A Sound Approach: Diagnostic Imaging for Pulsatile Tinnitus

Lilian White, MD
Posted on March 17, 2025

Although tinnitus is common and affects an estimated 10% to 25% of the population, pulsatile tinnitus is less common, affecting approximately 4% of the population. Given the relatively low prevalence of pulsatile tinnitus, studies tend to be small, which limits evidence for a generalizable diagnostic approach.

The differential diagnosis of pulsatile tinnitus includes intracranial vascular causes, cardiovascular disease, neoplasm, and increased intracranial pressure, among others. Evaluation of pulsatile tinnitus is critical given the significant risk of morbidity and mortality associated with some etiologies, such as hemorrhagic or ischemic stroke and blindness in patients with intracranial vascular causes or increased intracranial pressure. Venous malformations or variations tend to be the most common cause of pulsatile tinnitus. After headache and blurred vision, pulsatile tinnitus is one of the most common presenting symptoms of idiopathic intracranial hypertension, according to a small study. Glomus tumors are the most common tumor of the middle ear, with 80% of patients with glomus tumors presenting with pulsatile tinnitus. Approximately 30% to 50% of patients are found to have an etiology for their pulsatile tinnitus.

Findings on physical exam that support a vascular etiology of pulsatile tinnitus include bruits over the pre- or postauricular area, eyes, chest, or neck or pulsatile tinnitus that improves with gentle pressure over the ipsilateral jugular vein.

The American Academy of Otolaryngology-Head and Neck Surgery (AAO-HNS) recommends imaging studies for patients presenting with pulsatile tinnitus. Other red flag symptoms that should prompt further evaluation with imaging in patients with tinnitus include unilateral tinnitus, asymmetric hearing loss, and focal neurologic deficits. The AAO-HNS has a strong recommendation against obtaining imaging in the absence of these red flag symptoms. Watchful waiting may be considered in patients with intermittent symptoms, although this has not been studied. There is no consensus on “best” initial imaging test in the evaluation of pulsatile tinnitus because the accuracy of the test depends on the underlying etiology. Choice of imaging is guided by patient factors, the physical exam, and clinical suspicion.

CTA (computed tomography angiography) has a relatively high diagnostic yield (approximately 86%) compared with other imaging options; therefore, it is generally recommended for the initial evaluation of patients with pulsatile tinnitus in the family medicine setting in the absence of distinguishing features on history or exam. Specifically, CTA of the head and neck is a good initial imaging choice for detecting vascular lesions or increased intracranial pressure. A CT of the temporal bone without contrast is an alternative in patients with a lower suspicion of vascular etiology of pulsatile tinnitus and is best for detecting middle ear tumors or paragangliomas. A CT venogram may also routinely be considered concurrent with a CTA or in patients with a suspected venous etiology. Ultrasonography may be considered for further evaluation of suspected carotid stenosis.

In patients with contraindications to iodinated or gadolinium contrast, noncontrast magnetic resonance imaging and angiogram may be considered for suspected vascular lesions. Additionally, magnetic resonance imaging or angiogram is considered more sensitive for soft tissue masses and tends to be recommended for patients with unilateral and/or asymmetric hearing loss that is nonpulsatile.

A cerebral angiogram by a neurovascular interventionist may be considered if a high-risk cerebrovascular lesion is suspected in patients without a diagnosis after noninvasive imaging tests and has a relatively high diagnostic yield. A comprehensive audiological exam is recommended for all patients with tinnitus and may lead to a diagnosis in cases where imaging is nondiagnostic. Testing for systemic causes (eg, hyperthyroidism, anemia, pregnancy) of pulsatile tinnitus is generally not high-yield but may be considered for selected patients.