(76 days)
Long-term ventilator for intensive care. For adults, children, and neonates with a minimum body weight of 3 kg (6.6 lbs). For premature infants with a minimum body weight of 0.5 kg (1.1 lbs) with the NeoFlow option.
Evita XL is a time-cycled microprocessor-controlled intensive care ventilator intended to provide continuous ventilation for adults, children, neonates and infants with Neoflow option. The device can be also operated in the mask ventilation mode if the according option is installed. Software modification implements these new software options:
- . Option Mask Ventilation Plus (NIV Plus)
- Option Inspiratory Termination Criteria (Insp. Term. PIF) .
- . Option Proportional Pressure Support (PPS)
NIV Plus is a supplement to the existing Mask ventilation option (NIV), which improves ventilation performance (triggering and patient comfort) through higher leakage compensation and smooth ending of Assistant Spontaneous Breathing (ASB) strokes. NIV Plus implements Standby Plus an alternative for starting the active ventilation mode in mask ventilation when the patient takes the first spontaneous breath. Anti Air Shower detects disconnection during mask ventilation if the mask is deliberately removed from the face and reduces the flow supplied for the time of the disconnection.
Insp. Term. PIF is an adjustable stop criterion for pressure support strokes. At the end of the inspiratory phase the delivered inspiratory flow falls under a certain level of the maximum inspiratory flow. The ventilator cycles from inspiration phase to expiration phase when a fixed level was reached. The software modification allows it to adjust this inspiration termination criterion. The device can be configured by the user (adjustable inspiration termination criterion or fixed value is active).
The ventilation mode Proportional Pressure Support (PPS) is a software option for the ventilator Evita XL designed to assist the spontaneous breathing patient. In PPS, the device supports the patient's spontaneous breathing in proportion to the breathing effort. The degree of support in PPS mode can be set separately according to resistive and elastic components directly proportional to the patient effort.
Here's an analysis of the provided information, focusing on acceptance criteria and a study proving device performance, as requested.
Based on the provided 510(k) summary, the device is the Draeger Evita XL continuous ventilator with new software options. This document is a premarket notification for a medical device rather than a detailed study report. Therefore, it does not contain the typical structure of a clinical study with explicitly defined acceptance criteria and performance data in the format you've requested.
Instead, the document focuses on demonstrating substantial equivalence to predicate devices. This regulatory pathway typically relies on engineering data, performance testing (often bench testing), and risk analysis to show that the new device options are as safe and effective as existing legally marketed devices, rather than a large-scale clinical trial with specific performance metrics like accuracy, sensitivity, or specificity against a ground truth in a clinical setting.
Therefore, many of the specific questions you asked about sample sizes for test/training sets, expert qualifications, adjudication methods, MRMC studies, and standalone performance metrics are not addressed or present in this 510(k) summary.
However, I can extract information related to the device's intended performance and the general nature of the validation conducted:
1. A table of acceptance criteria and the reported device performance
As mentioned, the 510(k) summary does not present acceptance criteria and device performance in the typical format of a clinical study (e.g., sensitivity, specificity, AUC). Instead, it states that the improvements are related to "ventilation performance (triggering and patient comfort) through higher leakage compensation and smooth ending of Assistant Spontaneous Breathing (ASB) strokes" and enabling adjustable inspiration termination. The 'acceptance' in this context is based on demonstrating that these software changes function as intended and do not introduce new safety concerns, while maintaining or improving existing performance.
| Criterion Type (Inferred) | Stated Performance / Outcome |
|---|---|
| NIV Plus Option | Improves ventilation performance (triggering and patient comfort) through higher leakage compensation and smooth ending of ASB strokes. Implements Standby Plus (alternative for starting active ventilation in mask ventilation). Detects and reduces flow during disconnection (Anti Air Shower). |
| Insp. Term. PIF Option | Allows user to adjust inspiration termination criterion (previously fixed level): provides an adjustable stop criterion for pressure support strokes when inspiratory flow falls under a certain level of maximum inspiratory flow. |
| PPS Option | Assists spontaneous breathing proportionally to breathing effort; degree of support can be set separately according to resistive and elastic components. |
| Overall (Safety & Effectiveness) | Based on risk analysis and software validation measures, the new software options are "safe and effective and are substantially equivalent to the legally marketed predicate devices." |
2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)
- Not explicitly stated in the 510(k) summary. The document refers to "performed software validation measures" and "risk analysis," which would typically involve bench testing and simulations rather than clinical data sets. Therefore, there's no mention of a "test set" in the context of clinical data with human subjects or retrospective/prospective data collection from a specific country of origin.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience)
- Not applicable / Not explicitly stated. Since there is no mention of a clinical test set requiring human expert-established ground truth, this information is not provided. The validation would have involved engineering experts and possibly medical professionals advising on functional requirements and safety.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set
- Not applicable. As no clinical test set requiring expert adjudication is described, this information is absent.
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
- No MRMC study was mentioned or implied. The Evita XL is a ventilator, not an AI-assisted diagnostic imaging device with "human readers." The software options enhance the ventilator's function directly.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- The device is the algorithm/software operating the ventilator. The performance evaluation would be of the ventilator's output and responsiveness as controlled by the new software features. This isn't a "standalone AI" in the sense of a decision support system, but rather the core functionality of the device. The phrase "standalone" might not apply as neatly here, as the software is fundamental to the device's (ventilator's) operation. The validation would have assessed the ventilator's performance with the new software features, ensuring they meet specifications (e.g., accurate flow delivery, appropriate termination).
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
- For software-driven medical devices like ventilators, the "ground truth" for validation is typically defined by engineering specifications, physiological models, and recognized standards for ventilator performance (e.g., accuracy of delivered pressure/volume/flow, response time to patient effort, leakage compensation effectiveness). It's based on objective measurements against these predefined targets, often conducted in a lab or simulation environment. There would likely be no "pathology" or "outcomes data" ground truth in the clinical trial sense for this type of submission.
8. The sample size for the training set
- Not applicable / Not explicitly stated. This device is not described as involving machine learning or AI that requires a "training set" of data in the common sense (e.g., for image classification). The software development would follow traditional software engineering principles, with design, implementation, and verification/validation processes.
9. How the ground truth for the training set was established
- Not applicable. See point 8.
In summary: The provided 510(k) summary for the Draeger Evita XL ventilator focuses on establishing substantial equivalence for new software features. It indicates that "software validation measures" and "risk analysis" were performed, leading to the conclusion that the new options are safe, effective, and substantially equivalent to predicate devices. However, it does not present detailed clinical study data with specific acceptance criteria, sample sizes, expert involvement, or statistical analyses typical of AI/ML device submissions that rely on extensive clinical datasets.
§ 868.5895 Continuous ventilator.
(a)
Identification. A continuous ventilator (respirator) is a device intended to mechanically control or assist patient breathing by delivering a predetermined percentage of oxygen in the breathing gas. Adult, pediatric, and neonatal ventilators are included in this generic type of device.(b)
Classification. Class II (performance standards).