Search Results

The ADVIA® Chemistry Triglycerides 2 assay (TRIG 2) is intended for in vitro diagnostic use in the quantitative measurement triglycerides in human serum and plasma on the ADVIA® Chemistry systems. Measurements of triglycerides are used in the diagnosis and treatment of patients with diabetes mellitus, nephrosis, liver obstruction, other diseases involving lipid metabolism, or various endocrine disorders.

Device Description

The Triglyceride 2 reagent is a ready-to-use liquid reagent packaged for use on the automated ADVIA 1800 Chemistry system. It is supplied as a 358 tests/wedge. 4 wedges/Kit with a 38 mL fill in a 40 mL wedge size.

Existing Siemens Set-Point Chemistry calibrator (classified under CFR 862.1150 - calibrator, multi-analyte mixture, product code JIX), cleared under 510k K030169, is used to calibrate the assay on the ADVIA Chemistry systems.

AI/ML Overview

Here's a breakdown of the acceptance criteria and study information for the ADVIA® Chemistry Triglycerides 2 Reagents (TRIG 2) device, based on the provided text:

Acceptance Criteria and Device Performance

The acceptance criteria for the ADVIA® Chemistry Triglycerides 2 assay appear to be established by comparison with a predicate device and adherence to CLSI guidelines for various performance characteristics. Specific quantifiable acceptance criteria are often implied by "acceptable results" or meeting certain thresholds, but the document primarily reports the performance observed.

Performance Characteristic	Acceptance Criteria (Implied/Reference)	Reported Device Performance (ADVIA TRIG 2)
Precision	CLSI document EP5-A2, "Evaluation of Precision Performance of Quantitative Measurement Methods: Approved Guideline." (Specific CV/SD thresholds are not explicitly stated but are implied by the predicate device's expected performance and established laboratory practice for triglyceride assays.)	Serum Control (183 mg/dL): Total SD: 1.77 mg/dL, Total CV: 1.0%
Serum Control (92 mg/dL): Total SD: 0.82 mg/dL, Total CV: 0.9%
Serum Pool (254 mg/dL): Total SD: 2.04 mg/dL, Total CV: 0.8%
Serum Pool (503 mg/dL): Total SD: 2.00 mg/dL, Total CV: 0.4%
Linearity/Assay Range	Deviation from linearity of ≤ 5%. Range should be comparable to the predicate device or clinically appropriate.	Linear/measuring range: 10 - 550 mg/dL with a deviation from linearity of ≤ 5%. R-squared value of 0.9998. (Predicate range: 15-1000 mg/dL)
Limit of Blank (LoB)	Determined according to CLSI guideline EP17-A2. Proportions of false positives (a) less than 5%.	LoB: 5 mg/dL
Limit of Detection (LoD)	Determined according to CLSI guideline EP17-A2. Proportions of false negatives (β) less than 5%. Smallest amount reliably detected.	LoD: 8 mg/dL
Limit of Quantitation (LoQ)	Determined according to CLSI guideline EP17-A2.	LoQ: 10 mg/dL
Method Comparison	Demonstrated substantial equivalence through correlation and agreement with the predicate device. (Specific acceptable regression parameters like slope and intercept range are typical for such studies, but not explicitly stated as "acceptance criteria" here, rather implied by "acceptable results" and demonstrating substantial equivalence).	Serum Samples (n=101): ADVIA TRIG 2 = 0.94 (predicate) + 4.4 mg/dL. Slope 95% CI: 0.93 to 0.95. Intercept 95% CI: 2.88 to 5.85. Sample range: 20 - 540 mg/dL.
Plasma (Lithium Heparin) (n=60): ADVIA TRIG 2 = 1.01 (predicate) - 2.6 mg/dL. Slope 95% CI: 0.99 to 1.3. Intercept 95% CI: -5.95 to 0.73. Sample range: 34 - 509 mg/dL.
Plasma (Potassium EDTA) (n=60): ADVIA TRIG 2 = 1.02 (predicate) - 7.5 g/dL. Slope 95% CI: 1.00 to 1.03. Intercept 95% CI: -10.18 to -4.76. Sample range: 34 - 509 mg/dL.
Analytical Specificity	Interference criteria: Bias exceeding 10% is considered significant interference. (CLSI EP7-A2 referenced)	Bilirubin (conjugated): Interference at 22.5 mg/dL (-12.0%) with 148.0 mg/dL Trig concentration. NSI at other levels.
Bilirubin (unconjugated): Interference at 15.0 mg/dL (+12.4%) with 148.0 mg/dL Trig concentration. NSI at other levels.
Hemolysis: Interference at 750.0 mg/dL (+11.9%) with 138.0 mg/dL Trig concentration. NSI at other levels.
Ascorbic Acid: Interference at 6.0 mg/dL (-13.1%) with 144.0 mg/dL Trig concentration. NSI at other levels.
Stability	Reagent stability for opened and unopened products should meet manufacturer's claims. (No explicit criteria given, but the purpose of the study is to establish these values).	Opened reagent stable for 60 days on system. Shelf life of unopened product is 12 months at 2-8°C.
Traceability	Traceable to an accepted reference material.	Traceable to SRM909c from NIST.

Study Details

Sample size used for the test set and the data provenance:
- Precision: 80 replicates for each of the two serum controls and two serum pools. The data provenance is not explicitly stated (e.g., country of origin); however, the context of in vitro diagnostic testing in a 510(k) submission generally implies laboratory testing under controlled conditions, likely in the US by the manufacturer (Siemens Healthcare Diagnostics, Inc., Tarrytown, NY). The data is prospective as it was generated specifically for this submission.
- Linearity: Not explicitly stated how many samples/replicates were used, but it involved mixing low and high serum pools to create 12 levels (9 intermediate, 3 low-end).
- LoB, LoD: 320 determinations, comprising 160 blank and 160 low-level sample replicates.
- Method Comparison (Serum): 101 serum samples.
- Method Comparison (Plasma Matrices): 60 plasma samples (Lithium Heparin) and 60 plasma samples (Potassium EDTA).
- Analytical Specificity: Not explicitly stated, but involved testing samples with varying concentrations of interferents and triglycerides.
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):
- For this type of in vitro diagnostic device (chemical assay), the "ground truth" is typically established by the reference method (the predicate device) or by known concentrations of analytes in calibrators and control materials, rather than by human expert consensus or pathology. No human experts are mentioned for establishing ground truth in this context.
Adjudication method (e.g., 2+1, 3+1, none) for the test set:
- Adjudication methods like "2+1" or "3+1" are characteristic of studies involving human interpretation (e.g., imaging studies). This information is not applicable to a chemical assay where results are quantifiable values based on instrument measurements.
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 MRMC comparative effectiveness study was done. This type of study is not relevant for an automated chemical assay.
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Yes, this is a standalone device in the sense that the assay itself generates the quantitative results without direct human intervention in the result determination. The performance characteristics described are "algorithm only" or "device only" performance. Humans operate the instrument and interpret results, but the measurement itself is automated.
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- The ground truth for performance evaluation (e.g., method comparison) is based on the predicate device (Dimension Clinical Chemistry Triglycerides FLEX reagent cartridge, K010871) and known concentrations for controls, calibrators, and linearity pools. For traceability, the assay is referenced to NIST SRM909c, a certified reference material.
The sample size for the training set:
- This is a traditional IVD assay, not a machine learning or AI-based device. Therefore, there is no "training set" in the context of data used to train an algorithm. The development of the reagent and its parameters would have involved R&D studies, but these are not analogous to training data for AI.
How the ground truth for the training set was established:
- Not applicable, as there is no training set for an AI/ML algorithm.

Ask a Question

Ask a specific question about this device

Page 1 of 1