July 05, 2023

COVID-19 bacteremic co-infection is a major risk factor for mortality, ICU admission, and mechanical ventilation

Written by

featured discovery icon

Amit Gaggar, M.D., Ph.D.Amit Gaggar, M.D., Ph.D.Nathaniel Erdmann, M.D., Ph.D., Amit Gaggar, M.D., and Matthew Might, Ph.D., are the latest winners of the Heersink School of Medicine's Featured Discovery. This initiative celebrates important research from Heersink faculty members.

The team’s study, “COVID-19 bacteremic co-infection is a major risk factor for mortality, ICU admission, and mechanical ventilation," was recently published in Critical Care.

A collaborative study between UAB and the Ochsner Louisiana Health System (OLHS) has shown that bacterial co-infection acquired from the community plays a significant role in the mortality of COVID-19 patients.

“While bacteremic co-infections in COVID-19 patients are rare, accounting for less than four percent of hospital admissions, our findings indicate a significantly elevated risk of morality,” said Gaggar, co-senior principal investigator of the study and professor in the Division of Pulmonary, Allergy, and Critical Care Medicine. “Specifically, COVID-19 patients with these co-infections face an alarming 25 percent mortality rate within 30 days in UAB patients and a relative risk of 20 percent at Ochsner Louisiana State University Health Shreveport (OLHS).”

The study found that an elevated neutrophil-to-lymphocyte ratio (NLR) of 15 or higher within 48 hours of hospital admission strongly indicates co-infection. When co-infection was present, it significantly increased the risk of in-hospital mortality, ICU admission, and the need for mechanical ventilation in both cohorts.

"These findings underscore the significance of bacteria in SARS-CoV-2 mortality and underscore the potential of the neutrophil-to-lymphocyte ratio as a readily accessible prognostic biomarker for bacterial co-infection and, consequently, disease severity,” said Might, co-senior principal investigator, professor in the Department of Medicine, and director of Hugh Kaul Precision Medicine Institute.

The team found that the risk associated with bacterial co-infection surpassed previously known risk factors for COVID-19 mortality and remained consistent across different waves of SARS-CoV-2 variants during the pandemic.

Nathaniel Erdmann, M.D., Ph.D.Nathaniel Erdmann, M.D., Ph.D."These findings strongly indicate a significant interaction between bacterial pathogens, the SARS-CoV-2 virus responsible for COVID-19, and their collective influence on clinical outcomes”, said Erdmann, co-senior principal investigator and assistant professor in the Division of Infectious Diseases.

Michael John Patton, first author of this study, is an aspiring clinician and biomedical informatician enrolled in UAB's Medical Scientist Training Program for M.D.-Ph.D. trainees.

“The rapid adoption of electronic medical records since 2010 has opened new avenues in biomedical informatics by offering opportunities to discover actionable biomarkers for enhanced patient care,” said Patton. “Biomarkers such as elevated NLR can provide real-time guidance to clinicians for optimal patient care and enable researchers to collect valuable clinical samples for multi-omic translational experiments, unraveling the molecular mechanisms of bacterial co-infection and exploring potential therapeutics.”

Read more from UAB News about the team’s work.

The Heersink communications staff sat down with graduate student, Michael Patton, to gain insights about this study, UAB, and the science community.

Q: What compelled you to pursue this research?

In my previous research at the UAB Precision Medicine Institute (UAB-PMI) under the guidance of Dr. Matthew Might, my focus was primarily on developing artificial intelligence tools to aid in diagnosing and treating patients with rare or undiagnosed diseases. However, when it became evident in March 2020 that the COVID-19 pandemic was a persistent issue, I decided to shift the focus of my thesis research toward data science, machine learning, and outcome modeling using electronic medical record data from UAB Hospital.

Shortly after this shift, I had the opportunity to collaborate with a diverse group of physician-scientists (Dr. Nathan Erdmann and Dr. Amit Gaggar) and molecular biologists (Dr. Kevin Harrod and Dr. Carlos Orihuela) who were enthusiastic about contributing valuable clinical information to UAB's COVID-19 initiatives. As we discovered the association between bacterial co-infections in COVID-19 patients and adverse outcomes at UAB, my determination to validate our initial findings in an external healthcare system grew.

Q: What do you find makes the science community here unique?Matthew Might, Ph.D.Matthew Might, Ph.D.

With Dr. Might's and Dr. Gaggar's support, we collaborated with researchers at the Ochsner Louisiana Health System to validate our results using a dataset twice the size. Within a few months, we were able to confirm our findings externally. I am grateful for the supportive and collaborative environment cultivated by the UAB faculty, which made it all possible.

Q: What was your most unexpected finding?

The aspect of our research collaboration that surprised me the most was the remarkable reproducibility of our results between the UAB hospital dataset and the much more extensive statewide network within the Ochsner Louisiana Health System. We found remarkable consistency across both datasets, ranging from the outcome modeling and the elevated NLR biomarker to identifying pathogens in patient blood cultures. This level of agreement was unexpected and astonishing.

A few months after we published our findings, we received further validation when a research study conducted by the Centers for Disease Control and Prevention (CDC) involving 14 states reported an almost identical discovery. Witnessing the reproducibility of our results not only in our own hands but also by other respected researchers was undeniably the most gratifying aspect of this collaborative endeavor.

Q: How do you feel your research will impact the science community?

Michael Patton Michael John Patton The widespread adoption of electronic medical records since 2010 has opened up a new realm in biomedical informatics, offering immense potential for discovering innovative and clinically relevant biomarkers that can enhance patient care. Research studies like ours contribute in two significant ways. Firstly, biomarkers, such as the elevated NLR, can provide real-time information to clinicians by aiding them in making informed decisions about patient care. Secondly, these biomarkers provide researchers with valuable opportunities to gather clinical samples for multi-omic translational experiments to unravel the molecular mechanisms underlying bacterial co-infection and potentially identify novel therapeutic approaches. It combines the integration of medical knowledge, informatics, and cutting-edge research to improve patient outcomes and drive medical advancements.