Search Results

The Access NT-proBNP assay is a paramagnetic particle, chemiluminescent immunoassay for the quantitative determination of N-terminal pro B-type natriuretic peptide levels in human serum and plasma using the automated DxI Access Immunoassay Analyzers to aid in the following:

1. diagnosis of patients suspected of having acute heart failure in the Emergency Department
2. assessment of heart failure severity
3. risk stratification of patients with heart failure
4. risk stratification of patients with acute coronary syndrome

The Access NT-proBNP assay is a paramagnetic particle, chemiluminescent immunoassay for the quantitative determination of N-terminal pro B-type natriuretic peptide levels in human serum and plasma using the automated Dxl 9000 Access Immunoassay Analyzers to aid in the following: 1) diagnosis of patients suspected of acute heart failure in the Emergency Department, 2) assessment of heart failure severity, 3) risk stratification of patients with heart failure, 4) risk stratification of patients with acute coronary syndrome.

The Access NT-proBNP is a two-site immunoenzymatic (sandwich) assay. Paramagnetic particles coated with monoclonal anti-NT-proBNP antibody and monoclonal anti-NTproBNP antibody conjugated to alkaline phosphatase are added to a reaction vessel along with a surfactant-containing buffer and serum or plasma sample. The human NTproBNP binds to the anti-NT-proBNP antibody on the solid phase, while the anti-NTproBNP antibody-alkaline phosphatase conjugate reacts with a different antigenic site on the NT-proBNP molecule. After incubation, materials bound to the solid phase are held in a magnetic field while unbound materials are washed away. Then the chemiluminescent substrate is added to the vessel and light generated by the reaction is measured with a luminometer. The light production is directly proportional to the concentration of analyte in the sample. Analyte concentration is automatically determined from a stored calibration.

Other items required to use the assay include calibrators, Lumi-Phos PRO, and wash buffer. The Access NT-proBNP reagent packs. Access NT-proBNP calibrators, along with the Access wash buffer, and Lumi-Phos PRO are designed for use on the Dxl 9000 Access Immunoassay Analyzers in a clinical laboratory setting.

Here's a breakdown of the acceptance criteria and study details for the Beckman Coulter Access NT-proBNP device, based on the provided text:

Acceptance Criteria and Reported Device Performance

The acceptance criteria are implied by the clinical and non-clinical studies conducted, with the device performance needing to meet expected ranges and exhibit substantial equivalence to predicate devices. Specific quantitative acceptance criteria are given for imprecision, LoB/LoD/LoQ, and linearity.

Study Details

1. Sample Size and Data Provenance:

   • Test Set (Clinical Study for AHF Diagnosis): 2,384 patients presenting to the Emergency Department with clinical suspicion of acute heart failure.
     • Provenance: This was a "multicenter prospective study." While specific countries are not mentioned, regulatory submissions to the FDA typically involve studies conducted in the US or other regions following ICH GCP guidelines. The study type is explicitly prospective.
   • Test Set (Reference Interval Study): 675 apparently healthy adults for the overall ULN, broken down into 306 males and 369 females.
     • Provenance: This was a "multicenter prospective reference interval study." Again, specific countries are not mentioned but it was prospective.
   • Test Set (Imprecision): Number of runs (minimum 20 days) and replicates (duplicate) are specified for samples 1-7, with N ranging from 83 to 86 tests per sample.
   • Test Set (LoB, LoD, LoQ): Multiple reagents lots and 3 instruments over 3-5 days.
   • Test Set (Linearity): Native patient samples.
   • Test Set (Matrix Comparison): Sixty-eight (68) matched serum, lithium heparin, and EDTA plasma samples.
   • Test Set (Interfering Substances/Cross Reactivity): Lithium heparin plasma samples (concentrations approx. 125 ng/L and 1,800 ng/L) spiked with various substances. Number of individual samples not specified, but the number of substances tested is large.

2. Number of Experts and Qualifications for Ground Truth:

   • For the clinical sensitivity study (AHF diagnosis), "Final diagnoses were adjudicated by an independent committee of medical doctors."
   • Number of Experts: Not explicitly stated (e.g., "3 independent physicians"). It mentions "an independent committee," which implies more than one.
   • Qualifications: "Medical doctors." No specific specialties (e.g., cardiologists, emergency physicians) or years of experience are listed in the provided text.
   • For risk stratification, the basis for effectiveness comes from "an objective and systematic analysis of recent peer-reviewed literature and clinical practice guidelines," implying consensus from the broader medical community rather than a specific set of experts for this particular study.

3. Adjudication Method for the Test Set:

   • For the AHF diagnosis study, the text states: "Final diagnoses were adjudicated by an independent committee of medical doctors who decided on presence of acute heart failure. Adjudicators were blinded to the Access NT-proBNP assay results."
   • The specific method (e.g., 2+1, 3+1) is not explicitly detailed. It simply states "adjudicated by an independent committee," which could imply consensus, majority vote, or a specific tie-breaking rule, but the exact mechanism is not provided.

4. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:

   • No, this device is an in-vitro diagnostic (IVD) assay, not an AI/imaging device. Therefore, an MRMC study and human reader improvement with AI assistance are not applicable. The comparison made is against another IVD assay (Elecsys proBNP II) as a standalone algorithm/test.

5. Standalone Performance:

   • Yes, the primary performance shown is standalone (algorithm only/test only) performance. The Access NT-proBNP assay is a quantitative determination of NT-proBNP levels. Its performance (e.g., analytical performance, diagnostic accuracy via ROC curves) is measured and reported as the output of the assay itself. The clinical performance section directly presents the assay's ability to diagnose AHF and correlates its results with NYHA classification and risk stratification, without human interpretation of the assay's output as an input to a further AI system.

6. Type of Ground Truth Used:

   • Clinical Diagnosis (Expert Consensus/Adjudication): For the acute heart failure diagnosis study, the ground truth was established by an "independent committee of medical doctors" adjudicating the final diagnoses. This falls under expert consensus based on clinical information.
   • Clinical Outcomes/Literature Review: For risk stratification, the ground truth is based on the correlation of NT-proBNP with "increased incidence of cardiovascular events, mortality and composite outcomes" as established in existing peer-reviewed literature and clinical practice guidelines.
   • Reference Intervals for Healthy Population: Established by recruiting "apparently healthy adults" based on defined exclusion criteria and screening with additional biomarkers (eGFR, hsTnl) to ensure the health status.

7. Sample Size for the Training Set:

   • The provided document describes the validation studies (test set) for the Access NT-proBNP device. It does not mention a separate "training set" in the context of an AI/machine learning model. This is an IVD assay, not an AI device trained on data. The development process for such an assay involves R&D, optimization, and internal verification stages, but the term "training set" doesn't directly apply in the same way it would for AI products.

8. How Ground Truth for the Training Set Was Established:

   • As this is not an AI/machine learning device, the concept of a "training set" with established ground truth as per AI/ML typically doesn't apply. The development of the assay itself would involve internal studies and optimization based on known reference materials and clinical samples, but these are part of the assay's design and analytical verification, not a training phase for a learning algorithm.

Ask a specific question about this device