Search Results

The Hemoglobin A Ic assay is used in clinical laboratories for the quantitative in vitro measurement of percent hemoglobin A le (NGSP) or HbA Ic fraction mmobile blood and hemolysate on the ARCHITECT c 8000 System. Hemoglobin Alc measurements are used as an aid in the diagnosis of diabetes mellinus, as an aid to identify patients who may be at risk for developing diabetes mellitus, and for the monitoring of long-eem blood glucose control in individuals with diabetes mellitus.

The Hemoglobin Alc Calibrators are for use in the Hemoglobin Alc assay on the ARCHITECT c 8000 System.

The Hemoglobin Alc Controls are used for the estimation and the detection of systematic analytical deviations of the ARCHITECT c 8000 System.

The Hemoglobin A1c Reagent Kit contains Reagent 1 (R1), Reagent 2 (R2), and Diluent (A1cDIL). The Hemoglobin A1c Calibrator Kit contains Calibrator 1 (Cal 1) and Calibrator 2 (Cal 2), which are lyophilized human whole blood containing hemoglobin and glycated hemoglobin. The Hemoglobin A1c Control Kit contains Low Control (Control L) and High Control (Control H), which are lyophilized human whole blood containing hemoglobin and glycated hemoglobin. The assay utilizes an enzymatic method to measure glycated hemoglobin (HbA1c) and total hemoglobin (THb) concentrations, which are then used to calculate percent HbA1c (NGSP units) or HbA fraction (IFCC units). The assay is performed on the ARCHITECT c 8000 System.

Here's a summary of the acceptance criteria and the study details for the Hemoglobin A1c assay:

Acceptance Criteria and Device Performance

Study Details

1. Sample sizes used for the test set and data provenance:

   • Within-Laboratory Precision: Not specified for sample patients, but involved 3 reagent lots, 3 calibrator lots, 1 control lot (L/H), 1 lot of commercial controls (3 levels), human whole blood controls, and human whole blood panels. Tested in 2 replicates, twice per day for 20 days.
   • Limit of Blank (LoB) & Detection (LoD): Zero-level samples (minimum 3 replicates), low-level samples (minimum 2 replicates). Five runs over minimum 3 days.
   • Interferences (Endogenous, Drugs, Hb Variants, RF, Hb Derivatives): Minimum of 12 replicates for test and reference samples.
     • Hb Variants specific:
       • HbC: n=21
       • HbD: n=20
       • HbE: n=20
       • HbS: n=20
       • HbA2: n=26
       • HbF: n=19
   • Matrix Comparison: Specimens from a minimum of 43 different donors. Each sample tested in a minimum of 2 replicates.
   • Linearity: 9 samples (commercially available linearity sets + 5 prepared by combining them). Tested in a minimum of 2 replicates.
   • Method Comparison and Predicted Bias: Minimum of 120 human whole blood specimens. Tested internally in replicates of 2. Tested over a minimum of 5 days.
   • Data Provenance: Not explicitly stated for all studies (e.g., country of origin). However, the "human whole blood specimens" and "human whole blood controls" suggest these are human samples. Studies were performed in-house at Abbott Laboratories. All studies are retrospective in the sense that they are laboratory performance evaluations on collected samples, not prospective patient trials.

2. Number of experts used to establish the ground truth for the test set and qualifications of those experts (e.g. radiologist with 10 years of experience):

   • Not applicable as this is an in vitro diagnostic (IVD) device for quantitative measurement. The ground truth for the method comparison study was established using an NGSP secondary reference laboratory method (Tosoh HPLC analyzer), which is a recognized standardized method for HbA1c.

3. Adjudication method (e.g. 2+1, 3+1, none) for the test set:

   • Not applicable. The ground truth was established by comparison to a recognized reference method or by preparing samples with known concentrations/characteristics.

4. 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:

   • Not applicable. This is an IVD device, not an imaging AI device that involves human readers.

5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

   • Yes, the performance studies described (Precision, LoB/LoD, Interferences, Linearity, Method Comparison) are all standalone performance evaluations of the device and its reagents. The device operates as an automated system measuring HbA1c concentrations.

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

   • For Method Comparison: A National Glycohemoglobin Standardization Program (NGSP) secondary reference laboratory method (Tosoh HPLC analyzer) was used as the comparator (reference) method. This is a highly standardized and traceable ground truth for HbA1c measurements.
   • For Precision, LoB/LoD, Linearity, Interferences: Ground truth was established by:
     • Using calibrators traceable to the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) reference method.
     • Using controls with value-assigned concentrations (traceable to IFCC reference method and NGSP units).
     • Creating samples with known concentrations (e.g., "zero-level," "low-level," linearity sets, spiked interference samples, samples with known Hb variants).

7. The sample size for the training set:

   • Not applicable. This is an IVD device based on an enzymatic methodology, not a machine learning or AI algorithm that requires a "training set" in the traditional sense. The device's performance is based on its chemical and enzymatic reactions, and the calibration process uses a dedicated calibrator kit, not a "training set" of patient data for model learning.

8. How the ground truth for the training set was established:

   • Not applicable, as there is no "training set" for a machine learning model. The device's calibration involves calibrators that are value-assigned using secondary calibrators traceable to the IFCC reference method.

Ask a specific question about this device