Search Results

The ADVIA Centaur® High-Sensitivity Troponin I (TNIH) assay is for in vitro diagnostic use in the quantitative measurement of cardiac troponin I in human serum or plasma (lithium heparin) using the ADVIA Centaur XP system. The assay can be used to aid in the diagnosis of acute myocardial infarction (AMI).

Device Description

The ADVIA Centaur TNIH assay kit includes the ADVIA Centaur TNIH Primary Reagent ReadyPack and the ADVIA Centaur TNIH Calibrator. The Primary Reagent ReadyPack contains ADVIA Centaur TNIH Lite Reagent (Bovine serum albumin (BSA) conjugated to a recombinant monoclonal Fab anti-human cTnl (~0.2-0.4 µg/mL) labeled with acridinium ester in HEPES buffer with stabilizers and preservatives) and ADVIA Centaur TNIH Solid Phase Reagent (0.45 mg/mL streptavidin-coated magnetic latex particles with 2 biotinylated (mouse and sheep) monoclonal anti-troponin I antibodies in buffer with stabilizers and preservatives). The Calibrator kit contains ADVIA Centaur TNIH High Calibrator (Human serum with human cTnl and preservatives (lyophilized)) and ADVIA Centaur TNIH Low Calibrator (HEPES buffer with bovine serum albumin (BSA), surfactants, and preservatives (liquid)). The assay uses a chemiluminescence sandwich immunoassay methodology.

AI/ML Overview

The Siemens Healthcare Diagnostics Inc. ADVIA Centaur High-Sensitivity Troponin I (TNIH) assay is an in vitro diagnostic device used for the quantitative measurement of cardiac troponin I in human serum or plasma to aid in the diagnosis of acute myocardial infarction (AMI).

Here's an analysis of the acceptance criteria and the study that proves the device meets them:

1. Table of Acceptance Criteria and Reported Device Performance:

The document doesn't explicitly state "acceptance criteria" for each performance characteristic in a table format. However, it indicates expected outcomes based on standard guidelines (CLSI protocols) and high-sensitivity troponin definitions. The performance data presented demonstrates that the device meets these implied criteria.

Performance Characteristic	Implied Acceptance Criteria (Based on CLSI & IFCC)	Reported Device Performance
Precision	CVs within acceptable ranges for clinical assays; demonstrating consistent results. (Based on CLSI EP5-A3)	Repeatability (%CV): Serum samples ranged from 0.7% to 4.8%. Lithium Heparin Plasma samples ranged from 0.8% to 3.6%. Within-Lab (%CV): Serum samples ranged from 0.9% to 5.4%. Lithium Heparin Plasma samples ranged from 1.1% to 4.9%. These values are generally considered acceptable for clinical assays, especially within the context of troponin measurement.
Linearity	Deviation from linear fit within acceptable limits across the assay's measuring range. (Based on CLSI EP06-A)	Deviation from Linear Fit: For full range, Li Hep and Serum samples showed deviations up to 6.57%. For ~150 pg/mL range, deviations up to 3.5% were observed. While specific acceptance limits for deviation are not given, these results are presented as representative and implied to be acceptable.
Dilution Recovery	Recoveries for diluted samples should be within a reasonable percentage of expected values.	Recoveries for individual samples were all within 20%. Mean of 1:2 dilutions was 102.6%. Mean of 1:5 dilutions was 91.8%. This indicates acceptable recovery for diluted samples.
Hook Effect	No Hook Effect detected within the specified concentration range.	No hook effect observed up to 500,000 pg/mL. This is a positive finding, indicating reliability at high concentrations.
Detection Limit	LoB, LoD, and LoQ determined as per CLSI protocol EP17-A2.	LoB: 0.50 pg/mL. LoD: 1.60 pg/mL. LoQ: 2.50 pg/mL (at 20% total CV). These values describe the analytical sensitivity of the assay.
Endogenous Interference	Interferents should not cause >10% interference at tested concentrations. (Based on CLSI EP07-A2)	Most endogenous substances showed less than 10% interference. For substances causing >10% interference (none explicitly highlighted as such in the table, all shown are below), serial measurements were taken and analyzed by linear regression (though specific results for such cases are not provided in the table). The presented data indicates minimal endogenous interference.
Drug Interference	At tested concentrations, drugs should cause <10% interference. (Based on CLSI EP07-A2)	All listed therapeutic drugs caused <10% interference at the tested low and high concentrations. This is a strong positive finding for clinical utility.
Cross-Reactivity	Minimal to no cross-reactivity with similar or related substances.	0.00% cross-reactivity reported for Actin, Cardiac Troponin T, CK-MB, Myoglobin, Myosin Light Chain, Skeletal Troponin I, Tropomyosin, and Troponin C at tested levels. This indicates high specificity for cardiac troponin I.
Heterophile Interference	No significant interference from HAMA or RF.	No interference with HAMA or RF was observed. This ensures accurate results in the presence of these common interferents.
High-Sensitivity Designation	1. %CV at 99th percentile ≤10%. 2. Measurable concentrations above LoD for ≥50% of healthy individuals. (IFCC Task Force)	The assay meets both criteria: 1. Although individual %CV at 99th percentile is not explicitly stated, the overall precision data (1.3%-1.9% at 138-146 pg/mL) suggests it is well within specification, and the claim of meeting the criteria implies this. 2. The fact that the assay is designated "High-Sensitivity" and implied to meet this criterion, combined with its low LoD, supports this. This is a critical claim for modern troponin assays.
Clinical Performance (AMI Diagnosis)	Demonstrate clinical concordance (Sensitivity, Specificity, PPV, NPV) with adjudicated AMI diagnosis, showing utility across different time points.	Sensitivity: Generally good, especially at later time points (e.g., 90-100% after 2-3 hours). Lower at early time points (e.g., 78-82% at 0-1.5 hr). Specificity: Consistently high, typically ranging from 86-95%. PPV: Varies, generally lower than sensitivity/specificity, ranging from 41-68%. NPV: Consistently very high, typically ranging from 95-99.5%. The results support the assay's utility in aiding AMI diagnosis, particularly the high NPV which is useful for ruling out AMI.

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

Test Set (Clinical Study):
- The clinical study enrolled all patients presenting to the emergency department or ambulatory care center with signs or symptoms suspicious for an acute coronary syndrome (ACS) event.
- The total number of patients/samples isn't explicitly stated as a single number but can be inferred by summing the N values across different time points and matrices for the clinical performance tables. For instance, in the "overall 99th Percentile" table for Li Hep Plasma, the sum of N for Sensitivity and Specificity across all time points would provide an estimate of the number of positive and negative cases assessed:
  - Sensitivity N (true positive): Sums up to 1341
  - Specificity N (true negative): Sums up to 8853
  - This suggests a large test set of thousands of samples/patient time points.
- Provenance: Specimens were collected at 29 sites across different regions of the United States. Data is prospective, as patients were enrolled who presented with ACS symptoms.

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

Number of Experts: An "independent adjudication committee" was used. The exact number of experts on this committee is not specified.
Qualifications: The committee included cardiologists. No specific years of experience or board certifications are mentioned beyond "cardiologists."

4. Adjudication Method for the Test Set:

The ground truth for AMI diagnosis was established by an independent adjudication committee.
The adjudication was based on the Third universal definition of myocardial infarction consensus guideline endorsed by the European Society of Cardiology (ESC), the American College of Cardiology Foundation (ACCF), the American Heart Association (AHA), and the World Heart Federation (WHF).
The specific method (e.g., 2+1, 3+1) for how decisions were reached by the committee (e.g., how many members had to agree, or if a tie-breaker was used) is not explicitly stated. It just mentions "an independent adjudication committee."

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and Its Effect Size:

No, an MRMC comparative effectiveness study was not done.
This submission describes the performance of an in vitro diagnostic assay (a lab test), not an AI algorithm assisting human readers in interpreting medical images or data. Therefore, the concept of human readers improving with/without AI assistance does not apply here.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done:

Yes, this is a standalone performance study.
The ADVIA Centaur TNIH assay is a diagnostic laboratory test. Its performance characteristics (precision, linearity, detection limits, interference, and clinical concordance) are evaluated directly, without human interpretation of the assay's output being part of the device's measured performance in the way an AI algorithm for imaging would be. The clinical concordance described represents the algorithm's (assay's) performance compared to the adjudicated clinical diagnosis.

7. The Type of Ground Truth Used:

Expert Consensus / Clinical Outcomes Data (for Clinical Study): The ground truth for the clinical performance study was the adjudicated diagnosis of Acute Myocardial Infarction (AMI) established by an independent committee of cardiologists based on the Third universal definition of myocardial infarction consensus guideline. This is a form of expert consensus derived from clinical outcomes and medical guidelines.
Reference Standards/Characterized Samples (for Analytical Studies): For analytical performance characteristics (precision, linearity, etc.), the ground truth was established using characterized samples, reference materials, or spikes that conform to recognized standards (e.g., CLSI guidelines).

8. The Sample Size for the Training Set:

Not Applicable / Not Explicitly Stated: For an in vitro diagnostic assay like this, there isn't typically a "training set" in the sense of machine learning algorithms. The assay is developed and validated using various methods and samples.
However, if we consider assay development and optimization, a large number of samples would have been used internally by the manufacturer during research and development to optimize reagents, antibodies, and protocols to achieve the desired performance characteristics. This document does not detail those internal R&D sample numbers. The clinical study samples described are primarily for validation/testing, not "training."

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

Not Applicable: As there isn't a "training set" in the machine learning sense, the concept of establishing ground truth for it doesn't directly apply.
The development of such an assay relies on established biochemical principles, known concentrations of analytes, and recognized clinical definitions (e.g., for target analyte, cross-reactants, and interference substances). The "ground truth" during development involves ensuring the assay accurately measures the target molecule (cardiac troponin I) with high specificity and sensitivity based on these scientific and clinical standards.

Ask a Question

Ask a specific question about this device

Page 1 of 1