Search Results

The Cytovale IntelliSep test is a semi-quantitative test that assesses cellular host response via deformability cytometry of leukocyte biophysical properties and is intended for use in conjunction with clinical assessments and laboratory findings to aid in the early detection of sepsis with organ dysfunction manifesting within the first 3 days after testing. It is indicated for use in adult patients with signs and symptoms of infection who present to the Emergency Department. The test is performed on an EDTA anticoagulated whole blood sample.

The IntelliSep test generates an IntelliSep Index value that falls within one of three discrete interpretation bands based on the probability of sepsis with organ dysfunction manifesting within the first three days after testing. The IntelliSep test represents the probability of the clinical syndrome of sepsis and is intended to be used alongside other clinical information and clinical judgement. It does not identify the causative agent of infection and should not be used as the sole basis to determine the presence of sepsis. The IntelliSep test is intended for in vitro diagnostic use.

The Cytovale IntelliSep test is a short turn-around time (STAT) test, producing results in 10 minutes or less, to aid in the early detection of sepsis with organ dysfunction manifesting within the first 3 days after testing. It assesses the state of immune activation in patients with signs and symptoms of infection who present in the Emergency Department (ED).

The IntelliSep test is run on the Cytovale System, a laboratory benchtop analyzer depicted in Figure 1 and Figure 2.

To run a test, the laboratory operator transfers 100 uL of whole blood into the sample preparation tube which is then placed into the Cytovale System. The system automatically lyses red blood cells, and washes the purified leukocytes in a diluent, producing a total volume of approximately 1mL of prepared sample, which the operator then transfers to the IntelliSep cartridge for analysis on the Cytovale System. A microfluidic deformability cytometry technique is used to measure the biophysical properties of thousands of individual leukocytes in rapid succession. These properties have been shown to differ in quiescent white blood cell populations when compared to those in septic patients, enabling rapid assessment of the host response and the likelihood of having sepsis with organ dysfunction manifesting within the first 3 days after testing.

Based on these measurements, the test provides a single score, the IntelliSep Index (ISI), ranging from 0.1-10.0, stratified into three discrete interpretation bands (Band 2, Band 3) of increasing sepsis likelihood.

The provided document, K250513, focuses on a Special 510(k) submission for modifications to an already cleared device, the Cytovale IntelliSep test (K220991). As such, it does not detail a full clinical study with specific acceptance criteria and performance metrics for initial device clearance. Instead, it aims to demonstrate that the modifications made do not negatively impact the previously established performance of the device.

Therefore, the response below will describe the approach taken to prove the modifications meet the underlying performance requirements, rather than a new standalone clinical study with fresh acceptance criteria.

Key takeaway: The study proves that the modified device maintains identical analytical and clinical performance to the previously cleared device (K220991). The acceptance criteria, in this context, are implicitly tied to the performance claims and limitations established in the original K220991 clearance.

Acceptance Criteria and Study Proving Device Meets Criteria (for Modified Device)

Given that K250513 is a Special 510(k) for device modifications, the "acceptance criteria" here refer to demonstrating that the modified device performs identically to the original cleared device (K220991). The study's goal was to show non-inferiority and preservation of performance.

1. Table of Acceptance Criteria (Implicit) and Reported Device Performance

2. Sample Sizes Used for the Test Set and Data Provenance

The document explicitly states that "raw assay videos from the following performance studies were processed on both the modified device (K250513) and the baseline device (K220991)." This indicates that existing previously collected data from these studies served as the test set.

• Test Set Data Provenance:

  • IntelliSep Clinical Validation Study (CV-CLN-007): The original source of clinical data from K220991 clearance. The document does not specify the country of origin, but generally, clinical validation studies for FDA clearance are conducted in the United States or include diverse sites if international. It was a prospective study for the original clearance.
  • Within Laboratory Precision (CLSI EP05-A3): Data from analytical precision studies.
  • Reproducibility (CLSI EP05-A3): Data from analytical reproducibility studies.
  • Healthy Reference Range (CLSI EP28-A3C): Data for establishing a healthy population reference.
  • Sequestered Data Set: Processed on both devices for comparison. This implies a previously collected, retrospective dataset or a subset of data held separate from training/development.

• Sample Size for Test Set: The document does not specify the sample sizes for these re-analyzed datasets. It refers to the Decision Summary for K220991 for detailed performance claims and limitations, implying sample sizes for these studies would be found there. For a Special 510(k), the key is comparative performance on existing data, not necessarily new, large-scale clinical trials.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications

The document does not explicitly state the number or qualifications of experts used to establish ground truth for the test set in the context of this Special 510(k) re-analysis.

However, for the original K220991 clinical validation study (CV-CLN-007), the ground truth for sepsis with organ dysfunction would have been established by clinical experts based on:

• Standard clinical definitions (e.g., Sepsis-3 criteria).
• Review of electronic health records, lab findings, clinical assessments, and patient outcomes by qualified physicians (likely intensivists, emergency medicine physicians, or infectious disease specialists).
• This process would have involved their clinical judgment and interpretation of a comprehensive patient profile.

4. Adjudication Method for the Test Set

The document does not describe an adjudication method for the test set used in this Special 510(k). This is because the study re-ran previously established data (raw assay videos) through the modified and unmodified systems and compared the device's output directly. The focus was on whether the device's results changed, not on re-evaluating the ground truth diagnoses.

For the original clinical validation study that established the ground truth (CV-CLN-007), it is highly likely that an adjudication process (e.g., by expert clinicians, 2+1 or 3+1 consensus on patient outcomes/diagnosis) was used to establish the gold standard for sepsis with organ dysfunction. However, this is not detailed in the K250513 document.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done

No MRMC study was done or is described in this document. The IntelliSep test is an in vitro diagnostic (IVD) device that provides a quantitative index (IntelliSep Index) and an interpretation band. It is not an imaging AI system that aids human readers in interpretations. Its output is a score, not an image for human reading or analysis. Therefore, an MRMC study is not applicable to this type of device.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done

Yes, the primary evaluation for this device is standalone performance. The IntelliSep test is designed to generate an IntelliSep Index value and associated interpretation bands based on the biophysical properties of leukocytes. It's an automated measurement. For K250513, the core comparison was whether the modified algorithm/hardware produced the same ISI values and interpretation bands as the original algorithm/hardware when given the same raw data. The document explicitly states: "Raw assay videos from the following performance studies were processed on both the modified device (K250513) and the baseline device (K220991). The results were then directly compared to each other..." This confirms a standalone-like comparison of the device's output.

7. The Type of Ground Truth Used

For the foundational clinical performance validation (referenced from CV-CLN-007, originally for K220991), the ground truth for "sepsis with organ dysfunction manifesting within the first 3 days after testing" would have been established using expert consensus based on clinical assessments, laboratory findings, and patient outcomes data. This would align with standard clinical definitions of sepsis (e.g., Sepsis-3 criteria) and involved careful retrospective review of patient charts/records by clinicians and/or adjudication committees to determine the presence or absence of sepsis with organ dysfunction.

For the purpose of this Special 510(k), the "ground truth" was effectively the performance of the predicate device (K220991) on the re-analyzed data. The goal was to show that the modified device produced inputs/outputs identical to that ground truth.

8. The Sample Size for the Training Set

The document does not specify the sample size for the training set.

• The IntelliSep test's underlying algorithm (ISI calculation) and the deformability cytometry technique "remain unchanged from K220991." This suggests the core model was trained for the original K220991 submission.
• The modifications described in K250513 relate to software refinements (internal QC metrics, remote log export) and hardware changes (new IAM variant). These modifications would likely involve internal testing and verification/validation using previously established data or internal engineering samples, rather than a new "training set" in the machine learning sense, as the core algorithm itself was not re-trained.
• The comparison against a "sequestered data set" for internal quality control metrics indicates good machine learning practice, suggesting that if any internal model within the QC process was re-trained (which is not explicitly stated), this set would have been held out from that specific training.

9. How the Ground Truth for the Training Set was Established

Since the K250513 document does not describe new training, it relies on the previous K220991 submission. For the original development of the IntelliSep test (prior to K220991), the ground truth for training the algorithm would have been established through a similar process as the clinical test set: expert consensus based on clinical assessments, laboratory findings, and patient outcomes data according to established sepsis criteria. This clinical information would have been associated with the biophysical property measurements from the Cytovale system to enable the algorithm to learn the correlation between leukocyte deformability and sepsis severity/likelihood.

Ask a specific question about this device

The Cytovale IntelliSep test is a semi-quantitative test that assesses cellular host response via deformability cytometry of leukocyte biophysical properties and is intended for use in conjunction with clinical assessments and laboratory findings to aid in the early detection of sepsis with organ dysfunction manifesting within the first 3 days after testing. It is indicated for use in adult patients with signs and symptoms of infection who present to the Emergency Department. The test is performed on an EDTA anticoagulated whole blood sample.

The IntelliSep test generates an IntelliSep Index value that falls within one of three discrete interpretation bands based on the probability of sepsis with organ dysfunction manifesting within the first three days after testing. The IntelliSep test represents the probability of the clinical syndrome of sepsis and is intended to be used alongside other clinical information and clinical judgement. It does not identify the causative agent of infection and should not be used as the sole basis to determine the presence of sepsis. The IntelliSep test is intended for in vitro diagnostic use.

The Cytovale IntelliSep test is a short turn-around time (STAT) test, producing results in 10 minutes or less, to aid in the early identification of patients at risk for having or developing sepsis within three (3) days of testing. It assesses the state of immune activation in patients with clinical suspicion of infection who present in the Emergency Department (ED).

The IntelliSep test is run on the Cytovale System, a laboratory benchtop analyzer depicted in Figure 1. The Cytovale System is a closed system benchtop analyzer, comprised of three modules: Sample Preparation Module, Cell Imaging Module, and Imaging Analysis Module.

To run a test, the laboratory operator transfers 100 µL of whole blood into the sample preparation tube which is then placed into the Cytovale System. The system automatically lyses red blood cells, and washes the purified leukocytes in a diluent, producing a total volume of approximately 1mL of prepared sample, which the operator then transfers to the IntelliSep cartridge for analysis on the Cytovale System.

A microfluidic deformability cytometry technique is used to measure the biophysical properties of thousands of individual leukocytes in rapid succession. These properties have been shown to differ in quiescent white blood cell populations when compared to those in septic patients, enabling for rapid assessment of the host response and the likelihood of having or developing sepsis. Based on these measurements, the test provides a single score, the IntelliSep Index (ISI), ranging from 0.1-10.0, stratified into three discrete interpretation bands (Band 1, Band 2, Band 3) of sepsis likelihood.

Here's a breakdown of the acceptance criteria and study details for the Cytovale IntelliSep test, based on the provided document:

Acceptance Criteria and Device Performance

The document does not explicitly state pre-defined quantitative acceptance criteria in a pass/fail format for clinical performance endpoints (e.g., a specific PPV or likelihood ratio threshold). Instead, it focuses on demonstrating a clear relationship between the IntelliSep Index (ISI) and the likelihood of sepsis, and the statistical distinctness of the interpretation bands. For analytical performance, criteria are generally established for standard deviations within precision studies and confidence intervals for reproducibility and carry-over.

Here's a table summarizing the reported device performance, categorized by the relevant studies:

Clinical Performance

Implicit Acceptance Criterion: A clear relationship between ISI and the increasing likelihood of sepsis, and statistically distinct interpretation bands (non-overlapping 80% confidence intervals between bands).

Note: The clinical study explicitly states: "It is important to note that the algorithm for calculating the ISI ranges for the interpretation bands were defined based on prior sepsis-focused studies. No information from patients in this study influenced the calculations or interpretation bands of the ISI." and "Irrespective of the comparator scheme chosen, the probability of sepsis in the three ISI IntelliSep test Interpretation Bands was statistically distinct, defined as non-overlapping 80% confidence intervals between the bands." This suggests the demonstration of statistically distinct bands with appropriate clinical trends was the primary clinical performance criterion.

Analytical Performance

Study Details

1. Sample sizes used for the test set and the data provenance:

   • Test Set Sample Size: 572 evaluable subjects were included in the primary analyses for clinical performance.
   • Data Provenance: The study was a "blinded, prospective, observational, multicenter cohort study" conducted at five hospitals at four geographically dispersed sites in the United States. The subjects were "representative of the adult population (≥ 18 years of age) typical of those presenting to EDs across the United States."

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • Number of Experts: A multi-tiered adjudication process was used:
     • Initially, two independent adjudicators reviewed each case.
     • If there was disagreement, a third adjudicator was involved in a consensus meeting.
   • Qualifications of Experts: Physicians with board certification in Medical or Surgical Critical Care (CC), Infectious Diseases (ID), Emergency Medicine (EM), or Internal Medicine (IM) or related fields, from different institutions. They were not participating as an Investigator in the Study.

3. Adjudication method (e.g. 2+1, 3+1, none) for the test set:

   • The adjudication method was 2+1 (two independent adjudicators, with a third used for arbitration in case of disagreement).
   • Outcomes were categorized as 'unanimous' (both initial adjudicators agreed), 'consensus' (third adjudicator facilitated agreement), or 'forced' (majority two out of three if consensus couldn't be reached). The performance data is presented for all three schemes (forced, consensus, unanimous).

4. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:

   • No, an MRMC comparative effectiveness study was not explicitly stated or performed. The study evaluated the IntelliSep test as an aid in conjunction with clinical assessments and laboratory findings, but it did not directly compare human reader performance with and without AI assistance (i.e., IntelliSep test results). The test provides a quantitative index and interpretation bands, not an image for interpretation by human readers.

5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

   • Yes, a standalone performance evaluation was done. The IntelliSep test generates an "IntelliSep Index" (ISI) automatically from the measured biophysical properties of leukocytes. The clinical performance section directly assesses the predictive values and likelihood ratios of this automatically generated ISI, independent of real-time human interpretation during the study (treating physicians were unaware of study enrollment and IntelliSep test results). The test is intended for use in conjunction with clinical assessments, but its performance metrics are derived from the algorithm's output itself.

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

   • The ground truth was established through retrospective physician adjudication based on consensus definitions for Sepsis-2 and Sepsis-3. This involved a multi-tiered process by qualified physicians:
     • Collection of detailed patient information (demographics, labs, vitals, medical history, hospital encounter, infectious disease, medications, discharge disposition, SOFA scores, clinical impression).
     • Compilation into a Case Report Summary (CRS).
     • Review by independent adjudicators with arbitration for discordant results.
   • It's a form of expert consensus based on clinical data.

7. The sample size for the training set:

   • The document states: "It is important to note that the algorithm for calculating the ISI ranges for the interpretation bands were defined based on prior sepsis-focused studies. No information from patients in this study influenced the calculations or interpretation bands of the ISI."
   • This implies the training of the algorithm and definition of interpretation bands occurred prior to and separate from this specific clinical validation study. The sample size for the training set is not provided in this document.

8. How the ground truth for the training set was established:

   • Similar to the point above, the document only mentions that the ISI algorithm and interpretation bands were "defined based on prior sepsis-focused studies."
   • The method for establishing ground truth for those prior training studies is not detailed in this document.

Ask a specific question about this device