(80 days)
A glucose test system is a device intended to measure glucose quantitatively in blood and other body fluids. Glucose measurements are used in the diagnosis and treatment of carbohydrate metabolism disorders including diabetes mellitus, neonatal hypoglycemia, and idiopathic hypoglycemia, and of pancreatic islet cell carcinoma.
The Glucose assay is used for the quantitation of glucose in human serum, plasma, urine, or cerebrospinal fluid (CSF). Glucose is phosphorylated by hexokinase (HK) in the presence of adenosine triphosphate (ATP) and magnesium ions to produce glucose-6-phosphate (G-6-P) and adenosine diphosphate (ADP). Glucose-6-phosphate dehydrogenase (G-6-PDH) specifically oxidizes G-6-P to 6-phosphogluconate with the concurrent reduction of nicotinamide adenine dinucleotide (NAD) to nicotinamide adenine dinucleotide reduced (NADH). One micromole of NADH is produced for each micromole of glucose consumed. The NADH produced absorbs light at 340 nm and can be detected spectrophotometrically as an increased absorbance.
Here's a breakdown of the acceptance criteria and study information for the Glucose assay, as derived from the provided 510(k) summary:
Acceptance Criteria and Device Performance
| Acceptance Criteria Category | Specific Criteria | Reported Device Performance (Glucose Assay) |
|---|---|---|
| Correlation with Predicate Device (Hitachi 917 Glucose/HK assay) | Acceptable correlation with correlation coefficient as close to 1.0 as possible, slope as close to 1.0 as possible, and Y-intercept as close to 0 as possible. | AEROSET System vs. Hitachi 917: |
- Serum: r = 0.9995, slope = 1.09, Y-intercept = -5.50 mg/dL
- Urine: r = 0.9998, slope = 1.09, Y-intercept = -1.35 mg/dL
- CSF: r = 0.9998, slope = 1.09, Y-intercept = -4.29 mg/dL
ARCHITECT c8000 System vs. Hitachi 917: - Serum: r = 0.9993, slope = 1.06, Y-intercept = -4.54 mg/dL
- Urine: r = 0.9998, slope = 1.04, Y-intercept = -2.67 mg/dL
- CSF: r = 0.9997, slope = 1.04, Y-intercept = -3.89 mg/dL |
| Correlation between AEROSET and ARCHITECT c8000 Systems | Acceptable correlation with correlation coefficient as close to 1.0 as possible, slope as close to 1.0 as possible, and Y-intercept as close to 0 as possible. | ARCHITECT c8000 System vs. AEROSET System: - Serum: r = 0.9996, slope = 0.97, Y-intercept = 0.85 mg/dL
- Urine: r = 0.9999, slope = 0.96, Y-intercept = -1.36 mg/dL
- CSF: r = 0.9998, slope = 0.95, Y-intercept = 0.22 mg/dL |
| Precision (Total %CV) | Acceptable precision, typically low %CV values which often indicate less than 5% or 10% depending on the analyte and clinical context. (Specific numerical criteria not explicitly stated but implied by acceptable performance.) | AEROSET System: - Serum: Level 1: 1.06%, Level 2: 1.41%
- Urine: Level 1: 1.66%, Level 2: 1.35%
- CSF: Level 1: 1.55%, Level 2: 1.69%
ARCHITECT c8000 System: - Serum: Level 1: 2.15%, Level 2: 1.51%
- Urine: Level 1: 1.01%, Level 2: 0.80%
- CSF: Level 1: 1.07%, Level 2: 1.17% |
| Linearity (Quantitative Range) | Expected to be linear over a clinically relevant range. | - Serum/Plasma: Linear from 5 to 800 mg/dL - Urine/CSF: Linear from 1 to 800 mg/dL |
| Limit of Quantitation (Sensitivity) | Expected to be sensitive enough for clinical utility. | - Serum: 5.0 mg/dL - Urine/CSF: 1.0 mg/dL |
Study Details
-
Sample size used for the test set and the data provenance:
- The document does not explicitly state the sample size used for the comparative performance studies (method comparison and precision studies).
- Data Provenance: Not specified, but generally for medical device submissions in the US, data often originates from clinical sites within the US or from sites that adhere to similar regulatory and ethical guidelines. No country of origin is mentioned. The studies appear to be prospective in nature, as they involve testing the new Glucose assay in comparison to a predicate device and conducting precision studies.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- This is a clinical laboratory assay, not an imaging or diagnostic AI device that typically relies on human experts for ground truth. The "ground truth" for the test set is established by the predicate device (Hitachi 917 Glucose/HK assay) and the inherent chemical measurement of the analyte itself. There are no "experts" in the traditional sense involved in establishing ground truth for these types of assays; rather, it relies on accepted analytical methods.
-
Adjudication method for the test set:
- Not applicable. This is a quantitative chemical assay where the "adjudication" is the direct measurement and comparison of numerical results between the new device and the predicate device. There is no subjective human interpretation involved that would require an adjudication method like 2+1.
-
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:
- No, this is not an MRMC study. MRMC studies are typically performed for imaging or interpretive diagnostic AI systems where human readers evaluate cases with and without AI assistance. This device is a quantitative in vitro diagnostic assay.
-
If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
- Yes, the performance described is a standalone performance of the Glucose assay when run on the AEROSET and ARCHITECT c8000 systems. The device itself performs the measurement without human interpretation of the result, beyond reading the numerical value.
-
The type of ground truth used:
- The ground truth for the comparative studies is based on the measurements obtained from the predicate device, the Glucose/HK assay on the Hitachi 917 Analyzer. For linearity and sensitivity, the ground truth would be the known concentrations of calibrators or reference materials.
-
The sample size for the training set:
- The document does not mention a training set in the context of machine learning or AI. This device is a traditional chemical assay, not an AI/ML-based device.
-
How the ground truth for the training set was established:
- Not applicable, as there is no training set in the AI/ML sense.
§ 862.1345 Glucose test system.
(a)
Identification. A glucose test system is a device intended to measure glucose quantitatively in blood and other body fluids. Glucose measurements are used in the diagnosis and treatment of carbohydrate metabolism disorders including diabetes mellitus, neonatal hypoglycemia, and idiopathic hypoglycemia, and of pancreatic islet cell carcinoma.(b)
Classification. Class II (special controls). The device, when it is solely intended for use as a drink to test glucose tolerance, is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 862.9.