(35 days)
Antimicrobial Susceptibility Test Discs are for the semi-quantitative susceptibility testing by agar diffusion test procedure of rapidly growing micro-organisms. For Daptomycin these include Staphylococcus aureus (including MRSA), Streptococcus agalactiae, Streptococcus dysgalactiae subsp. equisimilis, Streptococcus pneumoniae, and Enterococcus faecalis (vancomycin-susceptible strains only).
Not Found
Here's an analysis of the provided text regarding the Oxoid Daptomycin Susceptibility Test Disc, focusing on the requested criteria for device acceptance and its supporting study:
The provided document is a 510(k) clearance letter from the FDA for the Oxoid Daptomycin Susceptibility Test Disc. This type of document declares "substantial equivalence" to a legally marketed predicate device, rather than a full de novo approval based on extensive clinical trials for a novel device. Therefore, the information typically available for novel medical devices regarding detailed acceptance criteria, MRMC studies, standalone performance, and ground truth establishment in a study report is not present in this 510(k) clearance letter.
However, based on the nature of antimicrobial susceptibility testing devices, we can infer some general principles and what would typically be expected, even if the specific data isn't in this document.
Inference Based on Standard AST Device Clearance
For Antimicrobial Susceptibility Test (AST) Discs, FDA clearance relies heavily on demonstrating performance (e.g., accuracy against a reference method) for identifying susceptible, intermediate, and resistant categories for specific microorganisms. The "study" here is typically a comparison to a recognized reference method for antimicrobial susceptibility.
1. Table of Acceptance Criteria and Reported Device Performance
Since this is a 510(k) premarket notification for an Antimicrobial Susceptibility Test (AST) Disc, the acceptance criteria and performance are typically related to the accuracy of categorizing isolates as Susceptible (S), Intermediate (I), or Resistant (R) when compared to a gold standard reference method (e.g., broth microdilution or agar dilution).
Inferred Table based on typical AST device clearance for a qualitative/semi-quantitative test:
| Performance Metric (Agreement with Reference Method) | Acceptance Criteria (Typical for FDA AST) | Reported Device Performance (Not explicitly stated in this document but would be in the full submission) |
|---|---|---|
| Essential Agreement (EA) | ≥ 90% (often ≥ 95% desired) | (Likely met, otherwise 510(k) would not be cleared) |
| Categorical Agreement (CA) | ≥ 90% (often ≥ 95% desired) | (Likely met, otherwise 510(k) would not be cleared) |
| Major Discrepancy (MD) | ≤ 3% (False Resistance, S incorrectly I/R) | (Likely ≤ 3%) |
| Very Major Discrepancy (VMD) | ≤ 1.5% (False Susceptibility, R incorrectly S) | (Likely ≤ 1.5%) |
| Minor Discrepancy (mD) | (No specific limit, but monitored) | (Monitored) |
Explanation:
- Essential Agreement (EA): Agreement within ±1 doubling dilution of the reference method's MIC value; often reflects the zone diameter correlation.
- Categorical Agreement (CA): Agreement in the final S, I, or R category assigned.
- Major Discrepancy (MD): The test device calls an isolate susceptible (S) or intermediate (I) while the reference method calls it resistant (R). (Incorrectly treats a resistant organism).
- Very Major Discrepancy (VMD): The test device calls an isolate resistant (R) while the reference method calls it susceptible (S) or intermediate (I). (Incorrectly treats a susceptible organism, leading to potential treatment failure).
- Minor Discrepancy (mD): One method calls an isolate susceptible (S) or resistant (R) while the other calls it intermediate (I), or vice-versa. (Less critical than MD or VMD but still tracked).
Note: The specific numerical acceptance criteria can vary slightly depending on the drug and organism, but the percentages listed above are broadly representative for AST devices.
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size for Test Set: Not specified in this document. For AST devices, typical studies involve testing a significant number of isolates (e.g., hundreds) for each organism/drug combination, encompassing various resistance mechanisms if relevant. This includes a balanced representation of S, I, and R categories.
- Data Provenance: Not specified in this document. AST studies are often conducted at multiple clinical microbiology laboratories to ensure generalizability. They are typically prospective or use a banked collection of retrospective isolates specifically chosen to represent clinical diversity and resistance profiles relevant to the drug. The country of origin would be global for a company like Oxoid, but specific study sites are not mentioned here.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Their Qualifications
- Number of Experts: Not applicable in the context of AST disc studies in the way it applies to image-based AI. The "ground truth" for AST is not established by expert visual interpretation/consensus.
- Qualifications of Experts: The "experts" in AST are typically trained laboratory personnel following standardized reference testing methodologies (e.g., CLSI or EUCAST guidelines). These individuals perform the reference broth microdilution or agar dilution tests, which serve as the gold standard.
4. Adjudication Method for the Test Set
- Adjudication Method: Not applicable. The "ground truth" in AST is a quantitative measurement (Minimal Inhibitory Concentration - MIC) obtained via a reference method, which then translates into categorical (S, I, R) results. Discrepancies between the investigational device and the reference method are analyzed numerically, not by expert adjudication of conflicting interpretations. Discrepant results, however, are often re-tested by both methods to confirm the initial findings.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done
- MRMC Study: No, highly unlikely. MRMC studies are specific to diagnostic tools where human interpretation of complex images or data is central, and the AI's role is to assist or augment that interpretation. AST discs are a semi-quantitative laboratory test and do not involve human "readers" interpreting images in a way that an MRMC study would be relevant. The performance is assessed against an objective reference method, not against human variability in interpreting the disc zone.
6. If a Standalone (Algorithm Only Without Human-in-the-Loop) Performance Was Done
- Standalone Performance: Yes, in essence. The entire premise of an AST disc is a standalone performance test. A technician prepares the plate, applies the disc, incubates, and then measures the zone of inhibition. This measurement is then interpreted against predefined breakpoints. The "device performance" relies solely on the physical disc's ability to diffuse the antimicrobial and inhibit bacterial growth, leading to a measurable zone that correlates with susceptibility. There isn't a separate "algorithm" in the AI sense, but rather a set of established breakpoints and measurement standards.
7. The Type of Ground Truth Used
- Ground Truth Type: For AST devices, the ground truth is established by a standardized reference method (e.g., broth microdilution, agar dilution, or a validated automated system) that quantitatively determines the Minimal Inhibitory Concentration (MIC) of the antimicrobial against the specific bacterial isolate. This MIC value is then categorized into Susceptible (S), Intermediate (I), or Resistant (R) based on established clinical breakpoints (e.g., CLSI M100 document). This is a form of objective laboratory measurement rather than subjective expert consensus or pathology.
8. The Sample Size for the Training Set
- Sample Size for Training Set: Not applicable in the typical AI sense. AST discs are not "trained" like machine learning algorithms. Their performance is inherent in the chemical composition of the disc and the diffusion properties of the antimicrobial. Development involves optimizing the drug concentration in the disc and correlating zone diameters with MIC values, which could involve empirical testing on many isolates, but it's not a "training set" for an algorithm.
9. How the Ground Truth for the Training Set Was Established
- Ground Truth for Training Set Establishment: Not applicable as described for AI. The "ground truth" for developing the correlation between disc zone diameters and MIC values (which guides optimal disc concentration and breakpoint setting) is established through extensive studies where MICs are determined by a reference method (e.g., broth microdilution) for a wide range of bacterial isolates, and simultaneously, the corresponding zone diameters are measured for the investigational disc. This allows for regression analysis and determination of appropriate zone diameter breakpoints that correlate with the clinical MIC breakpoints.
§ 866.1620 Antimicrobial susceptibility test disc.
(a)
Identification. An antimicrobial susceptibility test disc is a device that consists of antimicrobic-impregnated paper discs used to measure by a disc-agar diffusion technique or a disc-broth elution technique the in vitro susceptibility of most clinically important bacterial pathogens to antimicrobial agents. In the disc-agar diffusion technique, bacterial susceptibility is ascertained by directly measuring the magnitude of a zone of bacterial inhibition around the disc on an agar surface. The disc-broth elution technique is associated with an automated rapid susceptibility test system and employs a fluid medium in which susceptibility is ascertained by photometrically measuring changes in bacterial growth resulting when antimicrobial material is eluted from the disc into the fluid medium. Test results are used to determine the antimicrobial agent of choice in the treatment of bacterial diseases.(b)
Classification. Class II (performance standards).