(182 days)
The Envoy® 500 HDL Cholesterol Reagent and Envoy® 500 HDL Calibrator are intended for use with the Envoy® 500 Chemistry System as a system for the quantitative determination of high density lipoprotein (HDL) cholesterol in serum and plasma. HDL cholesterol measurements are used in the diagnosis and treatment of lipid disorders (such as diabetes mellitus), atherosclerosis, and various liver and renal diseases, and for the assessment of the risk of developing cardiovascular disease.
The Envoy® 500 HDL Cholesterol Reagent is a two-part reagent that is calibrated with the Envoy® 500 HDL Calibrator for use with the Envoy® 500 Chemistry System. This reagent determines high density lipoprotein cholesterol through the accelerator selective detergent methodology. This procedure measures HDL-cholesterol in a two step reaction sequence. In the first step, non-HDL cholesterol is rendered non-reactive. In the second step, HDL cholesterol is solubilized using a selective detergent and reacts to produce a red chromogen.
Here's an analysis of the acceptance criteria and study detailed in the provided text for the Envoy® 500 HDL Cholesterol Reagent and Envoy® 500 HDL Calibrator:
1. Table of Acceptance Criteria and Reported Device Performance
The provided text focuses on demonstrating the performance of the device without explicitly stating pre-defined "acceptance criteria" in a structured table. However, we can infer the acceptance criteria based on the reported performance for each study.
| Performance Characteristic | Stated Acceptance Criteria (Inferred) | Reported Device Performance |
|---|---|---|
| Usable Range (Linearity) | - Recovery of linearity related solutions across the expected range (5 to at least 150 mg/dL). | - Linear range from 5 to at least 150 mg/dL. |
- Linear regression: (Envoy Recoveries) = - 0.6 mg/dL + 19.284 x (Dilution Factor), r = 0.9998, syx = 1.11 mg/dL, n = 44, range = 0.1 to 173.5 mg/dL. |
| Limit of Detection (LoD) | - LoD with proportions of false positives (a) less than 5% and false negatives (B) less than 5%. | - Limit of Blank (LoB): 0.29 mg/dL. - Reported LoD: 0.46 mg/dL.
- Proportions of false positives (a) less than 5% and false negatives (B) less than 5%. |
| Precision | - Precision to be demonstrated by replicate assay of control sera. | Level 1 (mean 36.8 mg/dL): - Within Run: 1 SD = 0.52, %CV = 1.4%
- Total: 1 SD = 0.72, %CV = 2.0%
Level 2 (mean 71.1 mg/dL): - Within Run: 1 SD = 0.68, %CV = 1.0%
- Total: 1 SD = 1.25, %CV = 1.8% |
| Correlation | - Comparison to a commercially available method using least squares linear regression and Passing-Bablok regression. | Least Squares Linear Regression: - Envoy 500 = 0.7 mg/dL + 1.021 × Competitive Method
- r = 0.995
- S = 2.42 mg/dL
- 95% CI y-intercept: 0.07 to 1.38 mg/dL
- 95% CI slope: 1.010 to 1.032
Passing-Bablok Regression: - Envoy 500 = 0.7 mg/dL + 1.015 × Competitive Method
- 95% CI y-intercept: 0.2 to 2.0 mg/dL
- 95% CI slope: 1.000 to 1.029 |
| Stability | - Statistical estimates of total imprecision to be less than 1.5 mg/dL over claimed stability periods (14-day onboard reagent, 7-day calibration). | - Total imprecision less than 1.5 mg/dL for 14-day onboard reagent stability and 7-day calibration stability. |
2. Sample Sizes Used for the Test Set and Data Provenance
-
Usable Range:
- Sample Size: n = 44 (reference pools prepared by diluting human HDL cholesterol concentrate with stripped human serum pool).
- Data Provenance: Not explicitly stated (e.g., country of origin). The samples are referred to as "human HDL cholesterol concentrate" and "stripped human serum pool," suggesting they are derived from human biological materials.
- Retrospective/Prospective: Not specified.
-
Limit of Detection:
- Sample Size: 40 blank samples and 40 low-level samples.
- Data Provenance: Not explicitly stated.
- Retrospective/Prospective: Not specified.
-
Precision:
- Sample Size: Level 1: n = 45; Level 2: n = 48 (replicate assays of commercially available control sera).
- Data Provenance: "Commercially available control sera." Not explicitly stated (e.g., country of origin).
- Retrospective/Prospective: Not specified, but generally implies a prospective experimental setup for precision studies.
-
Correlation:
- Sample Size: n = 312 (mixed serum and plasma specimens).
- Data Provenance: "Collected from adult patients." Not explicitly stated (e.g., country of origin).
- Retrospective/Prospective: Not specified, but "collected from adult patients" often implies a retrospective collection or a prospective collection for this specific study.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
This device is an in-vitro diagnostic (IVD) reagent for determining a biochemical analyte (HDL cholesterol). The ground truth for such devices is established through reference methods or by the intrinsic properties of reference materials, not typically by human experts making diagnoses or interpretations like in imaging studies.
- For Usable Range, ground truth is based on known dilution factors of a reference standard.
- For Limit of Detection, ground truth is based on blank samples and low-level samples, with expected very low or zero concentrations.
- For Precision, ground truth is the "true" concentration of the control sera, established by the manufacturer of the control sera or a reference lab, not individual experts.
- For Correlation, the ground truth is the measurement obtained by a "commercially available method" (the predicate or another established method), not human experts.
Therefore, the concept of "number of experts" and "qualifications of those experts" as typically applied in AI/ML performance evaluation (e.g., radiologists interpreting images) is not directly applicable here.
4. Adjudication Method for the Test Set
As explained in point 3, the ground truth for chemical analyte measurements is typically established by reference methods or material properties, not through expert adjudication in the classic sense. Therefore, no adjudication method (like 2+1 or 3+1 consensus) is described or applicable for this type of device.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done
No, an MRMC comparative effectiveness study was not done. This device is an in-vitro diagnostic (IVD) reagent, not an imaging or diagnostic AI tool that assists human readers. Its performance is evaluated biochemically against reference methods and statistical parameters.
6. If a Standalone (i.e., Algorithm Only Without Human-in-the-Loop Performance) Was Done
Yes, the studies presented (Usable Range, Limit of Detection, Precision, Correlation, Stability) evaluate the performance of the Envoy® 500 HDL Cholesterol Reagent and Calibrator directly on serum/plasma samples using the Envoy® 500 Chemistry System. This represents a standalone performance evaluation of the assay system without human interpretation being a primary variable in the measurement outcome. The role of "human-in-the-loop" would be limited to operating the analyzer and interpreting the numerical results, not making subjective diagnoses from raw data.
7. The Type of Ground Truth Used
- Usable Range: The ground truth is based on the known concentrations of HDL cholesterol in reference pools prepared by diluting a human HDL cholesterol concentrate with stripped human serum. The dilution factors provide the expected "true" values.
- Limit of Detection: The ground truth is based on blank samples (expected zero concentration) and low-level samples with known, very low concentrations.
- Precision: The ground truth is the assigned value of commercially available control sera, which are typically established through rigorous inter-laboratory comparisons or against reference methods by the control manufacturer.
- Correlation: The ground truth is the measurement obtained by a "commercially available method" (the predicate device or another established method) used as a comparative standard.
- Stability: The ground truth is again the assigned value of serum controls.
In summary, the ground truth is established through a combination of:
- Reference materials with known concentrations.
- Established analytical methods.
- Commercially validated control materials.
8. The Sample Size for the Training Set
The provided summary describes performance evaluation studies. For an IVD reagent like this, the "training set" doesn't apply in the same way as it would for an AI/ML algorithm. Reagents and calibrators are developed through a formulation and optimization process that involves wet lab experimentation, but this experimental data is not typically referred to as a "training set" in the context of device submission. The studies detailed here are for validation of the finalized product.
Therefore, no specific sample size for a "training set" is mentioned because it's not a relevant concept for this type of device.
9. How the Ground Truth for the Training Set Was Established
Given that the concept of a "training set" is not applicable here as for AI/ML algorithms, the establishment of ground truth for such a set is also not relevant or described. The performance data presented demonstrates the analytical characteristics of the final reagent and calibrator.
§ 862.1475 Lipoprotein test system.
(a)
Identification. A lipoprotein test system is a device intended to measure lipoprotein in serum and plasma. Lipoprotein measurements are used in the diagnosis and treatment of lipid disorders (such as diabetes mellitus), atherosclerosis, and various liver and renal diseases.(b)
Classification. Class I (general controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to § 862.9.