(295 days)
The BinaxNOW® G6PD (Glucose-6-Phosphate Dehydrogenase) Test is an in vitro enzyme chromatographic test for the qualitative detection of G6PD enzyme activity in human venous whole blood, collected in heparin or ethylenediaminetetraacetic acid (EDTA). The BinaxNOW® G6PD Test is a visual screening test used for differentiating normal from deficient G6PD activity levels in whole blood and is intended to aid in the identification of people with G6PD deficiency. Samples which generate deficient results should be assayed using a quantitative G6PD test method to verify the deficiency.
The BinaxNOW® G6PD test device consists of a lateral flow test strip comprised of a white sample pad and a reaction pad, which is located at the top of the strip (2 U.S. patents pending). The reaction pad contains the reagents necessary for the G6PD enzymatic reaction and the subsequent reduction of a nitro blue tetrazolium dye into its concomitant blue formazan product. The resulting color change on the strip indicates enough G6PD activity is present to presume the sample is not deficient.
To perform the test, a whole blood sample is mixed with red blood cell (RBC) lysing reagent in a sample preparation vial and then transferred to the test device sample pad. The lysed blood sample migrates up the test strip, reconstituting reagents in the reaction pad. When the sample front (or liquid migration) covers the entire reaction pad, the device is closed.
Test results are read visually. If no change in the red color of the sample front is observed at the test read time, the sample is presumed to be deficient in G6PD enzyme activity. Samples normal in G6PD activity produce a distinct color change - the red sample color changes to a brown / black color on the upper half of the reaction pad.
Here's a breakdown of the acceptance criteria and the study details for the BinaxNOW® G6PD Test, based on the provided text:
Acceptance Criteria and Device Performance
The core acceptance criteria revolve around the agreement of the BinaxNOW® G6PD Test results with a commercially available quantitative G6PD test, particularly for identifying G6PD deficiency and normal activity.
| Acceptance Criteria (Implicit) | Reported Device Performance (BinaxNOW® G6PD Test) |
|---|---|
| Agreement with Quantitative G6PD Test for Deficient Results (Heparin Samples): High percentage of agreement for deficient results compared to the quantitative method. | Deficient result percent agreement (Heparin Samples): 98.0% (CI = 89.3 - 99.6%) |
| Agreement with Quantitative G6PD Test for Normal Results (Heparin Samples): High percentage of agreement for normal results compared to the quantitative method. | Normal result percent agreement (Heparin Samples): 97.9% (CI = 94.8 - 99.2%) |
| Overall Agreement with Quantitative G6PD Test (Heparin Samples): High overall percentage of agreement. | Overall percent agreement (Heparin Samples): 97.9% (CI = 95.3 - 99.1%) |
| Agreement with Quantitative G6PD Test for Deficient Results (EDTA Samples): High percentage of agreement for deficient results compared to the quantitative method. | Deficient result percent agreement (EDTA Samples): 98.0% (CI = 89.5 - 99.6%) |
| Agreement with Quantitative G6PD Test for Normal Results (EDTA Samples): High percentage of agreement for normal results compared to the quantitative method. | Normal result percent agreement (EDTA Samples): 97.4% (CI = 94.2 - 98.9%) |
| Overall Agreement with Quantitative G6PD Test (EDTA Samples): High overall percentage of agreement. | Overall percent agreement (EDTA Samples): 97.6% (CI = 94.8 - 98.9%) |
| Consistency between Heparin and EDTA Samples: High agreement between the two sample types. | Percent agreement between heparin and EDTA samples: 99% (for 240/243 subjects) |
| Reproducibility Across Operators and Sites: High agreement with expected test results across multiple operators, days, and sites. | Reproducibility Agreement: 98.4% (123/125) with no significant differences within run, between run, between sites, or between operators. |
| Precision (Single Operator): Consistent results by a single operator over multiple days. | Demonstrated 100% consistent results for both normal and deficient samples over ten successive days by a single operator. |
| Absence of Interference from Endogenous Substances: Test performance not affected by high levels of various endogenous blood components. | Interference Testing: None of the tested endogenous blood components (bilirubin, triglycerides, total cholesterol, lactic acid, lactate dehydrogenase, glucose, copper sulfate) affected test performance. |
| Performance with Abnormally Low/High Hematocrit Levels: Performance maintained within defined limits (though limitations existed as per package insert). | Test performance was affected by abnormally low and high hematocrit levels (17-18% and 54-65%), as described in the Limitations section of the package insert (details not provided in this summary). |
Study Details:
-
Sample Size used for the test set and the data provenance:
- Sample Size: 246 subjects (for the initial clinical sample performance study).
- Data Provenance: Prospective study conducted in 2007-2008 in the U.S.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- The document does not specify the number or qualifications of experts used to establish the ground truth. The "ground truth" was established by a "commercially available quantitative G6PD test," implying a quantitative laboratory assay rather than human expert interpretation of the BinaxNOW® test results.
-
Adjudication method for the test set:
- Adjudication method for the test set results is not explicitly mentioned. The BinaxNOW® G6PD Test is a visual screening test. The comparison was made against a "commercially available quantitative G6PD test," implying a direct numerical comparison against the quantitative assay's results.
-
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, an MRMC comparative effectiveness study was not done. This device is a rapid diagnostic test (lateral flow, enzyme chromatographic) for G6PD deficiency, read visually, not an AI-powered diagnostic imaging tool. It involves a single visual reading of the color change on the strip.
- The closest "multi-reader" equivalent mentioned is the Reproducibility Study, which involved "multiple operators" (6 operators) across "3 separate sites." However, this was to assess the consistency of the device's output and interpretation, not to compare human performance with/without AI assistance.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Yes, in effect, a standalone performance was done for comparing the device's output against the quantitative method. The BinaxNOW® G6PD Test is designed to be a standalone visual screening test. The "performance summary" directly compares its qualitative output (deficient/normal) to the quantitative method's results, indicating its performance as a standalone diagnostic tool. Human readers are involved in the interpretation of the visual result of the BinaxNOW® test, but the device itself is a qualitative biochemical assay, not an AI algorithm.
-
The type of ground truth used:
- Quantitative Assay Results: The ground truth was established using results from a "commercially available quantitative G6PD test." This serves as the reference standard. A specific cut-off value of 4.2 U/gHb from the comparative method was used to define "deficient" and "normal" for comparison purposes.
-
The sample size for the training set:
- The document does not explicitly mention a separate "training set" as would be typical for machine learning models. For this type of in-vitro diagnostic device, development and optimization often involve internal studies, but specific details on a "training set" size for a machine learning context are not provided because it's not an AI device. The clinical performance study used 246 subjects.
-
How the ground truth for the training set was established:
- As noted above, a distinct "training set" with established ground truth in the context of an AI algorithm is not applicable here. The development and validation of the BinaxNOW® G6PD Test would have involved establishing performance characteristics against known G6PD activity levels, likely using quantitative methods, but this is a product development process rather than an AI model training process.
§ 864.7360 Erythrocytic glucose-6-phosphate dehydrogenase assay.
(a)
Identification. An erythrocytic glucose-6-phosphate dehydrogenase assay is a device used to measure the activity of the enzyme glucose-6-phosphate dehydrogenase or of glucose-6-phosphate dehydrogenase isoenzymes. The results of this assay are used in the diagnosis and treatment of nonspherocytic congenital hemolytic anemia or drug-induced hemolytic anemia associated with a glucose-6-phosphate dehydrogenase deficiency. This generic device includes assays based on fluorescence, electrophoresis, methemoglobin reduction, catalase inhibition, and ultraviolet kinetics.(b)
Classification. Class II (performance standards).