When Silence
Scientific Knowledge
Research stage: Early human evidence

Central Neuroinflammation

Evidence level

Human (non-randomized) evidence, with supporting animal work

What is it?

This hypothesis proposes that tinnitus is driven not only by altered neural firing but by an immune response inside the brain itself. After hearing loss, the auditory cortex's resident immune cells (microglia) switch on and release the inflammatory signalling molecule TNF-α, which retunes cortical synapses, strengthening excitation and weakening inhibition, until the circuit tips into the hyperexcitable state heard as phantom sound. The evidence is unusually direct: in mice, deleting or blocking TNF-α prevents tinnitus, depleting microglia prevents it too, and simply infusing TNF-α into the auditory cortex creates tinnitus in animals with perfectly normal hearing, showing the inflammation is a cause, not just a bystander. In people the evidence is so far correlational: a meta-analysis of nearly thirty studies finds TNF-α and IL-1β elevated with microglia and astrocytes activated, and biomarker studies show inflammatory molecules tracking tinnitus. Because it points to an off-the-shelf class of drugs, anti-inflammatories and TNF-α blockers, this account also carries a clear therapeutic implication.

  1. Noise or hearing loss
    Damages the ear and reduces input to the cortex
  2. Microglia activate, release TNF-α
    Auditory-cortex immune cells switch to an inflammatory state
  3. Excitation, inhibition imbalance
    More excitation, less inhibition (PV neurons lost)
  4. Cortex becomes hyperexcitable
    The imbalance is heard as phantom sound
  5. Blocking TNF-α or microglia
    Prevented tinnitus in mice

Mechanism

Noise-induced hearing loss provokes a neuroinflammatory response in the primary auditory cortex: proinflammatory cytokines rise and microglia transition from a resting (ramified) to an activated (amoeboid) state. Tumour necrosis factor alpha (TNF-α) is the pivotal mediator, both necessary and sufficient for tinnitus in rodent models. Genetic knockout or pharmacological suppression of TNF-α prevents the neuroinflammation and the behavioural signs of tinnitus; pharmacological depletion of microglia likewise prevents tinnitus; and infusion of TNF-α directly into the auditory cortex induces tinnitus in normal-hearing animals. Mechanistically, TNF-α drives an excitatory-to-inhibitory synaptic imbalance in cortical pyramidal neurons, increased excitatory and decreased inhibitory currents, that blocking TNF-α fully prevents. A second study links this pathway to cortical inhibition directly: elevated brain TNF-α combined with noise reduces parvalbumin-positive (PV) interneuron density and impairs auditory processing, connecting neuroinflammation to the loss of cortical inhibition described by the PV Interneuron Dysfunction hypothesis. The glial response is not limited to microglia, reactive astrocytes are also implicated. A 2026 study extended this beyond cortex, finding microglial activation and cytokine elevation in the cochlear nucleus as well, and showed blood RNA-sequencing from tinnitus patients shares the same inflammatory signature as the mouse model, direct evidence for the blood-based biomarker approach raised in the open questions below. In humans the evidence remains correlational: a systematic review and meta-analysis of 29 studies found TNF-α and IL-1β elevated and microglia and astrocytes activated in tinnitus, and biomarker studies in older adults show inflammatory markers tracking tinnitus presence, duration and type. This inflammatory account is complementary to, and proposed as an upstream molecular driver of, the central hyperexcitability that the central gain and PV interneuron hypotheses describe.

Behind the science
What is neuroinflammation, and how can the brain's immune system cause a sound?

The brain has its own immune cells, called microglia, that normally sit quietly and survey their surroundings. When the ear is damaged, these cells switch into an activated state and release inflammatory signalling molecules, chief among them TNF-α. The hypothesis is that this chemical signalling changes how auditory-cortex neurons talk to each other, tipping the balance toward too much activity. In other words, the phantom sound isn't only a wiring problem; it is partly an inflammatory one, which is why anti-inflammatory drugs are of interest as a treatment.

How do we know the inflammation causes tinnitus rather than just accompanying it?

This is the strongest part of the evidence. In mice, researchers did three things. First, they removed TNF-α, genetically or with a drug, and the animals no longer developed tinnitus after noise. Second, they went the other way and infused TNF-α straight into the auditory cortex of animals with completely normal hearing, and those animals developed tinnitus. Third, they wiped out the microglia, and again tinnitus was prevented. Being able to switch tinnitus off by removing the inflammation and switch it on by adding it back is what lets scientists call the inflammation a cause, not a coincidence.

What does TNF-α actually do to the auditory cortex?

TNF-α retunes the balance between the brain's 'accelerator' and 'brake'. After hearing loss, cortical neurons receive more excitatory input and less inhibitory input, an excitatory-to-inhibitory imbalance that makes the cortex over-reactive. Blocking TNF-α prevents this imbalance entirely. A related study showed that raised TNF-α plus noise specifically reduces parvalbumin interneurons, the cells that provide much of the cortex's inhibitory 'brake', which is exactly the failure the PV Interneuron Dysfunction hypothesis describes, giving a molecular reason why that brake fails.

Is there any evidence in people, not just mice?

Yes, though it is weaker than the animal evidence and so far correlational. A systematic review and meta-analysis pooling nearly thirty studies concluded that inflammatory molecules like TNF-α and IL-1β are elevated in tinnitus and that microglia and astrocytes are activated. Biomarker studies in older adults have found blood inflammatory markers that differ measurably with tinnitus, for example, one Portuguese study found the anti-inflammatory marker IL-10 was lower in people with tinnitus and fell further the longer the tinnitus had lasted. What is still missing is a human trial showing that reducing inflammation reduces tinnitus.

Evidence (6)

Neuroinflammation mediates noise-induced synaptic imbalance and tinnitus in rodent models.

PLoS biology · 2019

The landmark causal study: noise-induced tinnitus is driven by TNF-α and microglial activation in the auditory cortex, and blocking either one prevents it in mice.

Animal study (in vivo)
Supports Central NeuroinflammationMouse & rat

Diffusible Tumor Necrosis Factor-Alpha (TNF-α) Promotes Noise-Induced Parvalbumin-Positive (PV+) Neuron Loss and Auditory Processing Impairments.

Frontiers in neuroscience · 2020

Links neuroinflammation to the PV-interneuron hypothesis: elevated brain TNF-α plus noise causes loss of cortical parvalbumin interneurons and auditory processing deficits.

Animal study (in vivo)
Supports Central NeuroinflammationMouse

The Role of Inflammation in Tinnitus: A Systematic Review and Meta-Analysis.

Journal of clinical medicine · 2022

Systematic review and meta-analysis of 29 studies (20 human, 9 animal) concluding that TNF-α and IL-1β are elevated and glia activated in tinnitus.

ReviewSupports Central NeuroinflammationHuman & animal

Tinnitus, hearing loss and inflammatory processes in an older Portuguese population.

International journal of audiology · 2020

Human evidence: blood inflammatory markers differ measurably with tinnitus in older adults, supporting an inflammatory component in people, not just animals.

Human study (non-randomized / first-in-human)
Supports Central NeuroinflammationHuman

Astrocytes and Tinnitus.

Brain sciences · 2024

Review extending the neuroinflammatory account of tinnitus to astrocytes, the brain's other major glial cell, alongside microglia.

ReviewSupports Central NeuroinflammationN/A

Inflammatory Signatures of Tinnitus in Patients and Mice.

Neuroscience bulletin · 2026

Blood RNA-sequencing shows the same inflammatory signature in tinnitus patients and a mouse noise-exposure model, and blocking microglial activation or TNF-alpha signaling reduces tinnitus-like behavior in mice.

Animal study (in vivo)
Supports Central NeuroinflammationMouse + Human

Timeline

  1. 2019

    The landmark causal study: noise-induced tinnitus is driven by TNF-α and microglial activation in the auditory cortex, and blocking either one prevents it in mice.

  2. 2020

    Links neuroinflammation to the PV-interneuron hypothesis: elevated brain TNF-α plus noise causes loss of cortical parvalbumin interneurons and auditory processing deficits.

  3. 2020

    Human evidence: blood inflammatory markers differ measurably with tinnitus in older adults, supporting an inflammatory component in people, not just animals.

  4. 2022

    Systematic review and meta-analysis of 29 studies (20 human, 9 animal) concluding that TNF-α and IL-1β are elevated and glia activated in tinnitus.

  5. 2024

    Review extending the neuroinflammatory account of tinnitus to astrocytes, the brain's other major glial cell, alongside microglia.

  6. 2026

    Blood RNA-sequencing shows the same inflammatory signature in tinnitus patients and a mouse noise-exposure model, and blocking microglial activation or TNF-alpha signaling reduces tinnitus-like behavior in mice.

Open Questions
  • Does blocking TNF-α or dampening microglial activation reduce established tinnitus in humans, not just prevent it in mice?
  • Is the neuroinflammation a primary cause of tinnitus, or a secondary response that amplifies damage from other mechanisms?
  • Which patients have an inflammatory tinnitus subtype that would respond to anti-inflammatory treatment, and can a blood or imaging biomarker identify them?