FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

The SERADYN OMS™ DIGOXIN assay is used for the quantitation of digoxin in human serum or plasma on the Abbott AEROSET® System.
Digoxin is a potent cardiac glycoside widely prescribed for the treatment of patients suffering from congestive heart failure, as well as some types of cardiac arrhythmias. Digoxin intoxication is a common and serious problem in the clinical setting. This is, in part, a result of the fact that cardiac glycosides have a low therapeutic ratio (a very small difference between therapeutic and tissue toxic levels). Coupled with the narrow therapeutic range is a marked patient variability in response to the same dosage of drug, often resulting in unpredictable serum drug levels. Intoxication symptoms are often indistinguishable from the original condition for which the drug was prescribed. It may not be immediately apparent whether the patient has been under or overdosed.
Monitoring serum digoxin levels combined with other clinical data can provide the physician with useful information to aid in adjusting patient dosage and achieving optimal therapeutic effect while avoiding useless sub-therapeutic or harmful toxic dosage levels.

Device Description

The Seradyn QMS™ Digoxin Assay is a homogeneous Particle Enhanced Turbidimetric Immunoassay used for the quantitation of digoxin in serum or plasma. The assay is intended for use on the Abbott AEROSET® System, using the Seradyn QMS Digoxin Calibrators (previously cleared as Digoxin Immunoassay K983323.
The reagent system components are 1) digoxin coated microparticle reagent, and 2) the antibody reagent which consists of a mouse monoclonal antibody specific for digoxin.
The technology is based on competition between the digoxin in the sample and digoxin coated onto the microparticles, for the antibody-binding sites of the anti-digoxin antibody reagent. In the absence of digoxin in the sample, the specific antibody in the antibody reagent binds the digoxin on the particle, and results in rapid agglutination of the microparticles. In the presence of digoxin in the sample, the digoxin in the sample competes for antibody binding sites of the specific antibody in the antibody reagent, and partially inhibits the agglutination of the microparticles. The rate of agglutination (turbidity) is directly proportional to the rate in absorbance change of incident light and is measured spectrophotometrically by the Abbott AEROSET® System at a wavelength of 700 nm.
A six level Seradyn QMS™ Digoxin Calibrator set, with known digoxin concentrations is used to quantitate the assay. An internal concentration-dependent calibration curve is generated by the AEROSET® System, by measuring the rate of absorbance change of each calibrator level. Maximum absorbance rate is at the lowest digoxin concentration and the lowest absorbance rate at the highest digoxin concentration.
By monitoring the change in rate of a specimen with unknown digoxin concentration, and comparing to the internal calibration curve, a sample's concentration can readily be obtained and reported as digoxin concentration in either ng/ml or nmol/L.

AI/ML Overview

Here's an analysis of the Seradyn QMS™ Digoxin assay's acceptance criteria and the studies performed, based on the provided text:

Acceptance Criteria and Device Performance for Seradyn QMS™ Digoxin on the Abbott AEROSET® System

1. Table of Acceptance Criteria and Reported Device Performance

Study Parameter	Acceptance Criteria	Reported Device Performance
Specificity	Comparable to predicate device (Roche Tina-quant Digoxin)	Similar % Cross-Reactivity for tested compounds (Digitoxigenin, Digitoxin, Digoxigenin, Digoxigenin bis-digitoxoside, Digoxigenin mono-digitoxoside, Spironolactone, Canrenone).
Accuracy by Recovery	100 ± 10% of theoretical value	94% - 102% recovery across four tested concentrations (1.21 - 4.85 ng/mL).
Sensitivity	Comparable to predicate (explicit criterion not stated but implied by comparison)	0.05 ng/mL (lower than predicate's 0.15 ng/mL)
Accuracy & Linearity by Dilution	100 ± 10% of theoretical value (recovery); Acceptable linearity	98% - 105% recovery across five dilutions (1.00 - 5.00 ng/mL); R² = 0.9955 (linear regression)
Precision	No explicit acceptance criteria stated, implied good precision from comparison to predicate.	Within Run CV(%): 0.98% - 2.17%; Between Day CV(%): 0.28% - 0.80%; Total Run CV(%): 0.91% - 2.31%
Method Comparison	No explicit acceptance criteria stated, implied good correlation to predicate.	Correlation Coefficient: 0.99346 (vs. Tina-quant on Hitachi 717)
Interfering Substances	100 ± 12% for hemoglobin; 100 ± 10% for bilirubin; 100 ± 20% for lipids (Note: Discrepancy in bilirubin and hemoglobin acceptance values in text)	99% for Bilirubin (20 mg/dL); 110% for Hemoglobin (1,000 mg/dL); 106% for Intralipid (2,000 mg/dL)
HAMA Interference	100 ± 10% recovery of control serum	101% for HAMA 1; 107% for HAMA 2
Instrument On-board Stability	Acceptable control recoveries, no re-calibration required	Demonstrated stability for 70 days, claimed on-board stability of at least 60 days.
Instrument Calibration Stability	No re-calibration required during the period under study	Claimed 60 days of stability based on on-board stability data.

Note on Interfering Substances: The document states "Acceptance criteria are recoveries of 100±12% for hemoglobin, 100±10% for hemoglobin、100 ±20% for lipids". There appears to be a typo, possibly intending 100±10% for bilirubin and 100±12% for hemoglobin, or vice-versa, given the reported values. I have used the written acceptance criteria as provided.

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

Specificity (Cross-reactivity): Not explicitly stated, samples were "spiked" with specific compounds. Data provenance: Not explicitly stated, but implies laboratory-prepared samples.
Accuracy by Recovery: Not explicitly stated how many samples were used, but "human serum" was spiked at four different concentrations. Data provenance: Not explicitly stated, but implies laboratory-prepared human serum.
Sensitivity: Not explicitly stated how many samples or what method, only the "lowest Digoxin concentration that can be distinguished from zero with 95% confidence."
Accuracy & Linearity by Dilution: The 5.0 ng/mL Digoxin Calibrator was diluted at 80%, 60%, 40%, and 20%. The "diluted samples, as well as the 5.0 ng/mL calibrator" were analyzed. Data provenance: Laboratory-prepared dilutions from a calibrator.
Precision: "A tri-level human serum based commercial control containing digoxin was assayed in duplicate twice a day for twenty days." This means n=80 (3 levels * 2 replicates/day * 20 days) for each level presented in the table. Data provenance: Laboratory testing using commercial control.
Method Comparison: "55" clinical specimens. Data provenance: Not explicitly stated, but "clinical specimens" suggests patient samples.
Interfering Substances: "Abnormal bilirubin levels were prepared by adding to human serum pool... Abnormal hemoglobin levels were prepared by addition of red blood cell lysate... Abnormal lipid levels were prepared by addition of Intralipid®". N=2 for each tested interfering substance. Data provenance: Laboratory-prepared human serum pools.
HAMA Interference: "A normal human serum pool (control), and HAMA type 1 and HAMA type 2 samples were spiked with the same amounts of digoxin." Each of these was assayed "in duplicate". Data provenance: Laboratory-prepared human serum and HAMA samples.
Instrument On-board Stability: Controls were assayed "twice weekly for a period of 60 days". Specific 'n' not given, but implies multiple measurements over time. Data provenance: Laboratory testing.
Instrument Calibration Stability: Data from on-board stability study was used.

In summary, the test set primarily consists of laboratory-prepared samples (spiked human serum, diluted calibrators, commercial controls, prepared interfering substance samples), with one study (Method Comparison) using "clinical specimens." The country of origin is not specified but is likely the US, given the submission to the FDA. All studies appear to be prospective experimental studies conducted by the manufacturer.

3. Number of Experts and Qualifications for Ground Truth

None stated. The studies described are analytical performance studies focused on the intrinsic characteristics and measurement accuracy of the device. They do not involve human interpretation or diagnostic decision-making that would typically require expert ground truth establishment. The ground truth for these studies is based on theoretical concentrations (for recovery, linearity, spiking) or established reference methods (for method comparison).

4. Adjudication Method for the Test Set

None applicable. As stated above, these are analytical performance studies, not studies involving human interpretation or subjective assessments that would require adjudication.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No, an MRMC comparative effectiveness study was not done. The studies are analytical performance assessments of the assay itself, comparing its measurements to known values or a predicate device. There is no human reader component to this type of device (an in-vitro diagnostic assay).

6. Standalone Performance Study (Algorithm only without human-in-the-loop)

Yes, all reported studies are standalone performance studies. The Seradyn QMS™ Digoxin assay is a fully automated immunoturbidimetric assay on the Abbott AEROSET® System. Its performance characteristics (specificity, accuracy, precision, etc.) are measured directly from the output of the instrument, without human intervention in the measurement process itself, beyond sample preparation and loading. The results are numerical concentrations.

7. Type of Ground Truth Used

The ground truth used varies by study:

Specificity (Cross-reactivity): Theoretical concentration of spiked cross-reactants.
Accuracy by Recovery: Theoretical concentrations of digoxin spiked into human serum.
Sensitivity: Defined statistically as the lowest concentration distinguishable from zero with 95% confidence.
Accuracy & Linearity by Dilution: Theoretical concentrations based on serial dilutions of a known calibrator.
Precision: Mean concentrations of commercial controls.
Method Comparison: Results from the predicate device (Roche Tina-quant assay on the Hitachi 717) using clinical specimens.
Interfering Substances: Theoretical concentrations of spiked interfering substances.
HAMA Interference: Mean concentration of the control normal human serum.
Instrument On-board Stability & Calibration Stability: Reference measurements at Day 0, against which subsequent measurements are compared.

8. Sample Size for the Training Set

Not applicable / Not explicitly stated. This device is an in-vitro diagnostic assay rather than an AI/ML algorithm that undergoes a distinct "training phase." Calibration for such assays involves using a multi-level calibrator set to establish a standard curve. The document mentions a "six level Seradyn QMS™ Digoxin Calibrator set" used to generate an "internal concentration-dependent calibration curve." This calibrator set essentially serves the function of "training" the instrument to interpret absorbance changes into digoxin concentrations.

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

For calibration: The ground truth for the "training set" (the six-level calibrator set) is established by the known digoxin concentrations assigned to each calibrator level. These concentrations are determined by the manufacturer (Seradyn, Inc.) through a rigorous process, typically involving reference methods (e.g., mass spectrometry) or traceability to international reference materials. While not detailed in this 510(k) summary, the previous clearance of "Digoxin Immunoassay K983323" for the QMS Digoxin Calibrators suggests their concentrations were already established and accepted by the FDA.

Summary

Regulation Number and Section

§ 862.3320 Digoxin test system.

(a)
Identification. A digoxin test system is a device intended to measure digoxin, a cardiovascular drug, in serum and plasma. Measurements obtained by this device are used in the diagnosis and treatment of digoxin overdose and in monitoring levels of digoxin to ensure appropriate therapy.(b)
Classification. Class II.