(155 days)
The Bayer ADVIA IMS C3 assay is an in vitro diagnostic device intended to measure Complement C3 (C3) in human serum. Such measurements are used as an aid in the diagnosis and treatment of immunologic disorders.
The Bayer ADVIA IMS C4 assay is an in vitro diagnostic device intended to measure Complement C4 (C4) in human serum. Such measurements are used as an aid in the diagnosis and treatment of immunologic disorders.
The Bayer ADVIA IMS IgA assay is an in vitro diagnostic device intended to measure Immunoglobulin A (IgA) in human serum. Such measurements are used as an aid in the diagnosis and treatment of abnormal protein metabolism and the body's inability to resist infectious agents.
The Bayer ADVIA IMS IgG assay is an in vitro diagnostic device intended to measure Immunoglobulin G (IgG) in human serum. Such measurements are used as an aid in the diagnosis and treatment of autoimmune diseases, chronic or recurrent infections and abnormal protein metabolism.
The Bayer ADVIA IMS Immunoglobulin M (IgM) assay is an in vitro diagnostic device intended to measure IgM in human serum. Such measurements are used as an aid in the diagnosis and treatment of chronic or recurrent infections and abnormal protein metabolism.
The Bayer ADVIA IMS Transferrin assay is an in vitro diagnostic device intended to measure transferrin (TRF) in human serum. Such measurements are used as an aid in the diagnosis and treatment of malnutrition, chronic infection, acute hepatitis, polycythemia, pernicious anemia, and red blood cell disorders, such as iron deficiency anemia.
The Bayer ADVIA IMS Vancomycin assay is an in vitro diagnostic device intended to measure vancomycin, an antibiotic drug, in human serum. Measurements of vancomycin are used as an aid in the diagnosis and treatment of vancomycin overdose and in monitoring therapeutic levels of vancomycin to ensure appropriate therapy.
The Bayer ADVIA IMS Valproic Acid Assay is an in vitro diagnostic device intended to measure valproic acid, an anti-epileptic drug, in human serum. Measurements of valproic acid are used as an aid in the diagnosis and treatment of valproic acid overdose and in monitoring therapeutic levels of valproic acid to ensure appropriate therapy.
Not Found
The provided text describes several immunoassay methods for the Bayer ADVIA IMS Systems, focusing on their performance compared to predicate devices. It does not explicitly state "acceptance criteria" for each test but rather presents performance data (precision, correlation, analytical range, interference) alongside a predicate device's performance. The study aims to demonstrate substantial equivalence to these predicate devices for FDA clearance.
Here's an analysis based on the available information:
1. Table of Acceptance Criteria and Reported Device Performance:
Since explicit acceptance criteria are not stated, I will infer them as the performance of the predicate device (the "similar device that was granted clearance of substantial equivalence") for correlation and precision, and generally acceptable performance within an analytical range with minimal interference. The reported device performance is that of the ADVIA IMS method.
| Metric (Implied Acceptance Criterion: Predicate Performance) | ADVIA IMS (Reported Performance) |
|---|---|
| C3 Method (Predicate: Dade Behring BNA Complement C3 method) | |
| Precision (Total) @ 72.4 mg/dL | 2.1% @ 79.7 mg/dL (Predicate: 5.9%) |
| Precision (Total) @ 237 mg/dL | 1.8% @ 154 mg/dL (Predicate: 2.1%) |
| Correlation (y=0.92x + 9.7, r=0.972, Syx=12.1 mg/dL) | Matched to Predicate |
| Interfering Substances: Bilirubin unconjugated (20 mg/dL) at 106 mg/dL C3 | +1% Change (Predicate: Not specified, but implied acceptable for clearance) |
| Interfering Substances: Bilirubin conjugated (20 mg/dL) at 113 mg/dL C3 | +4% Change (Predicate: Not specified) |
| Interfering Substances: Hemoglobin (500 mg/dL) at 114 mg/dL C3 | -2% Change (Predicate: Not specified) |
| Interfering Substances: Triglycerides (1000 mg/dL) at 107 mg/dL C3 | -8% Change (Predicate: Not specified) |
| Analytical Range (Normal: 27-360 mg/dL) | Matched to Predicate |
| C4 Method (Predicate: Dade Behring BNA Complement C4 method) | |
| Precision (Total) @ 15.7 mg/dL | 1.9% @ 19.3 mg/dL (Predicate: 2.9%) |
| Precision (Total) @ 34.8 mg/dL | 2.0% @ 34.8 mg/dL |
| Precision (Total) @ 49.3 mg/dL | 1.9% @ 49.3 mg/dL |
| Correlation (y=0.92x - 0.3, r=0.989, Syx=1.9 mg/dL) | Matched to Predicate |
| Interfering Substances: Bilirubin unconjugated (20 mg/dL) at 19 mg/dL C4 | 0% Change (Predicate: Not specified) |
| Interfering Substances: Bilirubin conjugated (20 mg/dL) at 21 mg/dL C4 | 0% Change (Predicate: Not specified) |
| Interfering Substances: Hemoglobin (500 mg/dL) at 20 mg/dL C4 | 0% Change (Predicate: Not specified) |
| Analytical Range (Normal: 7.2-96 mg/dL) | Matched to Predicate |
| IgA Method (Predicate: Dade Behring BNA Immunoglobulin A method) | |
| Precision (Total) @ 296 mg/dL | 2.3% @ 122 mg/dL (Predicate: 3.5%) |
| Precision (Total) @ 233 mg/dL | 1.6% @ 233 mg/dL |
| Precision (Total) @ 344 mg/dL | 1.4% @ 344 mg/dL |
| Correlation (y=1.00x - 1, r=0.994, Syx=75.6 mg/dL) | Matched to Predicate |
| Interfering Substances: Bilirubin unconjugated (20 mg/dL) at 180 mg/dL IgA | +2% Change (Predicate: Not specified) |
| Interfering Substances: Bilirubin conjugated (20 mg/dL) at 200 mg/dL IgA | -3% Change (Predicate: Not specified) |
| Interfering Substances: Hemoglobin (500 mg/dL) at 191 mg/dL IgA | +1% Change (Predicate: Not specified) |
| Interfering Substances: Triglycerides (1000 mg/dL) at 184 mg/dL IgA | -8% Change (Predicate: Not specified) |
| Analytical Range (Normal: 45-600 mg/dL) | Matched to Predicate |
| IgG Method (Predicate: Dade Behring BNA Immunoglobulin G method) | |
| Precision (Total) @ 2141 mg/dL | 1.8% @ 772 mg/dL (Predicate: 3.4%) |
| Precision (Total) @ 1317 mg/dL | 2.7% @ 1317 mg/dL |
| Precision (Total) @ 1416 mg/dL | 1.6% @ 1416 mg/dL |
| Correlation (y=0.97x - 3, r=0.994, Syx=118 mg/dL) | Matched to Predicate |
| Interfering Substances: Bilirubin unconjugated (20 mg/dL) at 876 mg/dL IgG | -3.0% Change (Predicate: Not specified) |
| Interfering Substances: Bilirubin conjugated (20 mg/dL) at 953 mg/dL IgG | -3.0% Change (Predicate: Not specified) |
| Interfering Substances: Hemoglobin (500 mg/dL) at 897 mg/dL IgG | +7.0% Change (Predicate: Not specified) |
| Interfering Substances: Triglycerides (1000 mg/dL) at 887 mg/dL IgG | -3.0% Change (Predicate: Not specified) |
| Analytical Range (Normal: 225-3000 mg/dL) | Matched to Predicate |
| IgM Method (Predicate: Dade Behring BNA Immunoglobulin M method) | |
| Precision (Total) @ 69.0 mg/dL | 3.6% @ 69.0 mg/dL (Predicate: 2.4% @ 128 mg/dL) |
| Precision (Total) @ 117 mg/dL | 1.9% @ 117 mg/dL |
| Precision (Total) @ 128 mg/dL | 2.4% @ 128 mg/dL |
| Precision (Total) @ 177 mg/dL | 1.5% @ 177 mg/dL |
| Correlation (y=1.05x - 13.4, r=0.990, Syx=123.5 mg/dL) | Matched to Predicate |
| Interfering Substances: Bilirubin unconjugated (20 mg/dL) at 77 mg/dL IgM | +3.9% Change (Predicate: Not specified) |
| Interfering Substances: Bilirubin conjugated (10 mg/dL) at 85 mg/dL IgM | +1.2% Change (Predicate: Not specified) |
| Interfering Substances: Hemoglobin (500 mg/dL) at 86 mg/dL IgM | 0.0% Change (Predicate: Not specified) |
| Analytical Range (Normal: 30-400 mg/dL) | Matched to Predicate |
| TRF Method (Predicate: Dade Behring BNA Transferrin method) | |
| Precision (Total) @ 127 mg/dL | 2.2% @ 127 mg/dL (Predicate: 2.7% @ 303 mg/dL) |
| Precision (Total) @ 303 mg/dL | 2.7% @ 303 mg/dL |
| Precision (Total) @ 268 mg/dL | 1.9% @ 268 mg/dL |
| Precision (Total) @ 400 mg/dL | 3.1% @ 400 mg/dL |
| Correlation (y=0.96x - 3, r=0.979, Syx=16.8 mg/dL) | Matched to Predicate |
| Interfering Substances: Bilirubin unconjugated (20 mg/dL) at 212 mg/dL TRF | +2% Change (Predicate: Not specified) |
| Interfering Substances: Bilirubin conjugated (20 mg/dL) at 231 mg/dL TRF | 0% Change (Predicate: Not specified) |
| Interfering Substances: Hemoglobin (500 mg/dL) at 261 mg/dL TRF | 0% Change (Predicate: Not specified) |
| Interfering Substances: Triglycerides (1000 mg/dL) at 215 mg/dL TRF | -4% Change (Predicate: Not specified) |
| Analytical Range (Normal: 54-720 mg/dL) | Matched to Predicate |
| Vancomycin Method (Predicate: Bayer Immuno 1 Vancomycin method) | |
| Analytical Range | 0.4-50 µg/mL (Predicate: Not explicitly stated, but implied similar operating range) |
| Precision (Total) @ 6.7 µg/mL | 2.8% @ 8.6 µg/mL (Predicate: 8.9%) |
| Precision (Total) @ 23.3 µg/mL | 3.4% @ 21.5 µg/mL (Predicate: 7.5%) |
| Precision (Total) @ 32.4 µg/mL | 3.9% @ 35.8 µg/mL (Predicate: 8.1%) |
| Correlation (y=1.02x + 0.68, r=0.987, Syx=1.6 µg/dL) | Matched to Predicate |
| Interfering Substances: Bilirubin unconjugated (25 mg/dL) at 15.3 µg/mL Vanco | +0.3% Change (Predicate: Not specified) |
| Interfering Substances: Bilirubin conjugated (25 mg/dL) at 15.9 µg/mL Vanco | +0.6% Change (Predicate: Not specified) |
| Interfering Substances: Hemoglobin (1000 mg/dL) at 16.1 µg/mL Vanco | -7.7% Change (Predicate: Not specified) |
| Interfering Substances: Lipemia (1000 mg/dL) at 16.0 µg/mL Vanco | -0.2% Change (Predicate: Not specified) |
| Valproic Acid Method (Predicate: Abbott TDx Valproic Acid method) | |
| Analytical Range | 0.3-150 µg/mL (Predicate: Not explicitly stated, but implied similar operating range) |
| Precision (Total) @ 37.5 µg/mL | 3.8% @ 20.3 µg/mL (Predicate: 3.4%) |
| Precision (Total) @ 75 µg/mL | 2.4% @ 60.2 µg/mL (Predicate: 3.4%) |
| Precision (Total) @ 125 µg/mL | 2.8% @ 108.5 µg/mL (Predicate: 3.7%) |
| Correlation (y=1.11x + 0.27, r=0.993, Sux=3.79 µg/dL) | Matched to Predicate |
| Interfering Substances: Bilirubin unconjugated (25 mg/dL) at 82.5 µg/mL VA | +6% Change (Predicate: Not specified) |
| Interfering Substances: Bilirubin conjugated (25 mg/dL) at 78.1 µg/mL VA | +1% Change (Predicate: Not specified) |
| Interfering Substances: Hemoglobin (1000 mg/dL) at 80.2 µg/mL VA | +4% Change (Predicate: Not specified) |
| Interfering Substances: Lipemia (500 mg/dL) at 80.8 µg/mL VA | +3% Change (Predicate: Not specified) |
2. Sample Sizes Used for the Test Set and Data Provenance:
- C3 Method: n = 101 samples for correlation. Data provenance is not specified (e.g., country of origin, retrospective or prospective).
- C4 Method: n = 94 samples for correlation. Data provenance is not specified.
- IgA Method: n = 97 samples for correlation. Data provenance is not specified.
- IgG Method: n = 97 samples for correlation. Data provenance is not specified.
- IgM Method: n = 89 samples for correlation. Data provenance is not specified.
- TRF Method: n = 106 samples for correlation. Data provenance is not specified.
- Vancomycin Method: n = 55 samples for correlation. Data provenance is not specified.
- Valproic Acid Method: n = 55 samples for correlation. Data provenance is not specified.
For precision and interfering substances studies, specific sample numbers are not given, but they are implied to be sufficient for the reported percentages. The studies appear to be clinical validation studies performed by the manufacturer. These are typically prospective studies using human serum samples.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts:
The concept of "experts" and "ground truth" as typically applied to image-based diagnostic AI algorithms (e.g., radiologists) is not directly applicable here. For these in vitro diagnostic (IVD) devices, the "ground truth" is established by:
- Predicate Device Measurements: The performance of the predicate device (BNA Complement C3 method, etc.) serves as the reference standard for correlation studies. The predicate devices themselves would have undergone rigorous validation.
- Reference Methods/Laboratory Standards: For precision, linearity, and interference studies, the "ground truth" for the concentration levels would be established using validated internal laboratory methods, certified reference materials, or gravimetric/volumetric standards, not expert readers.
Therefore, no information on "number of experts" or their "qualifications" is provided, as it's not relevant to this type of device comparison.
4. Adjudication Method for the Test Set:
Not applicable. Adjudication methods (e.g., 2+1, 3+1) are primarily used in studies where human interpretation of data (like medical images) is involved and discrepancies need to be resolved. For quantitative IVD devices measuring biomolecules, agreement is by quantitative comparison to a reference method or predicate device, not by expert consensus on interpretation.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done:
No, an MRMC comparative effectiveness study was not done. MRMC studies are specific to evaluating the impact of an AI system on human reader performance, particularly in diagnostic imaging. This document describes the performance of an automated laboratory assay, not an AI algorithm assisting human interpretation.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done:
Yes, the studies reported are essentially "standalone" performance studies for the Bayer ADVIA IMS systems. These are automated systems performing quantitative measurements. The performance metrics (precision, correlation, analytical range, interference) evaluate the device's inherent analytical capabilities, independent of human interaction during the measurement process, aside from loading samples and reagents.
7. The Type of Ground Truth Used:
The ground truth used for these studies is primarily:
- Predicate Device Measurements: For correlation studies, the measurements obtained from the predicate devices (e.g., Dade Behring BNA methods, Bayer Immuno 1, Abbott TDx) are considered the reference against which the ADVIA IMS device is compared. This demonstrates "substantial equivalence."
- Known Concentrations/Spiked Samples: For precision, analytical range, and interfering substance studies, the "ground truth" is established by using samples with known concentrations (e.g., control materials, diluted samples, samples spiked with interfering substances).
8. The Sample Size for the Training Set:
The concept of a "training set" applies to machine learning or AI models that learn from data. These documents describe the performance validation of traditional IVD assays. Therefore, there is no "training set" in the context of an AI algorithm, and no sample size is applicable or reported for such a purpose. The data presented are for validation/testing of the assay.
9. How the Ground Truth for the Training Set Was Established:
As there is no "training set" described for a machine learning model, this question is not applicable. The methods described are for chemical assays, not AI algorithms requiring training data.
§ 866.5240 Complement components immunological test system.
(a)
Identification. A complement components immunological test system is a device that consists of the reagents used to measure by immunochemical techniques complement components C1q , C1r , C1s , C2 , C3 , C4 , C5 , C6 , C7 , C8 , and C9 , in serum, other body fluids, and tissues. Complement is a group of serum proteins which destroy infectious agents. Measurements of these proteins aids in the diagnosis of immunologic disorders, especially those associated with deficiencies of complement components.(b)
Classification. Class II (performance standards).