(101 days)
In vitro test for the quantitative determination of glucose in serum, plasma, urine and cerebrospinal fluid (CSF).
Glucose measurements are used in the diagnosis and treatment of carbohydrate metabolism disorders including diabetes mellitus, neonatal hypoglycemia, idiopathic hypoglycemia and pancreatic islet cell tumors.
The cassette COBAS INTEGRA Glucose HK Gen. 3 contains an in vitro diagnostic reagent system intended for use on COBAS INTEGRA SYSTEMS for the quantitative determination of glucose in serum, plasma, urine, and cerebrospinal fluid (CSF).
The test principle is an enzymatic reference method with hexokinase.
The provided text describes the COBAS INTEGRA Glucose HK Gen. 3 Assay but does not explicitly state acceptance criteria in a pass/fail format. Instead, it presents performance characteristics (precision, linearity, analytical sensitivity, interference, method comparison) and compares them to a predicate device to establish substantial equivalence.
Based on the provided information, here's a structured response:
1. Table of Acceptance Criteria and Reported Device Performance
As explicit acceptance criteria are not stated for the device per se, the table below compares the performance of the COBAS INTEGRA Glucose HK Gen. 3 Assay to its predicate device, COBAS INTEGRA Glucose HK Gen. 3 (K061048), which served as the benchmark for substantial equivalence. The implication is that the new device must perform comparably to or better than the predicate.
| Feature / Performance Metric | Predicate Device (K061048) Performance | COBAS INTEGRA Glucose HK Gen. 3 Assay Performance | Implicit Acceptance Standard (vs. Predicate) | Meets Std? |
|---|---|---|---|---|
| Measuring Range (Regular Applications) | 2.16 - 720 mg/dL | 4.32 - 720 mg/dL | Should be comparable or improved. The lower limit is slightly higher for the new device. | Yes* |
| Measuring Range (Stat Applications) | Not specified (only Regular) | 4.32 - 541 mg/dL | New feature, performance is reported. | NA |
| Precision - Serum (Regular Application) | ||||
| Repeatability (Within-Run CV%) | Level I: 0.41%, Level II: 0.47% | HS1: 0.69%, HS2: 0.78%, HS3: 0.70%, PNU: 0.58%, PPU: 0.59% | Comparable CV% values. | Yes |
| Intermediate (Between Run CV%) | Level I: 1.09%, Level II: 0.90% | HS1: 1.26%, HS2: 1.41%, HS3: 1.32%, PNU: 1.24%, PPU: 1.21% | Comparable or slightly higher CV%. | Yes |
| Precision - Serum (Stat Application) | Not specified | Repeatability: 0.57-0.74%, Intermediate: 1.07-1.40% | New feature, performance is reported. | NA |
| Precision - Urine (Regular Application) | ||||
| Repeatability (Within-Run CV%) | Level I: 1.35%, Level II: 0.64% | HS1: 2.28%, HS2: 1.05%, HS3: 0.53%, PNU: 0.58%, PPU: 0.59% | Comparable CV% values. | Yes |
| Intermediate (Between Run CV%) | Level I: 0.75%, Level II: 0.83% | HS1: 2.83%, HS2: 1.37%, HS3: 1.01%, PNU: 1.24%, PPU: 1.21% | Comparable or slightly higher CV%. | Yes |
| Precision - Urine (Stat Application) | Not specified | Repeatability: 0.57-2.71%, Intermediate: 1.07-3.20% | New feature, performance is reported. | NA |
| Precision - CSF (Regular Application) | ||||
| Repeatability (Within-Run CV%) | Level I: 1.13%, Level II: 1.49% | Level I: 1.13%, Level II: 1.49% | Identical CV% values. | Yes |
| Precision - CSF (Stat Application) | Not specified | Repeatability: 0.39-0.63% | New feature, performance is reported. | NA |
| Analytical Sensitivity (Lower Limits) | Lower Detection Limit: 2.16 mg/dL | Limit of Blank: 2.16 mg/dL, Limit of Detection: 4.32 mg/dL | Comparable. Limit of detection is slightly higher for the new device. | Yes* |
| Method Comparison - Serum (new vs. Stat application) | Not applicable (no Stat application in predicate) | Passing Bablok: y=0.997x + 0.033 mmol/L, $\tau = 0.999$, SD (md 95) = 0.067 | High correlation between regular and Stat application. | NA |
| Method Comparison - Urine (new vs. Stat application) | Not applicable (no Stat application in predicate) | Passing Bablok: y=1.00x + 0.002 mmol/L, $\tau = 0.996$, SD (md 95) = 0.082 | High correlation between regular and Stat application. | NA |
Note on "Yes": While the lower measuring range and detection limits for the new device are slightly different (higher lower limit for measuring range, higher limit of detection), these values are still within clinically acceptable ranges for glucose measurements and are explicitly stated, indicating the device meets its own defined performance specifications, and the differences are not deemed substantial enough to prevent equivalence. The 510(k) cleared the device, implying these differences were acceptable.
2. Sample sizes used for the test set and data provenance
- Precision Studies (Serum, Urine, CSF): The tables provide "Sample" categories (e.g., HS1, HS2, HS3, PNU, PPU, Level I, Level II) with their Mean (mg/dL) and CV (%). The exact number of individual samples within each of these categories tested for precision is not explicitly stated. However, precision studies in IVD usually involve replicate measurements (e.g., 20 replicates over multiple days) of a few representative samples at different concentration levels.
- Method Comparison - Serum: n=79 human serum samples
- Method Comparison - Urine: n=50 human urine samples
- Data Provenance: Not explicitly stated (e.g., country of origin). Given the manufacturer is Roche Diagnostics (Indianapolis, IN, USA), it is likely the studies involved samples sourced within the US or under similar regulatory guidelines. The data is prospective in the sense that it was collected specifically for the purpose of demonstrating the device's performance characteristics for regulatory submission.
3. Number of experts used to establish the ground truth for the test set and their qualifications
This device is an in vitro diagnostic (IVD) assay for quantitative determination of glucose. For such devices, "ground truth" is typically established by comparing the device's results to a reference method or an established, legally marketed predicate device, as seen in the "Method Comparison" section. This does not involve human expert adjudication in the same way as, for example, image interpretation. Performance is assessed by comparing quantitative results against an accepted standard.
4. Adjudication method for the test set
Not applicable for this type of quantitative IVD assay. Performance is assessed through statistical comparison to a predicate device or reference method, and by demonstrating performance characteristics (precision, linearity, analytical sensitivity) meet established analytical requirements.
5. 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
Not applicable. This is an automated in vitro diagnostic assay, not an AI-assisted diagnostic tool that involves human readers interpreting cases.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done
Yes, the studies reported are assessing the standalone performance of the assay (reagent system and instrument). The measurements are quantitative chemical analyses performed by the COBAS INTEGRA system using the Glucose HK Gen. 3 reagent, without human interpretation influencing the numerical result.
7. The type of ground truth used
For this IVD device, the "ground truth" for evaluating its performance (beyond foundational analytical studies):
- Precision: Internal consistency of the device's own measurements.
- Method Comparison: Comparison against the same reagent (COBAS INTEGRA Glucose HK Gen. 3) on the same analyzer (COBAS INTEGRA 800) but using the Stat application instead of Regular, to demonstrate concordance between the different applications of the device. This implies the regular application (and the predicate device it supersedes) serves as a de facto reference. The initial predicate device would have been validated against a more universally accepted glucose reference method.
8. The sample size for the training set
Not applicable in the context of an IVD assay like this. There is no machine learning "training set" for an enzymatic assay. The "training" in this context refers to the development and optimization of the reagent formulation and instrument parameters. The studies presented are performance validation studies.
9. How the ground truth for the training set was established
Not applicable for this type of device. The "ground truth" (i.e., true glucose concentration) for developing and validating an assay relies on well-characterized samples and reference methods. The provided document focuses on the validation of the device against an existing predicate/application.
§ 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.