Search Results

The GBI TSH Neonatal Screening Kit is designed for the quantitative determination of Thyroid Stimulating Hormone (TSH) concentrations in neonatal blood samples that have been collected onto Whatman 903 specimen collection paper. The results are used to screen newborns for congenital hypothyroidism.

The GBI TSH Neonatal Screening Kit is an enzyme immunoassay. A highly specific polyclonal goat antihTSH (human) antibody has been immobilized onto each well of the 96-well microplates provided. To begin the assay, sample discs punched from dried whole blood spot standards, controls and neonate specimens are added to the coated wells. An elution buffer is also added. The plate is incubated to elute TSH from the sample disc and to allow capture of the eluted TSH by the antibody immobilized onto the microplate wells. Following incubation the plates are washed to remove the sample discs as well as the eluate. A second antibody, a D-specific anti-hTSH monoclonal that has been conjugated to the enzyme horseradishperoxidase (HRP), is then added to the wells and incubated. The eluted TSH of the sample already captured by the microplate-bound antibody is now also bound by the enzyme-conjugated monoclonal antibody added. An antibody-TSH-antibody bridge, or "sandwich", forms that is bound to the surface of the microplate wells. Any unbound complexes are removed with subsequent plate washings. The final stage of the assay is the detection of the microwell-bound complexes by the addition of a color developing reagent. The enzyme (HRP) portion of the bound "sandwich" reacts with the color developer, 3, 3', 5, 5'-Tetramethylbenzidine (TMB) in the presence of hydrogen peroxide (H2O2). The TMB/ H2O2 liquid is converted from colorless to blue. The degree of color change is directly proportional to the amount of TSH antigen that is bound in the well. The color development is terminated with the addition of a color stopper that converts the blue to yellow. The results are measured with a microplate reader at a wavelength of 450 mm. The absorbance measured is directly proportional to the concentration of TSH in the sample. A standard curve is generated by plotting the light absorbance of each standard versus its known TSH concentrations of TSH in the unknown samples are determined by interpolation from this standard curve.

Here's an analysis of the acceptance criteria and study proving the device meets them, based on the provided document:

1. Table of Acceptance Criteria and Reported Device Performance

Study Description and Details

The provided document describes a 510(k) premarket notification for the GBI TSH Neonatal Screening Kit, comparing it to a legally marketed predicate device (Accuwell Thyroid Stimulating Hormone (TSH) ELISA). The studies performed demonstrate the analytical performance of the GBI TSH EIA kit.

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

• Precision Study:
  • Sample Size: 40 replicates (2 aliquots per run, 1 run per day, over 20 days) for five different sample IDs (C1-C5).
  • Data Provenance: "in-house" testing, implying controlled laboratory conditions. The specific country of origin is not explicitly stated for the samples themselves beyond "human whole blood" used for standards. The study was conducted by Golden Bridge International, Inc. (Mukilteo, WA, USA). This appears to be a prospective study designed for this submission.
• Analytical Sensitivity Study:
  • Sample Size: 20 measurements of the zero standard within a single assay.
  • Data Provenance: "in-house" testing, likely from the manufacturer's lab.
• Linearity, Recovery and Assay Measurable Range Study:
  • Sample Size: 19 different TSH concentrations (expected values), with two 1/8 inch punches from each concentration tested in two runs (total of n=4 results per concentration level).
  • Data Provenance: Prepared dried whole blood spot samples for the study, likely in-house.
• Specificity (Cross-Reactivity) Study:
  • Sample Size: Not explicitly stated as a count of individual samples, but various concentrations of FSH, LH, and HCG were tested in TSH-free whole blood.
  • Data Provenance: Likely in-house prepared samples.
• Interfering Substances Study:
  • Sample Size: Not explicitly stated as a count of individual samples, but various concentrations of lipid, conjugated bilirubin, unconjugated bilirubin, and hemoglobin were tested. The main interference study involved "fresh whole blood" specimens at desired concentrations, and an "additional study" challenged the assay with higher hemoglobin concentrations using CDC TSH control samples C1-C4, with 6 measurements per condition ("Mean; n=6").
  • Data Provenance: Likely in-house prepared or spiked samples for the main study. For the additional hemoglobin study, CDC TSH control samples were used.
• Method Comparison Study:
  • Sample Size: 995 neonatal blood spot specimens (980 from a presumed normal population and 15 from patients confirmed as positive for hypothyroidism). A subset of N=832 samples was used for statistical analysis after excluding non-numeric results.
  • Data Provenance: "in-house" analysis using neonatal blood spot specimens. The origin of these specimens (e.g., specific country, retrospective/prospective collection) is not explicitly detailed beyond being "from a presumed normal population" and "from patients confirmed as positive for hypothyroidism." It is common for such samples to be retrospectively collected or part of a sample bank.

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

• No specific experts are mentioned for establishing ground truth for the test set samples.
• For the Method Comparison study, 15 samples were "from patients confirmed as positive for hypothyroidism." This implies a clinical diagnosis served as a ground truth, but the number and qualifications of the clinicians/experts making these diagnoses are not specified.
• For the Interfering Substances study, CDC TSH control samples were used, implying the CDC's established values serve as a reference, but no specific "experts" were consulted to establish the ground truth for the study samples.
• The "Expected Value" for linearity was a technical target, not an expert-derived ground truth from patient samples.
• The standards (calibrators) and controls are referenced against WHO 3rd IRP of human TSH 81/565, which is an international standard, not an expert consensus from individual case review.

4. Adjudication Method for the Test Set

• There is no mention of an adjudication method (like 2+1, 3+1, none) typically used for subjective endpoints or diagnostic imaging.
• For quantitative assays like this TSH EIA, the comparison is typically against a reference method (the predicate device) or known established values (e.g., CDC controls, spiked samples). Discrepancies are analyzed statistically rather than by expert adjudication.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and Effect Size of Human Readers Improve with AI vs without AI Assistance

• No, this is not applicable. The GBI TSH Neonatal Screening Kit is an in-vitro diagnostic (IVD) device, specifically an enzyme immunoassay (EIA). It's a laboratory-based test that provides a quantitative numerical result for TSH concentration. It does not involve human "readers" interpreting images or other subjective data, nor does it involve AI assistance in that context. Therefore, a multi-reader multi-case (MRMC) study or assessment of AI's effect on human reader improvement is not relevant to this type of device.

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

• Yes, this is effectively a standalone performance study. The GBI TSH Neonatal Screening Kit itself is the "algorithm" (the immunoassay process and signal detection). The performance metrics (analytical sensitivity, precision, linearity, specificity, interference, method comparison) are all measures of the device's intrinsic analytical capabilities without human intervention influencing the result generation, other than standard laboratory procedures for running the assay. The intent is for the device to quantitatively determine TSH levels.

7. The Type of Ground Truth Used

The ground truth for the various studies primarily relies on:

• Known concentrations: For precision, analytical sensitivity, linearity, cross-reactivity, and interference, the "ground truth" refers to the known concentrations of TSH (or lack thereof for specificity) in spiked samples, control materials (like CDC controls), or standards.
• Reference method comparison: For the method comparison study, the results from the predicate device (Accuwell TSH ELISA) served as the primary reference or "ground truth" to which the new device's results were compared.
• Clinical Diagnosis: For the 15 positive samples in the method comparison, they were "from patients confirmed as positive for hypothyroidism," implying a clinical diagnosis as ground truth.

8. The Sample Size for the Training Set

• The document does not explicitly describe a "training set" in the context of machine learning. This is an IVD device, not an AI/ML device that undergoes training.
• However, the standards (calibrators) used to generate the standard curve that unknowns are interpolated from can be considered analogous to training data for the assay itself. The kit includes "TSH Dried Blood Standards and Controls" described as 6 standards (zero, 7.5, 15.0, 30.0, 60.0, and 120.0 ulU/ml serum equivalent) and 3 controls (15, 30, and 80 µIU/m1 serum equivalent). These are used in every run to establish the curve and validate the assay, which is more akin to continuous calibration and quality control rather than a one-time training set for an algorithm.

9. How the Ground Truth for the Training Set Was Established

• As noted above, for an EIA kit, the "training set" is effectively the calibrators (standards).
• The ground truth for these standards is established by:
  • Preparation from human whole blood adjusted to 55% hematocrit and spotted onto Whatman 903 paper.
  • Addition of known concentrations of TSH analyte.
  • Referenced against WHO 3rd IRP of human TSH 81/565. This indicates traceability to an internationally accepted reference standard, ensuring the accuracy of the assigned TSH concentrations.

Ask a specific question about this device