Background and objective Pulsatile tinnitus (PT) can be debilitating and lead to significant morbidity. Cerebral venous sinus lesions, such as venous sinus stenosis, diverticula, and high-riding jugular bulb, are uncommon causes of PT, for which there is no standard treatment. Endovascular interventions have shown promising results for PT secondary to idiopathic intracranial hypertension, and may be a valid therapeutic option for isolated venous PT.
Methods We conducted a systematic literature review on the outcome and safety of endovascular treatment for patients with isolated, debilitating venous PT. The venous lesion characteristics, endovascular techniques, complications, and clinical outcomes were assessed. In addition, an illustrative case of endovascular stenting for PT caused by venous sinus stenosis was included.
Results A total of 41 patients (90.2% female) from 26 papers were included. The median age was 46 years (IQR 23; range 25–72 years). Focal venous sinus stenosis (20 patients) and sinus diverticula (14 patients) were the most common culprit lesions. Endovascular treatment included venous sinus stenting in 35 patients, 11 of whom had adjuvant coil embolization, and coil embolization alone in six patients. Complete resolution of the tinnitus was achieved in 95.1% of patients. There was one complication of cerebellar infarct, and no procedure-related mortality.
Conclusions In patients with debilitating PT secondary to venous sinus lesions, endovascular treatment by stenting and/or coil embolization appears to be safe and effective. Prospective randomized studies with objective outcome assessments are needed to confirm the treatment benefits.
Statistics from Altmetric.com
Tinnitus can affect up to 10–15% of the adult population, and the prevalence increases with age.1 It is frequently persistent, and negatively affects patients’ sleep and quality of life, leading to substantial morbidity, including insomnia, anxiety, and depression.2 Approximately 4% of all patients with tinnitus have pulsatile tinnitus (PT), and up to 92% of these have an underlying organic lesion.3
High-flow arterial and arteriovenous shunting lesions, such as carotid stenosis, arteriovenous malformations, dural arteriovenous fistulas, or glomus tumors, have been considered the most common vascular causes of PT. Recently, venous pathologies have been increasingly recognized as a significant cause of PT, in part owing to advancements in non-invasive CT and MR venography techniques. These lesions include high-riding jugular bulb, sinus diverticula, or venous sinus stenosis.4 Endovascular venous sinus stenting (VSS) has shown promising results in relieving PT associated with idiopathic intracranial hypertension (IIH) in patients with concurrent venous sinus stenosis.5 However, the efficacy and safety profile of endovascular treatment for patients with isolated debilitating PT secondary to venous pathologies remains unclear.
We conducted a literature review on the efficacy and safety of endovascular treatment for patients with isolated debilitating venous PT. In addition, an illustrative case of isolated debilitating PT secondary to venous sinus stenosis treated by VSS is provided to demonstrate the diagnostic and therapeutic considerations and technique.
Literature search and study selection
A comprehensive literature search was performed by the first and corresponding authors in PubMed, Medline, Scopus, and Web of Science for all articles in English published between January 2000 and October 2018 that reported endovascular treatment of tinnitus using the following keywords: pulsatile tinnitus, jugular bulb, sinus aneurysm, sinus diverticulum, sinus stenosis, diverticulum, endovascular treatment, venous sinus stenting, and coiling. This was supplemented by a reverse bibliography search from previously published case reports, systematic reviews, and/or meta-analyses.
Inclusion criteria were (1) studies reporting endovascular treatment of tinnitus and (2) individualized treatment outcome of tinnitus secondary to intracranial venous pathology. Studies were excluded in cases of (1) non-vascular etiology of tinnitus, (2) tinnitus secondary to arterial or arteriovenous shunting lesions, including vascular compression of the cochlear nerve or dural arteriovenous fistulas, and (3) concurrent indication other than tinnitus for endovascular VSS such as cases of IIH. Two reviewers independently screened the eligibility of articles according to the inclusion and exclusion criteria. The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) flow chart is shown in figure 1.
Data extraction and outcome variables
Individualized data on patient demographics, symptom characteristics and duration, etiology of the tinnitus, and endovascular treatment outcomes and complications were extracted. The corresponding authors of the included papers were contacted to provide additional information not published in the original manuscript. Special focus was put on describing the specific anatomical nature of venous sinus pathologies (stenosis versus aneurysm/diverticulum), and the technical detail and devices used for endovascular treatment. The tinnitus outcome and complication rates were studied. Descriptive statistics on demographics, symptom characteristics, venous sinus pathologies, and treatment outcomes were reported.
Literature search and study characteristics
The initial search yielded 256 articles for screening, of which, 215 were excluded based on abstract and title alone. Of the 41 full-text articles retrieved for evaluation, 17 were excluded for concurrent IIH or other causes of tinnitus, and three were excluded for incomplete demographic data and overlapping patients. A total of 21 papers were included in this review.
Eligible studies were all observational retrospective studies, including 19 case reports and two case series published between 2000 and 2018.
Patient demographics and clinical presentation
Patient demographics were summarized in table 1.6–26 Of the 41 patients included, 37 (90.2%) were women. The median age was 46 years (IQR 23; range 25–72 years). Pulsatile tinnitus was lateralized to the right in 52.1% of patients and the median duration of tinnitus before treatment was 15 months (IQR 33; range 3–240 months).
An internal jugular vein (IJV) compression test was performed on 27 patients. All but one patient tested positive, defined as the reduction or resolution of tinnitus when the ipsilateral IJV was compressed. Intrasinus manometry was performed in 22 patients with sinus stenosis. Concurrent headache and vertigo affected 5 (12.2%) and 2 (4.9%_ of patients, respectively.
Venous sinus pathologies
Among the 41 cases of venous tinnitus, all had ipsilateral transverse-sigmoid sinus or jugular bulb pathologies. Focal venous sinus stenosis was present in 20 patients, sinus diverticulum in 14, coexisting venous sinus stenosis and diverticulum in 4, coexisting high-riding jugular bulb and venous sinus stenosis in 2, and high-riding jugular bulb in one patient.
Endovascular treatment and outcome
All 20 patients with isolated venous sinus stenosis were treated through a transvenous approach with VSS across the stenotic segment using a Wallstent (Boston Scientific, Massachusetts, USA). For the 18 patients with sinus diverticulum with or without coexisting stenosis, nine were treated with stent-assisted coil embolization, six with coil embolization of the diverticulum, and three with VSS only. The three patients with high-riding jugular bulb were all treated with VSS, with adjuvant coil embolization in two patients.
The median clinical follow-up period was 7 months (IQR 14, range 1–84). Complete resolution of the tinnitus was achieved in 39 of the 41 patients (95.1%) after endovascular treatment. One patient reported 90% improvement with a mild residual non-pulsatile tinnitus, and one had no improvement owing to an erroneous diagnosis (subsequently established to be a subpetrosal internal carotid artery web). No deaths occurred, and only one procedure-related complication of focal cerebellar infarct, resulting in transient ataxia after stent-assisted coiling of a venous diverticulum.
A 57-year-old woman had right-sided, debilitating pulsatile tinnitus for 1 year. There was no associated hearing loss, headache, or visual disturbance. The cerebrospinal fluid opening pressure was normal. The tinnitus was relieved by gentle venous compression of right IJV at the neck. A magnetic resonance venogram and diagnostic catheter angiogram showed a dominant right transverse sinus with a focal high-grade stenosis of the distal segment due to prominent Pacchionian granulations (figure 2A–D). Intrasinus manometry demonstrated a minimal pressure gradient (2 mm Hg) across the stenotic segment. As the symptom was intolerable to the patient and other causes of tinnitus were excluded, a decision was made for endovascular treatment of the venous sinus stenosis.
Endovascular treatment was performed under general anesthesia through a femoral transvenous approach. A 6 F NeuronMAX long sheath (Penumbra, California, USA) and a 6F Navien intracranial support catheter (Covidien Vascular Therapies, Massachusetts, USA) were placed coaxially in the ipsilateral jugular bulb and the transverse sinus proximal to stenosis respectively. Endovascular VSS with a 7×40 mm Wallstent (Boston Scientific, Massachusetts, USA) was performed across the stenosis and completely opened the narrowed transverse sinus (figure 2E, F). The Wallstent was chosen for its closed-cell design and small free cell size to effectively exclude arachnoid granulations from the venous sinus. Balloon angioplasty was unnecessary, as the stenotic venous sinus was adequately expanded. Antiplatelet therapy with aspirin and ticagrelor was started 3 days before the procedure and continued for 3 months poststenting, after which aspirin monotherapy was continued. There was no complication and the PT completely resolved immediately after the intervention. The patient remained asymptomatic at the 3-month follow-up.
Vascular causes of PT can be broadly classified as arterial or venous. Common arterial lesions causing PT include aberrant courses of the internal carotid artery, glomus tumors, acquired carotid artery stenosis, and arteriovenous shunts.4 Venous PT is uncommon, but can be a debilitating symptoms nonetheless.27 In this review, venous sinus stenosis was the most common cause and accounted for half of the cases, followed by venous diverticulum, high-riding jugular bulb, or a combination of these. The vast majority of patients were middle-aged women who had a long duration of symptoms before treatment.
Using computational fluid dynamic models based on the venogram of a patient with transverse sinus stenosis and diverticulum, Han et al demonstrated that the high-velocity jet flow across the stenotic segment and the turbulent flow in the diverticulum were ameliorated after VSS.28 This change in the blood flow to a more uniform and laminar pattern was followed by PT resolution in the patients. Although similar results can be achieved by mastoidectomy and reconstruction of the bony defect overlying the venous diverticulum, endovascular stenting with or without adjuvant coil embolization is a less invasive option for these patients, especially those with venous sinus stenosis which is less amenable to open surgical repair.29 30
Our results showed that for all patients with venous sinus stenosis and/or diverticulum, endovascular stenting had a high success rate in resolving the PT, and a reasonable safety profile. The technique is illustrated in the case example, and is well described in the IIH literature.5 One complication of focal cerebellar infarct with transient deficit has been reported, which could be secondary to inadvertent occlusion of a cerebellar draining vein during the catheterization or coil deployment.1 Thus, during the intervention it is essential to visualize and protect the venous channels connecting to the transverse and sigmoid sinuses, such as the vein of Labbe, superior and middle cerebellar veins, and the superior and inferior petrosal sinuses, to avoid complications. Despite the low complication rate, endovascular treatment should be considered only for cases with severely debilitating PT that does not respond to more conservative management options.
It is important to recognize that not all venous sinus anomalies are symptomatic and require treatment. Incidental finding of high-riding jugular bulb is present in up to 6% of CT scans.7 9 The diagnosis of venous PT is established with comprehensive clinical and radiological assessments before treatment is contemplated.4 31 Unlike arterial PT, tinnitus of venous origin is diminished if the blood flow across the ipsilateral dural venous sinus and jugular system is reduced. This can be achieved by gentle ipsilateral IJV compression or by turning the head ipsilaterally, which was performed as a confirmatory test in two-thirds of the included cases.6 23 Conversely, tinnitus may be exacerbated by contralateral IJV compression that diverts the venous flow to the side of the lesion. Typical symptoms of increased intracranial pressure such as headache and visual disturbance should prompt a detailed evaluation for IIH, including a lumbar puncture to obtain the opening pressure.5 In contradistinction to IIH, where a pressure gradient commonly exist across the venous sinus stenosis, our review and the illustrated case showed that most non-IIH tinnitus was not associated with a pressure gradient at the venous sinus stenosis. The lack of a pressure gradient in itself should not exclude patients with debilitating tinnitus clearly attributable to venous sinus stenosis from receiving treatment.31 32 CT or MR venography can demonstrate most structural causes of venous PT, such as the venous sinus stenosis in our illustrative case, but high-resolution CT of the temporal bone, CT or MR angiography and occasionally conventional catheter angiography are needed to exclude other differential diagnoses of PT such as dural arteriovenous fistulas.33
When it is uncertain whether a patient’s PT symptom is fully attributable to the visualized venous sinus lesion, such as patients with negative IJV compression test, endovascular confirmatory tests can be performed with the patient awake under local anesthesia. This was performed in 2 of the 41 cases included. Park et al treated a case of venous diverticulum by coil embolization, and confirmed the resolution of PT before coil detachment.12 In a similar fashion, Du et al employed a retrievable and detachable stent (Solitaire AB, Medtronic, California, USA) for a venous sinus stenosis with debilitating PT.24 The stent was deployed in situ with the patient awake to assess symptom resolution, and a decision was made to detach the stent permanently only if there was substantial improvement of PT.
Our review has some limitations. All included articles are retrospective and most of them are case reports, with potential recall and publication bias. The assessment of tinnitus was subjective, and the follow-up period was relatively short to assess delayed symptom recurrence. The largest case series in the literature by Baomin et al, with 46 patients with PT secondary to venous sinus stenosis, was not included as some patients were treated for concurrent IIH and the results were not reported separately.34 Nevertheless, their outcome was similar to our findings and all patients had complete resolution of PT without procedure-related complications. A randomized controlled trial with more objective diagnostic criteria and outcome assessments is needed to better assess the treatment effect.
Venous sinus stenosis, sinus diverticula, and jugular bulb diverticula are uncommon causes of pulsatile tinnitus. The diagnosis is established by excluding other high-flow vascular, non-vascular, and neurological causes, and supported by a positive IJV compression test. In patients with debilitating venous tinnitus, endovascular treatment by VSS with or without coil embolization appears to be safe and effective. Prospective randomized studies with objective outcome assessments are needed to confirm the treatment benefit.
Contributors I-HY, ACOT collected and analyzed the data and prepared the manuscript. VMP, ACOT conceptualized and supervised the study. SL, PN, EO, JMK provided input in the study design and data interpretation. All authors critically reviewed the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Patient consent for publication Not required.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.