Sepsis has become a sneaky, yet lethal killer in the United States. The inflammatory disease kills more people than AIDS, breast cancer, and prostate cancer combined, and is a difficult one to pin down with diagnosis.

An early diagnosis, however, can mean the difference between life and death. This has made the need for swift recognition and treatment of sepsis a top medical issue, something researchers at Johns Hopkins University are making strides to address.

A new computer-based method has correctly predicted septic shock in 85 percent of cases, without increasing the false positive rate of current prediction methods. More than two-thirds of the time, the method could predict septic shock before any organ dysfunction — this is a 60 percent improvement over our existing screening methods.

"But the critical advance our study makes is to detect these patients early enough that clinicians have time to intervene," said Suchi Saria, an assistant professor of computer science and health policy at Johns Hopkins' Whiting School of Engineering, in a press release. She led the study, which was published Aug. 5 in the journal Science Translational Medicine.

The research is a sign of significant promise in treating sepsis, a condition that kills about a million Americans every year.

"We know a lot of those deaths would likely be preventable if sepsis were diagnosed well before it develops into septic shock and organ failure,” said Peter J Pronovost, who directs the Armstrong Institute for Patient Safety and Quality at Johns Hopkins Medicine in a press release. "Right now, much of sepsis is invisible until someone is on death's door.”

Every passing hour before sepsis patients receive antibiotics, he said, “correlates strongly with risk of death."

The study utilized the health records of 16,234 patients admitted to intensive care units including surgical, medical and cardiac units at Boston’s Beth Israel Deaconess Medical Center between 2001 and 2007. The researchers combined 27 different factors to come up with something called a Targeted Real-Time Early Warning Score (TREWScore), which measures the risk of septic shock.

"One strength of this approach," said Katharine Henry, a Ph.D. student in Saria's lab and first author of the study, "is that all of our inputs are routinely collected. You don't need specialized new measurements."

This method differs in several ways from previous attempts to predict septic shock — it takes account of more health indicators, factors in and controls for confounding elements and utilizes a larger data pool.

David Hager, one of the study’s co-authors, said that the algorithm could be used to alert medical professionals about a patient risk of septic shock by being programmed into electronic health records.

"The tricky issue is thinking about how the clinical team is provided with the information," Hager said. A hospital's electronic health records system could be set up to convey alerts to clinicians via pager or cell phone at regular intervals, he said.

Saria said that the methods are reaching a point at which they can be of real help to clinicians, especially when it comes to recognizing the subtle hints that sepsis could be developing.

Source: Henry K, et al. Computer Algorithm Can Forecast Patients' Deadly Sepsis. Journal of Translational Medicine. 2015.