(18 days)
The Piccolo Total Cholesterol - Capillary Test System used with the Piccolo xpress Chemistry Analyzer is intended for the in vitro quantitative determination of total cholesterol in capillary (fingerstick) heparinized whole blood in a clinical laboratory setting or point-of-care location.
Cholesterol measurements are used in the diagnosis and treatment of disorders involving excess cholesterol in the blood and lipid and lipoprotein disorders.
The Piccolo® Lipid Panel - Capillary Reagent Disc (which contains the Piccolo® Total Cholesterol - Capillary Test System) is designed to separate a heparinized whole blood sample into plasma and blood cells. The disc meters the required quantity of plasma and diluent, mixes the plasma with diluent, and delivers the mixture to the reaction cuvettes along the disc perimeter. The diluted plasma mixes with the reagent beads, initiating the chemical reactions that are then monitored by the analyzer.
Here's a summary of the acceptance criteria and the study that proves the device meets them, based on the provided text:
Device: Piccolo® Total Cholesterol - Capillary Test System
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria Category | Specific Metric | Acceptance Criteria (Implied/Stated) | Reported Device Performance |
|---|---|---|---|
| Linearity/Dynamic Range | Lower Limit | Not explicitly stated as an acceptance criterion, but established as 20 mg/dL | 20 mg/dL |
| Upper Limit | Not explicitly stated as an acceptance criterion, but established as 520 mg/dL | 520 mg/dL | |
| Slope (compared to reference method) | Not explicitly stated, but typically expected to be close to 1 | 0.989 | |
| Intercept (compared to reference method) | Not explicitly stated, but typically expected to be close to 0 | 19.7 | |
| Correlation Coefficient (r) | Not explicitly stated, but typically >0.99 for good linearity | 0.999 | |
| Precision | Within-Run %CV (Low Concentration e.g. Serum 1) | Not explicitly stated, but typically 0.95 or >0.98 for good correlation | Site 1: 0.991 (Linear & Deming), Site 2: 0.991 (Linear & Deming), Site 3: 0.987 (Linear & Deming), Combined: 0.989 (Linear & Deming). These very high R² values indicate excellent correlation. |
| Slope (Linear Regression) | Not explicitly stated, but expected to be close to 1 within a tight CI | Site 1: 0.97 (0.95 to 0.98 CI), Site 2: 0.96 (0.95 to 0.97 CI), Site 3: 0.96 (0.95 to 0.98 CI), Combined: 0.96 (0.95 to 0.97 CI). These are all very close to 1. | |
| Intercept (Linear Regression) | Not explicitly stated, but expected to be close to 0 within a tight CI | Site 1: 2.42 (-0.10 to 4.95 CI), Site 2: 2.83 (0.28 to 5.37 CI), Site 3: 1.30 (-1.73 to 4.33 CI), Combined: 2.20 (0.64 to 3.76 CI). These are all close to 0, often including 0 in their respective 95% CIs. | |
| Accuracy | CRMLN Certification | Certification by CRMLN (Cholesterol Reference Method Laboratory Network) | Accuracy was established by completing the certification process of the CRMLN. (This implies the device meets the accuracy standards required for CRMLN certification.) |
Note: The document implicitly defines successful performance by demonstrating strong correlation, linearity, and acceptable precision in comparison to the predicate device and established reference methods. Explicit numerical acceptance criteria (e.g., "R² must be >0.98") are not stated in this summary, but the reported results clearly meet the standards for substantial equivalence.
2. Sample Size Used for the Test Set and Data Provenance
- Linearity Test: Samples across the measuring range were used. The specific number of distinct samples isn't given, but testing was done on 20 Piccolo analyzers. The data provenance is not explicitly stated (e.g., country of origin, retrospective/prospective), but it would typically be prospective for device validation.
- Precision Test:
- Serum Samples for Within-Run and Total Precision: 160 measurements for each serum level (Serum 1 & Serum 2).
- Whole Blood Precision: 5 fresh whole blood samples were tested 7 times each on 4 analyzers over 3 hours. This means a total of 5 * 7 = 35 measurements per analyzer, and 35 * 4 = 140 measurements in total for whole blood precision (though the table states 28 for each sample). The document also states "a total of 20 analyzers were used" for this specific test, which introduces some ambiguity on the total number of data points if each sample was run on all 20 analyzers.
- The data provenance is not explicitly stated.
- Method Comparison Test:
- Site 1: 216 samples
- Site 2: 210 samples
- Site 3: 213 samples
- Combined Data: 639 samples
- The data provenance (e.g., country of origin, retrospective or prospective) for these clinical samples is not explicitly mentioned in the provided text. Based on the context of a 510(k) submission, these would typically be prospectively collected samples from a clinical laboratory or point-of-care setting.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
The ground truth for the device's performance (specifically for method comparison) was established by comparison to the Roche Cholesterol Test on the Cobas 6000 Analyzer.
For accuracy, the device was certified by the CRMLN (Cholesterol Reference Method Laboratory Network). This implies adherence to established reference methods and standards, which are overseen by expert bodies but doesn't involve individual experts establishing ground truth for each test case in the same way as, for example, image interpretation. The "experts" are essentially the established reference methods and organizations that define accurate cholesterol measurement.
4. Adjudication Method for the Test Set
Not applicable. This device is an in-vitro diagnostic (IVD) quantitative test system, not an interpretive device like an AI-powered diagnostic imaging tool that would require human expert adjudication of results. The "ground truth" is analytical (e.g., reference assay results, certified reference materials).
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done
No, an MRMC comparative effectiveness study was not done. This type of study design is typically used for interpretive diagnostic devices where human readers (e.g., radiologists) interpret cases with and without AI assistance to measure the AI's impact on human performance. This device is an automated quantitative test system.
6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, the studies presented (Linearity, Precision, Method Comparison) represent the standalone performance of the Piccolo® Total Cholesterol - Capillary Test System as an algorithm/device-only system. Its output is a quantitative value, not an interpretation requiring human input for its direct result.
7. The Type of Ground Truth Used
- For Linearity, Precision, and Method Comparison: The ground truth was established by comparison to:
- Reference materials or samples of known concentration (for linearity, precision).
- A legally marketed predicate device: The Roche Cholesterol Test on the Cobas 6000 Analyzer (for method comparison).
- For Accuracy: The ground truth was established by "completing the certification process of the CRMLN" (Cholesterol Reference Method Laboratory Network). CRMLN certification signifies that the method is traceable to the Centers for Disease Control and Prevention (CDC) reference method for cholesterol.
8. The Sample Size for the Training Set
The provided document describes validation studies (linearity, precision, method comparison) for a predicate-based 510(k) submission for an IVD device. It does not mention any "training set" in the context of machine learning. The device is a chemical analyzer using enzymatic endpoint reactions, not an AI/ML algorithm that requires a training set in the typical sense.
9. How the Ground Truth for the Training Set Was Established
As this is not an AI/ML device, the concept of a "training set" and its associated ground truth establishment is not applicable as described in the document. The device's operational parameters (e.g., reagent formulations, reaction conditions, calibration) would be developed and optimized through standard analytical chemistry and engineering practices, not machine learning training. Calibration is done via a barcode with factory-calibrated lot-specific data.
§ 862.1175 Cholesterol (total) test system.
(a)
Identification. A cholesterol (total) test system is a device intended to measure cholesterol in plasma and serum. Cholesterol measurements are used in the diagnosis and treatment of disorders involving excess cholesterol in the blood and lipid and lipoprotein metabolism disorders.(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.