Sigmoid Sinus Wall Abnormalities: A Treatable Cause, and the Trap of Assuming It Is the Cause

Most causes of tinnitus on this site are managed rather than fixed. This one is a genuine exception — an anatomical problem with an operation that usually works — which makes it worth knowing about, and makes the caveat attached to it worth knowing just as well.

What the anatomy is

The sigmoid sinus is not a sinus in the sense people mean when they say sinus problems. It is a large vein, one of the main drains carrying blood out of the skull, and it curves down through the temporal bone directly behind the inner ear. Between the blood and the ear is a thin plate of bone.

Two things can go wrong with that plate, and the literature treats them as one family:

• Dehiscence — the bone is absent, so the vein wall sits directly against the air spaces of the mastoid.
• Diverticulum — the vein wall bulges outward into or through the bone, sometimes described as an aneurysm of the sinus.

In the pooled surgical series, the split was 47.5% diverticulum, 35.3% dehiscence, and 17.3% both. Either way the physical effect is the same: turbulent venous flow ends up acoustically much closer to the ear than the design intends, and gets heard. That is why this produces pulsatile tinnitus rather than the steady kind, and why it is often the venous sort that changes when you turn your head or press on your neck.

How often it is the explanation depends on which study you read. The 2017 systematic review opens by noting that studies find sigmoid sinus anomalies in 4 to 20% of people with pulsatile tinnitus — a range wide enough to tell you that the answer depends heavily on who is being scanned and how carefully the images are read.

The people who reach an operation are a distinctive group. Across the 139 pooled surgical cases, 90.4% were women and the mean age was 39. That skew describes patients who got as far as surgery rather than everyone who has the finding, but it is close enough to the pattern in idiopathic intracranial hypertension — young and middle-aged women, often with a raised BMI — to be part of why the two conditions keep turning up together.

The finding that should change how you read a scan report

Here is the part that does not usually make it into the conversation.

A 2023 case-control study at Vanderbilt compared high-resolution CT in 141 adults with pulsatile tinnitus against 149 controls — patients being scanned for cochlear implant workup. The abnormality showed up in 34% of the tinnitus group and 9% of controls. Statistically clear-cut, p < 0.001.

Read the second number again. Nine per cent of people who were not complaining of a whooshing sound had the same anatomical finding. The correlation with having pulsatile tinnitus at all was described as weak — r = 0.354.

Then the study did the thing that makes it valuable. Among the 48 tinnitus patients who did have the abnormality, it checked which side each was on:

• 64.6% — the abnormality was on the same side as the sound.
• 25.0% — partial match, for instance bilateral abnormality with one-sided tinnitus.
• 10.4% — the abnormality was on the opposite side from the sound.

The authors' conclusion is the sentence to carry into an appointment: the abnormality is likely a contributing factor in roughly 65% of cases, and for about a third of patients the association is either not causative or not solely causative.

One honest limitation, because the study has one: the controls were people undergoing cochlear implant evaluation, not the general population. They have significant hearing loss by definition. That is a convenience sample, and it is a reason to hold the 9% loosely rather than treat it as a population rate.

Why that matters practically

A scan that finds something is enormously reassuring after weeks of not knowing. It supplies an answer, a name and a plan. The risk is that the answer arrives with more confidence than the evidence supports, and the plan is an operation.

The same case-control paper spells out the clinical consequence: surgeons counselling patients may be optimistic overall about resurfacing, but must appreciate the possibility of treatment failure, likely from untreated comorbid conditions. In other words, when the operation does not work it is often because something else was also driving the sound — most commonly raised pressure. The overlap with idiopathic intracranial hypertension is close enough that the two are routinely assessed together, and the demographic overlap is close too.

A reasonable question to ask, given all this: how confident are you that this finding is what I am hearing, and what else have you ruled out?

There is a sharper version of the same caution, and it names something worth ruling out specifically. A 2025 study of 80 people who had pulsatile tinnitus and a confirmed dural arteriovenous fistula found sigmoid sinus wall or jugular bulb anomalies in 40.8% of them. Its conclusion is that a sigmoid sinus wall anomaly on CT together with a positive jugular compression test should not be treated as conclusive for a venous cause. A fistula is the one entry on this differential that can bleed, and it is also the one that embolisation can cure outright, so it is the wrong thing to have attributed to a bone finding.

What treatment involves and how well it works

The 2017 systematic review in Otology & Neurotology pooled 139 patients from 21 studies. It is the best summary available, with the standing caveat that pooling published case series is not the same as running a trial.

• Sigmoid sinus wall reconstruction or resurfacing in 91.4% — rebuilding the missing bone or reinforcing the bulge, usually through the mastoid.
• Endovascular approaches in 7.9%.
• Postoperative recurrence: 3.5%, over a mean follow-up of 21.1 months.
• The review's own verdict: a rare but largely treatable condition.

Against the rest of this field that is a remarkable number. Almost nothing else in tinnitus reports resolution at that rate, which is the whole reason a treatable structural cause is worth finding.

Two details from the pooled analysis are worth carrying:

The material used mattered. Bone pate, soft-tissue graft and bone cement had the highest resolution rates. Temporalis fascia and autologous bone chips were the most commonly used materials and had significantly lower resolution rates than most of the alternatives — auricular cartilage and bone cement being the exceptions the review names. In other words, the two materials reached for most often were not the two that worked best. If you are being offered this operation, what the wall is rebuilt with is a fair thing to ask about.

The BMI finding is odd, and worth flagging as odd. The review reports that for every 1 kg/m² increase in body mass index the odds of resolution increased 9.2%, and that right-sided tinnitus resolved at a higher rate. Both come from a regression across pooled published cases rather than a designed study, and neither has an obvious mechanism. They are findings, not facts to plan around.

It is not necessarily something you were born with

The assumption in older writing was that these are congenital variants — you had the thin bone all along, and something eventually made it audible.

A 2024 study of 42 patients who each had at least two CT scans taken without surgery in between found that 12 of them, 28.6%, showed morphological progression over time. Within the diverticulum group, 7 of 23 enlarged. Two dehiscence cases showed progressive erosion of the sigmoid plate. Three progressed from dehiscence to a frank diverticulum. Tinnitus handicap scores differed significantly between the first and later visits.

The authors conclude this is a progressive clinical condition rather than a congenital one. It is a level-4 study, 42 patients, one centre, so it settles nothing on its own — but it is direct longitudinal evidence, which the congenital assumption never had, and it offers a plausible answer to the question people actually ask: why did this start when it did?

What to do with this

• If your pulsatile tinnitus has not been imaged, that is the gap. A dedicated CT of the temporal bone is what finds this; it is not visible on examination and does not have a distinctive sound.
• If a scan found a sigmoid sinus abnormality, it is likely relevant and worth acting on — and worth asking how the side and the sound line up, and what else was checked.
• If it found one and you also have headaches or visual symptoms, raise intracranial pressure explicitly. The two travel together, and treating only the bone can leave the sound.
• If you have been offered surgery, the reported recurrence rate is low and the reported failure mode is untreated comorbidity. Both are reasonable to discuss before agreeing.
• If nothing was found, that is common. Sigmoid sinus anomalies explain a minority of pulsatile tinnitus, and the parent article covers the wider range of causes.

For the wider context, can tinnitus be cured sets out why a 3.5% recurrence rate is so unusual: for chronic tinnitus without a structural cause, the honest answer is that treatment reduces distress rather than removing the sound, and spontaneous remission ran at 0.8% in a study of 388 patients. This is one of the exceptions, which is exactly why it is worth identifying correctly rather than approximately.