Search Results

The Dade Behring Dimension® PBNP Flex® reagent cartridge method is an in vitro diagnostic assay for the quantitative determination of N-terminal pro-brain natriuretic peptide (NT-proBNP) in human plasma. Measurements of NT-proBNP are used as an aid in the diagnosis of individuals suspected of having congestive heart failure.

The Dade Behring Dimension® PBNP Calibrator is intended to be used to calibrate the N-terminal pro-brain natriuretic peptide (PBNP) method for the Dade Behring Dimension® clinical chemistry system.

Device Description

The Dade Behring Dimension® NT-proBNP (PBNP) Flex® reagent cartridge method is an in vitro diagnostic test that consists of prepackaged reagents in a flexible plastic cartridge for use only on the Dimension® clinical chemistry system. The PBNP method is a one step enzyme immunoassay based on the "sandwich" principle. Sample is incubated with chromium dioxide particles coated with polyclonal antibodies which recognize an epitope located in the N-terminal part of proBNP, and a conjugate reagent [alkaline phosphatase (ALP)] labeled polyclonal antibody specific for a second independent epitope on NT-proBNP, to form a particle/NT-proBNP/conjugate sandwich. Unbound conjugate is removed by magnetic separation and washing. After separation and washing, the particle/NT-proBNP/coniugate sandwich is transferred to the cuvette where the sandwich-bound ALP triggers an amplification cascade. ALP dephosphorylates synthetic flavin adenine dinucleotide phosphate (FADP) to produce FAD. FAD binds to apo D-amino acid oxidase and converts it to active holo D-amino acid oxidase. Each molecule of holo D-amino acid oxidase produces multiple molecules of hydrogen peroxide (H2O2). H2O2 in the presence of horseradish peroxidase (HRP), converts 3,5-dichloro-2-hydroxybenzenesulfonic acid (DCHBS) and 4-aminoantipyrine (4-AAP) to a colored product that absorbs at 510 nm. The color change measured is directly proportional to the concentration of proBNP present in the patient sample.

The Dade Behring PBNP Calibrator is a frozen liquid product containing synthetic human NT-proBNP in a bovine albumin matrix with stabilizers and preservative. The kit consists of ten vials, two vials at each of five levels, 1.0 mL in each vial.

AI/ML Overview

Here's a breakdown of the acceptance criteria and study information for the Dade Behring Dimension® NT-proBNP (PBNP) Flex® reagent cartridge method, based on the provided text:

Acceptance Criteria and Device Performance

The provided text focuses on demonstrating substantial equivalence to a predicate device (Roche Diagnostics Elecsys® proBNP immunoassay) rather than explicitly stating pre-defined "acceptance criteria" for novel device performance. The performance metrics presented are comparative against the predicate.

Table of Acceptance Criteria and Reported Device Performance

Acceptance Criteria (Implied by Predicate Comparison)	Reported Device Performance (Dade Behring Dimension® PBNP)	Comparison to Predicate (Roche Elecsys® proBNP)
Clinical Performance:
Sensitivity for <75 years (Male)	84% (77 - 91% CI)	Predicate: 90% (85 - 96% CI)
Sensitivity for ≥75 years (Male)	91% (84 - 99% CI)	Predicate: 91% (84 - 99% CI)
Specificity for <75 years (Male)	94% (90- 99% CI)	Predicate: 92% (87 - 98% CI)
Specificity for ≥75 years (Male)	77% (67-88% CI)	Predicate: 73% (61 - 84% CI)
NPV (Male)	82 (<75 yrs), 92 (≥75 yrs)	Predicate: 88 (<75 yrs), 92 (≥75 yrs)
Sensitivity for <75 years (Female)	77% (64 – 89% CI)	Predicate: 81% (70–92% CI)
Sensitivity for ≥75 years (Female)	91% (79 – 100% CI)	Predicate: 95% (87 – 104% CI)
Specificity for <75 years (Female)	93% (89–98% CI)	Predicate: 91% (85–96% CI)
Specificity for ≥75 years (Female)	88% (80–96% CI)	Predicate: 87% (78–95% CI)
NPV (Female)	91 (<75 yrs), 96 (≥75 yrs)	Predicate: 92 (<75 yrs), 98 (≥75 yrs)
Analytical Performance:
Correlation Coefficient (vs. predicate)	0.985 (n=352)	Not applicable (comparison metric)
Slope (vs. predicate)	0.90	Not applicable (comparison metric)
Intercept (vs. predicate)	-15.4 pg/mL	Not applicable (comparison metric)
Reportable Range	10 - 30,000 pg/mL	Predicate: 5 - 35,000 pg/mL
Analytical Sensitivity	≤ 10 pg/mL	Predicate: 5 pg/mL
Functional Sensitivity	≤ 30 pg/mL	Predicate: < 50 pg/mL
Cut-off for < 75 years	125 pg/mL	Predicate: 125 pg/mL
Cut-off for ≥ 75 years	450 pg/mL	Predicate: 450 pg/mL
Within-Run Precision (Pool 1)	2.2% CV (159.0 pg/mL mean)	Not directly compared, shown to be low
Total Precision (Pool 1)	5.7% CV (159.0 pg/mL mean)	Not directly compared, shown to be low
Within-Run Precision (QC Pool 1)	1.8% CV (449.5 pg/mL mean)	Not directly compared, shown to be low
Total Precision (QC Pool 1)	3.7% CV (449.5 pg/mL mean)	Not directly compared, shown to be low
Within-Run Precision (QC Pool 2)	1.6% CV (956.7 pg/mL mean)	Not directly compared, shown to be low
Total Precision (QC Pool 2)	3.6% CV (956.7 pg/mL mean)	Not directly compared, shown to be low
Within-Run Precision (Audit™ Level 1)	1.1% CV (175.5 pg/mL mean)	Not directly compared, shown to be low
Total Precision (Audit™ Level 1)	3.8% CV (175.5 pg/mL mean)	Not directly compared, shown to be low
Within-Run Precision (Audit™ Level 2)	1.9% CV (3733.8 pg/mL mean)	Not directly compared, shown to be low
Total Precision (Audit™ Level 2)	3.1% CV (3733.8 pg/mL mean)	Not directly compared, shown to be low

Study Information

The study primarily supporting this 510(k) submission is a comparative study demonstrating substantial equivalence between the new Dimension® PBNP Flex® method and the predicate Roche Elecsys® proBNP immunoassay.

Sample size used for the test set and the data provenance:
- Clinical Performance Study:
  - Reference Study Group (without CHF): 308 individuals (163 women, 145 men).
  - Disease Study Group (with CHF): 227 patients (69 women, 158 men).
  - Data Provenance: The text states, "Data used to calculate the values are from the method comparison and reference interval data sets generated at the University of Maryland Medical Center." This implies prospective collection or a well-defined retrospective cohort from a specific medical center in the USA.
- Analytical Performance (Method Correlation):
  - Sample Size: 352 split patient heparinized plasma samples.
  - Data Provenance: Not explicitly stated, but likely from the same source as the clinical performance data (University of Maryland Medical Center).
- Reproducibility: Involved human plasma pools, internal QC pools, and commercial control materials. The number of individual patient samples in these pools is not specified, but the testing involved 20 days of duplicate analysis.
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- The ground truth for the clinical study was based on diagnosis of congestive heart failure (CHF). The document does not specify the number or qualifications of experts who established these diagnoses. It refers to individuals "without congestive heart failure" and "patients diagnosed with congestive heart failure (CHF)," implying standard clinical diagnostic procedures, likely by physicians, determined the CHF status.
Adjudication method for the test set:
- No information on a specific adjudication method (e.g., 2+1, 3+1) is provided for the clinical diagnoses of CHF or for resolving discrepancies. The diagnoses are assumed to be a clinical standard.
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, a MRMC comparative effectiveness study was not done. This device is an in vitro diagnostic assay, directly measuring a biomarker. It's not an image-based or interpretive AI device that involves human readers or their interpretation. Therefore, the concept of "human readers improve with AI vs without AI assistance" is not applicable here.
If a standalone (i.e. algorithm only without human-in-the loop performance) was done:
- This device is a standalone algorithm (an automated immunoassay method) without "human-in-the-loop performance" in the sense of human interpretation being required for each result. The device generates quantitative NT-proBNP values independently. Clinical interpretation of these values as an "aid in diagnosis" is then performed by a healthcare professional.
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- The ground truth for the clinical performance evaluation was the clinical diagnosis of congestive heart failure (CHF). This would typically be based on a combination of patient symptoms, physical examination, imaging (e.g., echocardiography), and other clinical findings, representing an "outcomes data" or "clinical diagnosis" type of ground truth.
- For the analytical performance (method correlation study), the ground truth was the measurements obtained from the predicate device (Roche Elecsys® proBNP immunoassay). This is a common approach for demonstrating substantial equivalence for new diagnostic assays.
The sample size for the training set:
- The document does not explicitly mention a separate "training set" in the context of device development. This type of immunoassay development typically involves calibration and verification steps rather than machine learning training. The information provided focuses on the validation (test) set for demonstrating performance against the predicate.
How the ground truth for the training set was established:
- As no explicit "training set" is described for this immunoassay (which is not an AI/ML device), this question is not applicable. The device's calibration is performed using the provided PBNP Calibrator, which contains synthetic human NT-proBNP at known concentrations.

Ask a Question

Ask a specific question about this device

Page 1 of 1