Performance of ear canal sound measurements in detecting dural arteriovenous fistulas

Sander W J Ubbink*, J Marc C van Dijk, Rutger Hofman, Pim van Dijk

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

2 Citations (Scopus)
17 Downloads (Pure)

Abstract

OBJECTIVE: Patients with pulsatile tinnitus (PT) are often referred for digital subtraction angiography (DSA) to exclude cranial dural arteriovenous fistula (DAVF). Because DSA is not without risk, the authors studied the sensitivity and specificity of an automated detection algorithm that analyses external ear canal sound measurements to evaluate the presence of DAVF in PT patients.

METHODS: Between 2015 and 2021, external ear canal sound measurements of 72 PT patients were collected prospectively at our tertiary tinnitus clinic preceding referral of these patients to the neurosurgical department for counseling about DSA. The measurements were analyzed with an algorithm that objectifies a pulsatile sound as a PT coherence index (PTCI) per frequency band.

RESULTS: Ultimately, DSA was performed in 49 PT patients. DAVF was revealed in 15 patients (31%). PTCI cutoff values of 0.7 (at 250 Hz) and 0.6 (at 1000 Hz) resulted in sensitivity of 100% (78%-100%) and specificities of 62% (44%-78%) and 68% (49%-83%), respectively, for detecting DAVF on DSA.

CONCLUSIONS: Analysis of external ear canal sound measurements in PT patients showed 100% sensitivity for detecting DAVF on DSA. This analysis can potentially be used as a screening tool to help clinicians and PT patients to decide on the necessity of DSA. Retrospectively, the use of sound measurement in our cohort would have narrowed the indication for DSA to 23 of 49 cases (47%) without missing a DAVF.

Original languageEnglish
Pages (from-to)1700–1706
Number of pages7
JournalJournal of Neurosurgery
Volume140
Issue number6
Early online date8-Dec-2023
DOIs
Publication statusPublished - Jun-2024

Fingerprint

Dive into the research topics of 'Performance of ear canal sound measurements in detecting dural arteriovenous fistulas'. Together they form a unique fingerprint.

Cite this