(126 days)
The SedLine Sedation Monitor is intended to monitor the state of the data acquisition and processing of EEG signals. The system includes the Patient State Index (PSI), a proprietary computed EEG variable that is related to the effect of anesthetic agents. The agents include: Alfentanil, Desflurane, Nitrous Oxide, Propofol, Remifentanil, and Sevoflurane. The Sedation Monitor is intended for use with adult patients (18 years of age and older) in the operating room (OR), intensive care unit (ICU), and clinical research laboratory.
SedLine® Sedation Monitor is a patient-connected, 4-channel processed Electroencephalograph (EEG) monitor designed specifically for intraoperative or intensive care use. It displays electrode status, EEG waveforms, Density Spectral Array (DSA), and Patient State Index (PSI), EMG Index, Suppression Ratio (SR) and Artifact (ARTF). The operator controls the unit using menus and dedicated buttons to select various display options. The system consists of 4 major components: Root, SedLine Module, SedLine Patient Cable, and SedLine Sensor.
Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided text:
Key Takeaway: This 510(k) submission is for a modification to an existing device (SedLine Sedation Monitor), specifically a modified Patient State Index (PSI) algorithm and an optional additional DSA display. The primary goal is to demonstrate substantial equivalence to the predicate device.
1. Table of Acceptance Criteria and Reported Device Performance
The document does not explicitly present a "table of acceptance criteria" with corresponding "reported device performance" in the typical format of a clinical trial results table that would define specific metrics and thresholds for success for the new PSI algorithm. Instead, the non-clinical testing section lists various standards and guidances it aimed to satisfy, and the clinical testing section describes how the new algorithm was compared to the predicate.
However, based on the context of a 510(k) submission for a device modification, the implicit acceptance criteria would revolve around demonstrating that the modified device performs at least as well as, or is substantially equivalent to, the predicate device. The general performance specifications for the SedLine Sedation Monitor are provided in Table 5.2.
Implicit Acceptance Criteria and Reported Performance (derived from document):
| Acceptance Criteria Category | Specific Areas (Implicit) | Reported Performance/Outcome (from text) |
|---|---|---|
| Non-Clinical Performance (Safety & Functionality) | Electrical Safety (IEC 60601-1) | Satisfied all requirements and performance specifications. |
| EMC (IEC 60601-1-2) | Satisfied all requirements and performance specifications. | |
| Alarm Testing (IEC 60601-1-8) | Satisfied all requirements and performance specifications. | |
| Usability (FDA Human Factors & Usability Draft Guidance) | Satisfied all requirements and performance specifications. | |
| Software Verification (FDA Software Guidance) | Validated DSA by comparing multi-taper DSA against the predicate's Hanning DSA for known input signals, testing: dynamic range, frequency range, spectral edge frequency, and high-contrast feature. Satisfied all requirements and performance specifications. | |
| Mechanical Testing (EN 60601-2-26) | Satisfied all requirements and performance specifications. | |
| Environmental Testing (EN 60601-2-26) | Satisfied all requirements and performance specifications. | |
| Clinical Performance (Equivalence of new PSI algorithm) | Comparison of subject PSI algorithm to predicate PSI algorithm | "The subject PSi algorithm was compared to the predicate PSi algorithm." (Implies the comparison satisfied the criteria for substantial equivalence, though specific statistical equivalence metrics are not detailed in this summary). The overall conclusion is that clinical testing "demonstrates that the subject device... is substantially equivalent to its predicate." |
2. Sample Size Used for the Test Set and the Data Provenance
- Sample Size for Test Set: 100 surgical patients
- Data Provenance:
- Country of Origin: Not explicitly stated, but the submission is to the FDA (USA), and typically, studies cited in such submissions are either US-based or explicitly noted if international.
- Retrospective or Prospective: Retrospective analysis of clinical data.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and the Qualifications of Those Experts
The document does not specify the number or qualifications of experts used to establish ground truth for the test set. For this type of device (Sedation Monitor), the "ground truth" for the PSI algorithm's performance is typically related to the administered anesthetic drug doses and observed physiological responses, rather than expert interpretation of EEG waveforms alone for classifying sedation depth. The study states "Clinical data used for the analysis includes continuous EEG, anesthetic drug dose information, and other physiological vital signs such as mean arterial blood pressure and heart rate," which would serve as the reference against which the PSI algorithm's output is compared.
4. Adjudication Method for the Test Set
The document does not mention any adjudication method for the test set. Given it's a retrospective analysis of clinical data including objective measurements (EEG, drug doses, vital signs), an adjudication process involving multiple human readers for "ground truth" might not have been applied in the same way as, for example, in an imaging study.
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, an MRMC comparative effectiveness study involving human readers and AI assistance was not performed or described. This study focused on the performance of the modified algorithm itself in comparison to its predicate, using retrospective clinical data. The SedLine Sedation Monitor is an monitoring device that provides a computed index (PSI) for clinicians to interpret, it's not a diagnostic AI intended to assist human interpretation of complex images in an MRMC setting.
6. If a Standalone (i.e., Algorithm Only Without Human-in-the-Loop Performance) Was Done
Yes, a standalone performance evaluation of the algorithm was done. The clinical study describes a "retrospective analysis of the clinical data" where "The subject PSi algorithm was compared to the predicate PSi algorithm" using collected physiological and drug administration data. This directly assesses the algorithm's output (PSI) based on its input (EEG signals, etc.) without human intervention in its calculation or interpretation to determine its output.
7. The Type of Ground Truth Used (Expert Consensus, Pathology, Outcomes Data, etc.)
The ground truth for evaluating the PSI algorithm's relation to anesthetic depth would be based on:
- Anesthetic drug dose information: The type and amount of anesthetic agents administered.
- Physiological vital signs: Mean arterial blood pressure and heart rate.
- Continuous EEG data: The raw EEG signals from which the PSI is derived, allowing for comparison of the algorithm's output against the expected EEG changes under anesthesia.
Combined, these elements serve as the reference for the "effect of anesthetic agents" that the PSI is designed to reflect. It's not a single "expert consensus" or "pathology" but rather a composite of objective clinical data related to the patient's state of anesthesia.
8. The Sample Size for the Training Set
The document does not provide the sample size for the training set. It only mentions the "validation" of the subject algorithm through retrospective analysis of clinical data in 100 surgical patients. This 100-patient dataset appears to be the test/validation set for assessing the modified algorithm, not necessarily a training set. Given that this is a modification of an existing algorithm, the original algorithm would have been developed and trained using prior data, but details about that are not included here.
9. How the Ground Truth for the Training Set Was Established
The document does not provide information on how the ground truth for the training set (if a separate training set was used for the modified algorithm) was established. It only describes the data used for the validation comparison of the subject algorithm against the predicate. For the original development of the PSI, ground truth would typically involve correlating EEG patterns with known states of consciousness/sedation induced by controlled anesthetic administration, likely established by expert assessment (e.g., Riker Sedation-Agitation Scale, Observer's Assessment of Alertness/Sedation Scale) and objective physiological markers during prospective studies. However, these details are absent for this specific submission's context.
§ 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).