

Experts commenting on the paper for TCTMD praised the quality and timeliness of the evidence, noting that it “essentially shows what we expected,” as Mohammad Madjid, MD (UTHealth, Houston, TX), put it to TCTMD. “It comes as no surprise that this virus, like other viruses, like influenza, can affect the heart.”

Likewise, James Januzzi Jr, MD (Massachusetts General Hospital, Boston), observed: “It is not all that surprising that cardiac injury is common among patients with COVID-19 and that it tracks with disease severity.” But of note, Januzzi continued, most of the troponin elevations seen in the COVID-19 setting come in at lower than what’s typically seen with acute MI. “The possibility of myocarditis as an explanation is one potential; some clear cases of COVID-19 myocarditis have been reported, however not all injury may be blamed on myocarditis,” he said in an email.

Frederick Welt, MD (University of Utah School of Medicine, Salt Lake City), also stressed that the study adds valuable information to the overall understanding of the COVID-19 disease process. He cautioned, though, that the current analysis is slightly skewed in that the researchers focused only on patients who had troponin tests and did not address why certain individuals had been tested. “Was there some underlying clinical suspicion? But having said that, I think that the finding in terms of [cardiac injury’s] relationship to mortality is striking,” Welt said.

The cardiac injury seen alongside COVID-19 likely has many causes, all agreed, and clinicians should be cautious in how they proceed. Whether this relationship is causative is impossible to know, Welt noted. As of now, he said, cardiac injury “is another marker of increased risk [but] doesn’t really inform treatment.”

Biomarkers, Complications, and Death

For their retrospective cohort study, Shi and colleagues looked at 786 laboratory-confirmed COVID-19 cases treated at their center between January 20 and February 10, 2020. Focusing on those who had clinical lab, radiological, and treatment data available, the researchers arrived at a total of 416 hospitalized patients in their final analysis. The median time from symptom onset to admission was 10 days.

Mean age in this group was 64 years, and 50.7% were women. Patients’ three most-common symptoms were fever (80.3%), cough (34.6%), and shortness of breath (28.1%).

Patients with cardiac injury tended to be older (median 74 vs 60 years), have more comorbidities, including hypertension (59.8% vs 23.4%), and have higher median leukocyte counts (9,400 vs 5,550 cells/µL). These patients also had higher median levels of C-reactive protein (10.2 vs 3.7 mg/dL), procalcitonin (0.27 vs 0.06 ng/mL), creatinine kinase-myocardial band (3.2 vs 0.9 ng/mL), myohemoglobin (128 vs 39 µg/L), high-sensitivity troponin I (0.19 vs < 0.006), N-terminal pro-B-type natriuretic peptide (1,689 vs 139 pg/mL), aspartate aminotransferase (40 vs 29 U/L), and creatinine (1.15 vs 0.64 mg/dL). On radiography, they had a greater proportion of multiple mottling and ground-glass opacity (64.6% vs 4.5%). All of the above differences were statistically significant.

In terms of treatment, noninvasive mechanical ventilation was used more often in patients with versus without cardiac injury (46.3% vs 3.9%), as was invasive mechanical ventilation (22.0% vs 4.2%; P < 0.001 for both). Cardiac injury also was associated with higher complication rates.

Complication Rates in COVID-19 Patients

ECGs were obtained for 22 (26.8%) of the patients with cardiac injury, with 14 (63.6%) performed when cardiac biomarkers were elevated. “All 14 ECGs were abnormal, with findings compatible with myocardial ischemia, such [as] T-wave depression and inversion, ST-segment depression, and Q waves,” Shi et al note.

Notably, 51.2% of people with cardiac injury died, as compared to 4.5% of those without cardiac injury (P < 0.001).

Using a multivariable-adjusted Cox proportional hazard regression model, the researchers found that cardiac injury was tied to a fourfold higher risk of death based on the time from symptom onset (HR 4.26; 95% CI 1.92-9.49) and a threefold higher risk based on the time from admission (HR 3.41; 95% CI 1.62-7.16). Acute respiratory distress syndrome (ARDS) also was linked to mortality, increasing risk by nearly eightfold (HR 7.89; 95% CI, 3.73-16.66).

‘The 30,000-Foot View’

The question is what to do with this information, particularly when it comes to troponin, Welt said. “The 30,000-foot view . . . is that checking troponin right now doesn’t really have an impact on what you’re going to do therapeutically” for COVID-19 patients.

The 30,000-foot view . . . is that checking troponin right now doesn’t really have an impact on what you’re going to do therapeutically. Frederick Welt

“Troponin is released in a variety of stressful circumstances,” Welt explained, pointing out that prior studies of shown troponin can rise in the setting of ARDS. “If you look at the patients in this particular study and their baseline characteristics, those who had evidence of cardiac injury were substantially more likely to have hypertension, diabetes, and other vascular diseases as well as chronic heart failure. So that does make you think that some of this could be stress-induced.”

It may also be that some patients have primary myocarditis caused by the virus itself. As Welt noted, “There’s a lot of interest in that because the way the virus gets into the cell is through the ACE2 receptor, and that is one that is apparently expressed in cardiac tissue.”

Januzzi further elaborated on the complex interplay between COVID-19 and the heart. “Mechanistically, most of the myocardial injury we see in COVID-19 patients is likely related to direct effects of disease severity, cytokines, vasopressors, hypoxia, and [disseminated intravascular coagulation],” he explained, adding that this is in line with earlier studies showing that cardiac injury is common in critically ill patients.

“Some clinicians have said they measure troponin in their COVID-19 patients to provide guidance on therapy. To be explicitly clear, there are no data yet generated to suggest an elevated troponin provides us with a signal to manage patients differently, but it may,” Januzzi stressed. “My hope would be we might see some data emerge to suggest there's utility here, but until then clinicians should be cautious in how they interpret the result and how they respond to it.”

Reports of “fake STEMI” are increasingly circulating in the cardiology community, and both Januzzi and Welt said it’s crucial to remember that cardiac injury alone isn’t enough to suspect acute MI.

To act on troponin findings, clinicians should make sure there is “corroborating evidence of other cardiac issues” before proceeding, Welt advised. For instance, “if you do see what looks like a STEMI on EKG, that’s a very reasonable starting point for beginning to investigate whether you should do something more. Some of those people will undoubtedly have true STEMIs. And it appears at least anecdotally that for some of those people it may be a manifestation of COVID-19.”

Specifically, globally reduced ejection fraction on echocardiography “might shift you more towards thinking this is a myocarditis picture. Signs of heart failure would obviously move you towards thinking about a more thorough assessment for a true cardiac issue,” Welt suggested.

“It’s standard clinical judgement, but the danger is that people reflexively start ordering tests,” he commented. In COVID-19 patients who are older and have a high burden of coronary disease, unnecessary testing could be especially risky.

As Januzzi noted, “a mistaken diagnosis of MI has ramifications for patients, along with risk for exposure of staff related to further downstream testing that might not be needed in most cases.”

To be explicitly clear, there are no data yet generated to suggest an elevated troponin provides us with a signal to manage patients differently, but it may. James Januzzi

What this study does indicate, Madjid said, is that clinicians should be vigilant not just for pulmonary and lung-related outcomes of COVID-19 but also for cardiac conditions. Through further research, it may be possible to find therapies to address myocardial injury.

“Our organs in the body work in very close unity with each other,” he said. “The more severe the condition becomes, the lungs and heart and liver and kidneys: everything gets affected. And the totality of multiorgan failure contributes to [death from COVID-19].”

Madjid, to TCTMD, marveled at the pace at which evidence is accumulating in this space. “More papers will be coming out and giving us a full picture. This is amazing—this is a disease that 3 months ago we didn’t know about. Now in such a short time, we know what causes it, the genetic code of the virus; we are learning more and more how it’s actually affecting [patients].”

Welt, too, emphasized that the knowledge base is shifting rapidly. “It’s moving so quickly. I think the important thing is going to be to take these individual pieces of data and try and form some sort of a larger picture of this over time.