The future of intraoperative neuromonitoring (IONM) in spinal surgery

On behalf of the NASS section on intraoperative neuromonitoring (IONM), we present a narrative perspective exploring the future of IONM in spine surgery in the US, drawing on current evidence and future projections.

IONM is used during hundreds of thousands of spinal procedures each year to enhance patient safety via real-time neurodiagnostic feedback. The most common service model is an in-room technologist and a remote supervising professional who interprets the neurophysiological data. The primary goal of IONM is to: (1) detect significant signal changes from baseline, (2) identify the cause-whether technical, positional, anesthetic, or iatrogenic, and (3) pinpoint the site of injury. This diagnostic process is time-sensitive, complex, and dependent on both the signal pattern change and patient and procedural factors that are dynamically variable.

Artificial intelligence (AI) and machine learning (ML) hold promise to enhance the accuracy in detecting and interpreting signal changes for IONM clinicians and be integrated into surgeon-directed software platforms. However, widespread AI/ML adoption depends on the availability of large, validated IONM datasets-currently hindered by practice variation, inconsistent perioperative documentation, and unharmonized IONM, anesthetic, surgical, and patient medical records.

IONM can improve in the consistency in which optimal IONM is delivered, how IONM is utilized with evidence-based planning for alerts, and the collection of harmonized and complete signal and clinical records. Most publications have focused on the diagnostic accuracy of IONM in predicting deficits, but more emphasis is needed on demonstrating the therapeutic impact of interventions to alerts and their role in preventing new deficits.