(249 days)
For in vitro diagnostic use in the quantitative determination of glucose in human serum, plasma, urine and CSF on the ADVIA 1650 Chemistry system. Such measurements are used in the diagnosis and treatment of carbohydrate metabolism disorders including diabetes mellitus, neonatal hypoglycemia, idiopathic hypoglycemia, and insulin overdose.
The ADVIA 1650 Chemistry Glucose Hexokinase 3 (GLUH 3) method uses a twocomponent reagent. Sample is added to Reagent 1, which contains the buffer, ATP, and NAD. Absorbance readings of the sample in Reagent 1 are taken and are used to correct for interfering substances in the sample. Reagent 2 is added, which initiates the conversion of glucose and the development of absorbance at 340/410 nm. The difference between the absorbance in Reagent 1 and Reagent 2 is proportional to the glucose concentration.
Here's an analysis of the acceptance criteria and the study that proves the device meets them, based on the provided text:
Device: ADVIA® 1650 Chemistry Glucose Hexokinase 3 (GLUH 3) method
1. Table of Acceptance Criteria and Reported Device Performance
| Category | Acceptance Criteria (Predicate Device K042015) | Reported Device Performance (ADVIA® GLUH 3) |
|---|---|---|
| Intended Use | Quantitative determination of glucose in human serum, plasma, urine, and CSF, for diagnosis and treatment of carbohydrate metabolism disorders. | Quantitative determination of glucose in human serum, plasma, urine, and CSF, for diagnosis and treatment of carbohydrate metabolism disorders (Same). |
| Sample Type | human serum, plasma (Li Heparin), urine, CSF | human serum, plasma (Li Heparin and K EDTA), urine, CSF |
| Instrument | ADVIA 1650 Chemistry | ADVIA 1650 Chemistry (Same) |
| Method Principle | Based on the method by Slein using hexokinase and glucose-6-phosphate dehydrogenase enzymes. | Same |
| Calibrators | Siemens Healthcare Diagnostics Chemistry Calibrator REF 09784096 | Same |
| Reportable Range | 0 - 700 mg/dL | 4 - 700 mg/dL (Slightly narrower low end, but the overall range overlaps significantly with the predicate.) |
| Traceability | Standard reference material 965a from NIST | Same |
| Format | Concentrate | Liquid |
| Reagents | Three: R1, R2, and R2 mix | Two: R1 and R2 |
| Interfering Substances (NSI) | Bilirubin-NSI to 25 mg/dL at glucose level of 80 mg/dL. | |
| Hemoglobin-NSI up to 500 mg/dL at glucose level of 80 mg/dL. | ||
| Lipemia (Intralipid)-NSI to 500 mg/dL at glucose level of 80 mg/dL. | Bilirubin-NSI to 30 mg/dL at glucose level of 54 mg/dL. | |
| Hemoglobin-NSI up to 1000 mg/dL at glucose level of 52 mg/dL. | ||
| Lipemia (Intralipid)-NSI to 1000 mg/dL. (Improved interference tolerance for Bilirubin, Hemoglobin, and Lipemia). | ||
| Precision (Total) | Serum: 2.2% at 74.7 mg/dL, 2.0% at 247.6 mg/dL. | |
| Urine: 4.1% at 45.8 mg/dL, 3.6% at 266.7 mg/dL. | ||
| CSF: 3.1% at 36.9 mg/dL, 2.7% at 60.2 mg/dL. | Serum: 0.8% at 87 mg/dL, 0.7% at 297 mg/dL. | |
| Urine: 1.1% at 49 mg/dL, 1.9% at 301 mg/dL. | ||
| CSF: 1.0% at 56 mg/dL, 0.8% at 97 mg/dL. (Significantly improved precision across all sample types and concentrations). | ||
| Accuracy / Correlation | Serum: Y = 1.02x - 1.84 (N=194, r=0.998) vs. ADVIA 1650 (ADVIA glucose-single reagent). | |
| Plasma (Li Heparin): Y = 1.00x - 0.09 (N=35, r=1.000). | ||
| CSF: Y = 1.03x - 1.25 (N=55, r=0.987). | ||
| Urine: Y = 0.97x - 7.44 (N=99, r=0.999). | Serum: Y = 1.001x + 0.3 (N=99, r=1.000) vs. ADVIA 1650 (ADVIA GLUH reagent). | |
| Plasma (Li-Heparin): Y = 1.001x + 0.2 (N=88, r=1.000). | ||
| Plasma (K-EDTA): Y = 1.002x - 0.0 (N=87, r=1.000). | ||
| CSF: Y = 1.005x - 0.1 (N=113, r=1.000). | ||
| Urine: Y = 0.989x - 0.3 (N=51, r=1.000). (Shows very strong correlation with the predicate, with regression lines close to ideal Y=X and correlation coefficients very close to 1). |
Study Proving Acceptance Criteria:
The study proving the device meets the acceptance criteria is a comparative testing study demonstrating substantial equivalence to the predicate device (ADVIA Chemistry Glucose Hexokinase II reagent, cleared under K042015). This is typical for a 510(k) submission where the new device is compared against a legally marketed predicate device.
2. Sample Size Used for the Test Set and Data Provenance:
- Accuracy/Correlation Test Sets:
- Serum: N=99
- Plasma (Li-Heparin): N=88
- Plasma (K-EDTA): N=87
- CSF: N=113
- Urine: N=51
- Precision Test Sets: The text provides precision percentages at different glucose levels for serum, urine, and CSF, but does not explicitly state the number of individual samples or replicates used for these tests. It's common for precision studies to involve multiple measurements of control materials or pooled patient samples over several days.
- Interference Test Sets: The text indicates the maximum concentrations of interfering substances where "No Significant Interference" (NSI) was observed. It does not specify the number of samples used for these interference studies.
- Data Provenance: The document does not explicitly state the country of origin of the data or whether the data was retrospective or prospective. However, given that Siemens Healthcare Diagnostics Inc. is based in Tarrytown, NY, and the submission is to the FDA, it is highly probable that the studies were conducted in the US and followed standard industry practices for in vitro diagnostic device validation. Assumed to be prospective validation studies as part of the device development process.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications:
The concept of "experts" and "ground truth" as typically defined in imaging or diagnostic AI studies (e.g., radiologists interpreting images) is not directly applicable here. For an in vitro diagnostic device like a glucose assay, the "ground truth" is established by a reference method.
- Reference Method: The predicate device itself (ADVIA 1650 (ADVIA GLUH reagent)) served as the reference method for the accuracy/correlation studies. This means the new device's readings were compared against the established, cleared method. In this context, the "qualification" of the predicate method is its prior FDA clearance (K042015) based on its own validation against accepted standards and potentially reference methods like isotope dilution mass spectrometry (IDMS) which are often traceable to NIST.
4. Adjudication Method for the Test Set:
Not applicable in the context of this type of in vitro diagnostic device study. Adjudication methods like "2+1" or "3+1" are typically used when there are subjective human interpretations (e.g., by multiple radiologists) that need to be resolved to establish ground truth for a diagnostic aid. For a quantitative chemistry analyzer, the output is a numerical value, and agreement is determined by statistical correlation and bias against a reference method.
5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study was done, and what was the effect size of how much human readers improve with AI vs without AI assistance:
Not applicable. This device is an in vitro diagnostic chemistry analyzer, not a diagnostic imaging AI tool designed to assist human readers. Therefore, an MRMC study and analysis of human reader improvement with AI assistance are not relevant.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
Yes, a standalone performance evaluation was done. The reported performance metrics (precision, accuracy/correlation, interference) are characteristics of the device itself, providing quantitative results directly from the instrument. While human operators are involved in running the analyzer and preparing samples, the core measurement and output are solely from the device's chemical reactions and detection system, without a human interpretation loop like in imaging diagnostics.
7. The Type of Ground Truth Used:
The ground truth for the comparative studies was established by comparison to the established, legally marketed predicate device (ADVIA 1650 (ADVIA GLUH reagent)). This is a common and accepted method for demonstrating substantial equivalence for 510(k) submissions in in vitro diagnostics. The "ground truth" of the predicate itself would have been established through correlation with recognized reference methods (e.g., standard reference materials from NIST, or other highly accurate laboratory methods) during its own clearance process.
8. The Sample Size for the Training Set:
The document does not specify a separate "training set" size. For traditional in vitro diagnostic devices like chemical reagents, there isn't typically an "algorithm training" phase in the same way there is for machine learning models. The development and optimization of the reagent formulation and assay parameters are guided by chemical and analytical principles, and then validated through analytical performance studies (like the ones presented). If there were any internal development studies that could be considered "training", their details are not provided in this 510(k) summary.
9. How the Ground Truth for the Training Set Was Established:
As there's no explicitly defined "training set" in the machine learning sense for this type of device, the concept of establishing ground truth for it doesn't directly apply. The method development likely involved iterative testing to achieve desired performance characteristics (e.g., sensitivity, specificity, linearity, interference profile) against known concentrations of glucose and interfering substances, often using reference materials or established laboratory methods as benchmarks during the development phase.
§ 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.