Early Sepsis Care Bundle Does Not Reduce Mortality

By Will Boggs MD

April 07, 2020

NEW YORK (Reuters Health) - A care bundle for early sepsis management might not reduce mortality or organ dysfunction in patients with community-onset or hospital-onset sepsis, a retrospective study suggests.

"The SEP-1 sepsis bundle, when administered in its complete form, was not associated with benefit in any of the cohorts we studied," Dr. Jonathan D. Baghdadi from University of Maryland School of Medicine in Baltimore told Reuters Health by email. "Considering the burden placed on medical centers to report on this measure and that performance on this measure is driving reimbursement, the lack of an associated benefit should lead us to rethink how we assess quality of care in sepsis."

The Early Management Bundle for Severe Sepsis/Septic Shock (SEP-1), a quality metric based on a care bundle for early sepsis management, is recommended for all patients with sepsis, despite the lack of evidence supporting its effectiveness, especially in hospital-onset sepsis, Dr. Baghdadi and colleagues note.

For the study, they used electronic medical records of four teaching hospitals to assess the association of SEP-1 with mortality and organ dysfunction (using days requiring vasopressor support as a proxy) in patients with community-onset or hospital-onset sepsis.

They evaluated the four components of the SEP-1 bundle required within three hours of arrival in the emergency department (for community-onset) or inpatient unit (for hospital-onset): blood culture results, broad-spectrum intravenous antibiotic treatment, serum lactate level testing, and intravenous fluid treatment if the blood pressure was low or if the lactate level was elevated Follow-up serum lactate level testing and initiation of vasopressor treatment is also required.

Among the 4,108 patients with community-onset sepsis, 1,654 (39.9%) received SEP-1-adherent care; 281 (12.2%) of the 2296 patients with hospital-onset sepsis received SEP-1-adherent care, according to the online report in JAMA Internal Medicine.

Receipt of the SEP-1 was not associated with reduced mortality or decreased vasopressin days in the total overall sample or in the community-onset or hospital-onset subgroups.

In the community-onset sepsis group, measurement of serum lactate level was associated with a significant 7.61% reduction in mortality risk, and blood cultures and broad-spectrum intravenous antibiotic treatment were associated with significantly fewer vasopressor days.

In the hospital-onset sepsis group, only the use of broad-spectrum intravenous antibiotic treatment was associated with a significant reduction in mortality risk (by 5.20%).

When the overall sample was rebalanced to account for differences in patient-level clinical variables, broad-spectrum intravenous antibiotic treatment was associated with a reduction in the risk of mortality and fewer vasopressor days. Collection of blood cultures was associated with significantly decreased vasopressor days, whereas serum lactate level testing and intravenous fluid treatment were not associated with either outcome.

"I would personally advocate that the SEP-1 bundle should be simplified, and antibiotics should be prioritized," Dr. Baghdadi said. "For instance, a trimmed-down bundle including timely antibiotics, blood cultures, serum lactate, and vasopressors would likely make more sense to clinicians and decrease the burden placed on hospital administrators."

"If the bundle itself is simple, more of the efforts related to adherence can focus on rapid recognition of sepsis," he said. "However, our study did not compare the existing SEP-1 protocol to any alternatives, so we cannot conclude from our data whether these modifications would really improve patient care."

Dr. Baghdadi added, "Hospital-onset sepsis and community-onset sepsis should be considered distinct clinical entities, and care protocols and quality metrics should reflect that distinction."

"The study shows that one size does not fit all when determining the best treatment for potentially septic patients," write Dr. Hannah Moreira and Dr. Richard Sinert from SUNY Downstate Medical Center and Kings County Hospital in New York City in a related editorial.

"We earnestly hope that studies such as the one performed by Baghdadi et al. demonstrating no mortality benefit with the use of SEP-1 bundle would influence CMS to consider the importance of individualized care for these critically ill patients," they conclude. "Sepsis care should be limited to only those who appear to be in septic shock, emphasizing early antibiotics and acknowledging physician judgment concerning the volume of fluids infused and the role of lactate measurements in the care of these patients."

But Dr. Joseph A. Carcillo from University of Pittsburgh School of Medicine, who has researched various aspects of sepsis and its treatment, told Reuters Health by email, "I am surprised that JAMA is publishing this article as is. It shows that individual components of a bundle are associated with favorable outcome but when all are used together they are not."

"This could mean that the sickest patients tended to receive the entire bundle of care because the physicians thought they were sick, and the less sick patients did not; however, the authors and the journal want the reader to believe they have used statistical techniques (score matching) which have controlled for this possibility," he said. "I am not buying it."

"Based on this paper, I do not think SEP-1 should be modified," Dr. Carcillo said. "Randomized trials are needed to determine what components of bundles are effective and (which) do not work. No more administrative data and registry data studies."

SOURCE: https://bit.ly/34gPz1Q and https://bit.ly/2JHKxlk JAMA Internal Medicine, online March 6, 2020.