Infection Site Predicts Death in Septic Shock

medpage

"Heterogeneity is a major problem," he said. "Some of these patients have cancer, some do not; some are older, some are younger. All these patients are in an intensive care unit. They are all critically ill."



PHILADELPHIA -- The anatomic site where an infection originates seems to predict mortality among patients diagnosed with septic shock, researchers reported at the American Thoracic Society conference.
In scouring a database that includes 8,000 septic shock patients in the U.S., Canada, and Saudi Arabia, those ICU patients with hydronephrosis, for instance, had a mortality of 21.1%, said Peter Dodek, MD, MHSc, professor of medicine at the University of British Columbia in Vancouver. But patients with ischemic bowel-caused infections experienced a mortality of 77.8%, he said.
"As many people realize, septic shock is a big issue in critical care medicine," Dodek said at a press conference sponsored by ATS. "Septic shock is associated with a high mortality -- in this study overall mortality was 52.4%."
He said that recently the mortality is decreasing, but "there are a lot of concerns because we haven't found a silver bullet to help treat septic shock."

A major problem in finding a suitable treatment is the heterogeneity of patients, which prevents researchers from focusing on a particular set of patients for clinical trials. With that in mind, Dodek said he and his colleagues asked the question: Are there meaningful differences in hospital mortality among patients who have septic shock, stratified by the location of the original infection?
The answer turns out to be Yes. Their research revealed the following sources and mortality rates:

• Disseminated infections, 84.5%
• Spontaneous bacterial peritonitis, 76%
• Toxic megacolon, 68.3%
• Other abdominal infections, 66.7%
• Pulmonary infections, 54%
• Pyelonephritis, 34.5%
• Enterocolitis/diverticulosis, 28%

Dodek's study used a dataset that was established to examine patients who have septic shock over the past 20 years, beginning in 1989. Even after using multiple regression analysis and adjusting for age, degree of illness at admission and multiple other factors, Dodek said the differences in infection sites still remained.
"There are meaningful differences in hospital mortality by source of infection," he told MedPage Today. "This may help us to stratify patients into groups for clinical trials of sepsis. Those with the highest risk may be the ones we want to try these disease-modifying agents on." But he suggested that it may be wise to think about scaling back on antibiotics for those who have a lower mortality.
"At the moment the tools we have to treat septic shock are early detection, early use of antibiotics which had been shown in some observational studies to make a difference and there have been a variety of drugs and other agents that have been tested but it is still a big conundrum," he said.
Gary Martin, MD, associate professor of medicine at Emory University in Atlanta, who served as the press conference moderator at which Dodek presented his work, told MedPage Today, "Septic shock is a condition that is difficult to study. You want to be able to target a population where you will have the best chance for intervention. This study may help us understand the mechanism and risk of septic shock."
"Heterogeneity is a major problem," he said. "Some of these patients have cancer, some do not; some are older, some are younger. All these patients are in an intensive care unit. They are all critically ill."