Search Results

The Accuwell" 17α-Hydroxyprogesterone EIA Kit is designed for the quantitative measurement of 17α-Hydroxyprogesterone (17-OHP) concentrations in neonatal blood samples that have been collected onto Whatman 903® specimen collection paper. The results are used to screen newborns for classical congenital adrenal hyperplasia (CAH).

Device Description

The Accuwell" 17a-Hydroxyprogesterone EIA Kit is a competitive solid phase enzyme immunoassay (ElA) for the quantitative measurement of 17«-Hydroxyprogesterone (17-OHP) concentrations in neonatal blood samples that have been collected onto Whatman S&S 903" specimen collection paper. Standard control, and unknown dried blood spot sample discs are added to specified wells in a 90-well microplate coated with antiserum specific for 17-OHP.

AI/ML Overview

Here's an analysis of the acceptance criteria and supporting study for the Accuwell 17-alpha-Hydroxyprogesterone EIA, based on the provided text:

Device: Accuwell 17-alpha-Hydroxyprogesterone (OHP) enzyme immunoassay (EIA) for neonatal screening.

Acceptance Criteria and Reported Device Performance

The acceptance criteria for this device are primarily demonstrated through comparison to a legally marketed predicate device (AutoDELFIA Neonatal 17a-OH-progesterone) and adherence to established internal precision and analytical sensitivity targets. While explicit "acceptance criteria" in terms of pass/fail thresholds for each metric are not always numerically stated as such, the document presents comparative data and targets that would have been used to demonstrate substantial equivalence.

Acceptance Criteria Category	Specific Metric	Acceptance Criteria (Implicit/Explicit)	Reported Device Performance (Accuwell 17-OHP EIA)
Intended Use Equivalence	Intended Use	To measure 17α-hydroxyprogesterone (17-OHP) levels in neonatal dried blood spots for screening newborns for classical congenital adrenal hyperplasia (CAH). (Identical to predicate)	Measures 17α-hydroxyprogesterone (17-OHP) concentrations in neonatal blood samples collected on Whatman 903® paper, used to screen newborns for classical CAH. (Meets)
Analytical Sensitivity	Limit of Detection	"Should be confirmed by each laboratory and appropriate concentrations defined." (Implies a low detection limit is desirable and comparable to predicate's 1.5 ng/ml)	Overnight Eluate: 2.2 ng/ml
3 Hour Eluate: 1.5 ng/ml
Overnight Direct: 2.4 ng/ml
Precision	Within-Run Precision (%CV)	Predicate: 8.8-11.3 (low), 8.9 (medium), 8.9 (high) %CV
Accuwell: 3.5% (low), 3.7% (medium), 2.7% (high) %CV for overall 'Proposed Device' table. More detailed data below from dedicated studies.	From Dedicated Precision Study:
ON Eluate: 10.7-11.4 %CV (total), within-run Sr provided
3 HR Eluate: 11.3-12.9 %CV (total), within-run Sr provided
ON Direct DBS: 10.0-14.1 %CV (total), within-run Sr provided
	Between-Run Precision (%CV)	Predicate: 4.0-3.3 (medium) %CV
Accuwell: 10.3-8.0 (medium), 11.2 (high) %CV for overall 'Proposed Device' table. More detailed data below from dedicated studies.	From Dedicated Precision Study:
ON Eluate: Between-day Sdd provided
3 HR Eluate: Between-day Sdd provided
ON Direct DBS: Between-day Sdd provided
	Total Precision (%CV)	Predicate: 12.0-9.5 (medium), 9.2 (high) %CV
Accuwell: 11.0-8.9 (medium), 11.6 (high) %CV for overall 'Proposed Device' table. More detailed data below from dedicated studies.	From Dedicated Precision Study:
ON Eluate: 10.7 (28.2 ng/ml), 8.8 (56.1 ng/ml), 11.4 (108.1 ng/ml) %CV
3 HR Eluate: 12.9 (26.2 ng/ml), 7.7 (53.0 ng/ml), 11.3 (104.6 ng/ml) %CV
ON Direct DBS: 14.1 (29.5 ng/ml), 10.0 (57.9 ng/ml), 12.2 (108.6 ng/ml) %CV
Linearity & Recovery	Range	Approximately 2.0 ng/ml to 250 ng/ml.	Linear range and measuring range is approximately 2.0 ng/ml to 250 ng/ml. Measured % Recovery between 85.4% and 119.6% across 12 concentration levels (10-300 ng/ml). Regression R^2 values: 0.9994 (ON Eluate), 0.992 (3 Hr Eluate), 0.99 (ON Direct DBS).
Method Comparison	Correlation with Predicate (R-value, Slope, Intercept)	No explicit threshold given. Implied to be strong correlation (e.g., R > 0.9, slope near 1, intercept near 0) demonstrating substantial equivalence.	R-values consistently > 0.90 across all populations and methods (range 0.8421 to 0.9785). Slopes generally between 0.970 and 1.234. Intercepts between +1.081 and +6.569. (Specific values vary by population, method, percentile).
Specificity	Cross-Reactivity	List of compounds with cross-reactivity 2500g b.w., 0-1 days old:** 133 infant blood spot specimens.

    *   **Population >2500g b.w., 2-3 days old:** 133 infant blood spot specimens.
    *   **Population >2500g b.w., >4 days old:** 197 infant blood spot specimens.
    *   **Population 1400-2500g b.w., 1-6 days old:** 30 infant blood spot specimens.
    *   **Total Samples for Method Comparison:** 133 + 133 + 197 + 30 = **493 infant blood spot specimens**.
    *   **Data Provenance:** Retrospective study. Test samples were **blinded neonatal dried blood spots collected in sequence under routine screening conditions from a U.S. department of public health laboratory.** Original screening results for each sample using the predicate test kit were also obtained from the submitting laboratory.

*   **Precision Study:**
    *   **Test samples:** CDC Control Lots 451, 452, and 453 (enriched with 25, 50, and 100 ng/ml serum equivalents, respectively). These are laboratory control materials, not patient samples.
    *   **Number of measurements:** 40 measurements for each control level across each of the three assay procedures (ON Eluate, 3 HR Eluate, ON Direct DBS).

*   **Analytical Sensitivity Study:**
    *   **Test Samples:** Zero standard (N=26).
    *   **Number of measurements:** 26 measurements in one assay for each of the two procedures (eluate and direct, spanning elution times).

*   **Linearity and Recovery Study:**
    *   **Test Samples:** Dried blood spots representing 12 concentrations (10, 24.5, 39, 68, 97, 126, 155, 184, 213, 242, 271, 300 ng/ml).
    *   **Number of measurements per sample:** Four aliquots of each of the 12 samples were tested in each of two runs using each of the three assay procedures. This implies 8 measurements per sample for each procedure.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
* Not applicable (N/A) in the AI sense. This is an in vitro diagnostic device measuring a biomarker. "Ground truth" for the test set in the method comparison study was established by the measurement of these same samples using the predicate device (AutoDELFIA Neonatal 17a-OH-progesterone kit) in a reference lab setting, combined with the fact they were routine screening samples from a public health lab. The device's performance is gauged against another validated assay, not against expert human interpretations of images or clinical cases.

Adjudication method (e.g., 2+1, 3+1, none) for the test set:
- None. This is a quantitative assay, comparing the numerical output of the new device against the numerical output of a predicate device. There is no human diagnostic interpretation to adjudicate.
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. This is an in vitro diagnostic device for laboratory measurement, not an AI-assisted diagnostic imaging or interpretation tool for human readers. Therefore, an MRMC study is not applicable.
If a standalone (i.e. algorithm only without human-in-the loop performance) was done:
- Yes, effectively. The Accuwell 17-OHP EIA kit itself is a standalone diagnostic tool. Its performance (accuracy, precision, linearity, sensitivity, specificity) is evaluated independently, and then its results are compared to those of another "standalone" diagnostic kit (the predicate). While a human technician operates the EIA and reads the microplate spectrophotometer, the assay itself generates the quantitative results without human cognitive interpretation of subjective data. The "human-in-the-loop" here is the lab professional following the procedural steps.
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Comparison to a legally marketed predicate device. In the context of 510(k) submissions for in vitro diagnostics, the predicate device's performance often serves as the "ground truth" for demonstrating substantial equivalence. The predicate device itself would have been validated against some form of clinical or analytical ground truth (e.g., confirmed CAH diagnoses, reference methods like GC-MS, or clinical outcomes), but for this submission, the predicate is the primary comparison.
The sample size for the training set:
- Not applicable (N/A) in the AI/machine learning sense. This is a chemical immunoassay kit, not a machine learning model that undergoes "training" on a dataset. The development of the assay would have involved various experimental iterations and optimizations, but there isn't a "training set" like in AI.
How the ground truth for the training set was established:
- Not applicable (N/A). As stated above, this is an immunoassay kit, not an AI model requiring a training set with established ground truth. The "training" of the device is through its manufacturing process and quality control, ensuring it produces accurate biochemical measurements based on validated reagents and procedures.

Ask a Question

Ask a specific question about this device

Page 1 of 1