(67 days)
THE BACT/ALERT 3D MICROBIAL DETECTION SESTEM WITH EXPRECTION OR MICORACTERIA IN IS USED IN QUALITATIVE PROCEDURES FOR GROWTH AND DETECTION OF MICOBACTERIA IN CLINICAL SPECIMENS OTHER THAN BLOOD.
BacT/Alert 3D System with expanded Mycobacteria indication is used in the qualitative procedures for growth and detection of Mycobacteria in clinical specimens other than blood.
Here's an analysis of the provided text regarding the acceptance criteria and the study that proves the device meets them:
Acceptance Criteria and Device Performance for BacT/Alert 3D with Mycobacteria Indication
Based on the provided 510(k) summary (K981736) for the BacT/Alert 3D System with expanded Mycobacteria Indication, the device aims to demonstrate substantial equivalence to the legally marketed MB/BacT Mycobacteria Culture System. The acceptance criteria are implicitly defined by the demonstration of comparable performance to the predicate device in several key aspects.
1. Table of Acceptance Criteria and Reported Device Performance
The core acceptance criterion is "substantial equivalence" to the MB/BacT Mycobacteria Culture System. This is evaluated through a comparison of performance characteristics.
| Acceptance Criterion | Reported Device Performance (BacT/Alert 3D) |
|---|---|
| Intended Use | Growth and detection of Mycobacteria in Clinical Specimens other than blood. (Identical to predicate) |
| Media | Utilizes MB/BacT Process bottles (K954468), Culture Media with suitable nutritional conditions to recover mycobacterial species. Containing Middlebrook 7H9 Broth, Pancreatic Digest of Casein, Bovine Serum Albumin, Catalase, Purified Water. (Identical to predicate) |
| Bottle Blocks | Sealed for spill control. (Identical to predicate) |
| Air Circulation | Shuts off when door is open. (Functionally equivalent to predicate's "shuts off when drawer is open") |
| Determination for | High enough amount of acceleration of CO2 production, High enough amount of long-term growth, and high enough total change (delta) in the growth curve. (Identical detection algorithms as predicate) |
| Clinical Performance | 22 isolates were positive on both the BacT/Alert 3D and the MB/BacT. |
| 14 isolates were positive on BacT/Alert 3D only. | |
| 8 isolates were positive on MB/BacT only. (The BacT/Alert 3D yielded test results comparable to that seen with the predicate device.) | |
| Non-clinical Performance (Time to Detection) | Analysis of times to detection showed no significant effect of detection system for either high (10^6 CFU/ml) or low (10^2 CFU/ml) inoculum levels when compared to the MB/BacT system. |
2. Sample Size Used for the Test Set and Data Provenance
-
Non-clinical Seeded Study:
- Sample Size: 10 organisms were tested. For low inoculum (10^2 CFU/ml), 6 bottles per organism were used (3 for current MB/BacT, 3 for BacT 3D), totaling 60 bottles. For high inoculum (10^6 CFU/ml), 2 bottles per organism were used (1 for current MB/BacT, 1 for BacT 3D), totaling 20 bottles.
- Data Provenance: In-house (Organon Teknika Corporation). Prospective.
-
Clinical Study:
- Sample Size: A total of 311 isolates were cultured from clinical specimens.
- Data Provenance: Clinical specimens. Retrospective or prospective is not explicitly stated, but "cultured from clinical specimens" generally implies real-world, likely retrospective collection, although a prospective collection of new specimens is also possible. The country of origin is not specified.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications
The document does not specify the number of experts or their qualifications used to establish ground truth for either the non-clinical or clinical studies.
4. Adjudication Method for the Test Set
The document does not specify any adjudication method (e.g., 2+1, 3+1, none) for the test set. For the clinical study, the results appear to be a direct comparison of positivity between the two systems, implying that a definitive "ground truth" might have been established prior to culturing, or that the systems were being compared against each other as equivalent methods.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done
No. This device is an automated microbial detection system, not a diagnostic imaging device typically evaluated with MRMC studies. The "readers" are the instruments themselves performing automated detection.
6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done
Yes. Both the non-clinical seeded study and the clinical study evaluated the performance of the BacT/Alert 3D system as a standalone algorithm/instrument, comparing its detection capabilities directly against the predicate device. There is no mention of human-in-the-loop performance in the context of aiding human interpretations.
7. The Type of Ground Truth Used
- Non-clinical Seeded Study: The ground truth was based on the known inoculation of specific organisms at defined concentrations (10^6 and 10^2 CFU/ml). The "truth" of an organism being present and viable was established by the experimental design.
- Clinical Study: The ground truth appears to be based on the comparative positivity between the BacT/Alert 3D and the MB/BacT Mycobacteria Culture System. While ideal ground truth for mycobacterial infection might come from further identification methods or clinical diagnosis, the study focuses on the agreement and discrepancies between the two culture systems in detecting something positive for mycobacteria. The document states "22 isolates were positive on both... 14 isolates were found to be positive on the BacT/Alert 3D only. 8 additional isolates were found to be positive on the MB/BacT only." This suggests the ground truth was the presence of a positive culture result according to each system's established criteria.
8. The Sample Size for the Training Set
The document does not provide information regarding a separate training set. Given the 1998 submission date and the nature of the device (an automated microbial growth monitor, not a complex AI model requiring extensive neural network training in the modern sense), it is highly probable that the "algorithm" refers to pre-programmed detection logic based on CO2 production, growth curves, and delta changes, which would typically be developed and validated rather than "trained" on a large dataset in the way contemporary machine learning models are. The seeded and clinical studies serve as verification and validation (V&V) of the pre-established algorithms.
9. How the Ground Truth for the Training Set Was Established
As no separate "training set" is described for algorithm development in the provided text, this question is not applicable. The algorithms for positivity determination (CO2 acceleration, long-term growth, total change in growth curve) are stated to be the same as the predicate device, implying they were already established and validated by the predicate device's prior regulatory submissions and development.
§ 866.2560 Microbial growth monitor.
(a)
Identification. A microbial growth monitor is a device intended for medical purposes that measures the concentration of bacteria suspended in a liquid medium by measuring changes in light scattering properties, optical density, electrical impedance, or by making direct bacterial counts. The device aids in the diagnosis of disease caused by pathogenic microorganisms.(b)
Classification. Class I. With the exception of automated blood culturing system devices that are used in testing for bacteria, fungi, and other microorganisms in blood and other normally sterile body fluids, this device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter.