Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K032951
    Manufacturer
    Date Cleared
    2004-04-02

    (193 days)

    Product Code
    Regulation Number
    864.5700
    Reference & Predicate Devices
    Predicate For
    N/A
    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    The Chrono-log WBA, Model 591A/592A is intended for determination of platelet function in a whole blood specimen, using ADP, Collagen and Ristocetin reagents.
    For Platelet Function testing of Whole Blood specimens using Impedance Aggregometry. For Prescription Use.

    Device Description

    The Chrono-log WBA, Model 591A/592A uses electrical impedance to measure platelet aggregation in a whole blood sample. The Impedance is measured using a Disposable electrode with two precious metal pins. A small voltage is applies across these two pins. When the electrode is placed into a diluted whole blood specimen and a monolayer is formed around the two pins. In the absence of an agonist, the platelet build-up stabilizes and a baseline is established. When an agonist is added to the specimen, the platelets begin to aggregate and collect on the electrode pins causing a change in impedance. The change of impedance is directly proportional to the amount of Aggregation in the specimen. This change of impedance is displayed on a front panel readout. The instrument has an analog output which produces an aggregation curve when connected to a strip chart recorder or AGGRO/LINK interface. Model 591 A is a single channel version; Model 592A is the duel channel version.

    AI/ML Overview

    This document describes the performance testing of the Chrono-log Whole Blood Aggregometer (WBA), Model 591A/592A, and its substantial equivalence to the predicate device, Model 591/592, with the primary difference being the introduction of disposable electrodes.

    1. Table of Acceptance Criteria and Reported Device Performance

    The acceptance criteria were primarily based on demonstrating substantial equivalence to the predicate device through correlation, statistical significance, and the ability to detect aggregation changes and patient conditions similarly. New normal ranges were also established.

    Acceptance Criteria / Performance MetricPredicate Device (Reusable Electrodes)New Device (Disposable Electrodes)Reported Performance (New vs. Predicate)Outcome / Acceptance
    Overall Correlation (Pearson's R)N/AN/AR = 0.84 (250 samples)Met
    Paired t-testN/AN/AP < 0.0001 (250 samples)Met
    Reproducibility (Collagen 2μg/mL)Mean: 13.9 ± 3.0 ΩMean: 14.2 ± 2.2 ΩP = 0.47, R = 0.24 (n=50)Comparable
    Reproducibility (Collagen 5μg/mL)Mean: 17.7 ± 2.8 ΩMean: 16.6 ± 2.3 ΩP = 0.03, R = 0.04 (n=50)Comparable
    Reproducibility (ADP 20μM)Mean: 9.94 ± 3.4 ΩMean: 13.2 ± 2.5 ΩP < 0.01, R = 0.61* (n=50)Comparable
    Ristocetin (vWD detection, low dose)Reduced/absent results for all patientsReduced/absent results for all patientsExpected for all vWD patients (8/8 for RP, 7/8 for DP showed reduced results)Met
    Ristocetin (vWD detection, high dose)Reduced/absent for vWD patientsReduced/absent for vWD patientsBetter performance for DP in identifying vWDMet
    Detecting changes with platelet count (Whole Blood)Positive correlation (R^2 = 0.4491)Positive correlation (R^2 = 0.9037)Both methods demonstrated ability to detect changesMet
    Detecting changes with platelet count (PRP)Positive correlation (R^2 = 0.9064)Positive correlation (R^2 = 0.8512)Both methods demonstrated ability to detect changesMet
    Normal Range (Collagen 5μg/mL)16-29 ΩN/ANew established: 12-23 ΩEstablished / Close to predicate
    Normal Range (Collagen 2μg/mL)N/AN/ANew established: 10-21 ΩEstablished
    Normal Range (ADP 20μM)9-14 Ω (for 10μM)N/ANew established: 9-18 ΩEstablished / Close to predicate
    Normal Range (Ristocetin 0.4mg/mL)17-38 Ω (for 1mg/mL)N/ANew established: >5 ΩEstablished / Close to predicate

    Note on R for ADP 20μM: The Pearson's R value of 0.61 is marked with an asterisk indicating p<0.05, suggesting a statistically significant correlation despite the moderate value. The p<0.01 for the paired t-test indicates a statistically significant difference between the means, but the purpose of this study is primarily to show comparable performance and the ability to detect aggregation, not identical readings.

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

    • Overall Correlation Study: 250 samples.
    • Reproducibility, Precision, and Accuracy (Collagen, ADP): n=50 for each agonist concentration.
    • Ristocetin Studies (Normal & vWD Patients): n=12 normal donors, n=8 vWD patients.
    • Platelet Count Correlation (Whole Blood & PRP): Specific numbers for each data point are provided in the tables (e.g., 13 data points for Whole Blood, 9 for PRP), but an overall sample size for these specific correlation analyses is not explicitly stated as a single number.
    • Data Provenance: The subjects were "normal, healthy, drug free subjects" and "patients known to have von Willebrands Disease (vWD)". The country of origin is not specified but implicitly assumed to be the United States, given the FDA filing. The data is prospective, as it was collected for the purpose of this 510(k) submission to compare the new device with the predicate.

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

    The concept of "ground truth" in this context is based on the measurements obtained from the predicate device (Chrono-log Whole Blood Aggregometer Model 591/592) and established medical understanding of platelet aggregation in normal individuals and those with conditions like von Willebrand's Disease. No external "experts" were explicitly used to establish a separate ground truth for the test set beyond the predicate device's readings and known clinical characteristics of patient groups.

    4. Adjudication Method for the Test Set

    No explicit adjudication method is mentioned. The comparison is objective, based on direct readings from two different devices (new vs. predicate) using the same blood samples and reagents. Statistical tests (Pearson Correlation, Paired t-test) were used to analyze the agreement and differences between the two devices.

    5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

    No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done. This study focuses on the performance of the device itself (new electrode vs. old electrode) rather than how human readers (users) interact with or interpret the results, or the impact of AI assistance.

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

    Yes, the study is essentially a standalone performance assessment of the new device (model 591A/592A with disposable electrodes) compared to the predicate device (model 591/592 with reusable electrodes). The output is a numerical measurement (aggregation in ohms), and the performance comparison is based on these objective measurements, demonstrating the device's inherent functionality.

    7. The Type of Ground Truth Used

    The ground truth is primarily:

    • Predicate Device Performance: The measurements obtained from the predicate Chrono-log Whole Blood Aggregometer (Model 591/592) served as the primary reference for comparison. The new device was expected to produce "comparable results."
    • Known Physiological Responses: The expected aggregation patterns in "normal, healthy, drug-free subjects" and patients with "von Willebrand's Disease (vWD)" were used as a clinical ground truth for assessing the device's ability to differentiate these conditions.
    • Clinical Understanding of Platelet Count Impact: The knowledge that platelet aggregation is influenced by platelet count (e.g., reduced aggregation below 50,000/uL) served as a physiological ground truth.

    8. The Sample Size for the Training Set

    No specific "training set" is mentioned in the context of machine learning. This device operates based on electrical impedance principles, not a machine learning algorithm that requires a training set. The data presented demonstrates the device's performance directly.

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

    Not applicable, as there is no machine learning model or "training set" in the context of this device's validation.

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1