Article: Could a blood test predict whether coronavirus will kill you? Researchers find five signs for inflammation and blood vessel problems that could tell doctors which COVID-19 patients are at risk for fatal infection

  • Researchers at George Washington University tested the blood of 299 COVID-19 patients hospitalized at the school’s facility 
  • Of those, 200 had all five biomarkers linked to inflammation and blood vessel problems 
  • Higher levels of any or all of the biomarkers were linked to greater risks of ICU admission or needing to be put on a mechanical ventilator 
  • Patients with levels the highest levels of two of the five biomarkers were at greatest risk of death, suggesting blood tests could indicate who gets sickest 


PUBLISHED: 19:22 EST, 11 August 2020 | UPDATED: 10:41 EST, 12 August 2020

Researchers at George Washington University (GWU) believe that a simple blood test could predict which coronavirus patients could become deathly ill. 

The scientists have identified five biomarkers that indicate risks of complications like inflammation and bleeding disorders that make someone more likely to die if they contract coronavirus. 

High levels of two of these blood indicators, in particular, are linked to far greater odds of dying from the infection. 

The GWU team believes a blood test for these biomarkers could give doctors a clearer picture of who might need ventilator support or early aggressive treatment with a more finely tuned tool than general risk factors like age and underlying conditions. 

A blood test for five biomarkers of inflammation and blood vessel malfunctions could help doctors predict which coronavirus patients are at greatest risk of death, a new study suggests

People over the age of 65 and those with underlying conditions are typically less able to fight off any infection, not just COVID-19. 

But coronavirus has proven deadly to scores of otherwise healthy, relatively young people too – and scientists are still not sure exactly why some COVID-19 patients quickly spiral downward and others have no symptoms at all. 

And knowing who might need the most aggressive care is critical for hospitals when the threat of drug and supply shortages looms. 

New York City narrowly avoided a shortage of ventilators when it became the global epicenter of the pandemic in March and April. 

Several hospitals in hard-hit parts of Texas completely ran out of beds for coronavirus patients (or others) as cases spiked there in June and July. 

Even though Texas and other sunbelt states are seeing fewer new cases per day, hospitalization rates remain high. Florida reported a new record high number of coronavirus deaths in a single day on Tuesday and cases continue to climb in states like Georgia, Alabama, Illinois and Illinois. 

And nationwide, there are still shortages of drugs needed for patients on mechanical ventilators and of the only FDA-authorized treatment, remdesivir. 

Treating any and all patients sick enough to need to be hospitalized as early as possible is the best course of action – but it’s also a luxury doctors may not have the needed supplies are so precious. 

Grim though it may sound, health care providers may have to choose for one patient to get treatment over another – and a blood test could make these decisions both easier and more likely to be the correct ones. 

‘When we first started treating COVID-19 patients, we watched them get better or get worse, but we didn’t know why,’ said Dr Juan Reyes, study co-author and assistant professor at GW School of Medicine.

‘Some initial studies had come out of China showing certain biomarkers were associated with bad outcomes. There was a desire to see if that was true for our patients here in the US.’ 

The data that Dr Reyes and his colleagues saw out of China inspired them to assess the blood levels of five biomarkers in COVID-19 patients at GW Hospital. 

The biomarkers they looked at were: 

  • IL-6, which is short for interleukin 6, one of several cytokine immune cells that raises the alarm to other parts of the immune cell and can indicate out of control inflammation. 
  • D-dimers, which are bits of degraded protein detectable in the blood after a clot disintegrates, and signal that the virus may be attacking blood vessels. 
  • CRP, or C-reactive protein, which is released by the liver in response to inflammation. 
  • LDH, or Lactate Dehydrogenase, an enzyme in lactic acid that the body sends to heal damaged tissues. 
  • Ferritin, a protein that helps the body’s cells store iron. Iron in turn, is crucial for healthy red blood cells that carry oxygen throughout the body. Too much or too little ferritin can signal anemia or an infection that’s impairing blood cell function.     

Of the 299 COVID-19 patients whose blood they tested, the researchers found all five biomarkers in 200. 

Patients with higher levels any or all of these biomarkers were more likely to need to be treated in the ICU or put on ventilators. 

High LDH levels (greater than 1200 units/l) or high D-dimer levels (greater than 3 μg/ml) predicted the greatest risks that patients would die of coronavirus. 

‘We hope these biomarkers help physicians determine how aggressively they need to treat patients, whether a patient should be discharged, and how to monitor patients who are going home, among other clinical decisions,’ said Dr Shant Ayanian, an assistant professor and the study’s first author.

Koco News: Local researcher forming team to study how genetics influences severity of COVID-19

Patrina Adger Reporter


There are so many unknowns when it comes to the COVID-19 coronavirus, one being why are so many people getting the deadly disease while others show no signs?

Daniel Brue, a local researcher with General Genomics, is forming a study on how genetics influences the severity of the virus, who is more likely to get it and why.

“We never had as much feedback on a pandemic as what we are having today,” Brue said. “We know there are certain factors that are influencing this.”

Brue is leading the research team from Oklahoma City. He said the information collected from genetic markers and family history is similar to what you’d obtain through or 23and Me matching.

“We can start tracking populations that have more susceptibility to COVID-19 and, ultimately, to other diseases as well,” Brue said.

Brue and his team also will work to figure out how viral patients respond to different treatments based on their genetics.

“That helps in the individual who may become ill, and it also helps their clinician in giving and describing a better treatment,” he said.

Brue told KOCO 5 that they are hoping to get federal support to help fund the study so results can come back quicker.