(701 days)
The BAR Monitor is intended to monitor the state of the brain by real-time data acquisition and processing of EEG signals. The system displays a patient's EEG and Composite Cortical State (CCS), a proprietary computed EEG index related to the effect of certain anesthetic or hypnotic agents include inhalation agents and propofol in combination with opioids.
The BAR Monitor is intended to assist medical professionals monitor adult patients (22 to 65 years of age) in the operating room (OR) and clinical research laboratory.
The Brain Anaesthesia Response Monitor (BARM) is a device designed to non-invasively monitor brain function in response to anesthetic agents including inhalation agents in the operating room (OR) and clinical research laboratory. The system consists of the following three components:
- . Brain Anaesthesia Response (BAR) Terminal - Provides a user interface to control the DAM, set configuration parameters and display the DAM output for users.
- . Data Acquisition Module (DAM) - Collects and processes the signal sent by the electrodes and calculates the Composite Cortical State (CCS) index and outputs the information to the BAR Terminal.
- Disposable cutaneous electrode sensor- The electrodes collect Brain electrical activity (EEG) signals and transfers them to the DAM. The electrodes are made from pre-gelled Ag-AgCl. The electrodes are connected with patient leads which is plugged into the DAM. The electrodes are applied to the patients' forehead and behind the ear and is held in place via biocompatible adhesives.
Here's a breakdown of the acceptance criteria and the studies proving the device meets those criteria, based on the provided text:
Acceptance Criteria and Reported Device Performance
The document doesn't explicitly list "acceptance criteria" in a formal table with pass/fail metrics. Instead, it demonstrates substantial equivalence to a predicate device (BIS EEG VISTA MONITOR SYSTEM) through a comparison of technological characteristics and then presents clinical performance data to support the device's intended use and show its comparable performance to existing technologies.
Here's a table synthesizing the implied acceptance criteria (based on the predicate device's features and the subject device's performance claims) and the reported device performance.
| Acceptance Criteria (Implied from Predicate/Claims) | Reported Device Performance (BAR Monitor) |
|---|---|
| Safety and Electrical Standards Compliance | - IEC 62133 (Battery Safety) met |
| - IEC 60601-1 (Electrical and Environmental Safety) met | |
| - IEC 60601-2-26 (Electrical Safety and Essential Performance) met | |
| - IEC 60601-1-2 (EMC Testing) met | |
| - ISO 14971 (Risk Assessment) met | |
| - FDA Software Guidance (Software V&V) met | |
| - IEC 60601-1-6 & IEC 62366-1 (Usability Evaluation) met | |
| - ISO 10993-1 (Biocompatibility) met | |
| EEG Index Related to Anesthetic/Hypnotic Agents | CCS (Composite Cortical State) index, 0-100 range, related to effect of certain anesthetic or hypnotic agents. |
| Accuracy in Predicting Loss of Response (Anesthetic Induction) | U111-1124-2523 Study: CCS equivalently predictive of loss of response (OAA/S 0) as BIS™ index. Prediction Probability (Pr) 0.9 for both devices. |
| Accuracy in Monitoring Emergence from Anesthesia | ACTRN12618000916246 Study: CCS observed to increase across signs of emergence with comparable values to BIS™. Prediction probabilities calculated across emergence endpoints for CCS were equivalent to or better than that for the BIS. |
| Correlation with OAA/S Levels (Anesthetic Depth) | 6001-004 Study: CCS observed to decrease with decreasing OAA/S level. Findings support a recommended range of CCS > 80 for awake and CCS between 40-60 for unresponsive. Prediction probabilities across all OAA/S levels were 0.85, 0.84, 0.78, and 0.83 (for 0, 2, 4 ng/ml remifentanil and all combined, respectively), equivalent to GE M-Entropy™. Prediction probabilities between OAA/S 5 and OAA/S 0 were ≥0.99. |
| Predictive of Response to Vocal Stimulus | 6001-004 Study: P ≥ 0.86. |
| Comparable EEG Processing & Display (e.g., Amplitude, Chart Speed, Analog Noise) | - EEG Waveforms Amplitude: 25 µV/division (default), with additional options. Equivalent to predicate. |
| - EEG Chart Speed: 32mm/sec (default), with additional options. Equivalent to predicate. | |
| - Analog Noise: 108 dB. Equivalent to predicate. | |
| - Bandwidth: 0.5 - 200 Hz. Equivalent to predicate. | |
| - Unprocessed EEG available: Yes. Equivalent to predicate. | |
| - Display of EEG time series: Yes. Equivalent to predicate. | |
| Appropriate Device Hardware & Environmental Specifications | (Various specifications listed in the comparison table, generally found "Equivalent" to predicate, e.g., power, battery runtime, operating temperature, humidity, storage, etc.) |
| Electrode Specifications | (Various specifications listed in the comparison table, generally found "Equivalent" to predicate, e.g., product code, where used, material, single-use, non-sterile, single cable). However, number of electrodes is different (3 vs 4). |
Study Details
The document describes three studies that support the performance of the BAR Monitor.
1. Study U111-1124-2523
- Sample Size: 20 subjects
- Data Provenance: Melbourne, Australia; Prospective, double-blinded, randomized controlled design.
- Number of Experts & Qualifications (Ground Truth):
- Number of Experts: Not explicitly stated how many experts established the OAA/S ground truth for each subject. Assuming standard clinical practice, it would likely involve the anesthesiologist(s) and/or trained research staff present.
- Qualifications: "Observer's Assessment of Alertness/Sedation (OAA/S) level 4" and "OAA/S 0" indicate clinical assessments by qualified personnel familiar with these scales. Specific qualifications (e.g., anesthesiologist experience) are not provided.
- Adjudication Method: Not explicitly stated.
- MRMC Comparative Effectiveness Study: The study compared the predictive ability of CCS with the BIS™ index for loss of response.
- Effect Size: CCS and BIS™ were found to be equivalently predictive, with a prediction probability (Pr) of 0.9 for both devices. This indicates no significant improvement by AI (CCS) over the established technology (BIS™) in this specific outcome, but rather a comparable performance.
- Standalone (Algorithm only) Performance: Yes, the prediction probability (Pr) of 0.9 for CCS indicates its standalone performance in predicting loss of response.
- Type of Ground Truth: Clinical assessment (Observer's Assessment of Alertness/Sedation - OAA/S levels).
- Training Set Sample Size: Not specified in this document. This study appears to be a clinical validation, not part of the algorithm's training.
- Ground Truth for Training Set: Not specified in this document.
2. Study ACTRN12618000916246
- Sample Size: 21 patients (10 sevoflurane, 11 xenon)
- Data Provenance: Melbourne, Australia; Prospective, double-blinded, randomized controlled design.
- Number of Experts & Qualifications (Ground Truth):
- Number of Experts: Not explicitly stated how many experts established the emergence criteria (first response to name, first eyes open, spontaneously ventilating). Likely clinical staff.
- Qualifications: Not explicitly stated, but implies clinical observation by trained medical professionals (e.g., anesthesiologists, nurses).
- Adjudication Method: Not explicitly stated.
- MRMC Comparative Effectiveness Study: The study compared CCS with the BIS™ index during emergence from anesthesia.
- Effect Size: Prediction probabilities for CCS were equivalent to or better than that for the BIS for the sevoflurane group, the xenon group, and both groups combined. This suggests comparable to slightly improved performance of CCS over BIS for emergence monitoring.
- Standalone (Algorithm only) Performance: Yes, prediction probabilities for CCS were calculated independently.
- Type of Ground Truth: Clinical assessments of emergence endpoints (first response to name, first eyes open, spontaneously ventilating).
- Training Set Sample Size: Not specified in this document.
- Ground Truth for Training Set: Not specified in this document.
3. Study 6001-004
- Sample Size: 42 subjects (14 subjects per remifentanil group: 0, 2, or 4 ng/ml)
- Data Provenance: Belgium; Retrospective re-analysis of previously recorded EEG data from a randomized, multi-arm study.
- Number of Experts & Qualifications (Ground Truth):
- Number of Experts: Not explicitly stated from the original study that collected the OAA/S data.
- Qualifications: OAA/S assessments imply clinical observation by qualified personnel. Specific qualifications are not provided.
- Adjudication Method: Not explicitly stated.
- MRMC Comparative Effectiveness Study: CCS was compared to GE M-Entropy™ state and response entropy measures.
- Effect Size: Prediction probabilities for CCS across all OAA/S levels were equivalent to values obtained using GE M-Entropy™ state and response entropy measures. This indicates comparable performance of CCS.
- Standalone (Algorithm only) Performance: Yes, prediction probabilities for CCS were calculated. E.g., Prediction probabilities between OAA/S 5 and OAA/S 0 were ≥0.99.
- Type of Ground Truth: Clinical assessment (Observer's Assessment of Alertness/Sedation - OAA/S levels).
- Training Set Sample Size: Not specified in this document. This study describes a re-analysis of previously recorded EEG data, suggesting it was used for validation purposes rather than initial model training.
- Ground Truth for Training Set: Not specified in this document.
General Note on Training Data: The document focuses entirely on clinical validation studies and non-clinical performance testing. It does not provide details regarding the training set sample size or how the ground truth for the training set was established for the BAR Monitor's algorithm (Composite Cortical State - CCS). This information would typically be found in a more detailed technical report on the algorithm development, which is not included in this FDA 510(k) summary. The studies provided demonstrate the performance of the already developed algorithm against clinical ground truth and predicate devices.
§ 882.1400 Electroencephalograph.
(a)
Identification. An electroencephalograph is a device used to measure and record the electrical activity of the patient's brain obtained by placing two or more electrodes on the head.(b)
Classification. Class II (performance standards).