Search Results

For the in vitro quantitative measurement of creatine kinase activity in serum and plasma on the SK500 Clinical Chemistry System. Measurements of creatine kinase are used in the diagnosis and treatment of myocardial infarction and muscle diseases such as progressive, Duchenne-type muscular dystrophy.

The SEKURE Creatine Kinase Assay (CK Assay) is a spectrophotometric, coupled enzyme assay for the quantitative measurement of creatine kinase (CK) activity. The assay consists of two working reagents, a buffer solution (R1) and a substrate reagent (R2). The SEKURE CK Assay employs the reverse reaction of CK, to produce adenosine triphosphate (ATP). The reaction is coupled to hexokinase and G6PDH which consumes ATP to generate NADPH. The rate of NADPH formation is monitored at 340 nm and is directly proportional to CK activity. Testing is performed on the SK500 in conjunction with calibrator and controls which are provided separately.

The SK500 Analyzer is manufactured as Clinical Chemistry Analyzer Tokyo Boeki Medisys Inc. Biolis 50i Superior. "SK500" is the Sekisui Diagnostics labelled name for the Tokyo Boeki Medisys Inc. Biolis 50i Superior instrument.

This document describes the SEKURE Creatine Kinase Assay, an in vitro diagnostic device, and its performance study to demonstrate substantial equivalence to a predicate device.

1. Acceptance Criteria and Reported Device Performance

The device performance is evaluated against various analytical metrics. While explicit "acceptance criteria" for each study are not individually listed as pass/fail thresholds in a table, the document reports the results of these studies, implying that the observed performance met internal or regulatory expectations for demonstrating substantial equivalence. The predicate device's characteristics serve as an implicit benchmark for similarity.

Here's a table summarizing the reported device performance, with implied acceptance based on the submission being cleared:

2. Sample Sizes Used for the Test Set and Data Provenance

• Precision:

  • Sample Size: 80 replicates per sample type (Serum Control 1, Serum Control 2, Serum High Pool, Plasma Low Pool, Plasma Med Pool, Plasma High Pool) for each of two reagent lots. Total of 480 measurements per lot.
  • Data Provenance: Not explicitly stated, but typical for in vitro diagnostic device studies would be laboratory-generated or purchased control materials and pooled human specimens. The study was conducted in-house by SEKISUI DIAGNOSTICS P.E.I. INC.
  • Retrospective/Prospective: Analytical studies like precision are typically prospective, performed under controlled laboratory conditions.

• Analytical Sensitivity (LoB, LoD):

  • Sample Size: 60 measurements for each level (low level samples, saline blanks, dilutions of pooled serum/plasma) for each reagent lot and matrix type.
  • Data Provenance: Not explicitly stated, but likely laboratory-generated materials.
  • Retrospective/Prospective: Prospective laboratory study.

• Limit of Quantitation (LoQ):

  • Sample Size: 40 replicates (8 linear serum samples, 8 linear plasma samples assayed in 5 runs over 3 days) for each reagent lot and matrix type.
  • Data Provenance: Not explicitly stated, but likely laboratory-generated materials.
  • Retrospective/Prospective: Prospective laboratory study.

• Linearity/Assay Reportable Range:

  • Sample Size:
    • Serum: 8 dilutions of pooled serum, tested in quadruplicate on two lots.
    • Plasma: 10 dilutions and admixtures of pooled plasma, tested in quadruplicate on two lots.
  • Data Provenance: Pooled serum and plasma likely from human donors, obtained through commercial biospecimen providers or internal collection (with ethics approval).
  • Retrospective/Prospective: Prospective laboratory study.

• Analytical Specificity (Interference):

  • Sample Size: Minimum of seven interferent concentrations in replicates of five on each of two reagent lots.
  • Data Provenance: Laboratory spiked samples using serum.
  • Retrospective/Prospective: Prospective laboratory study.

• Method Comparison with Predicate Device:

  • Sample Size: 112 serum specimens, tested in duplicate.
  • Data Provenance: Human serum specimens, source not further specified (e.g., country of origin).
  • Retrospective/Prospective: Likely prospective collection or use of banked retrospective samples under controlled laboratory conditions.

• Matrix Comparison (Serum vs. Lithium Heparin Plasma):

  • Sample Size: 75 matched serum and lithium heparin plasma specimens, assayed in duplicate.
  • Data Provenance: Matched human serum and plasma specimens, source not further specified.
  • Retrospective/Prospective: Likely prospective collection or use of banked retrospective samples under controlled laboratory conditions.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

This type of in vitro diagnostic device (IVDD) for Creatine Kinase measurement, which measures a biochemical analyte, does not typically rely on "experts" in the same way an imaging AI device would. The "ground truth" for these studies is established through:

• Reference Methods: For value assignment (e.g., DC-Cal Multi-Analyte Calibrator traceable to an IFCC reference method).
• Highly Characterized Materials: Pooled serum/plasma, control materials with known target values.
• Statistical Analysis: CLSI guidelines (e.g., EP05-A3, EP17-A2, EP06-A, EP07-A2, EP09-A3) provide the statistical framework for defining performance metrics like precision, linearity, and limits of detection/quantitation.
• Comparison to Predicate: The predicate device's established performance serves as a comparative benchmark.

Therefore, there were no specific "experts" (like radiologists interpreting images) establishing ground truth for individual test cases in the context of this 510(k) submission.

4. Adjudication Method for the Test Set

Not applicable. As described above, the ground truth for this IVDD study is based on quantitative measurements against reference methods, characterized materials, and statistical analysis, not on subjective interpretations requiring adjudication by multiple readers or experts.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done

No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done. MRMC studies are primarily relevant for imaging devices or AI tools where human interpretation of medical images is involved, and the AI's impact on reader performance is being evaluated. This 510(k) is for an in vitro diagnostic assay, which directly measures an analyte concentration.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done

Yes, the studies presented are all "standalone" in the sense that they evaluate the performance of the SEKURE Creatine Kinase Assay and the SK500 analyzer system based on their analytical capabilities to directly measure CK activity. There is no "human-in-the-loop" component in the direct measurement process or the evaluation of its analytical performance. The device provides a quantitative result without human interpretive input for the measurement itself.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

The ground truth used here is primarily analytical ground truth based on:

• Reference Methods: Specifically, the CK activity value assignment for DC-Cal is traceable to an IFCC reference method.
• Characterized Materials: Pooled serum and plasma with known or expected CK activity levels derived from rigorous analytical testing.
• Statistical Models: Ground truth for linearity (theoretical values) and detection limits (through statistical calculation as per CLSI guidelines).
• Comparison to a Legally Marketed Predicate Device: The predicate device's performance (Creatine Kinase-SL Assay on a Hitachi 717 Analyzer) serves as a comparative ground truth for demonstrating substantial equivalence.

8. The sample size for the training set

This submission pertains to the performance validation of a diagnostic assay (reagents and analyzer), not a machine learning or AI algorithm in the traditional sense that requires distinct "training sets" and "test sets" for model development and validation. Therefore, there is no explicit separate "training set" described for an AI model.

The "training" of such a system involves the development and optimization of the assay chemistry and instrument parameters. The data presented here are part of the verification and validation (V&V) studies conducted on the final product to demonstrate its performance characteristics and substantial equivalence, akin to a "test set" for a traditional product.

9. How the ground truth for the training set was established

As there isn't a "training set" in the context of an AI algorithm, the concept of establishing ground truth for it doesn't apply directly here. The development of the assay involves standard analytical chemistry principles, optimization of reagent concentrations, reaction conditions, and instrument calibration, with performance iteratively assessed against established analytical standards and clinical relevance. This is an engineering and chemistry development process rather than an AI model training process that relies on labeled datasets.

Ask a specific question about this device