Amid an uncertain time for the future of COVID-19, and whether new variants will appear, a recently published prediction formula — built using data from 1,102 hospitalized COVID-19 patients in 20 hospitals nationwide — now gives a better estimation about an individual’s projected disease outlook and could help the research community understand why some individuals have a level of natural immunity and why others are more susceptible to severe infection.
The paper, recently published in the journal Frontiers in Medicine, is the latest from the IMMuno Phenotyping Assessment in a COVID-19 Cohort (IMPACC), a research group studying immune responses of adult patients treated for COVID-19 at 20 sites nationwide to better understand what markers are associated with severe cases. With this knowledge, the group hopes to inform better treatments and care.
In the current work, the group studied 123,000 variables that might influence disease severity — including attributes from the patient’s genes, proteins in the body, amount of the COVID-19 virus present in the body and many other characteristics — to determine which factors best predict progression or death due to COVID-19 after 28 days in the hospital.
“Our findings may have clinical applications, but will likely have much greater implications for researchers looking to better understand SARS-CoV-2,” said senior author Mary Ann Comunale, EdD, an associate professor in the College of Medicine. “We developed formulas that improved on current methods of predicting COVID-19 severity and death, and our work enables accurate and rapid prediction using easily obtained clinical markers. In addition, our work identified biomarkers that provide direction for future research in understanding the cellular mechanisms behind the disease.”
As part of the study, data was collected on peripheral oxygen saturation (measured via pulse oximeter) and fraction of inspired oxygen. This ratio, called SpO2/FiO2, which is a demonstrated predictor of COVID-19 outcomes, measures the amount of oxygen an individual is inhaling and how much of that oxygen is actually getting into their blood. The authors found that SpO2/FiO2 was the most accurate predictor for severity and death from COVID-19.
Overall, SpO2/FiO2, age and body mass index (BMI) were the three attributes that most accurately predicted whether a patient’s case resulted in death within 28 days of being admitted to the hospital. As a prediction method, considering these three metrics together outperformed the commonly used sequential organ failure, or SOFA, score, which uses clinical data to predict risk of death in critically ill patients.
“The bottom line is that continued use of the non-invasive SpO2/FiO2 measure is still very effective at predicting severity and mortality from COVID-19,” said Comunale. “There are thousands of other factors that can be considered, but only increase that accuracy about 1-2% for each predictor. Generally, adding these other attributes to the formula isn’t significant enough to change an individual patient’s care plan. That said, the additional predictors offer the field of COVID-19 study important insights about what gives individual patients a better immune response than other patients.”
As patients selected for the study were hospitalized from May 2020 to March 2021, before the COVID-19 vaccine was available, the authors suggest these findings may help researchers understand what factors offer protection against the disease. For example, the authors show increased levels of plasma TNFRSF11B correlates with COVID severity. TNFRSF11B has also been linked to severity or progression of other lung diseases including COPD and asthma.
“This discovery is not likely to change how COVID-19 is treated in a clinical setting,” according to lead author Jintong Hou, PhD, who participated in the research while completing a PhD in biostatistics from Drexel. “But it is another step toward understanding a disease that killed millions worldwide and continues to afflict patients.”
The data may also be timely, the authors argue, as much of the U.S. population experiences waning immunity. Changing vaccination policy — including conflicting recommendations over whether those under 18 should receive a COVID vaccine — signals that many in the United States will have less-than-optimal protection against future variants through vaccination.
The American Academy of Family Physicians and the American Academy of Pediatrics recommend COVID-19 vaccination to children and pregnant women. Among the more recent studies showing benefits from COVID-19 vaccination in this population, a study last year in Pediatrics found that babies born to pregnant mothers who took the mRNA COVID-19 vaccine while pregnant were protected against the disease for at least their first six months of life.
COVID-19 is no longer considered a public health emergency by the World Health Organization. There are still hospitalizations and deaths from the disease every day in the United States, but as deaths and hospitalizations from the disease decline, the Centers for Disease Control and Prevention has shifted resources away from prevention strategies. Hospitals are no longer mandated to report COVID data to the CDC and the CDC has shifted resources from tracking into other areas. The CDC offers some insights through its wastewater monitoring, which may give clues about possible hotspots for SARS-CoV-2.
Although research like this keeps the focus on a pandemic the world might rather forget, the research team’s continued work offers better understanding of COVID-19 patient responses and may improve our response to future public health crises.
This research was supported by funding from the National Institutes of Health’s National Institute of Allergy and Infectious Diseases (NIAID), National Science Foundation and Drexel’s College of Medicine.