Dihexa for Long COVID Brain Fog: Neuroinflammation, Synaptic Pruning & the Neuroplasticity Hypothesis (2026 Review)

Brain fog is now the second most commonly reported symptom of Long COVID after fatigue, affecting an estimated one in four of the UK's roughly 2.1 million Long COVID sufferers. The 2026 picture of why it happens is increasingly clear — persistent microglial activation, excessive synaptic pruning, blood-brain barrier disruption, and loss of dendritic spines in cognitive control regions. That biology overlaps directly with what Dihexa is designed to do at the synapse. This 2026 review walks through the neuroscience, the NHS treatment gap, the mechanistic case for and against synaptogenic peptides, and an honest assessment of where Dihexa actually sits in the evidence hierarchy for post-COVID cognitive recovery.

Long COVID Brain Fog in 2026: Where the UK Actually Stands

Five years after the pandemic peak, Long COVID has not gone away. The most recent ONS-aligned estimates put the number of UK adults living with self-reported Long COVID at around 2.1 million, and the symptom that consistently ranks at or near the top of the patient experience — alongside fatigue and post-exertional malaise — is cognitive impairment. Patients describe it as brain fog: difficulty concentrating, slowed processing, word-finding problems, working memory lapses, and a sense of being mentally muffled that does not respond to rest.

Across systematic reviews and meta-analyses, the combined prevalence of mental health symptoms and brain fog in Long COVID cohorts sits at roughly 20%, with brain fog specifically reported by around 25% of Long COVID patients. That means an order-of-magnitude estimate for the UK alone is in the region of half a million adults currently dealing with post-COVID cognitive impairment.

The trajectory matters. In a multi-centre study following hospitalised COVID-19 survivors, the prevalence of memory and concentration difficulties rose progressively in vulnerable subgroups, with patients three to four times more likely to report brain fog at 24 months than during the acute infection itself. Brain fog is not, for many, a self-limiting nuisance. It can be a long, slow erosion of cognitive function.

Against this scale, the NHS response in 2026 is a network of more than 100 specialist Long COVID clinics in England with parallel services in Scotland, Wales and Northern Ireland. Pathways focus on multidisciplinary rehabilitation, pacing, cognitive therapy, and treatment of comorbid sleep, autonomic and mood problems. Wait times for an initial specialist appointment are commonly six months or longer. There is no NHS-prescribed pharmacological treatment specifically licensed for post-COVID cognitive impairment.

This is the gap that drives many self-experimenters to peptides like Dihexa. The question this article addresses is whether the science supports that move — or whether the gap is being filled with hope rather than evidence.

The 2026 Biology of Long COVID Brain Fog

Translational research over the last three years has converged on a coherent — if not fully unified — model of why Long COVID produces persistent cognitive dysfunction. Several mechanisms have moved from speculation to measurable, replicable findings:

Microglial Activation and Excessive Synaptic Pruning

Microglia are the brain's resident immune cells. In healthy adults they prune unneeded synapses, support neurons, and clear cellular debris. After SARS-CoV-2 infection, microglia in vulnerable brain regions can become and remain activated — producing a chronic, low-grade neuroinflammatory state. In animal models and post-mortem human studies, this activation drives excessive synaptic pruning: the brain's own immune machinery removes functional synapses faster than they can be replaced. The downstream consequence is reduced dendritic spine density, weaker synaptic connectivity in cognitive control regions, and the slow degradation of fluid cognitive performance that patients experience as brain fog.

Blood-Brain Barrier Disruption

2024 and 2025 imaging studies in Long COVID patients with cognitive impairment found measurable disruption of the blood-brain barrier: cerebral microvasculature becomes "leaky", allowing peripheral inflammatory mediators to enter brain parenchyma. This same vascular signature is correlated with brain fog severity and persistence.

Inflammatory Cytokines Crossing Into the Brain

Long COVID patients with brain fog often show elevated peripheral and CSF levels of inflammatory cytokines — notably IL-6, TNF-α, and IL-10. Once these cross the disrupted blood-brain barrier, they disrupt monoamine synthesis (particularly dopamine in the substantia nigra), interfere with synaptic transmission, and reinforce microglial activation in a self-sustaining loop.

Anterior Cingulate Cortex Dysfunction

Functional imaging studies have repeatedly localised post-COVID cognitive complaints to dysfunction in the anterior cingulate cortex — a hub for attention, error monitoring, and cognitive control. The anterior cingulate is one of the regions richest in HGF/c-Met expression, which is part of why this becomes mechanistically interesting in the context of Dihexa.

Possible Viral Persistence and Microclots

Two more contested but increasingly cited mechanisms are the possibility of viral RNA or antigen persistence in tissue reservoirs (gut, lymph nodes, possibly brain) and the formation of fibrinaloid microclots that impair small-vessel cerebral perfusion. Both are areas of active 2026 research and remain less consensus-supported than the microglia and BBB findings.

Reduced Neurotrophic Factor Signalling

Several studies have observed lower nerve growth factor (NGF) and altered BDNF profiles in Long COVID patients with cognitive symptoms. Reduced neurotrophic support means the brain has less of its own machinery available to rebuild the synapses that microglia are pruning. The neurotrophic case for and against Dihexa specifically is covered in the Dihexa vs BDNF deep-dive.

Where Dihexa Fits the Long COVID Picture

Dihexa is a small-molecule peptide analogue derived from angiotensin IV. Its central pharmacological action is amplification of hepatocyte growth factor (HGF) / c-Met receptor signalling — a system widely expressed in the brain that drives dendritic spine formation, synaptic remodelling, and neuroprotection. The full molecular detail is in the mechanism of action guide. The relevance to Long COVID brain fog rests on three points of overlap:

• Synaptogenesis vs synaptic pruning. Dihexa promotes the formation of new dendritic spines, which is mechanistically the inverse of the spine loss driven by Long COVID's microglial pruning. In cell-culture assays Dihexa induces spines within hours; the question is whether that translates to in vivo repair under conditions of ongoing inflammatory pruning.
• HGF/c-Met as a neuroprotective and anti-inflammatory pathway. Beyond direct synaptogenesis, HGF signalling has documented anti-inflammatory effects in CNS tissue and supports oligodendrocyte function. The HGF/MET system is endogenously upregulated in response to brain injury as part of the repair response, and pharmacologically activating it may add to that endogenous signal.
• The cingulate and hippocampal expression pattern. The brain regions most implicated in Long COVID cognitive impairment — cingulate cortex, hippocampus, and prefrontal regions — are also among the regions with the highest density of c-Met receptors. A pharmacological signal at c-Met therefore lands precisely where the deficit is.

This is why Dihexa — alongside fosgonimeton, cerebrolysin, and a small handful of HGF-related compounds — comes up in self-experimentation forums for post-viral cognitive symptoms. The mechanism is the right shape for the deficit. The remaining question, throughout this article, is whether shape is enough.

The Fosgonimeton Parallel: A Telling Stand-In for Direct Trials

Because no controlled human trial of Dihexa in Long COVID exists, the most informative clinical-stage comparator is fosgonimeton (ATH-1017): a small-molecule positive modulator of the HGF/MET system developed by Athira Pharma. Fosgonimeton is not Dihexa, but it shares the core mechanism of amplifying HGF/c-Met signalling, and unlike Dihexa it has been tested in humans.

What fosgonimeton has shown in preclinical and early clinical work is directly relevant to the Long COVID question:

• Procognitive effects in neuroinflammation-induced cognitive impairment. The active metabolite of fosgonimeton enhances the HGF/MET system and demonstrates measurable cognitive benefit in animal models of inflammation-driven cognitive deficit — the mechanistic neighbour of post-COVID brain fog.
• Working memory rescue after repeated mild TBI. A 2025 bioRxiv preprint reported that HGF/MET positive modulators dose-dependently rescued working memory deficits following repeated mild traumatic brain injury — another inflammatory, synaptic-disruption model.
• Phase 1 safety data in healthy volunteers and Alzheimer's patients. Fosgonimeton has acceptable safety and PK characteristics in early human studies, supporting the broad tolerability of HGF/MET modulation as a strategy.
• The Phase 3 LIFT-AD signal. The 2024 LIFT-AD Phase 3 trial of fosgonimeton in Alzheimer's did not meet its primary cognitive endpoint. The programme has been refocused, and the negative trial is a sobering reminder that mechanism alone — even well-validated, well-tolerated mechanism — does not guarantee clinical efficacy in cognition. The fosgonimeton story is covered in detail on the dedicated fosgonimeton page.

For the Long COVID question, fosgonimeton is encouraging on biology and cautionary on clinical translation. It tells us that pharmacologically modulating HGF/MET is feasible and tolerable in humans, that the mechanism does meaningful things in inflammation-driven cognitive models, and that translation to clinical endpoints in real patient populations is not automatic. Dihexa, which lacks even the limited human safety dataset that fosgonimeton has, sits one step further from the evidence.

The ME/CFS Overlap and Why It Matters

A substantial proportion of Long COVID patients meet diagnostic criteria for myalgic encephalomyelitis / chronic fatigue syndrome (ME/CFS). The clinical overlap is striking: post-exertional malaise, autonomic dysfunction, unrefreshing sleep, cognitive symptoms with the same character — sluggish thinking, word-finding difficulty, attention failure, and a distinctive post-exertional cognitive crash. The biology is increasingly understood to overlap as well: chronic neuroinflammation, microglial activation, autonomic and immune dysregulation.

This matters for the Dihexa question in two ways:

• The broader post-viral fatigue literature applies. Anything we know — or do not know — about cognitive intervention in ME/CFS provides the closest clinical-context analogue for Long COVID brain fog. The ME/CFS field has, after years of difficult experience, moved away from graded exercise therapy (GET) as a first-line intervention, with the 2021 NICE ME/CFS guideline (NG206) explicitly de-emphasising GET in favour of pacing within an energy envelope. Any pharmacological intervention — including a synaptogenic peptide — needs to respect this principle and not push patients past their post-exertional threshold.
• "Boosting cognition" can backfire. In a population vulnerable to post-exertional symptoms, an intervention that transiently enables more cognitive output without addressing the underlying load can produce a delayed crash. This is a real, repeatedly observed pattern in stimulant trials in ME/CFS. A synaptogenic peptide is not a stimulant, but the principle of caution about any compound that increases the brain's capacity for output without addressing energy metabolism applies.

The implication is not that Dihexa must be avoided in this population — the data do not exist to make that claim — but that anyone considering it for Long COVID or ME/CFS-pattern symptoms should be especially attentive to symptom flare patterns, not press through fatigue, and stop at any signal of post-exertional worsening.

Dihexa Compared to the Current Long COVID Options

Honest comparison helps put Dihexa in context. The interventions actually being used and studied for Long COVID brain fog in 2026 fall into several categories.

Cognitive Rehabilitation and Behavioural Approaches

Constraint-Induced Cognitive Therapy (CICT) and structured cognitive rehabilitation programmes have small but measurable effects in post-COVID cognitive impairment. They are accessible through specialist Long COVID clinics and through some private neurorehabilitation services. Effect sizes are modest, but the safety profile is excellent.

Brain Stimulation

Repetitive transcranial magnetic stimulation (rTMS) over the dorsolateral prefrontal cortex and transcranial direct current stimulation (tDCS) have both shown positive effects on cognitive function in early-phase post-COVID trials. tDCS combined with cognitive rehabilitation appears to act as an inducer of neuroplasticity, amplifying the training effects of rehabilitation. Access in the UK is patchy.

Off-Label Pharmacology

Low-dose naltrexone, low-dose aripiprazole, modafinil, antihistamines (for mast-cell-related symptoms), and beta-blockers (for POTS) are all used off-label by Long COVID clinicians depending on the symptom cluster. None has a UK licence specifically for post-COVID cognitive impairment, but several have growing observational evidence in Long COVID cohorts.

Sleep, Pacing, and Comorbidities

Treating undiagnosed sleep apnoea, optimising sleep hygiene, addressing nutritional deficiencies (B12, iron, vitamin D, folate), pacing within an energy envelope, and treating co-existing depression or anxiety probably account for more cumulative Long COVID symptom improvement in the UK than any pharmacological intervention. They are unglamorous, evidence-supported, and almost always the right first step.

Research Chemicals and Peptides (Dihexa, Cerebrolysin, Semax, BPC-157, P21)

A range of unlicensed compounds is used by self-experimenters. Cerebrolysin and P21 have been mentioned in post-viral fatigue community reports as anecdotally helpful for brain fog, with no formal trials in Long COVID. Semax and Selank have stronger pedigree in Russian-language clinical literature for post-stroke and cognitive applications but limited Western-language replication. Dihexa is in this category. None of these compounds has a published, registered trial in Long COVID.

Where Dihexa sits: The least clinically validated of the synaptogenic options being discussed in Long COVID self-experimentation, but with the most precise mechanistic targeting of the spine-loss element of the brain-fog model. Mechanistic precision and clinical evidence are not the same thing.

If Someone Were Considering It: Practical Realities

This section is descriptive, not prescriptive. There is no validated dosing protocol for Dihexa in Long COVID because there is no clinical trial. What follows is what self-experimenters in this space report and the practical realities they describe — together with the caveats that should temper any inference from those reports.

• No Long COVID-specific dose. Community dosing ranges (covered in detail in the Dihexa dosage guide) were derived from cognitive-enhancement use, not from inflammatory cognitive disorders.
• Short cycles, then assess. Self-experimenters in post-viral cognitive symptoms typically describe shorter Dihexa cycles than those used purely for cognitive enhancement, because the goal is to test for any signal without sustained c-Met activation.
• Sublingual or oral. Reported routes are oral and sublingual; injectable use is reported but is associated with substantially higher risk and is not recommended (see the Dihexa Review 2026).
• Stacking caution. Combining Dihexa with other peptides or with off-label medications used in Long COVID (LDN, modafinil, antihistamines) has no documented safety data. The general stacking guide cautions explicitly against complex combinations without clinician oversight.
• Monitor, document, stop fast. Anyone trialling an unlicensed compound for Long COVID should keep a written symptom diary, log post-exertional patterns, document any adverse events, and stop at the first sign of mood destabilisation, irritability, or post-exertional crash.

None of this should be read as endorsement. The strongest single piece of practical advice for Long COVID brain fog in 2026 is this: get a proper clinical assessment first — because the cause of cognitive symptoms in any individual case may be treatable through entirely conventional routes (sleep apnoea, anaemia, B12 deficiency, hypothyroidism, depression, POTS) that are missed when a peptide gets tried first.

Specific Risks in the Long COVID Context

The general Dihexa safety profile is covered in the side effects and risks guide. Several risks become more pointed in the Long COVID context.

Sustained c-Met Activation Over Long Periods

HGF/c-Met signalling is oncogenically relevant across many tumour types. Long COVID is a chronic, often years-long condition. Repeated, sustained pharmacological c-Met activation in a patient population that may use a compound for many months or years has a very different risk-benefit profile from a one-off cognitive cycle. This is the single most underweighted risk in the self-experimentation discussion of Dihexa for chronic conditions.

Post-Exertional Symptom Provocation

Anything that transiently increases the brain's capacity for cognitive output without addressing the underlying inflammatory or energy-metabolism deficit can precipitate a post-exertional crash in the ME/CFS-pattern Long COVID phenotype. Patients describe paying for cognitive "good days" with delayed symptom flares.

Masking a Treatable Diagnosis

An unlicensed peptide that produces a non-specific lift in mood, motivation or perceived clarity can delay a clinical assessment that would otherwise identify a treatable contributor — severe sleep apnoea, undiagnosed B12 deficiency, depression, POTS, or hypothyroidism. Each of these has reliable, evidence-based treatment available on the NHS.

Interaction with Long COVID Polypharmacy

Long COVID patients are often on combinations of LDN, antihistamines, beta-blockers, antidepressants, and supplements. None of these has a documented interaction profile with Dihexa, which means the interaction risk is undefined rather than demonstrably small.

Strong Placebo Susceptibility

Trials in chronic fatigue and post-viral cognitive conditions consistently show large placebo response rates — often 30%+. Self-experimenters cannot distinguish placebo from real biological effect, and uncontrolled forum reports systematically over-represent the responders. This is a methodological reality, not a criticism of patients.

Who Should Not Consider Dihexa for Long COVID

• Anyone who has not been clinically assessed for treatable Long COVID contributors (sleep apnoea, autonomic dysfunction, deficiencies, mood, hypothyroidism).
• Anyone with a personal or family history of cancer — particularly cancers with known c-Met involvement.
• Anyone with a diagnosed bipolar or psychotic-spectrum condition.
• Anyone pregnant, breastfeeding, or trying to conceive.
• Anyone in the acute or sub-acute phase of COVID-19 itself, or with active cardiopulmonary complications of Long COVID.
• Anyone taking immunomodulators, anticoagulants, or licensed cognitive-domain medications without clinician oversight.
• Under-18s.
• Anyone with an ME/CFS-pattern post-exertional malaise phenotype that has not been brought into a stable pacing baseline first.

What the Evidence Actually Supports for Long COVID Brain Fog in 2026

For balance — and because this is where most patients should start — here is what the 2026 evidence base actually supports:

• NHS Long COVID clinic referral via GP. Multidisciplinary assessment is the gateway to most other interventions.
• Pacing within an energy envelope. The single most consistent finding in post-viral cognitive symptom management. Avoid push-crash cycles.
• Sleep optimisation and apnoea screening. Untreated sleep apnoea is a frequent occult driver of fatigue and brain fog, and is highly treatable.
• Treat POTS and orthostatic intolerance. Increased fluid and salt intake, compression, and where indicated medication (under clinician care). Cerebral hypoperfusion from POTS produces brain fog directly.
• Address nutritional deficiencies. Ferritin, B12, folate, vitamin D, and thyroid function are the basics. Correcting deficiency can improve cognitive symptoms substantially in a meaningful subgroup.
• Cognitive rehabilitation. Structured programmes through the NHS or accredited private providers.
• Treat comorbid depression and anxiety. Depression after Long COVID is common and lowers cognitive performance independently of any organic brain change. The general mood discussion is in the Dihexa for Depression & Mood article, and the overlapping hormonal picture for midlife women is covered in Dihexa for Menopause & Perimenopause Brain Fog.
• Brain stimulation (rTMS/tDCS) where available. Modest but real effect sizes; access in the UK is variable.
• Trial-led pharmacology where appropriate. Discuss off-label options (LDN, modafinil, others) with a Long COVID clinician rather than self-prescribing.
• Patient advocacy and information. The NHS Long COVID information hub, the ME Association, and Long Covid SOS provide reliable UK-specific resources.

The Bottom Line in 2026

Long COVID brain fog has, in five years, gone from a contested complaint to a well-characterised neurological syndrome with a coherent biological model: chronic microglial activation, excessive synaptic pruning, blood-brain barrier disruption, and loss of dendritic spines in cognitive control regions. That biology is exactly the kind of damage a synaptogenic peptide is designed to repair. Dihexa, with its direct activation of the HGF/c-Met spine-formation pathway, looks like one of the few small molecules whose mechanism plausibly addresses the deficit.

The catch is that nothing in the previous paragraph has been demonstrated in a Long COVID patient under controlled conditions. There is no Phase 2 trial, no published open-label case series, no UK or international clinical study of Dihexa in post-COVID cognitive impairment, no biomarker data, and no long-term safety follow-up in chronic-condition use. The closest clinical-stage relative (fosgonimeton) confirms feasibility of HGF/MET modulation in humans but missed its primary endpoint in Phase 3 Alzheimer's. Mechanistic plausibility is necessary but not sufficient.

For the half-million UK adults living with Long COVID brain fog in 2026, the honest reading of the evidence is this. Dihexa is biologically interesting and clinically unproven for this indication. The first call should be a GP and an NHS Long COVID clinic referral, not a research-chemical vendor. Pacing, sleep, nutritional correction, treatment of POTS and mood, and structured cognitive rehabilitation are the interventions with measurable human effect sizes and a regulatory framework. If there is a place for synaptogenic compounds in Long COVID care in the future, it will come through proper trials — not through individual self-experimentation in a population already vulnerable to post-exertional crashes and polypharmacy interactions.

Frequently Asked Questions