(83 days)
For in vitro diagnostic use in the quantitative measurement of lipoprotein(a) (Lp(a)) in human serum or plasma on the ADVIA Chemistry systems. Measurement of Lp(a) may aid in the diagnosis of disorders of lipid (fat) metabolism and assessing persons at risk for cardiovascular diseases when used in conjunction with clinical evaluation and other lipoprotein tests.
The ADVIA® Chemistry Lipoprotein(a) calibrators is intended for use in the calibration of ADVIA® Chemistry systems for the ADVIA Chemistry Lipoprotein(a) (LPA) assay.
The Lipoprotein(a) reagents are ready-to-use liquid reagents packaged for use on the automated ADVIA 1650 Chemistry system. They are supplied as a 100 tests/wedge, 2 wedges/kit. ADVIA Chemistry Lipoprotein(a) calibrator is a single analyte, human serum based product containing human lipoprotein (a). The kit consists of 1 vial each of 5 calibrator levels which are lyophilized. The target concentrations of these calibrators are 7.5, 15, 30, 65, and 95 mg/dL. The volume per vial (after reconstitution with deionized water) is 1.0 mL. Deionized water is recommended to be used as a zero calibrator.
Here's a breakdown of the acceptance criteria and study details for the ADVIA® Chemistry Lipoprotein(a) Assay, based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance
The document doesn't explicitly state formal "acceptance criteria" in a definitive, quantified table for all performance characteristics. Instead, it describes studies that "gave acceptable results when compared to the predicate device" and lists the numerical outcomes of these studies. Therefore, this table is constructed by inferring the implied acceptance from the reported "acceptable results" and direct comparisons to the predicate device where available.
| Performance Characteristic | Implied Acceptance Criteria (from predicate similarity/acceptable results) | Reported Device Performance (ADVIA Chemistry Lipoprotein(a) Assay) |
|---|---|---|
| Precision | Within acceptable variability for a diagnostic assay, comparable to predicate. (Precise numerical criteria not explicitly stated but implied by acceptable results) | Within Run CV: 0.9% - 1.2% (lower concentrations), 0.8% (higher concentration) |
| Total CV: 1.3% - 1.6% (lower concentrations), 1.6% (higher concentration) | ||
| Linearity/Reportable Range | Linear performance across the intended measuring range. | Linear/measuring range: 10.00 mg/dL - 85.00 mg/dL |
| Limit of Blank (LoB) | Below 6.0 mg/dL (claim supported) | 5.35 mg/dL |
| Limit of Detection (LoD) | Below 9.0 mg/dL (claim supported) | 8.90 mg/dL |
| Limit of Quantitation (LoQ) | Below 10.0 mg/dL (claim supported) | 9.02 mg/dL |
| Method Comparison (Serum) | Strong correlation (R > 0.99) and agreement with predicate device. Slope and intercept confidence intervals encompassing 1 and 0 respectively for good agreement. | Correlation Coefficient: 0.99 |
| Linear Regression (y = mx + b): y = 1.01x - 1.02 mg/dL | ||
| Slope 95% CI: 1.00 - 1.02 | ||
| Intercept 95% CI: -1.47 - -0.57 | ||
| Matrix Comparison (Plasma) | Strong correlation (R > 0.99) and agreement with predicate device. Slope and intercept confidence intervals encompassing 1 and 0 respectively for good agreement. | Correlation Coefficient: 0.99 |
| Linear Regression (y = mx + b): y = 1.01x - 0.98 mg/dL | ||
| Slope 95% CI: 0.99 - 1.02 | ||
| Intercept 95% CI: -1.49 - -0.47 | ||
| Analytical Specificity | No significant interference (>10% variance) from common interferents (icterus, lipemia, hemolysis) at specified concentrations. | No significant interference found at: |
- Unconjugated bilirubin: 0-60 mg/dL
- Conjugated bilirubin: 0-60 mg/dL
- Intralipid: 0-1000 mg/dL
- Hemoglobin: 0-1000 mg/dL |
2. Sample Size Used for the Test Set and Data Provenance
- Precision: Not explicitly stated as a separate "test set" for precision, but studies used:
- Serum sample pools and serum-based controls.
- Each sample assayed 2 replicates per run, 2 runs per day, for at least 20 days.
- Linearity/Assay Reportable Range:
- Sample Size: Nine diluted samples (prepared from high and low serum pools).
- Limit of Blank, Limit of Detection, Limit of Quantitation:
- Sample Size: 160 replicates of "zero" serum pool and several serum pools (number not specified) with Lp(a) concentrations up to 4x LoD level.
- Method Comparison (Serum):
- Sample Size: 68 serum samples.
- Matrix Comparison (Plasma):
- Sample Size: 44 plasma samples.
- Analytical Specificity:
- Specific concentrations of interferents (unconjugated bilirubin 0-60 mg/dL, conjugated bilirubin 0-60 mg/dL, Intralipid 0-1000 mg/dL, hemoglobin 0-1000 mg/dL) were tested with samples at 3 specific Lp(a) levels (e.g., 14, 28, 47 mg/dL for bilirubin). The exact number of individual samples/replicates isn't specified beyond "in 14, 28, and 47 mg/dL lipoprotein (a) samples."
- Data Provenance: Not specified (e.g., country of origin, retrospective or prospective). The samples are referred to as "human serum" or "human plasma."
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
This type of blood assay does not typically use human experts to establish "ground truth" in the same way an imaging device or AI algorithm would. Instead, the ground truth for an in vitro diagnostic assay is established through:
- Reference Methods: Comparison against an established, legally marketed (predicate) device.
- Known Concentrations: Use of accurately prepared and characterized calibrators and controls with known concentrations of the analyte.
- Spiking Experiments: Addition of known amounts of interferents or analyte to assess impact.
The document implicitly relies on the predicate device (Randox Lipoprotein (a) assay on the Hitachi 717) as a "ground truth" reference for method and matrix comparison studies.
4. Adjudication Method for the Test Set
Not applicable. The "test set" here refers to clinical samples or control materials analyzed by the device, and their values are compared against a predicate device or expected values, not adjudicated by experts in a qualitative sense.
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 is an in vitro diagnostic assay for quantitative measurement of a biomarker, not an imaging device or an AI application that assists human readers.
6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, this entire submission describes the standalone performance of the ADVIA Chemistry Lipoprotein(a) assay. The device measures Lp(a) concentration directly on the ADVIA Chemistry systems without human interpretation or intervention in the measurement process itself, other than preparing samples and operating the instrument.
7. The Type of Ground Truth Used
The ground truth used for performance validation is primarily:
- Predicate Device Measurements: For method and matrix comparisons, the results from the legally marketed Randox Lipoprotein(a) assay on the Hitachi 717 are considered the comparative "ground truth" for demonstrating substantial equivalence.
- Known Concentrations/Prepared Samples: For linearity, LoB/LoD/LoQ determinations, and analytical specificity, samples with known concentrations (e.g., "zero" serum pool, serum pools with Lipoprotein (a) concentration up to 4 x LOD level, diluted samples with expected values, samples with spiked interferents) serve as the ground truth.
8. The Sample Size for the Training Set
The document does not describe a "training set" in the context of machine learning or AI. This device is a classical in vitro diagnostic immunoassay, not an AI algorithm that requires training data. The development and calibration of such assays involve extensive R&D, but the data used in those phases are not typically referred to as a "training set" in this manner.
9. How the Ground Truth for the Training Set Was Established
Not applicable, as there is no "training set" in the AI/machine learning sense for this device. The development of the assay's reagents and methodologies would have involved established chemical and immunological principles, and calibration of the assay (using the ADVIA Chemistry Lipoprotein(a) calibrator) uses products with assigned Lp(a) concentrations.
§ 866.5600 Low-density lipoprotein immunological test system.
(a)
Identification. A low-density lipoprotein immunological test system is a device that consists of the reagents used to measure by immunochemical techniques the low-density lipoprotein in serum and other body fluids. Measurement of low-density lipoprotein in serum may aid in the diagnosis of disorders of lipid (fat) metabolism and help to identify young persons at risk from cardiovascular diseases.(b)
Classification. Class II (performance standards).