That it’s available only on the Beckman Coulter platform is a limitation. “We use a Sysmex platform, so I can’t run this in my laboratory unless I have a dedicated platform just for that,” he said. It also has relatively low sensitivity and specificity compared with other emerging sepsis tools, and pediatric data are limited.
The Intellisep test by Cytovale is a rapid cellular host-response diagnostic that leverages the biochemical properties of white blood cells to stratify patients based on their probability of having sepsis.
Results are presented as a numerical value between 0.1 and 10.0, which is further stratified into three discrete bands. Patients with sepsis typically have higher scores, falling into band three. It’s indicated for use in adult patients who present to the ED with signs and symptoms of infection, and as an aid in the early detection of sepsis with organ dysfunction manifesting within the first three days after testing. “So now it’s not just what is the volume of the monocytes but what they are doing,” Dr. Farnsworth said.
The hallmark study for this marker evaluated the test in adults with signs and symptoms of infection presenting to seven U.S. EDs (O’Neal HR, et al. Acad Emerg Med. 2024;31[9]:883–893). To be considered positive for Sepsis-3, patients had to have infection at the time of presentation to the ED and organ dysfunction caused by a dysregulated host response to the infection, manifesting within three days of the ED visit. All cases were reviewed by at least two independent physicians for each of the required components of sepsis. Patients were included in the study if there was consensus in determining sepsis per the Sepsis-3 definition and severe sepsis per the Sepsis-2 definition between the independent panel and the site-level physician. “They then assessed clinical outcomes, including mortality and length of stay,” Dr. Farnsworth said.
Among patients with signs or suspicion of infection, the test achieved similar or better performance compared with other indicators in identifying patients at high risk for sepsis (specificity > 83 percent) and significantly superior performance in identifying those at low risk (sensitivity > 92 percent; zero percent sepsis-associated mortality). Almost 50 percent of the cases that fell into band three were adjudicated by physicians as septic, with an AUC of .87. The positive percent agreement with band one was 93.2 percent (88.7–96.3); the negative percent agreement with band three was 87 percent (81.4–91.4); the negative predictive value of band one was 97.5 percent (94.0–99.1); and the positive predictive value of band three was 55.9 percent (48.4–62.9). “It performed quite well in that patient population, with good sensitivity and specificity,” Dr. Farnsworth said.
“What was interesting was the potential to use it as a prognostic tool,” he said, though that’s not part of the current indication. For patients with a score in band one, 30-day in-hospital sepsis mortality was zero percent, compared with 1.2 percent and 8.8 percent in bands two and three, respectively. “If you’re in band three, the probability of survival is quite a bit lower than if you live in band one,” he said.
The same group conducted a follow-up implementation study and reported the results last year (Thomas CB, et al. Healthcare. 2025;13[11]:1273). Data from the electronic health record were analyzed across two groups: pre-implementation (April–July 2023) versus post-implementation (August 2023–July 2024). They assigned temporarily matched cohorts (April–July 2023 versus August 2023–July 2024) and assessed clinical outcomes and resource use.
Post-implementation, sepsis-associated mortality dropped from 10.9 percent to 6.6 percent in the temporarily matched cohort. Blood culture usage declined in patients who fell within band one, from 48.1 percent in August 2023 to a range of 23.8–37.0 percent from September 2023 to July 2024. At the same time, blood culture usage increased in the high-risk band three patients, from 91 percent to 98.6 percent.
Intellisep “could be one of the first technologies to have shown a reduction in mortality rates,” Dr. Farnsworth said, though there has been no randomized controlled trial of implementation impact. The assay has good sensitivity and specificity in the clinical studies, but it requires a dedicated instrument and physician heuristics or a specific triage algorithm. “In contrast to the previous study, where it was all-comers, now somebody has to be thinking sepsis,” he said.