How doctors can predict which patients are more likely to die of COVID Several studies have come out about the COVID prognosis.

As a result, we now have a powerful way to predict which patients with COVID pneumonia are more likely to die and require longer stays in the hospital. We already know that less than 15% of people who get COVID have a severe disease to the point of requiring hospitalization. Less than 5% of people who get coronavirus require ICU.

For those with COVID pneumonia who require ICU, and for some of the other patients in the hospital who might not necessarily need ICU, these are the patients who have the cytokine storm that we keep hearing about, meaning the immune system is going haywire.

There is a ton of inflammation taking place in the body….especially the lungs. But now, we have a more specific way of predicting who is more likely to develop cytokine storm, and therefore more likely to die, and who is more likely to require a longer stay in the hospital….and it’s based on the bloodwork we get when patients are hospitalized.

When we get bloodwork, we’re doing various lab tests. It is most commonly a complete blood count, CBC, and a complete metabolic panel or CMP. We often check other things like troponin, d-dimer, LDH, ferritin, and CRP. A complete blood count is specifically looking at 3 things: hemoglobin levels, platelets, and white blood cells. We can use the total number and the breakdown of the percentages of the types of white blood cells, meaning what percentage are neutrophils, monocytes, lymphocytes, eosinophils, and basophils. Lymphocytes generally make up about 15-45% of WBC. This is very important as it pertains to COVID, and you’ll see why in a little bit.

The CMP stands for the complete metabolic pan looking at sodium, potassium, chloride, bicarbonate, blood urea nitrogen, creatinine, glucose calcium, bilirubin, albumin, and liver enzymes like AST ALT alkaline phosphatase. Some of these also have important implications when it comes to COVID disease severity. Because In the recent study done at Temple University, they analyzed over 500 hospitalized COVID patients, all of which had inflammation in their lungs seen on a CT scan of the chest (show). They determined the criteria for COVID cytokine storm. For patients who met these criteria, their length of stay in the hospital was 15 days, compared to 6 days if they did not meet the criteria.

Mortality was 28.8% in the group that met the criteria, vs. 6.6% in those who did not. So if they met cytokine storm criteria, their likelihood of dying was 4 times higher. And how accurate was this prediction model? Pretty accurate, as this study had a specificity of 79% and a sensitivity of 85%, which is a lot of predictive power. Most of the patients who met the criteria for CS did so at the time of admission or shortly after. This suggests that there is an early and rapid progression in these patients and that there is a low likelihood of developing cytokine storms after 10 days of hospitalization.

So based on these specific lab parameters, this is how doctors can predict which patients are more likely to die of COVID. I find that this study correlates with what I’ve seen in my personal experience with hospitalized covid patients. And yes, the man whose CT scan I showed you earlier in this video met all of these criteria for cytokine storm, and he did so as soon as 24 hours of being admitted to the hospital.

Doctor Mike Hansen, MD Internal Medicine | Pulmonary Disease | Critical Care Medicine