(212 days)
The Xenex LightStrike™+ is a pulsed, broad-spectrum, high-intensity, germicidal UV light system intended to perform microbial reduction on non-porous, non-critical medical device surfaces, free from visual soiling, after manual cleaning and disinfection practices. LightStrike+ is intended for use in unoccupied operating rooms, hospital rooms, and other clinical settings where non-critical medical devices are present as an adjunct to existing manual cleaning and disinfection practices. The system is for over-the-counter (OTC) use.
LightStrike+ is a system that emits UV germicidal light to reduce the microbial load present on non-porous, non-critical medical device surfaces in healthcare environments, after manual cleaning and disinfection practices. Designed for microbial reduction cycles, the system will achieve the labeled log reductions through the utilization of Xenon lamp technology for the generation of UV germicidal light in the system's intended use environments. It is designed to be moved to targeted locations within a facility. The graphical user interface allows operators (trained cleaning staff) to select preprogrammed rooms and their associated protocols for cycle execution. The system is for use in unoccupied areas and includes design and redundant safety features to help prevent accidental exposure to the UV light.
The LightStrike+ is a whole room microbial reduction device that uses pulsed UV light to reduce microbial load on non-porous, non-critical medical device surfaces after manual cleaning and disinfection. The device is intended for use in unoccupied healthcare environments as an adjunct to existing cleaning practices.
Here's an analysis of the acceptance criteria and the study that proves the device meets them:
1. Table of Acceptance Criteria and Reported Device Performance
The device's performance is primarily assessed through its ability to achieve a sufficient microbial reduction and its safety. The acceptance criteria are largely derived from the "Special Controls" and the outcomes of the "Performance Testing - Bench" section.
| Acceptance Criteria (from Special Controls/Performance Testing) | Reported Device Performance |
|---|---|
| Microbial Log Reduction: Demonstrate microbial log reduction commensurate with the intended level, using the most resistant microorganism. | "The device has been appropriately evaluated for performance on the bench, demonstrating that is capable to achieve a microbial reduction of 2-log." |
| "The LightStrike+ device demonstrated 2-log average reduction in simulated use testing using worst-case microorganisms." | |
| Specific tests included: Optimized Kill Curve testing (determining dosage for 2-log reduction), Simulated-use tests (demonstrating 2-log reduction across inoculated surfaces with most UV resistant organism). | |
| Simulated Use Testing: Evaluate device performance under simulated worst-case use conditions (e.g., soiling, room objects and surfaces, distances). | "Simulated-use tests were provided, demonstrating microbial reduction on medical device surfaces, using the determined most UV resistant organism, achieving an average of 2-log reduction across all inoculated surfaces." (Implies worst-case conditions were incorporated in the "simulated use" methodology, as it's the primary test for confirming the 2-log reduction in a realistic setting). |
| In-use testing: Evaluate device performance under real-world use conditions. (iii) | Not explicitly stated as completed or presented. The document mentions "In-use testing must evaluate device performance under real-world use conditions," but the "Performance Testing - Bench" section primarily focuses on what appear to be laboratory-based simulations ("simulated-use tests"). The "Benefit/Risk Conclusion" specifically states: "The magnitude of the clinical benefit (i.e., transmission reduction) has not been established based on the non-clinical testing provided by the sponsor. Data was robust for 2-log average reduction on medical device surfaces, in the simulated environment." This strongly suggests a full real-world "in-use" study demonstrating human health outcomes was not part of this submission for determining clearance. This is a notable gap in reported data for this specific criterion. |
| Photobiological Safety: Demonstrate the photobiological safety of any lamps or lamp systems. | "Photobiological safety of lamps and lamp system test were provided, to evaluate the safety of the lamp that emits UV light when the robot is in use, conducted using a new lamp, because the light intensity is at its maximum when the lamp is new and that represents the worst-case scenario, per IEC 62471 'Photobiological Safety of Lamps and Lamp Systems.'" |
| "Photobiological safety of lamps and lamp system met IEC 62471 criteria." | |
| Safety Features: Evaluate safety features intended to prevent exposure and ensure that device operation can only occur in an unoccupied environment. | Device description details: "Tethered Motion Detection Cone" (detects motion, terminates cycle), "Remote Status/Stop Cone" (indicates in use, allows termination), "Door Placard" (warning sign). The system is designed to "automatically stop when the cycle time countdown expires." Operation only occurs "if the room is unoccupied." These design features fulfill the safety requirement. |
| Material Compatibility: Characterize the long-term material compatibility of the microbiocidal agent on clinically relevant surfaces and/or devices. | "Accelerated UV materials damage test were provided, to evaluate the impact of regular exposure to LightStrike+ with metallic and non-metallic materials commonly present in healthcare environments, including materials commonly used as medical device enclosures." |
| "Materials damage testing results met ISO 4582. ASTM D256-10. and ASTM A370-22 requirements." | |
| Biocompatibility/Chemical Byproducts: Demonstrate safe residual levels of chemicals on medical devices surfaces and/or gaseous byproducts in air. | "Testing was conducted to demonstrate that the operation of LightStrike+ generates Ozone at concentration levels that are below the maximum acceptable level of ozone in medical devices, according to 21 CFR 801.415." |
| "Additionally, test results demonstrate that the levels of toxic chemicals (other than ozone) generated during the use of LightStrike+... are either below the safe exposure limit or will dissipate to safe levels when a user enters the room after the completion of the cycle." | |
| "Specifically, the concentrations of Carbon monoxide (CO), Carbon dioxide (CO2), Nitric oxide (NO) and Nitrogen dioxide (NO2) are well below the safe exposure limits, per 29 CFR 1910.1000." | |
| "Ozone generation levels were below 21 CFR 801.415 acceptance levels of toxic chemicals were below 29 CFR 1910.1000 acceptance levels." | |
| Software Verification, Validation, and Hazard Analysis: for any software components. | "Adequate cybersecurity and software functionality were demonstrated according to FDA guidance 'Content of Premarket Submissions for Management of Cybersecurity in Medical Devices'." |
| "The effectiveness of the software security controls is determined through a threat model - risk assessment." "Xenex controls deployment of software updates as per change control and procedures within the Software Development Plan." | |
| EMC & Electrical Safety: Demonstrate electromagnetic compatibility (EMC) and electrical safety of the device. | "The LightStrike+ electrical safety is in conformance with IEC 60601-1-6. and the electromagnetic compatibility (EMC) is in accordance with IEC 60601-1-2." |
| Labeling: Include specific warnings, instructions, and material compatibility information. | "The labeling consists of a user manual, instructions for use, and packaging labels. The instructions for use include the indications for use, including the time to achieve a 2-log reduction; a description of the device, contraindications, warnings, precautions, including a door placard with caution signs advising 'DO NOT ENTER' to be affixed on entry ways of rooms/areas, warns of possible UV exposure during operation and warns of risk group 3 UV light generated by device; and instructions for the safe use of the device." "Labeling for this device is in accordance with the special controls listed below." |
2. Sample Size Used for the Test Set and Data Provenance
The document does not explicitly state a single dedicated "test set sample size" in the context of typical AI/ML model evaluation (i.e., a dataset held out for final performance assessment). The testing described is primarily bench-top, laboratory-simulated performance testing rather than a clinical study evaluating diagnostic accuracy or similar AI outcomes.
- Microbial Reduction Testing: The "simulated-use tests" involved "all inoculated surfaces" to achieve the 2-log reduction. The specific number of surfaces, microorganisms, and replicates is not provided.
- Safety Testing (Ozone, Chemicals, Photobiological): These involved specific test setups and measurements compliant with relevant standards (e.g., 21 CFR 801.415, 29 CFR 1910.1000, IEC 62471). The sample size here refers to the number of tests performed or measurements taken, which is not detailed.
- Material Compatibility: "metallic and non-metallic materials commonly present in healthcare environments" were tested. The exact number or types of materials is not specified.
Data Provenance:
- Given the nature of the tests (bench studies, simulated use, compliance with US and international standards), the data provenance is likely laboratory-generated data from Xenex's own testing facilities or accredited third-party labs.
- The data is retrospective in the sense that the studies were completed to support the De Novo application. There's no mention of prospective clinical trial data in human patients or real-world "in-use" data to assess patient outcomes or infection rates.
- Country of Origin: The sponsor is Xenex Disinfection Services, Inc., located in San Antonio, TX, USA. It's highly probable the testing was conducted in the USA or by labs commissioned by the US-based company.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications of Experts
This information is not applicable in the traditional sense for this device. Ground truth, in this context, is established by:
- Quantitative measurements: Log reduction achieved, chemical concentration levels, light intensity measurements, material degradation measurements.
- Compliance with established standards: IEC, ASTM, CFR regulations.
- Biological assay results: Live/dead microbial counts from inoculated surfaces.
The "experts" involved would be the scientists, microbiologists, engineers, and clinical staff who designed, executed, and interpreted these bench-top studies and ensured compliance with the relevant standards. Their specific number or qualifications are not detailed, but their expertise is implied by the successful completion and acceptance of the studies by the FDA. There is no human diagnostic interpretation involved that would require a panel of experts for ground truth establishment.
4. Adjudication Method for the Test Set
This is not applicable. Adjudication methods (like 2+1, 3+1 consensus) are typically used in clinical studies where human readers or multiple reviewers establish a ground truth for subjective interpretations (e.g., medical image diagnosis). For the LightStrike+, ground truth is based on objective, repeatable scientific measurements and compliance with quantitative thresholds defined by regulations and standards.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done
No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not performed.
- MRMC studies are relevant for evaluating the impact of AI on human performance in diagnostic tasks (e.g., radiologists reading images with or without AI assistance).
- The LightStrike+ is a disinfection device, not a diagnostic AI tool. Its effectiveness is measured by microbial reduction.
- The submission explicitly states: "The magnitude of the clinical benefit (i.e., transmission reduction) has not been established based on the non-clinical testing provided by the sponsor." This confirms the absence of studies evaluating its impact on human infection rates when used in a real-world setting.
Effect Size: Since no MRMC study was done, there is no effect size reported for human readers improving with AI vs. without AI assistance.
6. If a Standalone (i.e., Algorithm Only Without Human-in-the-Loop Performance) Was Done
Yes, in spirit. The performance tests described (e.g., "simulated-use tests demonstrating microbial reduction") represent the device's standalone performance in achieving microbial reduction. This is an "algorithm only" or "device only" performance in that it measures the physical effect of the UV light on microorganisms under controlled conditions, without human intervention in the disinfection process itself beyond setup and activation. The device is designed to operate autonomously in an unoccupied room.
7. The Type of Ground Truth Used
The ground truth for the LightStrike+ device's performance is established by:
- Quantitative Microbial Reduction: Direct measurement of viable microbial colonies before and after device operation on inoculated surfaces, yielding a calculated log reduction. This is a scientific and laboratory-based ground truth.
- Compliance with Regulatory Standards: For safety aspects (ozone, toxic chemicals, photobiological safety, electrical safety, EMC), the ground truth is defined by specific quantitative limits set forth in regulations (e.g., 21 CFR 801.415, 29 CFR 1910.1000) and international standards (e.g., IEC 62471, IEC 60601-1-6, IEC 60601-1-2).
- Material Integrity: Adherence to material degradation standards (ISO 4582, ASTM D256-10, ASTM A370-22).
This is primarily a laboratory-based, direct measurement, and standards-compliance ground truth, rather than clinical outcomes or expert consensus on subjective data.
8. The Sample Size for the Training Set
The concept of a "training set" in the context of artificial intelligence/machine learning is not directly applicable here. The LightStrike+ is a physical medical device that emits UV light, not an AI model that learns from a dataset.
- While it has "Software" components, these appear to be primarily for control, user interface, and safety features (e.g., "room guide," "cycle execution," "motion detection," "wireless communications") rather than data-driven learning or prediction for its primary function of microbial reduction.
- The "training" of the device is analogous to its engineering design and optimization to achieve the desired UV output and safety features.
Therefore, there is no "training set" in the common AI/ML sense.
9. How the Ground Truth for the Training Set Was Established
As there is no AI/ML training set, this question is not applicable. The "ground truth" for the device's design and optimization would be the physical and biological principles of UV light inactivation of microorganisms, safety standards, and engineering specifications. This is established through established scientific methods, engineering principles, and iterative design/testing, rather than data labeling for machine learning.
N/A