(524 days)
The Urea Nitrogen2 assay is used for the quantitation of Urea Nitrogen in human serum, plasma, or urine on the ARCHITECT c System.
The Urea Nitrogen2 assay is to used as an aid in the diagnosis and treatment of certain renal and metabolic diseases.
The Urea Nitrogen2 assay is an automated clinical chemistry assay. The Urea Nitrogen2 assay is a modification of a totally enzymatic procedure. The test is performed as a kinetic assay in which the initial rate of the reaction is linear for a limited period of time. Urea in the sample is hydrolyzed by urease to ammonia and carbon dioxide. The second reaction, catalyzed by glutamate dehydrogenase (GLDH), converts ammonia and a-ketoglutarate to glutamate and water with the concurrent oxidation of reduced nicotinamide adenine dinucleotide (NADH) to nicotinamide adenine dinucleotide (NAD). Two moles of NADH are oxidized for each mole of urea present. The initial rate of decrease in absorbance at 340 nm is proportional to the urea concentration in the sample.
The provided document is a 510(k) premarket notification for a new in vitro diagnostic device, the Urea Nitrogen2 assay. This type of submission focuses on demonstrating substantial equivalence to a legally marketed predicate device rather than comprehensive clinical effectiveness studies as might be seen for novel devices. Therefore, the information provided primarily concerns non-clinical performance characteristics of the assay itself, rather than human-in-the-loop performance or diagnostic accuracy evaluated in a clinical setting with patient outcomes.
Based on the provided document, here's an analysis of the acceptance criteria and the study that proves the device meets them:
Core Purpose of the Submission: To demonstrate that the Urea Nitrogen2 assay is substantially equivalent to the predicate device (Urea Nitrogen assay, K981918) for the quantitative measurement of urea nitrogen in human serum, plasma, and urine on the ARCHITECT c System. This means proving the new device performs similarly and is as safe and effective as the predicate.
Acceptance Criteria and Reported Device Performance
The acceptance criteria are implicitly derived from established clinical laboratory standards (CLSI guidelines) and comparison to the predicate device's known performance. The performance metrics evaluated are standard for in vitro diagnostic assays.
1. Table of Acceptance Criteria and Reported Device Performance
| Performance Characteristic | Acceptance Criteria (Implicit) | Reported Device Performance (Urea Nitrogen2) |
|---|---|---|
| Reportable Interval | Must cover clinically relevant range and be supported by analytical performance (LoD, LoQ, linearity, imprecision, bias). It should be comparable to or improve upon the predicate device. Predicate: Serum 2-125 mg/dL, Urine 2-1991 mg/dL. | Serum/Plasma: AMI: 2-125 mg/dL, EMI: 125-625 mg/dL, Reportable Interval: 2-625 mg/dL. |
| Urine: AMI: 16-1991 mg/dL, Reportable Interval: 11-1991 mg/dL. | ||
| (Meets/Exceeds predicate in serum/plasma upper limit, comparable in urine.) | ||
| Within-Laboratory Precision (Imprecision) | Based on CLSI EP05-A3 guidelines. Low %CV (Coefficient of Variation) and SD (Standard Deviation) demonstrating consistent results. Comparable to or better than predicate. Predicate: Serum %CV 1.8-2.0% (15.5-48.0 mg/dL), Urine %CV 3.1-3.8% (504.8-896.4 mg/dL). | Serum/Plasma: Samples (4-102 mg/dL) demonstrated SDs ≤ 0.4 mg/dL and %CV ≤ 2.7%. |
| Urine: Samples (55-1605 mg/dL) demonstrated SDs ≤ 11.7 mg/dL and %CV ≤ 2.1%. | ||
| (Meets/Exceeds predicate with broader range tested and generally lower %CV.) | ||
| Accuracy (Bias) | Low percentage bias relative to a recognized standard reference material (NIST SRM 912b). Bias should be within acceptable limits for clinical utility. | Serum: Bias ranged from 1.6% to 4.2%. |
| Urine: Bias ranged from -1.3% to 3.0%. | ||
| (Demonstrates acceptable accuracy.) | ||
| Lower Limits of Measurement (LoB, LoD, LoQ) | Determined by CLSI EP17-A2 guidelines. LoB, LoD, and LoQ should be low enough to be clinically useful and comparable to the predicate. Predicate Serum: LoD 0.7 mg/dL, LoQ 1.4 mg/dL. Predicate Urine: LoD 15.0 mg/dL, LoQ 40.0 mg/dL. | Serum: LoB 1 mg/dL, LoD 2 mg/dL, LoQ 2 mg/dL. |
| Urine: LoB 6 mg/dL, LoD 11 mg/dL, LoQ 16 mg/dL. | ||
| (Comparable to predicate's analytical sensitivity; slight differences but within typical ranges for this type of assay.) | ||
| Linearity | Device should show a linear response across its stated analytical measuring interval as per CLSI EP06-A. | Serum: Linear across 2 to 125 mg/dL. |
| Urine: Linear across 16 to 1991 mg/dL. | ||
| (Demonstrates linearity across its Analytical Measuring Interval.) | ||
| Interference | No significant interference (typically defined as ±10% bias) from common endogenous and exogenous substances at specified levels. The scope and levels tested should address potential clinical interferences. | Serum/Plasma (Endogenous): No significant interference (within ±10%) for Bilirubin (60 mg/dL), Hemoglobin (2000 mg/dL), Triglycerides (1500 mg/dL). Interference was observed for Total Protein (11 g/dL at 10 mg/dL analyte level: 11% (9%, 14%) CI). |
| Serum/Plasma (Exogenous): No significant interference (within ±10%) for a broad list of drugs. Interference was observed for Cefoxitin (6600 mg/L at 10 mg/dL analyte level: 10% (6%, 14%) CI). | ||
| Urine (Endogenous): No significant interference (within ±10%) for Ascorbate (200 mg/dL), Glucose (1000 mg/dL), Protein (50 mg/dL). | ||
| Urine (Exogenous): No significant interference (within ±10%) for a broad list of drugs/substances. (Generally good performance, with identified interferences clearly reported.) | ||
| Method Comparison | High correlation and acceptable agreement (slope and intercept near 1 and 0, respectively) when compared to the predicate device on the same platform. Expected correlation coefficient near 1.00. | Serum: N=124, Correlation Coefficient=1.00, Intercept=0.74, Slope=1.02 (Concentration Range 4-123 mg/dL). |
| Urine: N=121, Correlation Coefficient=1.00, Intercept=8.95, Slope=1.03 (Concentration Range 41-1754 mg/dL). | ||
| (Excellent correlation and agreement with the predicate.) | ||
| Tube Type Suitability | Demonstration that the device performs acceptably with specified blood collection tube types. | Serum: Serum tubes, Serum separator tubes. |
| Plasma: Lithium heparin tubes, Lithium heparin separator tubes, Sodium heparin tubes. | ||
| (Acceptable for specified tube types.) | ||
| Dilution Verification | Demonstration that the automated dilution protocol yields results comparable to manual dilution. Accuracy of diluted results compared to undiluted or expected values. | % difference values for automated dilution vs. manual dilution ranged from -2.8% to -1.3%, demonstrating acceptable performance. |
| (Acceptable performance for automated dilution.) |
Study Details
The studies described are primarily analytical performance studies, characteristic of a 510(k) submission for an in vitro diagnostic device, especially a chemical analyzer assay. They demonstrate the device's technical specifications and how it performs compared to a reference method or the predicate device.
2. Sample Size Used for the Test Set and Data Provenance
- Precision Studies (Within-Laboratory):
- Serum/Plasma: 3 human serum panels + 2 controls. Each tested in duplicate, twice per day, for 20 days on 3 reagent lot/calibrator lot/instrument combinations. For a representative combination, n=80 per sample/control.
- Urine: 3 human urine panels + 2 controls. Each tested in duplicate, twice per day, for 20 days on 3 reagent lot/calibrator lot/instrument combinations. For a representative combination, n=80 per sample/control.
- Data Provenance: Not explicitly stated but inferred to be laboratory-based analytical studies, likely from the manufacturer's R&D facilities. No country of origin is specified. The studies are retrospective analytical evaluations of manufactured samples and controls.
- Accuracy (Bias): 3 concentrations of standard across 3 reagent lots, 2 calibrator lots, and 1 instrument. (NIST SRM 912b is the standard).
- Lower Limits of Measurement (LoB, LoD, LoQ): n ≥ 60 replicates of zero-analyte samples for LoB, n ≥ 60 replicates of low-analyte level samples for LoD/LoQ. Conducted using 3 reagent lots on 2 instruments over a minimum of 3 days.
- Linearity: Not explicitly stated sample count, but typically involves preparing a dilution series of samples across the range.
- Interference: "Each substance was tested at 2 levels of the analyte (approximately 10 mg/dL and 30 mg/dL for serum/plasma; 700 mg/dL and 1500 mg/dL for urine)." No specific N for how many replicates or individual samples are run per interferent level, but implied to be sufficient for statistical analysis (e.g., 95% CI).
- Method Comparison:
- Serum: n=124 samples.
- Urine: n=121 samples.
- Data Provenance: Not explicitly stated, but these would be clinical or proficiency samples analyzed side-by-side with the predicate.
- Tube Type: Samples collected from a minimum of 40 donors.
- Dilution Verification: 5 human serum samples (spiked with urea). Each sample tested with automated dilution and 3 manual dilutions (by 2 technicians). Tested in replicates of 5.
3. Number of Experts Used to Establish Ground Truth and Qualifications
- For this type of in vitro diagnostic device (a quantitative chemical assay), "ground truth" is established by:
- Reference materials: e.g., NIST SRM 912b for accuracy. This is a primary standard, not established by human experts.
- Predicate device measurements: For method comparison, the predicate device provides the comparative 'truth' (or established method performance).
- Clinical laboratory professional consensus/guidelines: Standards like CLSI (Clinical and Laboratory Standards Institute) guidelines (EP05-A3, EP17-A2, EP06-A, EP07, EP09-A3, EP34) serve as the "expert consensus" on how to conduct and interpret these analytical studies. These are published by committees of experts in laboratory medicine, clinical chemistry, and statistics.
- No "expert readers" in the traditional sense (e.g., radiologists interpreting images) are involved in establishing ground truth for this type of device. The validation is based on metrological traceability to standards and comparison to an established analytical method.
4. Adjudication Method for the Test Set
- Not applicable as this is an analytical performance study of a chemical assay, not a diagnostic accuracy study relying on human interpretation of subjective data (like imaging or pathology). Results are quantitative measurements read by the instrument.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done
- No. An MRMC study is relevant for diagnostic devices that involve human interpretation (e.g., AI-assisted image interpretation) and aim to show an improvement in human reader performance. This device is a quantitative chemical assay that provides a numerical result; there is no human "reader" to assist in the primary measurement. The comparison is between the new assay's performance and the predicate assay's performance, as well as against analytical standards.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done
- Yes, effectively. The entire submission details the standalone analytical performance of the Urea Nitrogen2 assay system (reagents + ARCHITECT c System instrument). Its performance (precision, accuracy, linearity, etc.) is evaluated independently of human interpretation of the final numerical result, beyond the standard operation and quality control typically performed in a clinical lab.
7. The Type of Ground Truth Used
- Analytical Standards and Reference Methods/Predicate Device:
- NIST SRM 912b: Used as the true value for accuracy determination.
- Predicate Device (Urea Nitrogen assay): Used as the comparative reference for method comparison studies, demonstrating substantial equivalence.
- CLSI Guidelines: Act as the "ground truth" for the methodologies and acceptance criteria applied to evaluate the analytical performance (e.g., how LoQ is defined and determined, how precision is calculated).
8. The Sample Size for the Training Set
- For this type of in vitro diagnostic assay, there isn't a "training set" in the machine learning sense (where an algorithm learns from data). The "training" for such an assay primarily refers to:
- Reagent formulation and optimization: This involves extensive R&D to achieve desired chemical reactions and stability.
- Instrument calibration: The ARCHITECT c System is calibrated using specific calibrators (Consolidated Chemistry Calibrator mentioned, which is itself traceable to standards) to ensure accurate measurement across the range.
- The "training" is inherent in the chemical and engineering development of the assay and the instrument, rather than an algorithmic learning process on a large dataset. Therefore, a specific "training set sample size" as one would discuss for an AI model is not applicable.
9. How the Ground Truth for the Training Set Was Established
- As explained above, there isn't a "training set" with ground truth in the AI/ML context for this device. Instead, the assay's performance characteristics (calibration, linearity, reaction kinetics, etc.) are optimized and confirmed through:
- Chemical principles and R&D: The enzymatic reaction (urease, GLDH kinetics) is based on established biochemical mechanisms.
- Quality control materials and calibrators: These materials have assigned values, often traceable to international standards (like NIST SRM), and are used to "train" or calibrate the instrument system.
- Iterative laboratory testing and optimization: The assay's components and instrument parameters are refined through repeated experiments to meet performance specifications.
§ 862.1770 Urea nitrogen test system.
(a)
Identification. A urea nitrogen test system is a device intended to measure urea nitrogen (an end-product of nitrogen metabolism) in whole blood, serum, plasma, and urine. Measurements obtained by this device are used in the diagnosis and treatment of certain renal and metabolic diseases.(b)
Classification. Class II.