Various new “sepsis tests” have come to market or will soon, claiming to solve the problem of diagnostic imprecision in the early management of suspected sepsis.
Could they?
The lack of a reliable diagnostic test or universally accepted criteria to diagnose sepsis leads to significant challenges in clinical practice and research. Overly general case definitions and sloppy EMR algorithms result in a high rate of overtreatment with antibiotics due to false positives, as well as delayed treatment from false negatives.
Clinical research in sepsis is plagued by the same basic problem: it’s unclear whether a significant proportion of enrolled patients even had infections in the first place (much less sepsis).
The current gold standard test, blood cultures, takes days to produce results, making it impractical for the rapid clinical decision-making that’s essential in suspected sepsis.
Over the past decade, a variety of new diagnostic tests have emerged, claiming to reliably identify severe infections within minutes. These technologies seem to offer the potential to revolutionize standard care for sepsis by making faster, more accurate diagnoses.
Of course, they also have limitations you won’t likely learn from the marketers and hospital administrators eager to implement the latest advancements.
The most fundamental is that most of the tests rely on adjudicated sepsis cases to validate their performance—cases that were defined in the same flawed circular method (i.e., organ dysfunction plus suspected infection with or without evidence).
Still, it seems likely the addition of objective biological measurements could at least improve the accuracy of sepsis diagnoses.
This series of posts will take a closer look at the new tests for the rapid diagnosis of infection that are commercially available today or on the horizon.
Septicyte Rapid™
What Is It?
Septicyte Rapid™ is a PCR-based test that identifies immune system activity consistent with severe infection. It does not identify pathogens.
Septicyte was cleared by FDA in 2021 for use in hospitalized patients with suspected sepsis. Immunexpress, a molecular diagnostic company based in Brisbane (Australia) and Seattle, had announced one U.S. health system using Septicyte as of January 2025.
How Does It Work?
Septicyte uses genetic amplification testing (PCR) to analyze mRNA expression by white blood cells (a process called transcriptomics).
Blood is introduced into a cartridge that is inserted into a Biocartis IdyllaTM PCR machine. Two RNA transcripts (PLAC8 and PLA2G7) are amplified from whole blood using quantitative real-time PCR. These two transcripts are elevated in patients with more severe sepsis.
An algorithm then returns a clinical risk score of 0 to 15, in one of four probability bands for sepsis.
How Fast Are The Results?
About one hour, according to the manufacturer.
How Was It Validated?
Septicyte Rapid was validated using banked blood from 356 critically ill patients in multiple observational studies on sepsis (in 2011-2016), and prospectively in 63 critically ill patients in 2021. All were enrolled upon entry to the ICU, from the ED or wards, with suspected sepsis (i.e., they looked very ill).
Sepsis cases were adjudicated by three blinded physicians, at least two of whom agreed 90% of the time. About 40% of patients were ultimately adjudicated as having sepsis. The 10% of indeterminate cases were excluded (cherry-picked out) of the analysis.
(A legacy product, Septicyte LAB tested four mRNA sequences and was validated in other studies, but appears to have been discontinued in favor of Septicyte Rapid.)
What Were Its Performance Characteristics?
Performance characteristics are highly influenced by the prevalence of sepsis in the population tested, and the cut-off values selected for positivity. The authors chose a cut-off score that would optimize sensitivity.
For a low-risk (band 1, score 0.1 to 5) result, they reported 91% negative predictive value and 94% sensitivity. This was seen in about 1/4 of the patients.
Band 4 had a positive predictive value of 81% (specificity 90%): about one-third of the patients.
Bands 2 and 3 did not discriminate well and did not provide useful or actionable information.
Table 3 of their paper shows as high as a 12% false negative rate with a low-risk result, even after excluding the indeterminate cases.
What Does It Cost?
The price per test has not been publicly reported.
Septicyte Rapid: Takeaway
It’s hard to see the use case for this test. If patients are ill enough to be admitted to the ICU with suspected sepsis, they should get antibiotics. If and when they improve, antibiotics can be discontinued. We would need to see much higher sensitivity in the lower-risk bands before relying on this test to withhold antibiotics in critically ill patients.
Only 25% of patients were in the low-risk band; the others would all be getting empiric antibiotics anyway, irrespective of the test result.
Sepsis risk stratification makes the most sense in the ED, not on the way to the ICU. If Septicyte were validated in patients with suspected sepsis in the ED or general wards, it might provide more value.
Intellisep™
What Is It?
Intellisep is a point-of-care system that infers a probability of sepsis based on changes in the deformability of a patient’s white blood cells.
Marketed by San Francisco-based Cytovale, Intellisep received FDA 510k clearance in January 2023 for use in U.S. emergency departments. Two health systems were reported by the company to be using it as of January 2025, according to a press release and their website.
How Does It Work?
A blood sample as small as 100 µL is loaded onto a cartridge containing tiny channels, and a countertop scanner analyzes individual leukocytes as they squeeze through the channel under “computer vision”. Activated white blood cells (elevated in severe infections) exhibit recognizable deformability patterns that are indexed by an algorithm trained on patterns seen in WBCs of patients with sepsis.
The algorithm then classifies patients as low, intermediate, or high risk for sepsis mortality, referenced to its validation studies.
How Fast Are The Results?
Intellisep results can return in under 10 minutes, according to Cytovale.
How Was It Validated?
Intellisep was validated in one thousand ED patients with suspected infection in five prospective observational cohort studies; 192 (19%) were adjudicated to have sepsis by blinded assessors.
Similarly to the Septicyte cohorts, they were highly selected: “Only patients with high confidence in the adjudicated label [of sepsis or no sepsis] were included (n = 1002), defined as patients for whom there was consensus in the determination of sepsis per the Sepsis-3 and severe sepsis per the Sepsis-2 definitions between both the independent adjudication panel and the site-level physician.” This simplified the analysis, and also a priori would be expected to boost the test’s performance.
What Were Its Performance Characteristics?
Intellisep proved most useful when it generated a low-risk (0% expected mortality, in about half the sample) or a high-risk (~13% mortality) result, in a quarter of the sample.
A low-risk result had a 97.5% negative predictive value (i.e., the likelihood that a low-risk result agreed with the adjudicators’ ruling of “no sepsis”).
A high-risk result had a 56% positive predictive value (i.e., about half the patients with a positive test were adjudicated to have sepsis).
An intermediate test result (found in a quarter of patients) provided little actionable information. Sepsis was considered present by adjudicator assessments in 46 of 243 intermediate-risk tests, producing either a 19% positive predictive value or an 81% negative predictive value (depending on your point of view).
Performance characteristics like these are are highly influenced by the prevalence of sepsis in the population tested and the cut-off point for positivity.
What Does It Cost?
Intellisep’s price is not publicly reported. It requires purchase of a countertop point of care system and single use cartridges for the microfluidic blood analysis.
Intellisep: Takeaway
Intellisep rapidly identifies changes in white blood cells to suggest a probability of sepsis. It appears promising as an adjunct to clinical judgment in identifying patients at low risk for sepsis in the ED (e.g., the 25% of patients with a lactate >2.4 in the low risk cohort, who would all get overtreatment with fluids and antibiotics under SEP-1).
However, its real world performance characteristics are unknown and it requires extensive further validation in real world settings.
References
Balk, R.; Esper, A.M.; Martin, G.S.; Miller, R.R., III; Lopansri, B.K.; Burke, J.P.; Levy, M.; Opal, S.; Rothman, R.E.; D’Alessio, F.R.; et al. Validation of SeptiCyte RAPID to Discriminate Sepsis from Non-Infectious Systemic Inflammation. J. Clin. Med. 2024, 13, 1194.
It reproduces the fundamental mistake: defining a disease by the prognosis.
https://thethoughtfulintensivist.substack.com/p/sepsis-research-is-scary
Not impressed. It's admirable that we are trying to find sepsis early, and I think we are not going to be very successful. Sepsis is the ultimate chameleon of disease, and we have to be ever vigilant against. it.