(50 days)
The Sensititre Haemophilus/Streptococcus pneumoniae (HP) MIC Susceptibility plate is an in vitro diagnostic device for the quantitative determination of susceptibility of Streptococcus pneumoniae and Haemophilus influenzae. This 510(k) is for the addition of Linezolid in the dilution range of 0.25 - 32 µgml to the Sensititre Haemophilus/Streptococcus pneumoniae MIC panel for testing Streptococcus pneumoniae isolates. The approved primary "Indications for Use" for the Sensititre Haemophilus/Streptococcus pneumoniae MIC panel for testing Streptococcus pneumoniae and Haemophilus influenzae isolates remains unchanged. Clinical correlation is provided for: Streptococcus pneumoniae (penicillin-susceptible strains) and Streptococcus pneumoniae (penicillin-resistant strains).
Not Found
This is a 510(k) premarket notification for the addition of Linezolid to the Sensititre Haemophilus/Streptococcus pneumoniae (HP) MIC Susceptibility Plates. The document describes the device, its indications for use, and a summary of the performance data to support substantial equivalence.
Here's an analysis of the provided information, focusing on the acceptance criteria and the study that proves the device meets them:
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria for antimicrobial susceptibility testing (AST) devices, particularly for broth microdilution methods like the Sensititre plates, are typically based on agreement with a reference method (e.g., NCCLS (now CLSI) broth microdilution). The key metrics are:
- Essential Agreement (EA): The MIC value obtained with the device is within +/- 1 twofold dilution of the reference method.
- Categorical Agreement (CA): The interpretation of susceptible, intermediate, or resistant (SIR call) from the device matches the reference method.
- Major Discrepancies (MD): The device calls susceptible, but the reference calls resistant. (Considered a serious error).
- Very Major Discrepancies (VMD): The device calls resistant, but the reference calls susceptible. (Considered a very serious error).
The document states, for "Linezolid vs. Streptococcus pneumoniae" and "Linezolid vs. Haemophilus influenzae" that the "overall Essential Agreement (EA) values were >90%" and "Categorical Agreement (CA) values were >90%". It also provides specific rates for VMD and MD.
Here's the table of acceptance criteria and reported device performance based on the executive summary:
| Metric | Acceptance Criteria (Typically) | Reported Device Performance (Linezolid vs. S. pneumoniae) | Reported Device Performance (Linezolid vs. H. influenzae) |
|---|---|---|---|
| Essential Agreement (EA) | ≥ 90% | 98.4% | 99.1% |
| Categorical Agreement (CA) | ≥ 90% | 98.3% | 99.1% |
| Very Major Discrepancy (VMD) | ≤ 1.5% | 0.0% | 0.0% |
| Major Discrepancy (MD) | ≤ 3.0% | 1.7% | 0.9% |
2. Sample Size Used for the Test Set and Data Provenance
- Test Set Sample Size:
- Streptococcus pneumoniae: 298 isolates (including 116 challenge isolates)
- Haemophilus influenzae: 200 isolates (including 50 challenge isolates)
- Data Provenance: The document does not explicitly state the country of origin. However, the study involved both "reference organisms" and "clinical isolates." The clinical isolates are typically collected from diverse patient populations within the context of clinical microbiology laboratories, but a specific country is not mentioned. The study is retrospective, as it used pre-existing isolates (clinical and challenge) to test the new drug-device combination.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
The document does not specify the number of experts or their qualifications for establishing the ground truth. However, the ground truth for AST is typically established by performing the NCCLS (now CLSI) reference broth microdilution method. This method is standardized and performed by trained microbiologists following strict protocols. The "expert" in this context is the consensus interpretation of the reference method's results.
4. Adjudication Method for the Test Set
The document does not describe an "adjudication method" in the sense of multiple human experts reviewing conflicting interpretations. For AST studies, discrepancies (Major or Very Major) are typically investigated to determine the root cause, which might involve retesting or further analysis using other methods if the reference method itself is believed to be inaccurate. The primary adjudication is essentially the comparison to the established NCCLS reference method.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done
No, an MRMC comparative effectiveness study was not done. This type of study is more common for diagnostic imaging or subjective interpretation tasks where human readers' performance is directly measured and compared with and without AI assistance. For AST devices, the primary goal is typically to show accurate and reproducible results compared to a standardized reference method, not to evaluate human reader improvement.
6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done
Yes, a standalone performance study was done. The Sensititre system, in this context, is an automated or semi-automated system that reads and interprets the MIC values. The performance statistics (EA, CA, VMD, MD) directly reflect the device's ability to accurately determine MICs and SIR calls without human intervention in the interpretative step (beyond setting up the test and potentially validating the results). The "device performance" reported is the algorithm's performance compared to the reference method.
7. The Type of Ground Truth Used
The type of ground truth used is the NCCLS (now CLSI) reference broth microdilution method. This is a well-established, standardized, and internationally accepted method for determining antimicrobial minimum inhibitory concentrations (MICs). It serves as the gold standard for comparing new AST devices.
8. The Sample Size for the Training Set
The document does not explicitly state a separate training set sample size. For an AST device like the Sensititre plates, the "training" (or development and calibration) of the system for a new antimicrobial like Linezolid often involves a different approach than typical machine learning models. The system's underlying methodology for reading MICs from growth patterns and applying interpretive breakpoints is generally pre-established. The introduction of a new drug primarily requires:
- Validation that the specific drug lot performs correctly on the plate.
- Verification that the established interpretive breakpoints (e.g., from NCCLS/CLSI) are correctly applied by the system.
- Confirmation of agreement with the reference method across a range of concentrations and diverse isolates.
While there might have been internal development data used to optimize the plate formulation or reading algorithms for Linezolid, it's not typically referred to as a "training set" in the context of this type of 510(k) submission, which focuses on validation data against a reference standard.
9. How the Ground Truth for the Training Set Was Established
Given that a distinct "training set" is not explicitly mentioned as per common AI/ML terminology in this context, the ground truth for any internal development or calibration would also have been established based on:
- NCCLS (CLSI) reference broth microdilution method: To ensure the system correctly measures MICs of Linezolid against relevant organisms.
- Known interpretive breakpoints: These are established by independent expert bodies (like CLSI) based on clinical outcomes, pharmacokinetic/pharmacodynamic data, and surveillance data.
In essence, the "ground truth" used throughout the entire development and validation process for AST devices revolves around agreement with the established reference method and adherence to recognized interpretive guidelines.
§ 866.1640 Antimicrobial susceptibility test powder.
(a)
Identification. An antimicrobial susceptibility test powder is a device that consists of an antimicrobial drug powder packaged in vials in specified amounts and intended for use in clinical laboratories for determining in vitro susceptibility of bacterial pathogens to these therapeutic agents. Test results are used to determine the antimicrobial agent of choice in the treatment of bacterial diseases.(b)
Classification. Class II (performance standards).