Can All Sepsis Deaths Be Prevented?

Author: Karen Blum

Anesthesiology News

Many people think that most sepsis deaths could be preventable if hospitals were better at treating the condition, but that may not be the case (JAMA Open Network 2019 Feb 15).

A records review of 568 adult patients who died in the hospital or were discharged to hospice in 2014-2015 from six U.S. academic and community hospitals found that sepsis was present in 300 cases (53%) and the immediate cause of death in 199 cases (35%). However, only 4% of these sepsis deaths were found to be definitely or likely preventable, with an additional 8% considered possibly preventable with optimal clinical care.

Sepsis appears to be the most common cause of death in U.S. hospitals, said Chanu Rhee, MD, MPH, an assistant professor of population medicine at Harvard Medical School and the Harvard Pilgrim Health Care Institute, and an assistant professor of medicine at Brigham and Women’s Hospital, in Boston. However, he said, “our study indicates that most sepsis-related deaths may not be preventable through better hospital-based care because most cases occur in very medically complex patients with severe chronic illnesses.”

Overall, 395 (70%) patients died in the hospital and 173 (30%) died after being discharged to hospice. The majority of those who died and had sepsis—219 patients (73%)—had sepsis on admission.

The median age of those who died was 73 years. Terminal conditions were present in 121 (40%) of the 300 sepsis deaths, most commonly metastatic or progressive solid cancers (20%), refractory hematologic cancer and severe debilitating dementia (5% each), or severe debilitating stroke or severe chronic lung disease (4% each). Underlying causes of death among patients with sepsis included solid cancers (21%), chronic heart disease (15%) and hematologic cancers (10%).

One in eight sepsis deaths were found to be potentially preventable with better hospital-based care, but only one in 25 were judged to be likely preventable.
The findings “do not diminish the importance of trying to prevent as many sepsis deaths as possible, nor the tragedy of having even one preventable death from sepsis,” Dr. Rhee said, “but it emphasizes that to really achieve major reductions of sepsis deaths, we need more innovations in the prevention and care of underlying conditions.”

A logical target for future research would be to look at aspects before hospitalization, he added, particularly whether better preventive care or faster referral to the hospitals could have prevented some of these deaths.

In his own experience in the medical ICU, Dr. Rhee said, a lot of patients with sepsis “are very, very sick. Even though they may receive guideline-compliant care, many of them continue to worsen, which may be a function of underlying chronic illnesses. Sometimes they are so sick, and because they have had these chronic illnesses for so long, the patients and families don’t even want anymore aggressive treatment once they develop sepsis.”

“In people with complicated illnesses, it is quite common not to be sure why they die, whether it was the infection, the inflammatory response to the infection, or their comorbidities,” Dr. Masur said. “That is a huge problem unless you meticulously go over cases one by one.” Even in cases where he and his colleagues had known ICU patients and followed them, they weren’t always sure why the patients had multiorgan failure, he said.

The Society of Critical Care Medicine and other societies have endorsed the Surviving Sepsis campaign to promote evidence-based guidelines to improve the care of septic patients. But there has been some controversy over whether items lumped in a bundle of activities to be completed within the first hour to six hours of presentation—like measuring blood lactate levels or giving “early goal-directed therapy”—make a difference, Dr. Masur said.

“For treating the infectious triggers of bad outcomes, I think the ones that really make a difference are prompt administration of antibiotics, fluids and vasopressors,” he said. “All the rest are unsubstantiated or conjectural.”