The Epidermal Footprint of Long COVID

John Murphy, CEO, The COVID-19 Long haul Foundation

A narrative review

Abstract

Background
Post-COVID-19 condition, commonly termed long COVID, is a multisystem disorder in which symptoms persist, recur, or newly develop after acute SARS-CoV-2 infection. Although neurological, cardiopulmonary, vascular, autonomic, and constitutional manifestations have received substantial attention, the skin and its appendages constitute an important and incompletely characterized component of the syndrome. Cutaneous manifestations range from transient inflammatory eruptions and pernio-like lesions to persistent pruritus, eczematous disease, pigmentary alterations, edema, nodules, altered sweating, dysesthesia, alopecia, and histologically demonstrable small-fiber neuropathy involving epidermal and autonomic nerve fibers. The biological heterogeneity of these manifestations raises the possibility that the epidermis is not merely an external marker of systemic disease but a biologically informative tissue in which viral persistence, immune dysregulation, endothelial injury, neuroimmune dysfunction, and altered epithelial homeostasis may converge.

Methods
Evidence was reviewed from peer-reviewed clinical, pathological, immunological, genomic, dermatological, and neuropathological literature concerning SARS-CoV-2 infection, post-COVID condition, cutaneous manifestations, skin biopsy findings, epidermal nerve fibres, viral persistence, host genetics, autoimmunity, vascular injury, and treatment. Particular emphasis was placed on systematic reviews, cohort studies, tissue studies, genome-wide association studies, and contemporary mechanistic investigations.

Findings
Cutaneous disease associated with COVID-19 is phenotypically diverse. Acute disease is characterised by urticarial, maculopapular, vesicular, pernio-like, livedoid, purpuric, and vasculopathic eruptions. Some patients subsequently develop persistent or recurrent dermatological abnormalities. A 2024 systematic review of 44 studies reported post-COVID dermatological manifestations including nonspecific rashes, alopecia, pruritus, nodules, dermatitis, edema, and pigmentary changes, although substantial heterogeneity and methodological limitations prevent precise prevalence estimates. Viral antigen and RNA have been demonstrated in skin tissue from patients with long COVID, including detection of nucleocapsid protein in association with macrophages. SARS-CoV-2 entry biology is biologically plausible in keratinocytes because ACE2 is expressed in these cells, although direct productive infection of keratinocytes appears limited and remains incompletely established. Microvascular injury, complement activation, endothelial dysfunction, immune-cell trafficking, and persistent inflammatory signalling provide alternative or complementary mechanisms.

Skin punch biopsy has emerged as particularly important in a subset of patients with neuropathic long COVID. In a cohort of 977 patients, 55% reported neuropathic symptoms; among a biopsied subgroup, small-fibre abnormalities were frequent and involved epidermal and autonomic fibres. Anti-ganglioside antibodies were detected in approximately one quarter of patients with long-COVID neuropathy, suggesting that at least a subset of cutaneous sensory and autonomic abnormalities may be immune mediated.

Host genetics are also beginning to illuminate long COVID. A 2025 international genome-wide association study involving up to 15,950 individuals with long COVID and more than 1.8 million controls identified a reproducible association involving FOXP4, a transcription factor with important expression in pulmonary and immune tissues. No convincing evidence currently establishes a skin-specific long-COVID susceptibility locus. Rather, epidermal disease is likely to reflect interaction between systemic genetic susceptibility, tissue-specific gene regulation, viral and immune factors, vascular biology, and environmental triggers.

Interpretation
The skin should be regarded as a potentially informative peripheral tissue in long COVID rather than a collection of unrelated cosmetic manifestations. The epidermis, dermal microvasculature, sensory nerves, sweat glands, hair follicles, and immune cells form an integrated neurovascular-immune organ that can register systemic disturbances. Nevertheless, causal inference remains limited. Future research should combine longitudinal dermatological phenotyping with serial skin biopsies, spatial transcriptomics, single-cell sequencing, viral antigen/RNA detection, endothelial and complement profiling, autoantibody mapping, and host-genetic analysis. Such studies could establish whether distinct cutaneous phenotypes represent biomarkers of defined long-COVID endotypes and whether the skin can serve as a minimally invasive window into persistent systemic disease.

Introduction

The skin is often treated as the visible boundary of the human organism. Biologically, however, it is a highly dynamic immunological, neurological, vascular, endocrine, and epithelial organ. Its outermost compartment, the epidermis, is not an inert shell. Keratinocytes participate in innate immunity; melanocytes respond to inflammatory signals; Langerhans cells survey the cutaneous environment; sensory nerve fibres terminate within and immediately beneath the epidermis; adnexal structures regulate thermoregulation and barrier function; and the dermal microvasculature supplies a metabolically active tissue whose integrity depends on endothelial homeostasis.

This complexity makes the skin an unusually sensitive recorder of systemic disease.

SARS-CoV-2 infection exposed that property early in the pandemic. Dermatologists reported eruptions ranging from urticaria and morbilliform exanthems to vesicular lesions, acral pernio-like lesions, livedoid change, and purpura. A systematic review encompassing 86 studies and 2560 patients found pernio-like lesions to be the most frequently reported cutaneous manifestation, followed by erythematous maculopapular eruptions and viral exanthems.[1] These observations established that COVID-19 could affect the integumentary system.

The subsequent emergence of long COVID raised a more difficult question.

If SARS-CoV-2 can produce acute cutaneous disease, can infection also leave behind persistent injury or altered function within the skin?

The answer increasingly appears to be yes—but with an important qualification. The evidence does not support a single dermatological disease called “long-COVID skin disease”. Instead, cutaneous manifestations appear to constitute a heterogeneous collection of phenotypes generated by several interacting biological processes.

Some lesions are probably consequences of the acute inflammatory illness and gradually resolve. Some may represent immune-mediated disease triggered by infection. Some may reflect vascular or endothelial dysfunction. Some may be manifestations of autonomic dysfunction. Others may be secondary to systemic illness, medications, nutritional disturbances, immobilization, altered hormonal physiology, or psychological stress. In a subset of patients, persistent viral antigen or RNA may contribute.

This distinction is essential.

The presence of a rash after COVID-19 does not establish that SARS-CoV-2 continues to infect the skin. Conversely, the absence of demonstrable viral replication does not exclude a biologically meaningful postinfectious process. The modern concept of long COVID is increasingly one of persistent or dysregulated host biology rather than simply prolonged viral infection.

WHO defines post-COVID condition as a syndrome occurring in individuals with probable or confirmed SARS-CoV-2 infection, usually three months after onset, with symptoms lasting at least two months and not explained by an alternative diagnosis. Symptoms may persist from acute illness or emerge after apparent recovery, and they may fluctuate or relapse.[2] Contemporary WHO estimates suggest that approximately 6% of people infected with SARS-CoV-2 develop post-COVID condition, although estimates vary considerably according to population, variant, vaccination status, case definition, and follow-up interval.[3]

Within that broad syndrome, the skin has received disproportionately little mechanistic attention.

The purpose of this review is therefore not simply to catalogue rashes. It is to examine the epidermis as a potential biological interface between SARS-CoV-2, the immune system, the microvasculature, peripheral nerves, and the systemic mechanisms of long COVID.

1. The epidermis as an immunological and neurovascular organ

The epidermis is composed predominantly of keratinocytes arranged in a stratified architecture extending from the basal layer through the spinous and granular layers to the cornified surface. The basal layer contains proliferating keratinocytes, melanocytes, and Merkel cells. Langerhans cells provide antigen surveillance. The stratum corneum creates the principal physical permeability barrier.

Keratinocytes are also immunologically active cells.

They express pattern-recognition receptors and can produce cytokines, chemokines, antimicrobial peptides, and other mediators of innate immunity. Interactions among keratinocytes, dendritic cells, lymphocytes, mast cells, endothelial cells, and sensory neurons allow the skin to convert local and systemic inflammatory signals into clinically visible changes.

The epidermis is additionally innervated by unmyelinated C fibres and thinly myelinated Aδ fibres. Intraepidermal nerve fibres transmit nociceptive and thermal information and participate in neurogenic inflammation. Their integrity can be quantified through skin punch biopsy, making the skin unusually useful for investigating peripheral small-fibre disease.

This becomes important in long COVID because neuropathic symptoms—including burning, electric sensations, paresthesias, dysesthesia, altered temperature perception, and pruritus—are common in some patient populations.

The skin therefore represents several biological systems simultaneously:

an epithelial barrier;
an immune surveillance organ;
a vascular interface;
a sensory organ;
an autonomic target;
an endocrine and metabolic compartment;
a reservoir for resident immune cells and microbial communities; and
a readily accessible tissue for molecular and histological investigation.

A persistent alteration in any one of these systems could generate cutaneous symptoms. A disorder affecting several simultaneously could produce the complex phenotype observed in some patients with long COVID.

2. SARS-CoV-2 and the skin: what was established during acute infection?

The initial dermatological literature was characterised by case reports and small series. As larger registries and systematic reviews emerged, several recurring phenotypes became apparent.

The principal acute cutaneous categories include:

urticarial eruptions;
morbilliform or maculopapular exanthems;
papulovesicular eruptions;
pernio-like acral lesions;
livedoid and necrotic lesions;
purpuric eruptions;
petechiae;
vasculitic or vasculopathic lesions;
nonspecific erythema;
vesicles and pustules; and
exacerbation or emergence of inflammatory dermatoses.

A comprehensive systematic review identified 2560 patients with dermatological manifestations and found pernio-like lesions to be the most frequently reported phenotype.[1]

These categories are clinically useful but biologically heterogeneous.

A urticarial eruption is fundamentally different from a thrombotic vasculopathy. A pernio-like lesion is different from telogen effluvium. A neuropathic burning sensation without visible inflammation is different from eczema. Grouping them under the single label “COVID rash” obscures rather than clarifies pathogenesis.

This lesson is equally important in long COVID.

3. From acute dermatological disease to persistent epidermal dysfunction

The central question is whether persistent cutaneous disease represents:

(A) continuation of acute SARS-CoV-2 infection;

(B) a postinfectious immune disorder;

(C) persistent vascular dysfunction;

(D) autonomic or small-fibre neuropathy;

(E) secondary consequences of systemic long COVID;

(F) unrelated coincidental dermatological disease; or

(G) some combination of all six.

Current evidence favours the final model.

A 2024 systematic review of 44 studies identified substantial dermatological morbidity after COVID-19. Reported manifestations included nonspecific skin rash, alopecia, pruritus, subcutaneous nodules, dermatitis, edema, and pigmentary abnormalities. The review also identified reports of autoantibodies directed against epidermal targets.[4]

However, the reported prevalences should not be interpreted as population-level estimates of long-COVID skin disease. The underlying studies differed greatly in case definition, ascertainment, follow-up duration, severity of acute infection, vaccination status, variant, and whether dermatological findings were clinician-confirmed.

The evidence is therefore stronger for existence of persistent cutaneous phenotypes than for their precise prevalence.

4. Viral persistence: a possible epidermal reservoir?

One of the most provocative hypotheses in long COVID is that SARS-CoV-2 antigen or RNA can persist within tissues after apparent clearance from the respiratory tract.

This hypothesis is supported by several tissue studies, although the presence of residual viral material should not automatically be equated with productive infection.

A particularly relevant study examined tissues from two patients with long COVID and detected SARS-CoV-2 nucleocapsid protein in appendix, skin, and breast tissue 163 and 426 days after symptom onset. Viral RNA was also detected, and nucleocapsid protein colocalised with CD68-positive macrophages.[5]

This observation has several possible interpretations.

First, skin can contain residual viral material long after acute disease.

Second, the detected material may reside within immune cells rather than actively infected keratinocytes.

Third, persistent antigen could theoretically sustain local immune activation even without productive viral replication.

Fourth, antigen persistence elsewhere in the body could generate systemic immune signals capable of altering epidermal function.

The distinction between viral persistence and viral replication is critical.

Detection of RNA by sensitive molecular methods can reflect fragmented or noninfectious nucleic acid. Immunohistochemical detection of nucleocapsid protein can reflect residual antigen. Demonstration of negative-strand RNA, subgenomic RNA, replication complexes, infectious virions, or increasing viral burden over time would provide substantially stronger evidence of active replication.

At present, persistent SARS-CoV-2 material in skin should therefore be regarded as a plausible biological observation rather than proof that the epidermis constitutes a chronic infectious reservoir.

5. Can SARS-CoV-2 directly infect keratinocytes?

The question of direct epidermal infection has produced apparently conflicting observations.

ACE2, the principal SARS-CoV-2 receptor, has been detected in human keratinocytes. Early work demonstrated substantial ACE2 expression and proposed that the skin could represent a viral target.[6]

More recent experimental work provides a more nuanced picture. Human keratinocytes express ACE2, and ACE2 expression increases during differentiation and Toll-like receptor 3 activation. However, TMPRSS2, an important protease for plasma-membrane fusion, appears absent from keratinocytes in the experimental systems studied, while mature cathepsin L is expressed.[7]

This suggests that keratinocytes possess some molecular machinery required for viral entry but may not support efficient infection through the canonical TMPRSS2-dependent pathway.

Thus, the epidermal model is unlikely to be as simple as:

SARS-CoV-2 → keratinocyte infection → chronic rash.

A more plausible model is:

SARS-CoV-2 exposure → epithelial sensing ± limited cellular infection → innate immune activation → endothelial and neural effects → altered epidermal homeostasis.

The distinction matters because it changes therapeutic strategy. If persistent cutaneous disease is predominantly antigen-driven immune pathology, prolonged antiviral therapy might have limited value once replication has ceased. If replication persists in a defined tissue reservoir, however, antiviral therapy becomes biologically attractive.

This is one of the most important unanswered questions in long-COVID dermatology.

6. Endothelial dysfunction and the epidermal microcirculation

The skin is exceptionally dependent on its microcirculation.

The dermal vascular network supplies oxygen and nutrients, regulates temperature, supports wound healing, and communicates continuously with epidermal and immune cells.

SARS-CoV-2 infection can produce profound endothelial activation and microvascular injury. In severe COVID-19, skin biopsies from patients with purpuric lesions demonstrated pauci-inflammatory thrombogenic vasculopathy with deposition of C5b-9 and C4d. Spike protein was found to colocalise with complement components in the cutaneous microvasculature in some cases.[8]

These findings established a mechanistic connection between COVID-19 and cutaneous microvascular injury.

Whether the same mechanism persists for months or years in long COVID is much less certain.

Nevertheless, long COVID has been associated with abnormalities involving endothelial function, platelet activation, coagulation, and complement pathways.[9,10] A persistent microvascular state could theoretically produce:

altered epidermal oxygenation;
impaired thermoregulation;
delayed tissue repair;
livedoid change;
acral discoloration;
edema;
altered sensation;
hair follicle dysfunction; and
increased susceptibility to inflammatory injury.

The epidermis is therefore potentially a visible endpoint of a systemic vascular disorder.

7. Complement activation

Complement is increasingly recognised as an important component of COVID-19 vascular pathology.

In severe acute disease, complement activation can damage endothelial cells, promote leukocyte recruitment, increase vascular permeability, and amplify thrombosis. C5b-9 deposition in cutaneous microvessels provides direct pathological evidence of complement activation in skin.[8]

Long COVID studies have also identified persistent abnormalities involving complement and coagulation systems, although these abnormalities vary substantially between patient groups.

A mechanistic sequence can therefore be proposed:

SARS-CoV-2 infection
→ endothelial activation
→ complement activation
→ microvascular injury
→ platelet and coagulation activation
→ impaired tissue perfusion
→ epithelial and neural dysfunction.

This sequence is biologically plausible but should not yet be regarded as a universal explanation for long-COVID skin disease.

8. Immune dysregulation and autoimmunity

A second major mechanism is immune dysregulation.

Long COVID is associated in different studies with altered T-cell populations, B-cell abnormalities, persistent inflammatory signalling, autoantibodies, and impaired immune regulation.[9]

The skin is particularly susceptible to immune-mediated injury because it contains a dense network of resident and recruited immune cells.

Potential mechanisms include:

molecular mimicry;
epitope spreading;
bystander activation;
persistent antigenic stimulation;
aberrant B-cell activation;
autoreactive T-cell responses;
mast-cell activation;
complement activation; and
altered cytokine signalling.

Recent research has strengthened the broader concept of tissue-specific autoimmunity in long COVID. A 2026 study of 114 long-COVID patients identified tissue-directed autoantibodies more frequently than in historical controls, with particularly notable vascular autoreactivity and persistent IgM-dominant responses.[11]

Although this study did not specifically establish epidermal autoimmunity, its significance for cutaneous disease is substantial: it demonstrates that conventional ANA testing may fail to detect clinically relevant tissue-directed immune responses.

The implication is that a normal ANA does not exclude an immune-mediated post-COVID disorder.

9. Epidermal autoantibodies

Reports of antibodies directed against epidermal structures deserve particular attention.

The 2024 systematic review of post-COVID dermatological disease identified studies reporting epidermal-targeting autoantibodies, with substantial variation by viral era.[4]

These findings should not be overinterpreted.

An autoantibody can be:

pathogenic;
an epiphenomenon;
a marker of tissue injury;
a consequence of altered immune tolerance; or
unrelated to the patient’s cutaneous phenotype.

Demonstrating pathogenicity requires more than detecting an antibody. Ideally, investigators should establish:

the antigenic target;
antibody specificity;
tissue localisation;
temporal relationship to disease;
correlation with clinical phenotype;
functional effects on target cells;
transferability in experimental models; and
improvement after selective antibody depletion or immune intervention.

This distinction is essential before epidermal autoantibodies can become clinical biomarkers.

10. The neurocutaneous dimension: epidermal small-fibre injury

Perhaps the most important development in long-COVID dermatology is the recognition that the skin can reveal peripheral nerve injury.

Small-fibre neuropathy affects thinly myelinated Aδ fibres and unmyelinated C fibres. These fibres mediate pain, temperature, itch, and autonomic function.

Because their distal endings extend into the epidermis, their density can be measured with immunohistochemical skin biopsy.

A large 2025 cohort involving 977 patients with long COVID found neuropathic symptoms in 55% of participants. Among 85 patients who underwent skin biopsy, small-fibre abnormalities were common, affecting both epidermal and autonomic nerve fibres. Approximately 25% of patients with long-COVID neuropathy had anti-ganglioside antibodies, and a small pilot treatment group receiving intravenous immunoglobulin reported improvement in neuropathic symptoms.[12]

These observations substantially broaden the meaning of an “epidermal lesion”.

A patient may have no visible rash whatsoever while the epidermis contains measurable structural evidence of neurological disease.

This produces at least three distinct dermatological phenotypes:

10.1 Visible inflammatory disease

The skin is visibly abnormal.

Examples include:

erythematous plaques;
papules;
urticaria;
vesicles;
dermatitis;
pernio-like lesions;
purpura;
livedoid change.

10.2 Structural epidermal disease

The skin may appear relatively normal but biopsy reveals:

reduced intraepidermal nerve fibre density;
abnormal nerve morphology;
inflammatory infiltrates;
vascular alterations;
other microscopic changes.

10.3 Neurocutaneous disease

The patient experiences:

burning;
prickling;
electric sensations;
abnormal temperature perception;
neuropathic itch;
dysesthesia;
allodynia;
altered sweating.

The absence of a visible rash therefore cannot be used to exclude clinically significant cutaneous involvement.

11. Autonomic dysfunction and the sweat apparatus

The skin is a major autonomic organ.

Sweat glands, arrector pili muscles, cutaneous vessels, and thermoregulatory mechanisms are under autonomic control. Long COVID frequently includes autonomic symptoms such as orthostatic intolerance, altered sweating, temperature dysregulation, palpitations, and episodic flushing or chills.

Skin biopsy can evaluate autonomic nerve fibres around sweat glands.

The large neuropathy cohort described above found abnormalities in autonomic fibres as well as epidermal sensory fibres.[12]

This raises an important conceptual possibility: some apparently dermatological symptoms of long COVID may actually be manifestations of systemic autonomic dysfunction.

For example:

excessive sweating;
reduced sweating;
heat intolerance;
cold intolerance;
episodic flushing;
altered skin temperature;
dependent discoloration; and
neuropathic itch

may share a common autonomic-neurovascular substrate.

This interpretation prevents misclassification of autonomic manifestations as primary inflammatory dermatoses.

12. Pruritus: inflammatory itch versus neuropathic itch

Pruritus is one of the most frequently reported persistent cutaneous symptoms.

The term, however, describes a sensation rather than a disease mechanism.

Itch can arise through:

histaminergic pathways;
non-histaminergic inflammatory pathways;
cytokine signalling;
mast-cell activation;
peripheral neuropathy;
central sensitisation;
xerosis;
cholestatic disease;
renal disease;
hematological disease;
medication effects; or
psychophysiological mechanisms.

Consequently, a patient with persistent post-COVID itch should not automatically be labelled as having an allergic eruption.

Neuropathic itch is particularly important.

If small-fibre injury is present, antihistamines may provide little benefit because the fundamental mechanism is neural rather than histaminergic.

Clinical clues to neuropathic itch include:

burning or electric quality;
dysesthesia;
absence of primary lesions;
sharply localised or unusual distribution;
temperature-triggered exacerbation;
coexistence of numbness or paresthesia;
poor response to conventional antihistamines.

Skin biopsy may be appropriate in selected patients when small-fibre neuropathy is suspected.

13. Eczematous and inflammatory dermatoses

Persistent or newly developed dermatitis has been described after COVID-19.

Possible mechanisms include immune dysregulation, altered epidermal barrier function, cytokine signalling, psychological stress, medication exposure, environmental factors, and disruption of the skin microbiome.

It is important to distinguish:

true postinfectious inflammatory disease;
exacerbation of pre-existing atopic dermatitis;
irritant contact dermatitis;
allergic contact dermatitis;
seborrhoeic dermatitis;
psoriasis;
drug eruption; and
nonspecific xerosis.

COVID-era behavioural changes—particularly frequent hand washing and alcohol-based sanitisation—also produced substantial irritant dermatitis and cannot simply be attributed to SARS-CoV-2 biology.

A Lancet-quality analysis should therefore resist the temptation to assign every new eczematous eruption to long COVID.

The correct approach is clinical phenotyping followed by conventional dermatological diagnosis.

14. Pernio-like lesions and microvascular disease

“COVID toes” became one of the most recognisable dermatological manifestations of the pandemic.

Pernio-like lesions typically present as erythematous-to-violaceous macules, papules, plaques, edema, burning, or pruritus involving acral surfaces.

Histological investigations demonstrated vascular alterations and strong type I interferon responses in some patients.[13]

Interestingly, many patients with pernio-like lesions were not systemically ill and had negative conventional viral testing.

The phenomenon remains biologically complex.

A strong type I interferon response could theoretically suppress viral replication while simultaneously producing vascular and inflammatory skin changes. This could partly explain why some patients with striking cutaneous lesions had little systemic disease.

Whether persistent pernio-like lesions constitute a long-COVID phenotype remains uncertain. Most acute lesions resolve. Chronic or recurrent acral vascular symptoms should prompt evaluation for other causes, including autoimmune disease, cryoglobulinemia, antiphospholipid syndrome, cold exposure, vasculitis, and peripheral vascular disease.

15. Livedoid, purpuric, and necrotic injury

Livedoid and purpuric lesions deserve particular caution because they may signify vascular disease rather than benign inflammation.

Severe COVID-19 can produce endothelial injury, platelet activation, complement activation, and microvascular thrombosis. Skin may reveal this systemic vascular process.

The pathology of COVID-associated purpura has included thrombogenic vasculopathy and complement deposition.[8]

In long COVID, persistent livedoid change should not automatically be interpreted as proof of ongoing SARS-CoV-2 vascular injury.

Alternative diagnoses include:

antiphospholipid syndrome;
vasculitis;
cryoglobulinemia;
cholesterol embolisation;
peripheral arterial disease;
venous insufficiency;
medication-associated purpura;
hematological disorders; and
systemic autoimmune disease.

The appearance of necrosis, rapidly progressive purpura, painful retiform lesions, or tissue loss warrants urgent medical assessment.

16. Alopecia: the most recognisable persistent appendageal manifestation

Hair is technically an appendage rather than epidermis, but its biology is intimately connected to the epidermal unit and makes alopecia one of the most important long-term cutaneous consequences of COVID-19.

Telogen effluvium is particularly well established.

In telogen effluvium, systemic stress causes a disproportionate number of follicles to transition prematurely from anagen into catagen and telogen. Hair shedding then becomes apparent weeks to months later.

A systematic review identified 465 cases of acute telogen effluvium after COVID-19, with onset averaging approximately 74 days after symptom onset. Most patients recovered, although some experienced persistent shedding.[14]

Another systematic scoping review identified telogen effluvium as the dominant form of post-COVID hair loss and reported a median onset around two months and substantial subsequent recovery.[15]

The mechanisms are likely multifactorial:

fever;
systemic inflammation;
cytokine excess;
metabolic stress;
hypoxia;
nutritional deficiency;
psychological stress;
medication exposure;
endocrine disruption; and
direct effects on the follicular microenvironment.

Importantly, telogen effluvium is generally non-scarring.

Persistent hair loss should therefore trigger evaluation for alternative or superimposed diagnoses, including:

androgenetic alopecia;
alopecia areata;
iron deficiency;
thyroid disease;
nutritional deficiency;
androgen excess;
chronic systemic illness;
medication-induced alopecia; and
scarring alopecia.

17. Other forms of alopecia

COVID-19 has also been temporally associated with alopecia areata and other forms of hair loss.

Alopecia areata represents an autoimmune attack on anagen hair follicles and therefore provides a particularly interesting model of postviral immune dysregulation.

Whether SARS-CoV-2 directly triggers follicular autoimmunity remains uncertain.

The distinction between telogen effluvium and alopecia areata is clinically crucial:

Feature	Telogen effluvium	Alopecia areata
Pattern	Diffuse	Focal or patchy
Scarring	No	No
Hair pull	Often positive	Variable
Exclamation-point hairs	No	Characteristic
Follicular inflammation	Minimal	Peribulbar lymphocytic
Typical course	Reversible	Variable
Treatment	Usually trigger correction/support	Dermatological immune therapy

Long COVID may therefore act as a systemic trigger rather than directly destroying hair follicles.

18. Pigmentary changes

Postinflammatory hyperpigmentation and hypopigmentation have been reported after COVID-19.

Pigmentary change may arise from:

inflammatory injury;
melanocyte dysfunction;
epidermal turnover abnormalities;
medication-related photosensitivity;
postinflammatory melanogenesis;
vascular changes; or
pre-existing pigmentary disorders.

The biological significance of pigmentation abnormalities in long COVID remains unclear.

Nevertheless, melanocytes are neuroectodermal cells and interact closely with keratinocytes and immune mediators. Future spatial studies may reveal whether persistent pigmentary abnormalities represent a local footprint of inflammatory signalling.

19. Edema and cutaneous fluid imbalance

Edema has been reported in post-COVID dermatological cohorts.[4]

The causes are likely heterogeneous.

Potential contributors include:

endothelial permeability;
venous insufficiency;
lymphatic dysfunction;
immobility;
autonomic dysfunction;
cardiac disease;
renal disease;
hepatic disease;
medications; and
systemic inflammatory states.

Edema should therefore never be labelled “long-COVID skin disease” without evaluating systemic causes.

This is particularly important because lower-extremity edema can be a manifestation of cardiovascular, renal, hepatic, venous, or lymphatic disease.

20. Epidermal barrier dysfunction

The stratum corneum depends on an organised mixture of corneocytes, ceramides, cholesterol, free fatty acids, natural moisturising factors, and intercellular lipid structures.

Disruption of this system causes increased transepidermal water loss, xerosis, scaling, fissuring, irritation, and increased susceptibility to dermatitis.

There are several plausible routes through which post-COVID illness might disrupt barrier function:

inflammatory cytokines altering keratinocyte differentiation;
systemic nutritional deficiency;
endocrine disruption;
autonomic changes affecting sweating;
altered skin microbiota;
repeated washing and sanitisation;
medication exposure;
reduced physical activity;
chronic stress;
immune-mediated epidermal injury.

Direct evidence for a persistent long-COVID-specific epidermal barrier defect remains limited.

This is a major research gap.

Objective measurement of transepidermal water loss, corneometry, lipid composition, filaggrin expression, involucrin, loricrin, claudins, and epidermal transcriptomic signatures should become part of future mechanistic studies.

21. Keratinocyte biology and interferon signalling

Keratinocytes respond strongly to interferons and viral pattern-recognition pathways.

SARS-CoV-2 exposure can therefore affect epidermal biology without requiring extensive productive infection.

Type I interferon signalling may:

increase antiviral gene expression;
alter keratinocyte differentiation;
recruit immune cells;
modify chemokine production;
affect vascular signalling;
influence melanocyte behaviour; and
alter sensory-neural interactions.

This provides an attractive explanation for why persistent skin disease could occur despite limited direct viral replication.

The epidermis may be responding to a systemic cytokine environment rather than serving as a primary viral target.

22. Genomics of long COVID

The genetic architecture of long COVID is beginning to emerge, but it is not yet dermatologically specific.

A 2025 international genome-wide association study analysed up to 6450 long-COVID cases and more than one million population controls in the discovery analysis and replicated the principal association in an independent dataset. The study identified a genome-wide significant association involving FOXP4, a transcription factor implicated in lung physiology and immune biology.[16]

The larger international analysis ultimately incorporated 15,950 individuals with long COVID and 1,892,830 controls across 33 cohorts and 19 countries.[16]

The lead FOXP4 association was not simply explained by acute COVID-19 severity. The risk allele was associated with long COVID even after analyses accounting for hospitalization and disease severity, suggesting that susceptibility to persistent disease may involve biological pathways distinct from those determining acute severity.[16]

This is important for dermatology.

It demonstrates that long COVID has a measurable host-genetic component.

But FOXP4 should not presently be described as a gene for long-COVID skin disease.

There is no convincing evidence that FOXP4 specifically determines epidermal manifestations.

Rather, it illustrates a broader principle: long COVID likely reflects genetically influenced variation in immune, epithelial, vascular, pulmonary, and metabolic responses to infection.

23. The need for skin-specific genomics

The next stage of research should move beyond blood-based GWAS.

The skin has its own epigenomic and transcriptional landscape.

A patient may possess a systemic susceptibility allele whose clinical expression depends on:

epidermal chromatin state;
keratinocyte differentiation;
local cytokine exposure;
tissue-resident immune cells;
endothelial phenotype;
neural innervation;
microbiome;
ultraviolet exposure;
age; and
sex hormones.

Consequently, a conventional GWAS may identify susceptibility but not explain phenotype.

Future research should combine:

Genome-wide association

with

single-cell RNA sequencing

plus

single-cell ATAC sequencing

plus

spatial transcriptomics

plus

proteomics

plus

epigenomics

plus

skin microbiome analysis

plus

viral antigen mapping.

The ultimate objective would be to identify molecularly defined cutaneous endotypes.

24. Candidate genomic pathways

Although skin-specific long-COVID loci remain undefined, several biological pathways deserve investigation.

24.1 ACE2 pathway

Variation in ACE2 expression or regulation could theoretically alter epithelial susceptibility and renin–angiotensin signalling.

24.2 TMPRSS2 and cathepsin pathways

Differences in viral entry machinery could influence tissue susceptibility.

24.3 Interferon pathways

Genes controlling type I and III interferon responses could alter viral clearance and postinfectious inflammation.

24.4 HLA and antigen presentation

HLA variation could affect viral peptide presentation and subsequent immune persistence.

24.5 Complement genes

Genetic variation in complement components and regulators could influence vascular injury.

24.6 Autoimmune susceptibility loci

Variants affecting B-cell activation, T-cell regulation, and immune tolerance could predispose to persistent immune-mediated disease.

24.7 Keratinocyte differentiation genes

Genes involving filaggrin, loricrin, involucrin, keratins, desmosomes, and epidermal lipid metabolism may modify the response to systemic inflammatory stress.

24.8 Neuroimmune genes

Genes regulating small-fibre function, ion channels, neuropeptides, and autonomic signalling may influence neuropathic cutaneous manifestations.

These are research hypotheses rather than established clinical biomarkers.

25. Epigenetic mechanisms

An important frontier is epigenetics.

Viral infection and inflammation can alter:

DNA methylation;
histone modification;
chromatin accessibility;
noncoding RNA;
enhancer activity; and
transcription-factor networks.

Because keratinocytes continuously differentiate and renew, persistent epigenetic alterations could theoretically generate prolonged changes after the original inflammatory trigger has subsided.

A hypothetical model is:

SARS-CoV-2 infection
→ inflammatory signalling
→ altered chromatin state
→ persistent keratinocyte transcriptional changes
→ altered barrier, immune, or repair function.

Evidence sufficient to establish this mechanism in long-COVID epidermis does not yet exist.

But it is experimentally testable.

26. Clinical diagnosis

There is no validated laboratory test that establishes long-COVID skin disease.

Diagnosis is therefore clinical and exclusionary.

The clinician should establish:

documented or probable previous SARS-CoV-2 infection;
timing of onset of skin symptoms;
persistence or recurrence;
morphology;
distribution;
associated systemic symptoms;
medication exposures;
previous dermatological history;
systemic disease;
laboratory abnormalities;
evidence for autoimmune disease;
evidence for vascular disease;
neurological symptoms; and
alternative dermatological diagnoses.

The diagnosis should be framed as:

post-COVID dermatological manifestation

rather than assuming a single pathogenesis.

27. Morphological examination

The basic dermatological examination remains indispensable.

Documentation should include:

primary lesion;
secondary lesion;
colour;
scale;
border;
surface;
induration;
tenderness;
warmth;
blanching;
ulceration;
necrosis;
distribution;
symmetry;
mucosal involvement;
nail involvement;
scalp involvement;
hair density; and
temporal progression.

High-resolution photographs can provide valuable longitudinal evidence.

Dermoscopy may assist in:

alopecia;
pigmentary lesions;
inflammatory disease;
vascular lesions; and
subtle structural abnormalities.

28. Laboratory evaluation

Laboratory testing should be phenotype-directed rather than indiscriminate.

Potential studies include:

complete blood count;
comprehensive metabolic panel;
liver enzymes;
renal function;
ferritin;
iron studies;
thyroid function;
vitamin B12;
folate;
vitamin D where clinically indicated;
inflammatory markers;
autoimmune testing when indicated;
complement studies;
serum protein electrophoresis where appropriate;
infectious testing when clinically indicated; and
targeted coagulation evaluation for suspected vascular disease.

Testing should be adapted to the patient.

A patient with diffuse hair loss requires a different investigation from one with purpura or neuropathic burning.

29. Skin biopsy

Skin biopsy is not required for routine long-COVID skin disease.

It becomes particularly useful when:

vasculitis is suspected;
inflammatory dermatosis is uncertain;
autoimmune blistering disease is possible;
lesions are persistent or atypical;
malignancy must be excluded;
cutaneous vasculopathy is suspected; or
small-fibre neuropathy is being evaluated.

For suspected small-fibre neuropathy, a standardized punch biopsy with quantification of intraepidermal nerve fibre density can provide objective evidence.

The biopsy can also be used experimentally for:

immunohistochemistry;
viral antigen detection;
RNA analysis;
complement deposition;
endothelial markers;
immune-cell phenotyping;
spatial transcriptomics; and
proteomics.

This makes skin an unusually attractive tissue for mechanistic long-COVID research.

30. Differential diagnosis

The diagnosis of long-COVID skin disease requires disciplined exclusion of other causes.

Important mimics include:

Inflammatory

atopic dermatitis;
contact dermatitis;
psoriasis;
lichen planus;
urticaria;
drug eruption.

Autoimmune

lupus;
dermatomyositis;
vasculitis;
antiphospholipid syndrome;
autoimmune blistering disease.

Infectious

fungal infection;
bacterial infection;
herpesviruses;
scabies;
secondary syphilis.

Vascular

venous insufficiency;
peripheral arterial disease;
cholesterol embolisation;
cryoglobulinemia;
thrombotic disease.

Neurological

small-fibre neuropathy;
diabetic neuropathy;
nutritional neuropathy;
medication-induced neuropathy.

Systemic

thyroid disease;
renal disease;
hepatic disease;
hematological disease;
nutritional deficiency.

This differential is particularly important because long COVID can coexist with conventional disease.

31. Treatment: principles rather than a single regimen

There is currently no established disease-modifying therapy specifically approved for long-COVID epidermal disease.

WHO continues to emphasise individualized symptomatic care and rehabilitation for post-COVID condition.[3]

Treatment should therefore be phenotype-driven.

The therapeutic hierarchy is:

identify the lesion → identify the mechanism → treat the mechanism → reassess objectively.

32. Treatment of inflammatory dermatitis

For eczematous disease:

emollients;
barrier repair;
avoidance of irritants;
topical corticosteroids when indicated;
topical calcineurin inhibitors;
treatment of secondary infection;
targeted systemic therapy in severe conventional dermatoses.

There is no evidence that all post-COVID dermatitis requires systemic immunosuppression.

The dermatologist should treat the actual dermatosis rather than the presumed viral trigger.

33. Treatment of urticaria

Persistent urticaria should generally be managed according to established chronic urticaria principles.

Second-generation H1 antihistamines are usually first-line.

For refractory chronic spontaneous urticaria, established specialist therapies such as omalizumab may be considered according to standard criteria.

The presence of previous COVID-19 does not by itself justify prolonged systemic corticosteroid treatment.

34. Treatment of pruritus

Treatment depends on mechanism.

For histaminergic itch:

H1 antihistamines may help.

For eczematous itch:

anti-inflammatory topical treatment;
moisturisation;
barrier repair.

For neuropathic itch:

gabapentinoids;
selected antidepressants;
topical/local approaches;
specialist neurological management.

For systemic itch:

treatment of the underlying renal, hepatic, hematological, endocrine, or inflammatory disease.

Long-COVID-associated itch should therefore be treated as a symptom with multiple possible mechanisms rather than a single disease.

35. Treatment of telogen effluvium

The principal intervention is correction of the triggering systemic condition.

Patients should be reassured that telogen effluvium is usually non-scarring and often reversible.

Evaluation may include:

iron status;
thyroid function;
nutritional status;
systemic inflammatory disease;
medication review.

Minoxidil may be considered selectively, particularly when another form of alopecia coexists, but it should not be represented as a proven disease-modifying therapy for post-COVID telogen effluvium.

36. Treatment of alopecia areata

If alopecia areata is established, treatment should follow conventional dermatological practice.

Depending on severity, options may include:

topical corticosteroids;
intralesional corticosteroids;
topical immunotherapy;
systemic therapies;
JAK inhibitors in appropriate patients.

Whether post-COVID alopecia areata responds differently from conventional disease remains unknown.

37. Treatment of small-fibre neuropathy

This is one of the most challenging areas.

Treatment should include:

identification of metabolic causes;
evaluation for autoimmune disease;
control of diabetes when relevant;
medication review;
neuropathic pain therapy;
autonomic management;
physical and occupational rehabilitation where appropriate.

Intravenous immunoglobulin has generated interest.

The 977-patient cohort found anti-ganglioside antibodies in approximately one quarter of long-COVID neuropathy patients, and eight patients in a pilot treatment group reported improvement after IVIG.[12]

These findings are provocative but insufficient to establish IVIG as standard treatment.

A randomized controlled trial with objective neuropathy endpoints is needed.

38. Treatment of vascular manifestations

Patients with suspected vascular disease should undergo appropriate vascular and systemic evaluation.

Anticoagulation should not be prescribed simply because a patient has long COVID and a rash.

Antiplatelet or anticoagulant therapy should be based on an established indication.

This distinction is critical because the bleeding risks of unnecessary anticoagulation are substantial.

39. Antiviral therapy

Persistent viral antigen has generated interest in antiviral treatment.

However, detection of residual RNA or protein does not prove active viral replication.

The appropriate clinical question is therefore not:

“Is viral material present?”

but:

“Is replication-competent virus driving the patient’s disease?”

Answering that question requires carefully designed trials incorporating:

tissue viral burden;
replication markers;
serial sampling;
symptom phenotyping;
inflammatory biomarkers; and
randomized treatment allocation.

Until such evidence exists, prolonged antiviral treatment for isolated cutaneous long-COVID manifestations should not be considered established practice.

40. Immunomodulatory treatment

Immune-targeted therapies are scientifically attractive because immune dysregulation is increasingly evident in subsets of long COVID.

Potential therapeutic categories under investigation include:

corticosteroids;
immunoglobulin;
B-cell-directed therapy;
cytokine-targeted agents;
JAK inhibition;
mast-cell-directed therapy;
complement inhibition.

But the central principle should be endotype-directed treatment.

An antiviral will not necessarily correct autoimmunity.

An immunosuppressant will not necessarily eradicate a persistent viral reservoir.

Anticoagulation will not necessarily correct neuropathy.

The heterogeneous biology of long COVID makes indiscriminate treatment particularly hazardous.

41. The epidermis as a biomarker tissue

One of the most promising implications of this field is that skin may become a biomarker tissue for systemic long COVID.

A skin biopsy can simultaneously interrogate:

small nerve fibres;
autonomic fibres;
blood vessels;
immune cells;
complement;
keratinocytes;
viral antigen;
RNA;
gene expression;
epigenetic state.

Compared with muscle, nerve, or visceral tissue, skin is accessible and relatively low-risk to sample.

The epidermis may therefore provide a “window” into otherwise inaccessible systemic biology.

42. A proposed cutaneous long-COVID classification

A clinically useful framework would divide manifestations into six mechanistic groups.

Type I: Inflammatory epidermal phenotype

Features:

eczema;
urticaria;
inflammatory papules;
persistent erythematous eruptions.

Likely mechanisms:

immune dysregulation;
cytokines;
mast cells;
barrier dysfunction.

Type II: Vascular phenotype

Features:

pernio;
livedoid lesions;
purpura;
edema;
acral discoloration.

Likely mechanisms:

endothelial dysfunction;
complement;
platelet activation;
microvascular injury.

Type III: Neurocutaneous phenotype

Features:

burning;
dysesthesia;
neuropathic itch;
altered temperature;
allodynia.

Likely mechanisms:

small-fibre neuropathy;
autonomic dysfunction;
neuroimmune signalling.

Type IV: Appendageal phenotype

Features:

telogen effluvium;
alopecia areata;
nail abnormalities.

Likely mechanisms:

systemic stress;
immune dysregulation;
follicular-cycle disruption.

Type V: Pigmentary phenotype

Features:

hyperpigmentation;
hypopigmentation;
postinflammatory change.

Likely mechanisms:

melanocyte injury;
inflammatory signalling;
altered epidermal turnover.

Type VI: Persistent-antigen phenotype

Features:

chronic or recurrent skin abnormalities with tissue evidence of SARS-CoV-2 antigen or RNA.

Likely mechanisms:

residual antigen;
immune-cell sequestration;
potentially persistent replication.

These categories are not mutually exclusive.

A patient could simultaneously have neurocutaneous, vascular, and inflammatory phenotypes.

43. Prognosis

The prognosis depends strongly on phenotype.

Telogen effluvium

Generally favourable.

Most patients experience substantial regrowth as the follicular cycle normalises.[14,15]

Urticaria and inflammatory eruptions

Often improve, although chronic disease can persist.

Pernio-like lesions

Usually self-limited, although persistent acral vascular symptoms require evaluation for alternative causes.

Pigmentary abnormalities

Often improve slowly but may persist.

Small-fibre neuropathy

Potentially prolonged.

Nerve-fibre regeneration is slow and incomplete in some patients. Symptoms can fluctuate, improve, or remain chronic.

Autoimmune disease

Prognosis depends on the specific autoimmune phenotype.

Persistent viral-antigen-associated disease

Unknown.

This represents one of the largest gaps in knowledge.

44. Why some patients recover and others do not

The variability of long-COVID skin disease probably reflects interactions among:

viral burden;
viral variant;
vaccination status;
acute disease severity;
age;
sex;
host genetics;
immune phenotype;
autoimmunity;
vascular susceptibility;
autonomic dysfunction;
metabolic health;
microbiome;
environmental exposures; and
repeated SARS-CoV-2 infections.

The 2025 long-COVID GWAS demonstrates that genetic susceptibility exists, but its contribution is modest relative to the complexity of the syndrome.[16]

Recovery may therefore depend not on a single genetic switch but on whether multiple biological systems successfully return to homeostasis.

45. The importance of reinfection

Repeated SARS-CoV-2 infection creates another unresolved problem.

Each infection can theoretically:

reactivate immune responses;
increase antigenic burden;
induce another inflammatory episode;
disturb autonomic regulation;
produce another episode of telogen effluvium;
trigger autoimmune phenomena; or
exacerbate pre-existing disease.

Longitudinal dermatological studies following patients across repeated infections are scarce.

This should be a priority.

46. Age and the epidermal reserve

Age is likely to modify cutaneous long COVID.

Aging produces:

thinner epidermis;
reduced barrier function;
slower wound healing;
reduced collagen synthesis;
altered immune surveillance;
reduced microvascular reserve;
reduced nerve-fibre density;
altered melanocyte function;
increased prevalence of alopecia.

Consequently, an inflammatory or vascular insult that would be clinically trivial in younger skin may produce substantial persistent symptoms in older adults.

Age-stratified research is therefore essential.

47. Sex differences

Long COVID overall appears to show sex-related differences, with women frequently reporting higher prevalence of persistent symptoms.

The reasons are probably multifactorial.

Potential mechanisms include:

sex hormones;
immune-response differences;
autoimmune susceptibility;
X-chromosome biology;
vascular biology;
differences in acute disease;
differential healthcare utilisation.

Dermatological manifestations may also differ by sex. The post-COVID literature has frequently identified women as disproportionately represented among patients with alopecia and some immune-mediated cutaneous conditions.[4,15]

This is a hypothesis requiring prospective validation rather than a universal rule.

48. The microbiome

The skin microbiome is another underexplored component.

The epidermis hosts complex communities of bacteria, fungi, viruses, and other microorganisms that interact with keratinocytes and immune cells.

Systemic infection, inflammation, antibiotic exposure, altered sweating, topical products, and behavioural changes can alter this ecosystem.

A persistent microbiome disturbance could theoretically contribute to:

dermatitis;
follicular disease;
barrier dysfunction;
itch;
infection susceptibility.

Future long-COVID studies should include paired skin microbiome and epidermal transcriptomic analysis.

49. The epidermis and systemic metabolic dysfunction

Long COVID frequently coexists with metabolic abnormalities.

Hyperglycaemia, nutritional deficiency, weight change, endocrine disturbances, and altered mitochondrial metabolism can all affect skin.

Because the epidermis undergoes continuous renewal, it is metabolically demanding.

Mitochondrial dysfunction could therefore theoretically alter:

keratinocyte differentiation;
barrier repair;
wound healing;
hair-follicle cycling;
sensory-neural function.

The mitochondrial hypothesis is biologically attractive but remains insufficiently tested specifically in epidermal long COVID.

50. Wound healing

Persistent inflammation, vascular dysfunction, neuropathy, metabolic disease, and impaired mobility can converge on wound healing.

A normal wound requires:

haemostasis;
inflammation;
proliferation;
angiogenesis;
re-epithelialisation;
extracellular matrix deposition;
remodelling.

Long COVID could theoretically interfere with several stages.

Endothelial dysfunction could impair angiogenesis.

Immune dysregulation could prolong inflammation.

Neuropathy could reduce protective sensation.

Metabolic dysfunction could impair keratinocyte proliferation.

Autonomic dysfunction could alter perfusion and sweating.

This hypothesis warrants direct study in patients with persistent wounds after COVID-19.

51. Clinical red flags

Certain cutaneous findings should not be attributed casually to long COVID.

Urgent evaluation is warranted for:

rapidly spreading purpura;
skin necrosis;
severe blistering;
mucosal involvement;
extensive skin detachment;
fever with systemic deterioration;
rapidly progressive edema;
painful retiform lesions;
new neurologic deficits;
signs of severe infection;
unexplained bruising or bleeding;
suspected vasculitis;
rapidly progressive alopecia with scarring.

Long COVID is a diagnosis that should sharpen clinical reasoning, not replace it.

52. Major limitations of the current literature

The evidence base has several weaknesses.

52.1 Heterogeneous definitions

Studies use different definitions of long COVID.

52.2 Inconsistent dermatological ascertainment

Some studies rely on patient questionnaires; others use physician diagnosis.

52.3 Lack of controls

Many studies lack appropriately matched non-COVID controls.

52.4 Variant effects

Early studies disproportionately reflect ancestral SARS-CoV-2 and Alpha-era disease.

52.5 Vaccination

Vaccination changes both acute disease and long-COVID risk but is inconsistently incorporated.

52.6 Medication confounding

Antivirals, steroids, antibiotics, anticoagulants, and other drugs can produce cutaneous effects.

52.7 Publication bias

Striking dermatological cases are more likely to be reported.

52.8 Lack of serial tissue sampling

Most biopsy studies are cross-sectional.

52.9 Limited tissue genomics

Skin-specific transcriptomic and epigenomic datasets remain sparse.

52.10 Causal uncertainty

Association remains substantially stronger than causal evidence.

53. A research programme for epidermal long COVID

A definitive research programme should recruit patients immediately after acute infection and follow them longitudinally.

At baseline:

clinical dermatological examination;
standardized photography;
blood biomarkers;
immune profiling;
viral testing;
host genotyping.

At 1, 3, 6, 12, and 24 months:

repeat dermatological examination;
symptom phenotyping;
hair assessment;
autonomic testing;
neurological assessment where indicated;
skin barrier measurement;
sweat testing;
skin biopsy in predefined subgroups.

Biopsy material should undergo:

routine histopathology;
immunohistochemistry;
SARS-CoV-2 antigen detection;
RNA sequencing;
spatial transcriptomics;
single-cell RNA sequencing;
single-cell ATAC sequencing;
proteomics;
complement mapping;
nerve-fibre quantification;
immune-cell phenotyping.

Blood should undergo:

autoantibody profiling;
cytokine analysis;
complement studies;
coagulation markers;
metabolomics;
proteomics.

The resulting dataset could determine whether clinical phenotypes correspond to molecular endotypes.

54. The critical experiment

The most important unanswered question can be reduced to a simple experimental design.

Recruit patients with persistent cutaneous symptoms after COVID-19.

Obtain:

lesional skin;
non-lesional skin;
blood;
appropriate non-COVID controls.

Determine:

Is SARS-CoV-2 RNA present?
Is nucleocapsid protein present?
Is there evidence of replication?
Which cells contain antigen?
Are keratinocytes infected?
Are macrophages infected or merely carrying antigen?
Is complement activated?
Are endothelial cells injured?
Are epidermal nerve fibres reduced?
Are autoantibodies present?
Which genes are activated?
Which chromatin regions are altered?

Then follow the patient longitudinally.

If viral material disappears while the phenotype persists, a postviral immune or neurovascular mechanism becomes more likely.

If viral material persists and tracks with symptoms, tissue persistence becomes more compelling.

If different patients show different patterns, long COVID skin disease should be divided into molecular endotypes.

That experiment would move the field beyond descriptive dermatology.

55. Toward precision dermatology for long COVID

The ultimate objective should not be to create another broad diagnostic label.

It should be to identify biologically defined subgroups.

For example:

Endotype A: immune-epidermal

Dominant findings:

inflammatory rash;
cytokine activation;
immune-cell infiltration.

Endotype B: vascular

Dominant findings:

endothelial activation;
complement deposition;
livedoid or acral disease.

Endotype C: neurocutaneous

Dominant findings:

reduced epidermal nerve-fibre density;
neuropathic symptoms;
autonomic abnormalities.

Endotype D: antigen-persistent

Dominant findings:

tissue viral antigen/RNA;
macrophage-associated antigen;
persistent immune activation.

Endotype E: follicular

Dominant findings:

telogen effluvium;
alopecia areata;
follicular immune changes.

Endotype F: mixed

Multiple biological systems are simultaneously affected.

Such classification could allow treatment to become mechanistic rather than symptomatic.

56. Treatment trials should incorporate dermatological endpoints

Long-COVID clinical trials frequently focus on fatigue, cognition, exercise tolerance, pulmonary function, or global symptom scores.

This risks missing substantial biological information.

Dermatological trials should incorporate:

standardized lesion severity;
photographic scoring;
pruritus scales;
hair density;
dermoscopy;
transepidermal water loss;
epidermal nerve-fibre density;
autonomic fibre density;
vascular imaging;
skin transcriptomics;
tissue antigen burden.

These endpoints could reveal whether an intervention actually changes tissue pathology rather than merely altering subjective symptoms.

57. Prognostic biomarkers

The future may allow clinicians to identify patients at risk of persistent cutaneous disease early after infection.

Potential candidates include:

inflammatory cytokines;
complement markers;
endothelial markers;
autoantibodies;
host genetic variants;
interferon signatures;
viral antigen burden;
epidermal nerve-fibre density;
autonomic dysfunction.

No such biomarker is currently validated for routine clinical use.

The distinction between promising research marker and clinical biomarker should remain explicit.

58. The epidermis as a longitudinal record

An unusual advantage of skin is that it changes continuously.

Hair records systemic stress through follicular cycling.

Pigmentation records previous inflammation.

Nails record physiological disturbances through growth.

Epidermal nerve fibres can be quantified.

The skin therefore provides something approaching a biological timeline.

In a disease as temporally complex as long COVID, that property could be invaluable.

59. A broader model of long COVID

The epidermal evidence supports a broader understanding of long COVID.

The syndrome may be less analogous to a single chronic disease than to a collection of persistent biological states generated by a common initiating infection.

Different patients may develop:

persistent antigen;
autoimmunity;
endothelial dysfunction;
autonomic dysfunction;
small-fibre neuropathy;
mitochondrial dysfunction;
altered metabolism;
tissue-specific inflammation.

These mechanisms can coexist.

The skin is particularly valuable because several can be observed within the same small tissue sample.

60. Conclusions

The epidermis occupies an unusual position in long COVID.

It is visible enough to reveal disease clinically, yet sophisticated enough to record systemic immune, vascular, neural, and epithelial abnormalities.

The dermatological manifestations associated with COVID-19 are now well established, but persistent cutaneous disease after SARS-CoV-2 infection remains incompletely characterised. Existing evidence supports several important conclusions.

First, long COVID can be associated with persistent dermatological manifestations including rash, pruritus, dermatitis, pigmentary changes, edema, nodules, and particularly alopecia.[4]

Second, the biology is heterogeneous. Not every post-COVID rash is caused by the same mechanism.

Third, SARS-CoV-2-related cutaneous disease can involve the microvasculature. Complement deposition and thrombogenic vasculopathy have been demonstrated in severe acute disease.[8]

Fourth, residual SARS-CoV-2 antigen and RNA have been identified in skin tissue from patients with long COVID, demonstrating that viral material can persist in the integumentary system well beyond the acute phase.[5] Whether this represents active replication, immune-cell sequestration, or inert residual material remains unresolved.

Fifth, keratinocytes express ACE2, demonstrating biological plausibility for direct interaction with SARS-CoV-2, but experimental evidence suggests that productive infection of keratinocytes is limited and may differ from infection of respiratory epithelial cells.[6,7]

Sixth, the skin can reveal neurological disease. Large contemporary cohorts demonstrate that small-fibre neuropathy involving epidermal and autonomic fibres occurs in a substantial subset of patients with neuropathic long COVID.[12]

Seventh, immune dysregulation and autoimmunity are increasingly plausible contributors. Tissue-specific autoantibody studies suggest that routine serological tests may underestimate the extent of immune abnormalities in some patients.[11]

Eighth, host genetics contributes to susceptibility to long COVID. The discovery of a reproducible FOXP4 association establishes an important genetic foothold, although no skin-specific long-COVID susceptibility gene has yet been established.[16]

Finally, the most important conclusion is methodological.

The field should move from asking whether COVID-19 “causes a rash” to asking which biological systems are persistently abnormal in the skin of individual patients.

The epidermis may contain evidence of viral persistence, immune activation, endothelial dysfunction, autonomic disturbance, and peripheral neuropathy. It may therefore serve not simply as an organ affected by long COVID, but as a readily accessible experimental window into the syndrome itself.

The next generation of research should combine clinical dermatology with molecular pathology, spatial transcriptomics, host genetics, immunology, virology, and neurobiology. Serial skin sampling could determine whether persistent cutaneous manifestations are driven primarily by residual antigen, immune memory, autoimmunity, microvascular disease, neural injury, altered epithelial homeostasis, or combinations thereof.

The therapeutic implications are substantial.

A patient whose skin contains persistent viral antigen may require a fundamentally different intervention from one whose principal abnormality is autoimmune vascular disease. A patient with reduced epidermal nerve-fibre density may require neurological rather than dermatological treatment. A patient with telogen effluvium may require reassurance and correction of systemic stressors rather than immunosuppression. A patient with true inflammatory dermatitis should receive evidence-based dermatological therapy irrespective of whether SARS-CoV-2 initiated the process.

Long COVID therefore demands a precision approach.

The skin may be where that precision begins.

Footnotes

1. Acute versus post-acute disease.
Cutaneous manifestations documented during acute COVID-19 should not automatically be classified as long-COVID manifestations. Persistence, recurrence, or delayed onset after apparent recovery is necessary before attributing a lesion to post-COVID condition.

2. Viral persistence versus viral replication.
Detection of SARS-CoV-2 RNA or protein in tissue does not establish replication-competent virus. Future studies should distinguish residual antigen from active infection using complementary molecular and virological methods.

3. Epidermis versus skin.
The epidermis is only one compartment of the skin. Many relevant long-COVID abnormalities involve dermal vessels, sensory nerves, sweat glands, hair follicles, or immune cells rather than keratinocytes themselves.

4. Skin biopsy for neuropathy.
Intraepidermal nerve-fibre density is a validated tool for assessing small-fibre neuropathy, but interpretation requires age-, sex-, site-, and laboratory-specific normative values.

5. Autoantibodies.
Detection of an autoantibody does not establish pathogenicity. Functional studies and longitudinal correlations are required before an antibody can be considered mechanistically causal or clinically actionable.

6. Alopecia.
Telogen effluvium is a well-established consequence of systemic physiological stress and is not specific to SARS-CoV-2. Temporal association with infection is supportive but not definitive evidence of causality.

7. Treatment evidence.
Most therapies discussed for long COVID remain investigational or phenotype-specific. No pharmacological treatment has been established as a universal disease-modifying therapy for long COVID.

8. Genomics.
The FOXP4 association is a genetic association with long COVID as a systemic syndrome. It should not currently be interpreted as a genetic determinant of epidermal disease.

9. Diagnostic exclusion.
Long COVID can coexist with conventional disease. The diagnosis should not prevent investigation of renal, hepatic, endocrine, hematological, vascular, autoimmune, infectious, medication-related, or neurological causes of skin symptoms.

References

Mirza FN, Malik AA, Omer SB, Sethi A. Dermatologic manifestations of COVID-19: a comprehensive systematic review. Int J Dermatol. 2021;60:418–450.
World Health Organization. A clinical case definition of post COVID-19 condition by a Delphi consensus. Geneva: WHO; 2021.
World Health Organization. Post COVID-19 condition (long COVID). WHO; 2025.
Cayón Figueroa BA, Mendoza Rojas W, Jiménez DT. Dermatological complications due to post-COVID-19 syndrome: a systematic review. Mol Infect. 2024;5:9.
Gaebler C, et al. Case report: persistence of residual antigen and RNA of the SARS-CoV-2 virus in tissues of two patients with long COVID. PubMed-indexed report.
Xue X, Mi Z, Wang Z, Pang Z, Liu H, Zhang F. High expression of ACE2 on keratinocytes reveals skin as a potential target for SARS-CoV-2. J Invest Dermatol. 2021;141:206–209.e1.
Human keratinocytes exhibit limited potential for SARS-CoV-2 infection despite ACE2 and mature cathepsin L expression. J Invest Dermatol. 2026.
Magro C, Mulvey JJ, Berlin D, et al. Complement associated microvascular injury and thrombosis in the pathogenesis of severe COVID-19 infection: a report of five cases. Transl Res. 2020.
Davis HE, McCorkell L, Vogel JM, Topol EJ. Long COVID: major findings, mechanisms and recommendations. Nat Rev Microbiol. 2023.
Gianni P, Goldin M, Ngu S, et al. Complement-mediated microvascular injury and thrombosis in the pathogenesis of severe COVID-19: a review. World J Emerg Med. 2022;12:53–67.
Tatai O, Nagy S, Nguyen THT, et al. Tissue-specific autoantibody signatures reveal immune alterations undetected by routine serology in long COVID. GeroScience. 2026;48:3623–3647.
Analysis of 977 Long COVID Patients Reveals Prevalent Neuropathy and Association with Anti-Ganglioside Antibodies. PubMed. 2025.
Frumholtz L, Bouaziz J-D, Battistella M, et al. Type I interferon response and vascular alteration in chilblain-like lesions during the COVID-19 outbreak. Br J Dermatol. 2021;185:1176–1185.
Hussain N, Agarwala P, Iqbal K, et al. A systematic review of acute telogen effluvium, a harrowing post-COVID-19 manifestation. J Med Virol. 2022;94:1391–1401.
Characteristics of hair loss after COVID-19: a systematic scoping review. PubMed. 2022.
Lammi V, Nakanishi T, Jones SE, et al; Long COVID Host Genetics Initiative. Genome-wide association study of long COVID. Nat Genet. 2025;57:1402–1417.
Jamshidi P, Hajikhani B, Mirsaeidi M, et al. Skin manifestations in COVID-19 patients: are they indicators for disease severity? A systematic review. Front Med. 2021;8:634208.
Perna A, Passiatore M, Massaro A, Terrinoni A. Skin manifestations in COVID-19 patients, state of the art: a systematic review. Int J Dermatol. 2021;60:547–553.
Cutaneous manifestations of COVID-19: a systematic review. J Drugs Dermatol. 2021.
Sharquie KE, Jabbar RI. COVID-19 infection is a major cause of acute telogen effluvium. Ir J Med Sci. 2022;191:1677–1681.
Post-COVID telogen effluvium. PubMed. 2023.
SARS-CoV-2 infection—a trigger factor for telogen effluvium: review of the literature with a case-based guidance for clinical evaluation. PubMed. 2023.
World Health Organization. Coronavirus disease (COVID-19): post COVID-19 condition. WHO; 2023.
World Health Organization. Post-COVID-19 condition: clinical management and rehabilitation guidance. WHO.
Long COVID: major findings, mechanisms and recommendations. Nat Rev Microbiol. 2023.
Falco P, et al. Autonomic small fiber involvement in painful long COVID: a histological and clinical study. Front Hum Neurosci. 2026.
Tissue-specific autoantibody signatures reveal immune alterations undetected by routine serology in long COVID. GeroScience. 2026.
Genome-wide association study of long COVID. Nature Genetics. 2025;57:1402–1417.
Human keratinocytes exhibit limited potential for SARS-CoV-2 infection despite ACE2 and mature cathepsin L expression. 2026.
Type I interferon response and vascular alteration in chilblain-like lesions during the COVID-19 outbreak. Br J Dermatol. 2021;185:1176–1185.