(344 days)
INNOVANCE® Heparin Assay
In vitro diagnostic automated chromogenic assay for the quantitative determination of unfractionated heparin (UFI) and low molecular weight heparin (LMWH) activity in human plasma collected from venous blood samples in 3.2% sodium citrate tubes on the BCS® XP System in the clinical laboratory. For use with plasma from patients undergoing heparin anticoagulant therapy with either UFH or LMWH. The performance of this device has not been established in neonate and pediatric patient populations.
INNOVANCE® Heparin Calibrator
For calibration of the INNOVANCE® Heparin assay for the quantitative determination of the activity of unfractionated heparin (UFH) and low molecular weight heparin (LMWH) in citrated human plasma.
INNOVANCE® Heparin UF and LMW Controls
For quality control of the INNOVANCE® Heparin assay for the quantitative determination of the activity of unfractionated heparin (UFH) and low molecular weight heparin (LMWH) in citrated human plasma.
The INNOVANCE® Heparin assay is a one stage chromogenic assay. The reagent kit consists of two components. One component (INNOVANCE Heparin Reagent) contains Coagulation Factor Xa (Xa), the other (INNOVANCE Heparin Substrate) a chromogenic substrate specific for Xa. Upon mixing of INNOVANCE Heparin Reagent and INNOVANCE Heparin Substrate, Xa converts the chromogenic substrate into two products, one of them is paranitroaniline. The formation of paranitroaniline can be quantified by the coaqulation analyzer employing light absorption at a specific wavelength (405 nm). In the presence of a heparin containing sample the formation of paranitroaniline will be reduced in a time dependent manner. This is due to inhibition of Xa by the heparin/antithrombin (AT) complex. This complex is formed in the patient's plasma and competes with the substrate conversion by Xa. The concentration of the complex is not only dependent on the concentration of heparin but also on the availability of the patient's endogenous antithrombin. By comparison to a reference curve the heparin activity of the sample can be quantified. To reduce the influence from heparin antagonists, such as platelet factor 4 (PF4), dextran sulfate is included in the reaction mixture.
The INNOVANCE® Heparin Calibrator consists of 5 calibrator levels. INNOVANCE® Heparin Calibrator 1 represents plasma containing no heparin. INNOVANCE® Heparin Calibrator 2, 3, 4 and 5 contain defined activities of LMWH and are calibrated against the World Health Orqanization (WHO) International Standards for UFH and LMWH. The calibrator levels are used to establish a reference curve (calibration curve) which then can be employed to quantify the heparin activity of UFH and LMWH containing plasmas.
The INNOVANCE® Heparin Controls consist of plasmas containing defined activities of either UFH or LMWH. Recovery of these controls within their assigned ranges indicates proper functionality of the assay system.
This document describes the performance data for the INNOVANCE® Heparin Assay, INNOVANCE® Heparin Calibrator, and INNOVANCE® Heparin UF and LMW Controls. The information provided focuses on the analytical performance of the device rather than a clinical study involving human readers and AI assistance. Therefore, some requested information related to human expert involvement, MRMC studies, and AI-specific ground truth establishment cannot be extracted from this document.
Here's an analysis of the acceptance criteria and study proving the device meets them, based on the provided text:
1. A table of acceptance criteria and the reported device performance
The document provides extensive data on linearity, specificity (interference), reproducibility, and repeatability. Specific acceptance criteria are mentioned as "pre-established acceptance criteria" or inferred from the "no interference up to..." statements. Since the exact numerical acceptance criteria are not explicitly listed in a consolidated table, I will present the reported performance and note that the studies "met pre-established acceptance criteria" or "were found to cause no interference."
Acceptance Criteria and Reported Device Performance for INNOVANCE® Heparin Assay System
| Performance Metric | Acceptance Criteria (Implicit/Explicit) | Reported Device Performance |
|---|---|---|
| Measuring Range (Linearity & LoQ) | - Established Limit of Quantitation (LoQ) | - Assay range established as 0.10 - 1.50 IU/mL. |
| - Linearity verified across specified range | - Linearity study performed with 21 different dilutions for UFH and LMWH (0.00 to 1.90 IU/mL). | |
| Specificity (Interference) | - No interference up to specified concentrations for various endogenous and exogenous substances. | No interference observed up to: |
| - Ascorbic Acid: 176 mg/dL | ||
| - Bilirubin (unconjugated): 60 mg/dL | ||
| - Bilirubin (conjugated): 40 mg/dL | ||
| - Hemoglobin: 347 mg/dL | ||
| - Platelet Factor 4: 0.7 µg/mL | ||
| - Triglycerides: 807 mg/dL | ||
| - Rheumatoid Factor: 220 IU/mL | ||
| - Apixaban: 4.1 ng/mL | ||
| - Danaparoid Sodium: 0.03 IU/mL | ||
| - Fondaparinux: 35.5 ng/mL | ||
| - Rivaroxaban: 7.1 ng/mL | ||
| Reproducibility & Repeatability (Reagent) | - Meets CLSI EP05-A2 guidelines. | - Reproducibility (Table 1): Total %CVs for LMW controls ranged from 2.97% to 4.40%, for UFH controls from 3.29% to 7.30%. For LMWH plasma pools, total %CVs ranged from 3.47% to 21.15%. For UFH plasma pools, total %CVs ranged from 4.30% to 20.60%. |
| - Repeatability (Table 2): Total %CVs for LMW controls ranged from 1.85% to 3.12%, for UFH controls from 3.58% to 3.71%. For LMWH plasma pools, total %CVs ranged from 2.83% to 9.30%. For UFH plasma pools, total %CVs ranged from 3.21% to 7.58%. | ||
| Reproducibility & Repeatability (Calibrator) | - Total %CV for reproducibility (Table 3) and repeatability (Table 4) within specified limits. | - Reproducibility (Table 3): Total %CVs for calibrator levels ranged from 2.23% to 3.36%. |
| - Repeatability (Table 4): Total %CVs for calibrator levels ranged from 2.53% to 3.74%. | ||
| Reproducibility & Repeatability (Control) | - Total %CV for reproducibility (Table 5) and repeatability (Table 6) within specified limits. | - Reproducibility (Table 5): Total %CVs ranged from 2.97% to 7.30%. |
| - Repeatability (Table 6): Total %CVs ranged from 2.45% to 3.67%. | ||
| Frozen vs. Fresh Method Comparison | - Passing-Bablok regression slope, intercept, and correlation coefficient meet pre-established criteria. | - Slope: 1.00, Intercept: -0.01 IU/mL, r: 0.996, r2: 0.993. Predicted bias for MDP (0.3, 0.4 IU/mL) was -0.01 IU/mL. Percent bias for MDP (0.6, 0.7, 1.0 IU/mL) ranged from -1.00% to -1.67%. All acceptance criteria were met. |
| Method Comparison (vs. predicate device) | - Passing-Bablok regression slope, intercept, and correlation coefficient meet pre-established criteria. | - UFH samples: Slope: 0.93, Intercept: 0.03, Correlation Coefficient (r): 0.98. |
| - LMWH samples: Slope: 1.06, Intercept: -0.01, Correlation Coefficient (r): 0.99. | ||
| - Results met the pre-established acceptance criteria. |
2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)
- Measuring Range (Linearity): 21 different dilutions per heparin type (UFH and LMWH). Uses plasma pools spiked with standards.
- Specificity (Interference): Base pools of UFH and LMWH were prepared. Specific sample count not provided, but implies multiple samples to establish "no interference up to..." levels.
- Reproducibility and Repeatability:
- Reagent (reproducibility): 240 measurements (N=240) per sample type (LMW/UF controls, LMWH/UFH plasma pools).
- Conducted at three external sites and one internal site (Siemens Marburg site prepared pools). Data provided is multicenter.
- Plasma pools were prepared at Siemens Marburg and sent frozen to external sites, implying retrospective analysis of these prepared samples.
- Reagent (repeatability): 240 measurements (N=240) per sample type.
- Conducted at one site (Bad Oeynhausen).
- Calibrator (reproducibility): 120 measurements (N=120) per calibrator level.
- Conducted at three internal sites.
- Calibrator (repeatability): 240 measurements (N=240) per calibrator level.
- Conducted at one site.
- Controls (reproducibility): 240 measurements (N=240) per control level.
- Conducted at three sites.
- Controls (repeatability): 240 measurements (N=240) per control level.
- Conducted at one site.
- Reagent (reproducibility): 240 measurements (N=240) per sample type (LMW/UF controls, LMWH/UFH plasma pools).
- Frozen vs. Fresh Method Comparison: 69 samples (N=69).
- Conducted at one study site in Germany.
- This study involved both fresh samples (measured within 4 hours) and frozen aliquots of the same samples (measured after at least one week of -70°C storage), making it a comparison of sample handling rather than purely retrospective vs. prospective patient data collection.
- Method Comparison (vs. predicate device):
- UFH samples: 165 samples (N=165).
- LMWH samples: 155 samples (N=155).
- Conducted at three sites, including one external US site. All sites used the same protocol.
- Frozen samples were thawed and measured, indicating a retrospective use of collected samples.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)
This document describes the analytical performance of an in-vitro diagnostic device for quantifying heparin activity. The "ground truth" in this context is established by reference methods, international standards (WHO International Standards for UFH and LMWH), and analytical accuracy/precision measurements, not by human expert interpretations of images or clinical outcomes. Therefore, there are no human experts involved in establishing the "ground truth" as described for medical imaging AI devices.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
Not applicable. This is an analytical performance study for an in-vitro diagnostic assay. Adjudication methods like 2+1 or 3+1 are typical for subjective human interpretation tasks, such as radiology reads, which are not relevant here.
5. 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
Not applicable. This document is about an in-vitro diagnostic device, not an AI-assisted diagnostic tool that involves human readers interpreting cases.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, the studies presented (linearity, specificity, reproducibility, repeatability, frozen vs. fresh comparison, method comparison) are effectively "standalone algorithm" performance evaluations, as they assess the device's analytical measurement capabilities directly, without human interpretation of results as a variable. The "algorithm" here refers to the biochemical reactions and instrument logic that produce the quantitative heparin activity measurement.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
The ground truth for this device is based on:
- International Standards: The INNOVANCE® Heparin Calibrators and Controls are traceable to the World Health Organization (WHO) International Standards for UFH and LMWH. This is the primary reference for accuracy and calibration.
- Reference Methods/Predicate Devices: The method comparison studies demonstrate equivalence to an established, legally marketed predicate device (Coamatic Heparin assay on the IL ACL TOP®), which itself would have been validated against accepted reference methods and standards.
- Known Concentrations: Linearity and interference studies use samples spiked with known concentrations of UFH, LMWH, and potential interferents.
- Statistical Analysis: CLSI guidelines (e.g., EP06-A, EP07-A2, EP05-A2, EP09-A3) are used for statistical evaluation of analytical performance, ensuring robust and validated methodologies for assessing precision and accuracy.
8. The sample size for the training set
This document does not specify a "training set" in the context of machine learning. The device is a traditional in-vitro diagnostic assay, not an AI/ML algorithm that undergoes a distinct training phase on a large dataset. The "training" for such devices typically involves the development and optimization of the reagent formulations, instrument parameters, and calibration procedures, which is an engineering and chemistry development process rather than a data-driven training of a model.
9. How the ground truth for the training set was established
As there is no "training set" in the AI/ML sense, this question is not applicable. The development and calibration of the assay components (reagents, calibrators, controls) are established through rigorous analytical chemistry and metrology principles, tracing back to international reference standards and validated against known concentrations and predicate devices.
§ 864.7525 Heparin assay.
(a)
Identification. A heparin assay is a device used to determine the level of the anticoagulant heparin in the patient's circulation. These assays are quantitative clotting time procedures using the effect of heparin on activated coagulation factor X (Stuart factor) or procedures based on the neutralization of heparin by protamine sulfate (a protein that neutralizes heparin).(b)
Classification. Class II (performance standards).