(610 days)
The Capnostream®20p combined capnograph/pulse oximeter monitor and its accessories are intended to provide professionally trained health care providers with continuous, non-invasive measurement and monitoring of carbon dioxide concentration of the expired and inspired breath and respiration rate, and with continuous non-invasive monitoring of functional oxygen saturation of arterial hemoqlobin (SpO2) and pulse rate. It is also indicated for continuous noninvasive monitoring of carboxyhemoglobin saturation (measured by an SpCO/SpMet/SpHb sensor), methemoglobin saturation (measured by an SpCO/SpMet/SpHb sensor) and total hemoglobin concentration (measured by an SpCO/SpMet/SpHb sensor). It is intended for use with neonatal, pediatric, and adult patients in hospital-type facilities, intra-hospital transport and home environments.
Capnostream®20p is to be operated by qualified healthcare personnel only. The Capnostream®20p monitor provides the clinician with an integrated pulmonary index (IPI). The IPI is based on four parameters provided by the monitor: end tidal carbon dioxide, respiration rate, oxygen saturation and pulse rate. The IPI is a single index of an adult or pediatric patient's ventilatory status displayed on a scale of 1 - 10, where 10 indicates optimal pulmonary status. IPI monitoring displays a single value that represents the patient's pulmonary parameters and alerts clinicians to changes in the patient's pulmonary status. The IPI is an adjunct to, and is not intended to replace, vital sign monitoring.
The Capnostream20p HiFi mode provides a measurement of expired carbon dioxide and detects spontaneous breaths for infant/neonatal patients ventilated using high frequency oscillatory ventilation (HFOV).
The Capnostream20p bedside monitor is a two parameter monitor consisting of a microMediCO2 capnography module and a pulse oximetry module implemented in a host device. The host device displays parameters received from the respective modules and generates alarms when preset alarm thresholds are crossed.
The HiFi capnography software feature, presented in this submission, is intended to enable measurement of airway CO2 for infant/neonatal patients during high frequency oscillatory ventilation (HFOV) and detects spontaneous breaths for infant/neonatal patients ventilated using high frequency oscillatory ventilation (HFOV).
Here's an analysis of the acceptance criteria and study information for the Capnostream® 20p with HiFi CO2 monitoring, based on the provided document:
1. Table of Acceptance Criteria and Reported Device Performance
| Parameter | Acceptance Criteria | Reported Device Performance |
|---|---|---|
| CO2 Measurement Accuracy (HiFi Mode) | Not explicitly stated in the provided text as a quantitative acceptance criterion. However, the implicit acceptance is that the device accurately measures CO2. | Demonstrated accuracy of CO2 measurements in HFOV mode up to 900 BPM (15 Hz). |
| CO2 Range (HiFi Mode) | Not explicitly stated as a quantitative acceptance criterion. | Tested range: 16 mmHg to 76 mmHg. |
| Spontaneous Respiratory Rate Detection (HiFi Mode) | Not explicitly stated as a quantitative acceptance criterion. | Tested at spontaneous respiratory rates of 30, 45, and 60 BPM. |
| Safety | Meets safety standards (IEC/EN 60601-1, IEC/EN 60601-1-2, IEC 60601-1-8, ISO 14971, EN 980). | Hazard analysis concluded residual risks were acceptable. No new potential safety risks identified. |
| Performance Standards | Meets ISO 21647, ISO 9919. | Meets performance standards. |
| Substantial Equivalence | To predicate device: Capnostream20p with Smart A/hr & ODI (K112368). | Functionally and in intended use, substantially equivalent. Minor differences raise no safety/effectiveness issues. |
2. Sample Sizes and Data Provenance for the Test Set
- Sample Size for Test Set: Not explicitly stated as a number of patients or cases. The testing was performed on a "lung simulator system."
- Data Provenance: The testing was conducted on a "lung simulator system," indicating it was not performed on human subjects. This suggests a controlled laboratory environment rather than retrospective or prospective clinical data from a specific country of origin.
3. Number of Experts and Qualifications for Ground Truth Establishment (Test Set)
- Not applicable as the ground truth was established through a "lung simulator system" rather than expert interpretation of biological data. The "ground truth" would be the known, controlled CO2 and respiratory rate inputs of the simulator.
4. Adjudication Method for the Test Set
- Not applicable. Since the testing was done on a lung simulator, comparison would have been to the known, programmed values of the simulator, not requiring expert adjudication.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- No, an MRMC comparative effectiveness study was not done. This device is a measurement instrument, not an AI-assisted diagnostic tool that would involve human readers interpreting output. The study focused on the accuracy of the device's measurements.
6. Standalone Performance Study
- Yes, a standalone performance study was done. The performance testing on the lung simulator system assessed the algorithm's (Capnostream20p with HiFi) ability to accurately measure CO2 and detect breaths independently of human interpretation.
7. Type of Ground Truth Used (Test Set)
- The ground truth used was the known, controlled parameters established by the "lung simulator system." This would include the programmed CO2 concentrations and respiratory rates (both passive and spontaneous).
8. Sample Size for the Training Set
- The document does not explicitly state a training set sample size. Given the nature of a capnograph/pulse oximeter, its core algorithms are typically based on physiological models and physical principles (e.g., infrared absorption for CO2). While there might be internal tuning or calibration processes that could be considered "training" in a broad sense, it's not described as a machine learning model with a distinct training dataset in the same way as an image recognition AI. The focus is on the validation of the HiFi software feature.
9. How the Ground Truth for the Training Set was Established
- Not explicitly described in the document. As mentioned above, for this type of device, "training" might not involve a distinct ground truth dataset in the way a modern AI algorithm would. Its development would likely rely on physical laws, engineering specifications, and possibly internal testing to refine signal processing and detection algorithms within the device.
§ 868.1400 Carbon dioxide gas analyzer.
(a)
Identification. A carbon dioxide gas analyzer is a device intended to measure the concentration of carbon dioxide in a gas mixture to aid in determining the patient's ventilatory, circulatory, and metabolic status. The device may use techniques such as chemical titration, absorption of infrared radiation, gas chromatography, or mass spectrometry.(b)
Classification. Class II (performance standards).