(457 days)
The Alc GEAR System is intended for in vitro diagnostic use only for the quantitative measurement of the percent hemoglobin A 1 (%HbA1c) from finger-stick blood or venous whole blood collected in either EDTA or sodium fluoride (NaF) for clinical laboratory and point of care use. The measurement of HbAlc is recommended to monitor long-term glycemic control of persons previously diagnosed with diabetes mellitus. This test is not for screening or diagnosis of diabetes.
The Alc GEAR instrument is a fully automated desktop electric spectrophotometer that measures %HbAlc in human whole blood using a dedicated reagent (MEDIDAS HbA1c). The system illuminates a 660 nm LED (Light Emitting Diode) through the test material and quantitatively measures the percent of hemoglobin Alc in the total hemoglobin (%HbA1c) by means of light absorbance changes and a non-linear calibration curve. The system includes the Hemoglobin A Ic Analyzer (Alc GEAR), thermal printer, barcode reader, power cable, and fan filter. MEDIDAS HbAIc is composed of a test cartridge, capillary, pipette tip and master calibration card. The cartridge is pre-filled with reagent; latex (reagent R1), antibody (reagent R2), and sample dilute solution.
Here's a summary of the acceptance criteria and the study that proves the device meets them, based on the provided text:
Acceptance Criteria and Device Performance for Alc GEAR System
Note: The provided document is a 510(k) Summary, which typically focuses on demonstrating substantial equivalence to a predicate device rather than explicitly stating pre-defined "acceptance criteria" in a pass/fail format for each performance study. Instead, the document presents performance data to support the claim of substantial equivalence. The table below interprets the "acceptance criteria" as the performance demonstrated by the device, which is considered sufficient for regulatory clearance by showing equivalence to established methods/devices.
1. Table of Acceptance Criteria and Reported Device Performance
| Performance Metric | Acceptance Criteria (Inferred from study design and context of substantial equivalence) | Reported Device Performance |
|---|---|---|
| Linearity | High linearity (r^2 near 1.00) and acceptable recovery rate for the specified analytical range. (e.g., within 90-110% recovery) | Range: 4.0 - 13.1% HbA1c |
| Regression line: y = 0.98x + 0.19 | ||
| r^2: 1.00 | ||
| Recovery (%): 98 - 103 | ||
| Method Comparison (Venous to Venous) | Strong correlation (high r^2) and close agreement (slope near 1, intercept near 0) with established HPLC methods and predicate POC analyzer. | HPLC I (N/A): r^2 = 0.98, y = 1.03x - 0.33 (Range: 4.6-10.6% HbA1c) |
| HPLC 2 (N=40): r^2 = 0.98, y = 0.99x + 0.31 (Range: 4.2-9.8% HbA1c) | ||
| DCA Vantage (N=60): r^2 = 0.99, y = 0.95x - 0.12 (Range: 4.7-11.7% HbA1c) | ||
| Matrix Comparison | Strong correlation (high r^2) and close agreement (slope near 1, intercept near 0) between different sample matrices. | Finger (y) vs. EDTA-venous (x) (N=78): r^2 = 0.99, y = 0.96x + 0.15 (Range: 4.3-9.0% HbA1c) |
| NaF-venous (y) vs. finger (x) (N=46): r^2 = 0.99, y = 1.04x - 0.06 (Range: 4.8-8.8% HbA1c) | ||
| NaF-venous (y) vs. EDTA-venous (x) (N=81): r^2 = 0.99, y = 1.01x + 0.01 (Range: 5.3-10.9% HbA1c) | ||
| Precision (Internal) | Acceptable within-run, between-day, and total CVs for control materials and patient samples across different %HbA1c levels. | Control L (5.2%): Within-run CV=1.26%, Between-day CV=0.51%, Total CV=1.36% |
| Control H (9.0%): Within-run CV=0.85%, Between-day CV=0.26%, Total CV=1.06% | ||
| Sample 1 (5.5%): Total CV=1.12% | ||
| Sample 2 (11.1%): Total CV=1.37% | ||
| Sample 3 (12.1%): Total CV=1.52% | ||
| Precision (External) | Similar acceptable CVs to internal site, demonstrating reproducibility in different settings. | Control L (5.0%): Within-run CV=1.08%, Between-day CV=0.75%, Total CV=1.31% |
| Control H (8.9%): Within-run CV=0.65%, Between-day CV=0.53%, Total CV=0.90% | ||
| Sample 1 (5.2%): Total CV=1.34% | ||
| Sample 2 (8.8%): Total CV=1.05% | ||
| Reproducibility (Overall) | Acceptable between-site and total CVs. | Control L (5.1%): Between-site CV=2.55%, Total CV=2.81% |
| Control H (9.0%): Between-site CV=0.56%, Total CV=0.94% | ||
| POC Precision | Acceptable within-site and overall reproducibility CVs across multiple POC sites and multiple levels of controls and patient samples. | Control 1 (approx 5.2%): Reproducibility Total CV=2.26% |
| Control 2 (approx 7.0%): Reproducibility Total CV=2.11% | ||
| Control 3 (approx 11.0%): Reproducibility Total CV=2.55% | ||
| Sample Low (approx 5.8%): Reproducibility Total CV=3.12% | ||
| Sample Middle (approx 8.0%): Reproducibility Total CV=4.16% | ||
| Sample High (approx 10.8%): Reproducibility Total CV=5.25% | ||
| POC Method Comparison | Strong correlation (r near 0.99) and close agreement (slope near 1, intercept near 0) with a reference HPLC method. | Site 1 (N=47): r=0.995, Slope=0.968, Intercept=0.04 (Range: 4.9-11.9% HbA1c) |
| Site 2 (N=41): r=0.990, Slope=0.976, Intercept=0.12 (Range: 5.4-10.8% HbA1c) | ||
| Site 3 (N=46): r=0.990, Slope=0.989, Intercept=0.08 (Range: 5.0-9.6% HbA1c) | ||
| Interference | No significant interference from common substances within specified concentrations. | No significant interference observed for Free-bilirubin, Conjugated-bilirubin, Rheumatoid factor, Chyle (including triglycerides, phospholipids, free fatty acids), additional Triglycerides, Acetaminophen, Ibuprofen, Glibenclamide, Metformin, Ascorbic acid at tested concentrations. |
| Analytical Specificity (Hb Variants) | Limited or no interference from common hemoglobin variants. (Note: The study revealed significant interference here, which is important for labeling/limitations) | Interference observed: Samples with Hemoglobin C elevated by 24%, Hemoglobin D by 16%, Hemoglobin E by 13%, Hemoglobin S by 14%. Samples with >10% Hemoglobin F decreased by 32%. All variants tested interfered. This results in a limitation specified in the Indications for Use. |
| Analytical Specificity (Modified Hb) | No significant interference from common modified hemoglobins. | No effect found for Carbamylated hemoglobin, Acetylated hemoglobin, Labile hemoglobin. |
| Limit of Detection (LOD) | A detectable range appropriate for HbA1c measurements. | LOD: 2.6% HbA1c. LOB: 2.3% HbA1c. |
| Stability - Real-Time | Reagent cartridge maintains performance for a specified shelf life. | Reagent cartridge can be stored for up to one year at 2-8 °C (36-46 °F). |
2. Sample Sizes and Data Provenance
- Test Set Description / Data Provenance:
- Linearity: Two whole blood samples (low and high HbA1c) mixed in different proportions to obtain 11 samples. Measured in triplicate. Provenance not specified, but implied to be in-house.
- Method Comparisons (Venous): N/A (HPLC 1), N=40 (HPLC 2), N=60 (DCA Vantage). Venous whole blood collected into EDTA tubes from donors. Provenance not specified, but implied to be in-house.
- Matrix Comparison: N=78 (finger vs. EDTA-venous), N=46 (NaF-venous vs. finger), N=81 (NaF-venous vs. EDTA-venous). Samples collected from each donor (finger-stick, EDTA-venous, NaF-venous). Provenance not specified, but implied to be in-house.
- Precision (Internal/External): N=80 for each control material and patient sample (measured for 20 days, in duplicate, twice a day). Tested at an internal site and an external site. Provenance not specified.
- Reproducibility (Overall): N=160 (combining internal and external site data) for each control material. Provenance not specified.
- POC Precision: Control samples (3 levels): N=120 per site (3 sites = 360 total per level). Patient samples (3 levels): N=60 per site (3 sites = 180 total for low), N=120/128 per site (3 sites = 368 total for middle), N=120/128 per site (3 sites = 368 total for high). Provenance not specified, but implied to be across multiple external POC sites.
- POC Method Comparison: N=47 (Site 1), N=41 (Site 2), N=46 (Site 3). Finger-stick blood from donors. Provenance not specified, but implied to be across multiple external POC sites.
- Interference, Analytical Specificity (Hb Variants & Modified Hb), LOD, Stability: Specific sample sizes are not detailed for each substance/variant, but commercial samples or prepared samples were used. Provenance not specified.
3. Number of Experts and Qualifications for Ground Truth
- The document primarily describes analytical performance studies against reference methods or established standards. It does not mention the use of human experts to establish "ground truth" in the way it might for image-based diagnostics.
- For the method comparisons, the "ground truth" or reference values were established by:
- Ion-exchange HPLC methods (specific models not always named, but generally considered gold standard for HbA1c).
- The predicate device, DCA Vantage.
- Tosoh G8 ion-exchange HPLC reference method (for POC method comparison).
- Personnel performing these reference tests are described as "qualified laboratory technicians at a reference laboratory" for the POC method comparison.
- For precision and other analytical studies, the "ground truth" relates to the inherent properties of the control materials or patient samples as measured by the reference techniques.
4. Adjudication Method
- Not applicable. The studies are analytical performance assessments against reference methods or internal calibration, not clinical interpretation requiring expert adjudication.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- No, a MRMC comparative effectiveness study was not done. This type of study is typically associated with diagnostic imaging or other human interpretation tasks, not an automated quantitative assay like HbA1c.
6. Standalone Performance (Algorithm Only)
- Yes, the entire set of performance data represents the standalone performance of the Alc GEAR System. It is an automated instrument that directly measures %HbA1c without a human-in-the-loop interpretation step that would significantly influence the result. Its performance is compared to other analytical methods (like HPLC) which are also standalone assays.
7. Type of Ground Truth Used
- Reference Methods / Gold Standards: Ion-exchange HPLC methods (e.g., Primus Ultra Boronate Affinity HPLC, Tosoh G8, and other unnamed HPLC systems) are frequently used as the reference "ground truth" for HbA1c measurements (e.g., in linearity, method comparison, and analytical specificity studies).
- Predicate Device: The DCA Vantage was used as a comparison method in some studies.
- Expected Values: For linearity, expected values were created by mixing known low and high samples.
- Commercial Controls: Commercial control materials were used in precision studies, for which expected values would be established by the manufacturer through rigorous testing.
8. Sample Size for the Training Set
- The document describes the performance testing of the device. It does not provide information about a "training set" in the context of machine learning algorithms. The Alc GEAR System is described as using light absorbance changes and a non-linear calibration curve, typical of an immunoassay/photometric system rather than a machine learning model that would require a distinct training set. If "calibration" is considered training, the master calibration card (mentioned in Section 11) would set the initial parameters, but the data used to develop that calibration curve is not provided.
9. How the Ground Truth for the Training Set was Established
- As noted above, no "training set" in the machine learning sense is described. The device's operation relies on a "non-linear calibration curve" and a "master calibration card". The ground truth for this calibration would be established by measuring a series of known HbA1c concentration standards (likely traceable to a primary reference method like HPLC) to define the relationship between absorbance and %HbA1c. The details of this calibration process and the data used are not provided in this 510(k) summary.
§ 864.7470 Glycosylated hemoglobin assay.
(a)
Identification. A glycosylated hemoglobin assay is a device used to measure the glycosylated hemoglobins (A1a , A1b , and A1c ) in a patient's blood by a column chromatographic procedure. Measurement of glycosylated hemoglobin is used to assess the level of control of a patient's diabetes and to determine the proper insulin dosage for a patient. Elevated levels of glycosylated hemoglobin indicate uncontrolled diabetes in a patient.(b)
Classification. Class II (performance standards).