(569 days)
The SensiTox C. difficile Toxin Test is an immunofluorescence assay intended for the qualitative detection of Clostridioides difficile toxins A and/or B in human stool specimens. The test is intended as an aid in the diagnosis of C. difficile infection (CDI) in patients exhibiting symptoms of CDI. Negative results do not preclude toxigenic C. difficile infection. The SensiTox C. difficile Toxin Test should not be used as the sole basis for treatment or other management decisions. The test can only be used with the MultiPath platform.
The SensiTox C. difficile Toxin Test detects toxins A and B in stool samples using an immunofluorescence assay and the proprietary MultiPath detection technology. The assay is performed on the proprietary MultiPath Analyzer. A stool sample is added to Stool Specimen Diluent, processed through a spin column, and the filtrate is added to the SensiTox C. difficile Cartridge. The Cartridge is loaded onto the MultiPath Analyzer for processing. The Analyzer reads barcodes, heats the cartridge, splits the sample into aliquots, mixes with antibody conjugated fluorescent and magnetic particles, and incubates. Magnetic particles and tethered fluorescent particles are drawn to the bottom imaging surface by magnets and imaged and quantified using non-magnified digital imaging. Results are interpreted by the MultiPath applications software.
Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided text:
Device: SensiTox C. difficile Toxin Test
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria are implied by the performance metrics reported, as these are the benchmarks the device aims to meet.
| Performance Metric | Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|---|
| Limit of Detection (LoD) | Detection rate ≥95% at specified concentrations for Toxins A and B | Toxin A LoD: 3.5 ng/mL |
| Toxin B LoD: 50 ng/mL | ||
| Reproducibility (Overall) | High reproducibility (implied by 99.2% accuracy reported) | 99.2% (373 correct results out of 376 total samples) |
| Reproducibility (Negative Samples) | High accuracy for negative samples | 98.9% (89/90 correct) |
| Reproducibility (Low Positive Samples) | High accuracy for low positive samples | 98.9% (89/90 correct) |
| Reproducibility (Moderate Positive Samples) | High accuracy for moderate positive samples | 98.9% (89/90 correct) |
| Reproducibility (High Positive Samples) | High accuracy for high positive samples | 100% (100/100 correct) |
| Analytical Reactivity (Inclusivity) | Ability to detect multiple ribotypes of toxins A and B | All 6 tested toxin A ribotypes detected |
| All 8 tested toxin B ribotypes detected | ||
| Analytical Specificity | No cross-reactivity with common organisms; no negative interference from common organisms on toxin detection | None of 31 tested organisms cross-reacted. |
| None of 31 tested organisms interfered with toxin A/B detection (except Vancomycin at 50 mg/mL). | ||
| Interfering Substances | No negative impact on performance from specified substances at tested concentrations | None of 13 tested substances negatively impact performance at specified concentrations (except Vancomycin at 50 mg/mL). |
| Clinical Sensitivity | High sensitivity (benchmark not explicitly stated, but 90.6% is achieved) | 90.6% [95% CI: 83.1% - 95.0%] |
| Clinical Specificity | High specificity (benchmark not explicitly stated, but 95.7% is achieved) | 95.7% [95% CI: 94.2% - 96.8%] |
| Positive Predictive Value (PPV) | (Benchmark not explicitly stated) | 68.0% [95% CI: 59.5% - 75.4%] |
| Negative Predictive Value (NPV) | (Benchmark not explicitly stated) | 99.0% [95% CI: 98.1% - 99.5%] |
2. Sample Size Used for the Test Set and Data Provenance
- Clinical Study Test Set Sample Size: 1046 human stool specimens.
- Data Provenance: Prospective clinical study performed at three geographically diverse sites in the US. The samples were "left over de-identified, unpreserved, stool specimens from patients suspected of having C. difficile infection." This indicates a prospective collection for the purpose of the study, though using "left over" specimens.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
This information is not provided in the document. The ground truth for the clinical study was established by the Cellular Cytotoxicity Neutralization Assay (CCNA), which is a laboratory method, not human expert interpretation.
4. Adjudication Method for the Test Set
This information is not applicable as the ground truth was established by a laboratory assay (CCNA), not by human adjudication of clinical images or interpretations.
5. 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
This information is not applicable. The SensiTox C. difficile Toxin Test is an in vitro diagnostic device that directly detects toxins in stool samples using an immunofluorescence assay run on an automated analyzer. It is not an AI-assisted imaging device or a tool that directly assists human readers/interpreters in a diagnostic workflow where their performance would be measured with and without AI.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done
Yes, the performance data presented is for the device operating in a standalone capacity (algorithm only). The MultiPath Analyzer interprets the results using its applications software, reporting "toxin detected" or "toxin not detected" automatically. There is no human interpretative step described for the SensiTox C. difficile Toxin Test's output.
7. The Type of Ground Truth Used
- Bench Studies (LoD, Reproducibility, Analytical Reactivity, Analytical Specificity, Interfering Substances): The ground truth was established using known spiked samples (purified toxins, cultured organisms, interfering substances) in negative pooled stool.
- Clinical Performance Evaluation: The ground truth was established by the Cellular Cytotoxicity Neutralization Assay (CCNA). CCNA is a traditional laboratory method for detecting C. difficile toxins and is considered a gold standard for toxin activity.
8. The Sample Size for the Training Set
This information is not provided in the document. The document describes a "test set" for clinical performance evaluation but does not specify a separate "training set" with its sample size for the development of the device's analytical interpretation software. Given this is an in vitro diagnostic device, the "training" would likely involve optimizing the assay chemistry and the MultiPath Analyzer's ability to detect the fluorescent signals, rather than a machine learning model trained on a large dataset of patient samples.
9. How the Ground Truth for the Training Set Was Established
Since the document does not mention a distinct "training set" in the context of machine learning, the establishment of ground truth for any internal development or optimization of the device would implicitly involve the same methods used for the bench studies: controlled experiments with known concentrations of purified toxins and known negative samples, and potentially comparisons to established laboratory methods like CCNA during early development. The document focuses on the validation studies, not necessarily the development phase details.
§ 866.2660 Microorganism differentiation and identification device.
(a)
Identification. A microorganism differentiation and identification device is a device intended for medical purposes that consists of one or more components, such as differential culture media, biochemical reagents, and paper discs or paper strips impregnated with test reagents, that are usually contained in individual compartments and used to differentiate and identify selected microorganisms. The device aids in the diagnosis of disease.(b)
Classification. Class I (general controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to § 866.9.