Scientists have revealed that secondary bacterial infections in the lungs are extremely common among those infected with COVID-19, affecting nearly half of the patients who require mechanical ventilation. By applying machine learning to data from medical records, researchers at Northwestern University Feinberg School of Medicine found that secondary bacterial pneumonia (pneumonia), which has not been well understood until now, was a primary driver of death among COVID-19 patients, exceeding even the mortality rates from the viral infection itself.
The researchers also found evidence that COVID-19 does not cause the "cytokine storm" that was often believed to cause death. Senior researcher Benjamin Singer, an assistant professor of medicine at Northwestern Feinberg, stated in a recently published paper: "Our study highlights the importance of preventing, identifying, and aggressively treating secondary bacterial pneumonia in patients with severe illness and acute pneumonia, including those with COVID-19." They discovered that nearly half of COVID-19 patients develop secondary bacterial pneumonia related to mechanical ventilation.
Singer added, "Those who recover from secondary pneumonia are more likely to survive, while those whose pneumonia went untreated were more prone to death. Our data suggest that the mortality rate directly associated with the virus itself is relatively low, but other factors occurring during an ICU stay, such as secondary bacterial pneumonia, counterbalance that."
Singer noted that the study's findings also disprove the cytokine storm theory, saying that the term "cytokine storm" refers to severe inflammation leading to organ failure in the lungs, kidneys, brain, and other organs. He explained, "If this were true, if a cytokine storm were behind the prolonged hospital stays we observe in COVID-19 patients, we would expect to see frequent transitions to conditions characterized by multi-organ failure. That is not what we have seen."
The study analyzed 585 patients in the ICU at Northwestern Memorial Hospital suffering from acute pneumonia and respiratory failure, of which 190 had COVID-19. The scientists developed a new machine learning approach that aggregates the days of ICU patients similar to clinical cases based on electronic health record data. This new approach allowed for researching how complications such as bacterial pneumonia affect the course of the disease.
