A vascular loop on an MRI report is usually a normal anatomical variant rather than the cause of tinnitus: a 2025 meta-analysis found loops no more common in ears with ordinary tinnitus than in ears without it.
The phrase tends to appear at the end of a scan ordered for one-sided tinnitus, often alongside the more important news that there is no tumour. It names something real: a small artery, such as the anterior inferior cerebellar artery, curving near or into the internal auditory canal, the bony channel that carries the hearing and balance nerve. Whether that loop matters depends much less on the picture than on what you hear and feel.
What a vascular loop on an MRI is
The nerve involved is the eighth cranial nerve, also called the vestibulocochlear or cochleovestibular nerve. Small arteries in the space beside the brainstem can loop close to the nerve or into its canal, and high-resolution MRI shows them in detail.
Radiologists can grade a loop by how far it reaches. One UK study used the Chavda classification: type 1 sits in the cerebellopontine angle, outside the canal; type 2 extends into the canal; type 3 reaches its far end. A report may also say the loop is in contact with the nerve, which is described as neurovascular contact or conflict.
None of those words, on their own, means the loop is causing symptoms.
How often loops turn up without causing symptoms
This is the core of the incidental-finding problem, and very different studies point the same way.
- The Bárány Society's 2016 criteria for vestibular paroxysmia state that MRI findings of vascular compression are seen in about 30% of healthy people without symptoms, and so are not diagnostic of the disease or of which side is affected.
- A multicentre study of 2,622 MRI scans of the cerebellopontine angle found a loop in the internal auditory canal in 18–20% of patients. Of the 118 with direct loop-nerve contact, the contact was asymptomatic in 41.5%.
- In a UK hospital, 64 of 6,978 people scanned for one-sided ear symptoms had loops (0.9%). In 25 of them the loop was on the opposite side to the main symptom. Loop grade did not predict the symptomatic side, and all 64 were reassured and discharged with no return visits over three years of follow-up.
The prevalence figures differ widely because the studies used different scanners, sequences and definitions. The direction of the evidence does not.
Do vascular loops cause tinnitus?
For ordinary tinnitus, the best evidence says no.
- A 2025 meta-analysis of 16 studies and 3,455 ears found no significant association between tinnitus and either a loop in the canal (odds ratio 0.90) or contact with the nerve (odds ratio 1.15). Its conclusion is that these findings are unrelated to undefined tinnitus, hearing loss or vertigo.
- A 2022 systematic review of 15 studies covering 11,788 patients found that 70% of the patients were in studies showing no correlation with symptoms. It concluded that loops are probably anatomical variants in most cases, with an uncommon subset causing symptoms. It adds that most authors advise MRI to look for one only when the symptoms suggest compression, not routinely.
- A study of 44 people with one-sided tinnitus and normal hearing found loops in 14, with no relationship between having a loop and scores on the Tinnitus Handicap Inventory.
Two signals stand out from the same literature. The 2025 meta-analysis found loops in the canal modestly more often in ears with sudden hearing loss (odds ratio 1.34), and a 2008 meta-analysis found contacting loops twice as often on the side of one-sided hearing loss (odds ratio 2.0). Both are associations, not proof of cause. And a multicentre MRI study reported that loops pressing directly on the nerve, with a calibre over 0.85 mm, in the middle part of the canal, were associated with tinnitus and vertigo. Loop shape may one day help separate a bystander from a culprit, but that is not yet a validated test.
When nerve compression is taken seriously
Compression becomes a credible explanation when the symptoms have a recognisable pattern, and in both of the recognised patterns the diagnosis rests on history and drug response rather than on the scan.
Vestibular paroxysmia
Vestibular paroxysmia is a disorder of frequent, very short attacks of vertigo, attributed to a blood vessel irritating the balance part of the nerve. Its Bárány Society diagnostic criteria (Strupp and colleagues, 2016) require, for a definite diagnosis:
- at least ten attacks of spontaneous spinning or non-spinning vertigo
- each lasting less than one minute
- the same pattern of attack each time in that person
- a response to carbamazepine or oxcarbazepine
- no better explanation from another diagnosis
A probable diagnosis needs at least five attacks lasting under five minutes, occurring spontaneously or with certain head movements. The presumed mechanism is abnormal electrical cross-talk between nerve fibres in the stretch of nerve closest to the brainstem, which is insulated by the same kind of cells as the brain itself. The criteria list Ménière's disease, vestibular migraine, benign positional vertigo and transient ischaemic attacks among the conditions to rule out.
A 2025 systematic review adds that attacks can come with one-sided tinnitus or sound sensitivity, that vestibular paroxysmia is diagnosed in about 3% of patients at a specialist vertigo centre, and that the average age at onset is 47 to 51. It calls a good response to sodium channel blockers the most reliable clinical sign; MRI can support the diagnosis. One randomised crossover trial has been published: in Vestparoxy, oxcarbazepine cut the number of attacks by about half compared with placebo (ratio 0.53), but only 18 of 43 randomised patients gave usable data, and the most common documented reason for dropping out was side effects. The broader picture of dizziness alongside tinnitus covers the more common causes.
Typewriter tinnitus
Typewriter tinnitus is the hearing-side version: intermittent, staccato bursts in one ear, described as a typewriter, popcorn or Morse code. In the original 2006 series, all six patients had their tinnitus fully suppressed by carbamazepine, and compression on the tinnitus side was found in four of the five imaged. In a 2023 study, 17 of 18 responded to carbamazepine or oxcarbazepine, and 7 of those 17 relapsed after stopping.
Here too the scan is the weakest clue. A 2017 MRI study of 15 carbamazepine-responding patients found nerve compression more often on the symptomatic side, but with considerable false positives and false negatives. A second series of 22 patients from the same group found no difference in loops or contact between the affected and unaffected sides, yet all 22 responded immediately to a short course of carbamazepine. Both studies conclude that careful history and the response to a first trial of the drug are more reliable than imaging.
That drug response applies to these defined syndromes only. The US clinical practice guideline recommends against anticonvulsants for the routine treatment of persistent, bothersome tinnitus; drugs for tinnitus covers that wider picture.
Vascular loops and pulsatile tinnitus
Older and newer evidence disagree. The 2008 meta-analysis found contacting loops about 80 times more often in people with pulsatile tinnitus than in those with non-pulsatile tinnitus, though the confidence interval ran from 10.9 to 821.8. A 2025 study of 337 people with unexplained one-sided pulsatile tinnitus compared the pulsing ear with the quiet one and found loops in 19.3% versus 25.2%, and nerve contact in 59.9% versus 65.6%, with no significant difference. Its authors argue against reporting such loops in pulsatile tinnitus at all.
If your tinnitus pulses, a loop should not end the search. The causes that do matter, from venous sinus narrowing to aneurysms, are covered in pulsatile tinnitus.
Microvascular decompression: what surgery offers
Microvascular decompression is an operation, usually through an opening behind the ear, that separates the vessel from the nerve. It is well established for trigeminal neuralgia and hemifacial spasm; for the eighth nerve the evidence is thinner.
- A 2017 systematic review with individual patient data pooled 35 studies and 572 patients. The authors rated the level of evidence in those studies as low. Complete relief was reported in 28% of those with tinnitus, 32% with vertigo and 62% with both. At least one complication was reported in 11%. Their conclusion: because of low success rates, the operation cannot be considered a standard treatment for tinnitus or vertigo, though people with both symptoms had better odds (odds ratio 3.8).
- Small recent series read more warmly: in one 2026 report of seven patients, five had complete resolution of tinnitus and vertigo over 1 to 10 months of follow-up. Seven patients with short follow-up cannot outweigh the pooled data.
The risks fall on the ear itself. In 183 patients having the operation mostly for trigeminal neuralgia, 17.7% had at least one ear-related complication, including permanent hearing loss in 4.17% and tinnitus after surgery in 5.21%. In a Dutch group of 55 people operated on for hemifacial spasm, 36% reported tinnitus about five years later, against 16% before surgery. An operation meant to silence tinnitus can also start it.
The 2025 vestibular paroxysmia review places surgery as the last resort for classical cases when medication is not tolerated.
What to do with the finding
- Read the whole report. A loop mentioned beside an otherwise normal scan usually means the thing the MRI was looking for, such as an acoustic neuroma, was not there. The reasons for scanning in the first place are in do you need a scan for tinnitus and the asymmetric hearing loss MRI criteria.
- Describe the rhythm. Steady ringing points away from compression. Staccato bursts, or spinning spells lasting seconds, point towards it, and are worth naming to an ENT or neurologist.
- Keep the ordinary approaches. If the loop is incidental, tinnitus in one ear and the ordinary management of tinnitus still applies, including sound therapy. The diagnosis overview sets out the usual workup.
When to get urgent help
A loop does not change the red flags. Vestibular paroxysmia attacks last seconds to minutes; severe vertigo or loss of balance that will not settle, or sudden weakness or drooping in the face or an arm, or trouble speaking or seeing, is different. For those, call your local emergency number or go to an emergency department, as stroke and tinnitus explains. Hearing that drops in one ear over hours to three days needs to be seen the same day; see sudden sensorineural hearing loss. The when to seek care tool walks through the tiers.