(95 days)
The NVM5® System is a medical device that is intraoperative neurophysiologic monitoring during spinal surgery. The device provides information directly to the surgeon, to help assess a patient's neurophysiologic status. NVM5 provides this information by electrically stimulating nerves via electrodes located on surgical accessories and monitoring electromyography (EMG), transcranial or lumbar motor evoked potential (MEP), or somatosensory evoked potential (SSEP) responses of nerves. The System also integrates Bendini® software used to locate spinal implant instrumentation for the placement of spinal rods.
· XLIF (Detection) - The XLIF (Detection) function allows the surgeon to locate and evaluate spinal nerves, and is used as a nerve avoidance tool.
· Basic & Dynamic Screw Test - The Screw Test functions allow the surgeon to locate and evaluate spinal nerves by providing proximity information before, during or after bone preparation and placement of bone screws.
· Free Run EMG - The Free Run EMG function identifies spontaneous EMG activity of spinal nerves by continually displaying a live stream waveform of any mechanically induced myotome contractions.
· Twitch Test (Train of Four) = The Twitch Test Function allows the surgeon to assess moderate degrees of neuromuscular block in effect by evaluating muscle contraction following a train of four stimulation pulses.
· MEP - Transcranial or lumbar (i.e., conus in region of LI-L2) stimulation techniques for motor evoked potentials are used to assess for acute dysfunction in axonal conduction of the corticospinal tract and peripheral nerves. The MEP function provides an adjunctive method to allow the surgeon to monitor spinal cord and motor pathway integrity during procedures with a risk of surgically induced motor injury.
· SSEP - The SSEP function allows the surgeon to assess sensory spinal cord function in surgical procedures during which the spinal cord is at risk.
· Remote Reader - The Remote Reader function provides real time remote access to the NVM5 System for a monitoring physician outside of the operating room.
· Guidance - The Guidance function is intended as an aid for use in either open or percutaneous pedicle cannulation procedures in the lumbar and sacral spine (LI-S1) of adult patients, and when used in conjunction with radiographic imaging and EMG, allows the surgeon to assess the angulation of system accessories relative to patient spinal anatomy for the creation of a cannulation trajectory for bone screw placement.
· Bendini - The Bendini Spinal Rod Bending function is used to locate spinal implant system instrumentation (screws, hooks) to determine their relative location to one another to generate bend instructions to shape a spinal rod. A surgeon is able to use those instructions and bend a rod using the Bendini Bender, a mechanical rod bender.
The NVM5 System is a medical device that is intended for intraoperative neurophysiologic monitoring during spinal surgery. The device provides information directly to the surgeon, to help assess a patient's neurophysiologic status. NVM5 provides this information by electrically stimulating nerves via electrodes located on surgical accessories and monitoring electromyography (EMG), motor evoked potential (MEP) or somatosensory evoked potential (SSEP) responses of nerves. Moreover, a Twitch Test function is utilized to test the ability of the nerve to respond, or contract, following four stimulation pulses to determine the presence of neuromuscular block.
Additionally, the NVM5 System includes an integrated stereotactic guidance system (NVM5 Guidance) to support the delivery of pedicle screws during EMG monitoring. The System also integrates Bendini software used to locate spinal implant instrumentation for the placement of spinal rods. Lastly, the system also offers an optional screen sharing application to allow a secondary physician to remotely view the events represented on the NVM5 user interface. In summary, the NVM5 System includes the following six (6) software functionalities / modalities:
- Electromyography (EMG)
- Motor Evoked Potential (MEP)
- Somatosensory Evoked Potential (SSEP)
- Remote Reader
- Guidance
- Bendini
The NVM5 System hardware consists of a Patient Module (PM) and computer, as well as accompanying accessory components which consist of an assortment of disposable conductive probes, electrodes, and electrode leads.
The provided document is a 510(k) Premarket Notification for the NuVasive® NVM5® System, a medical device for intraoperative neurophysiologic monitoring during spinal surgery. The document focuses on demonstrating substantial equivalence to a predicate device (NuVasive NVM5 System - K141968) rather than presenting a detailed study proving the device meets explicit acceptance criteria in the format typically used for clinical performance studies of AI/ML devices.
However, based on the nonclinical testing section (Section G: Performance Data), we can infer the type of acceptance criteria and the nature of the study conducted to demonstrate the device's performance.
Here's an attempt to structure the information based on your request, with inferred details where explicit information is missing from the provided document:
1. Table of Acceptance Criteria and Reported Device Performance
The document does not explicitly state quantitative acceptance criteria for clinical performance (e.g., sensitivity, specificity, accuracy). Instead, it focuses on demonstrating that the subject device performs similarly to its predicate and meets design specifications. The key performance aspects mentioned are related to the functional capabilities of the device.
| Acceptance Criteria (Inferred from testing scope) | Reported Device Performance (Summary from Section G) |
|---|---|
| Verification of stimulation parameters (pulse width, amplitude, current polarity, rates) | Met or exceeded the performance of the predicate device. |
| Verification of response detection ranges | Met or exceeded the performance of the predicate device. |
| Validation of effectiveness of boundary conditions, extreme values, and nominal entries on GUI | Met or exceeded the performance of the predicate device. |
| Verification of point acquisition for Guidance function | Met or exceeded the performance of the predicate device. |
| Verification of user-defined inputs for Guidance function | Met or exceeded the performance of the predicate device. |
| Verification of rod bending instructions for Bendini function | Met or exceeded the performance of the predicate device. |
| Validation that user-defined inputs, point acquisition, and measurements result in proper bend instructions and/or calculated offsets for Bendini function | Met or exceeded the performance of the predicate device. |
| Angular tolerance for Guidance function (explicitly mentioned as performance requirement for Guidance) | The subject device, like the predicate, is designed to meet an angular tolerance of ±2° and confirm alignment to pre-planned trajectory. The studies "met or exceeded" this performance. |
2. Sample Size for the Test Set and Data Provenance
The document describes nonclinical testing, which includes:
- Laboratory bench top testing: This suggests no human patient data was used.
- Cadaveric testing: This implies the use of cadaver specimens.
Therefore:
- Sample Size for Test Set: Not specified in terms of number of cadavers or bench tests. No human patient test set was explicitly described for performance validation.
- Data Provenance: Laboratory bench top and cadaveric testing. No country of origin is specified, but typically this would be conducted within the manufacturer's R&D facilities or contracted labs. The data is retrospective in the sense that it's performed on non-living subjects or simulated conditions rather than live clinical cases for this specific submission.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications
Given that the testing was nonclinical (bench-top and cadaveric), the concept of "experts establishing ground truth for a test set" in the context of clinical interpretation (like radiologists for imaging) does not directly apply here. Instead, ground truth would be established by engineering specifications, known physical properties, and controlled experimental conditions.
- Experts: Not applicable in the context of clinical ground truth. The "effectiveness of boundary conditions, extreme values, and nominal entries" and "proper bend instructions and/or calculated offsets" would be evaluated against established engineering principles and the predicate device's known performance.
- Qualifications: Not specified. It would likely involve engineers, technicians, and potentially surgeons for cadaveric testing evaluating the functional aspects of the device.
4. Adjudication Method for the Test Set
Not applicable for this type of nonclinical verification and validation testing. Adjudication methods like 2+1 or 3+1 are used for human expert consensus on clinical data. Performance here would be measured against engineering tolerances and functional outputs.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No MRMC study was described. This document focuses on demonstrating substantial equivalence of the device's functional performance to its predicate through nonclinical testing, not on comparing human reader performance with and without AI assistance. The device itself is a neurophysiological monitoring system, which aids a surgeon, but the submission doesn't detail a study comparing surgeon performance.
6. Standalone (Algorithm Only Without Human-in-the-loop Performance)
The document describes a standalone device performing its intended functions (monitoring EMG, MEP, SSEP, providing guidance, and bend instructions). The tests are likely evaluating the system's output accuracy and reliability against a known truth or a predicate's output. While a human (surgeon or technician) interacts with and interprets the information from the device, the validation described is of the device's output itself, independent of the human 'decision-making' aspect. So, in essence, the nonclinical testing demonstrates the "algorithm only" performance (i.e., the system's ability to fulfill its specified functions).
7. Type of Ground Truth Used
The ground truth used would be based on:
- Engineering specifications: For pulse width, amplitude, frequency response, detection ranges, angular tolerance.
- Direct measurement: For physical verification on bench-top and cadaveric models.
- Predicate device performance: The "met or exceeded the performance of the predicate device" implies the predicate device's known and validated performance serves as a benchmark for substantial equivalence.
8. Sample Size for the Training Set
Not applicable. This device is a neurophysiological monitoring system with specific algorithms described (EMG, MEP, SSEP, Guidance, Bendini). The document does not indicate that this system uses machine learning or deep learning algorithms that require a "training set" in the conventional sense. The algorithms listed are likely rule-based or signal processing algorithms, not models that are "trained" on large datasets.
9. How the Ground Truth for the Training Set Was Established
Not applicable, as there is no indication of a machine learning training set. The algorithms are either identical to the predicate or have modified stimulation parameters or additional baseline algorithms/views (as noted for MEP and SSEP), suggesting design changes rather than a new machine learning model requiring a training phase.
§ 874.1820 Surgical nerve stimulator/locator.
(a)
Identification. A surgical nerve stimulator/locator is a device that is intended to provide electrical stimulation to the body to locate and identify nerves and to test their excitability.(b)
Classification. Class II.