Neurosurgery 97:1397–1405, 2025
This multicenter prospective study evaluates a wearable, wireless thermal anisotropy sensor for noninvasive detection of cerebrospinal fluid (CSF) shunt flow in symptomatic hydrocephalus patients. Using a random forest algorithm on data from 112 validated subjects, the device showed high sensitivity (88.9%) and a very high negative predictive value (96.8%), though specificity was modest (49.2%) and a 37.5% no-result rate occurred.
Methods, usability, and comparative analyses are detailed: device design, algorithm development, blinded validation, and 7-day surgical follow-up with intraoperative confirmation as ground truth. Ease-of-use was rated highly, adverse events were minor skin reactions, and combined use with CT/MRI improved specificity and overall accuracy, supporting adjunctive clinical application.
Device Purpose: A novel, noninvasive, wireless wearable device was developed to assess cerebrospinal fluid (CSF) shunt flow in hydrocephalus patients presenting with possible shunt failure, using thermal anisotropy measurements to detect flow through the shunt tubing.
Study Design: The prospective, blinded, multicenter study included both pediatric and adult patients with existing ventriculoperitoneal shunts and symptoms suggestive of shunt failure; subjects underwent device evaluation alongside standard-of-care assessments.
Algorithm Performance: The device’s random forest algorithm demonstrated high sensitivity (88.9%) and negative predictive value (NPV, 96.8%) but moderate specificity (49.2%) for detecting complete shunt failure, indicating strong ability to rule out failure when flow is confirmed.
Clinical Utility: The high NPV suggests the device could help identify patients whose shunts are functioning, potentially reducing unnecessary surgical interventions or additional invasive tests.
Comparison with Standard Tests: Standard imaging tests (CT/MRI) showed moderate-to-high sensitivity and specificity but require radiation or sedation, whereas the device is noninvasive, wireless, and does not require such resources.
Combined Diagnostic Approach: When device results were combined with CT or MRI findings, specificity increased to 93.3% and NPV to 96.6%, supporting a potential workflow where the device is used to complement imaging for risk stratification.
Safety and Usability: The device was rated as easy to use by clinicians, with a low rate of mild, self-resolving skin redness and no serious adverse events reported.
Limitations: The device is not suitable as a stand-alone diagnostic tool due to moderate specificity and a substantial rate of incomplete measurements; further studies are ongoing to optimize performance and generalizability.
You must be logged in to post a comment.