Rook's Textbook of Dermatology

Centre for Dermatology Research, Institute of Inflammation and Repair, University of Manchester, Manchester, UK; and Department of Dermatology, University of Münster, Germany

Introduction and overview

When classical endocrine glands and their systemically secreted products (hormones) were originally recognized, it did not take long to realize that excessive or insufficient circulating levels of these hormones could affect the skin. Only much later, when the specific receptors for these steroid and peptide hormones were identified and the molecular basis of interactions between hormonal ligands and their receptors were better understood, was it realized that cutaneous responses to endocrine abnormalities reflect the fact that all constituent cell populations of human skin express multiple cognate receptors not just for these hormones but also for many other neuromediators. Though the mechanisms involved are still insufficiently understood, the visible effects of abnormal ligand–receptor interactions on the skin certainly provide important diagnostic clues to underlying endocrine disease. Yet, the importance of hormones in dermatology goes well beyond this.

The field of dermatoendocrinology has undergone a major revolution over the past two decades: human skin and its appendages are now recognized not only as prominent hormone target tissues, but also as major endocrine organs themselves. Human scalp hair follicles have provided a model for exploring this relatively recent research frontier in investigative dermatology and for identifying ‘novel’ functions of classical neurohormones, particularly as they represent exquisitely hormone-sensitive mini-organs [1, 2, 3]. Therefore, where appropriate, hair follicles will be used as models for illustrating and exploring general principles.

This chapter reviews the key elements of the biological basis of dermatoendocrinology to help clinicians understand how general skin signs and symptoms such as pruritus, skin dryness, hair loss, hypertrichosis, hirsutism, hyperhidrosis and/or hyperpigmentation result from defined endocrine disorders. Also discussed are the characteristic skin manifestations of underlying endocrine disease and a practical approach to skin patients with suspected underlying hormonal problems, with special emphasis on thyroid and pituitary disorders and mention of dermatologically relevant drugs which may cause endocrine abnormalities that affect the skin. Diabetes (see Chapter 64) and hormonally active vitamins (hyper- or hypovitaminosis A or D) (see Chapter 63) are discussed elsewhere.

This chapter concludes by examining why the fact that skin operates as a complex (neuro-) endocrine organ has important practical implications for future dermatological therapy: the clinical significance and biomedical fascination of dermatoendocrinology extend beyond providing diagnostic clues to the identification of underlying endocrine disease.

Biological basis of dermatoendocrinology

Principles of endocrinology

Enshrined as a concept by Bayliss’ and Starling's famous Croonian Lecture [4], ‘hormones’ have generally come to be understood as relatively stable secreted molecules that are released into the bloodstream to reach and modulate the function of distant tissue targets. This ‘endocrine’ secretory activity is distinguished from ‘paracrine’ hormone secretion that targets adjacent cells, ‘autocrine’ signalling which describes the autostimulation of a cell with a hormone released by that cell and ‘intracrine’ signalling whereby the hormone in question does not leave the cell that has produced it and acts intracellularly [5].

Classical endocrine signalling operates along several central axes through which the brain controls key functions of ‘professional’ endocrine glands. The most familiar of these are the hypothalamopituitary–adrenal (HPA) and the hypothalamopituitary–thyroid (HPT) axes; central nervous system (CNS) controlled signalling axes dominated by prolactin, catecholamines or acetylcholine are also well known (Figure 149.1). If hormone imbalances within these central axes lead to excessive or insufficient serum levels of key hormones, peripheral tissues such as the skin will be affected provided that they express cognate hormone receptors. Such imbalances can result from a wide range of disorders including hormone-secreting tumours, autoimmune attack directed at endocrine glands resulting in either failure of or excessive hormone secretion, biochemical abnormalities, environmental or nutritional signals, drugs, psychological and emotional stress, or the physiological consequences of puberty, menopause or ageing on systemic hormone levels.

Image described by caption. — **Figure 149.1** Schematic representation of three hypothalamopituitary axes that impact on human skin: key elements of the central hypothalamopituitary–adrenal (HPA) and the hypothalamopituitary–thyroid (HPT) axes, and the central control of prolactin (PRL) release from the pituitary gland. 5-HT, 5-hydroxytryptamine; ACTH, adrenocorticotrophic hormone; CRH, corticotropin-releasing hormone; E2, oestrogen; EGF, epidermal growth factor; FGF, fibroblast growth factor; TSH, thyrotropin; TRH, thyrotropin-releasing hormone; VIP, vasoactive intestinal peptide. (Redrawn from Paus *et al. Trends Mol Med* 2014 [3].

Reproduced with permission from Elsevier.)

The term endocrine was originally coined to distinguish the control of organ function by secreted hormones from those exerted more directly by the nervous system. With increased knowledge, however, the borders between endocrinology, neuroendocrinology, neuropharmacology and other neuroscience disciplines have become indistinct. Hormones can act both locally and at far distant body sites as well as on the nervous system, and the distribution of high-affinity hormone receptors is far more widespread than previously thought. Unsurprisingly, all hormones and neuromediators exert many more biological activities than had been apparent during the early years following their discovery. For example, classical steroid hormones generated outside the CNS or administered therapeutically (e.g. glucocorticoids, retinoids, androgens, oestrogens) exert profound neurochemical effects in the brain, while classical neurohormones, neuropeptides and neurotransmitters have an impact not only on the growth, regeneration and other specific functions of epithelial and mesenchymal tissues but are also produced by them. Essentially, all tissues and organs, including human skin and its appendages, can generate a wide range of hormones and neuromediators.

Human epidermal and hair follicle keratinocytes, sebocytes and subcutaneous adipocytes are recognized as potent sources of steroid and peptide hormones as well as of a steadily growing list of neuromediators (Table 149.1 and see later). In situ, many of these were first discovered in human scalp hair follicles [3] (Figure 149.2).

Table 149.1 Key components in the human cutaneous (neuro-) endocrine signalling mechanism. For a comprehensive review and additional relevant ligands and receptors see [3, 8, 9, 15].

Ligand	Intracutaneous receptor	Selected activities	References
Generated within human skin and/or its appendages
CRH	CRHR1, CRHR2	Stress response coordinator, controls skin HPA axis equivalent and both mast cell activation and local maturation	[8, 9, 26]
ACTH	MC2R (MC1R)	Stimulates intracutanous cortisol synthesis, activates mast cells	[9, 26]
α-MSH	MC1R	Stimulates pigmentation, maintains hair follicle immune privilege, multiple immune-inhibitory, anti-allergic, tolerogenic and oxidative damage-protective functions, regulates collagen synthesis; possibly also has antimicrobial activities	[42-46]
β-endorphin	μ-opioid receptor	Key role in itch modulation, mast cell secretagogue, stimulates pigmentation	[6, 8, 47, 125]
Cortisol	GR	Regulates cell metabolism, epidermal barrier function, skin and pilosebaceous immune homeostasis (maintenance of relative immunoinhibition?)	[3, 8, 9, 26, 57]
Androgens (DHT)	AR	DTH synthesis controlled by 5-α-reductase, AR stimulation of hair follicles results in either TGF-β or IGF-1 secretion (location dependent); stimulate sebaceous gland	[40, 50, 51, 52]
Oestrogens (17-β-oestradiol)	ER	17-β ER synthesis controlled by aromatase, large gender-dependent differences in response to ER stimulation, promote wound healing; inhibit sebaceous gland	[53-56]
Vitamin D3 and its metabolites	VDR	Pleiotropic effects (e.g. inhibit keratinocyte proliferation, stimulate keratin expression, anti-inflammatory)	[39, 41, 57]
Retinoids	RAR, RXR	Pleiotropic effects
PPARγ ligands (small lipids)	PPARγ	Inhibit keratinocyte proliferation and hair growth, but maintain epithelial stem cells; anti-inflammatory; promote melanogenesis	[59, 60, 85]
Endocannabinoids (e.g. anandamide, 2-AG)	CB1, CB2	CB1 stimulation inhibits hair growth and excessive mast cell degranulation/maturation, but stimulate sebogenesis	[61, 62, 63, 64, 65]
Catecholamines (also derived from intracutaneous nerve fibres)	α-/β-adrenergic receptors	Regulate skin perfusion and sweating; inhibit mast cell degranulation, retard keratinocyte migration and wound healing; regulate melanocyte functions	[30, 33, 34, 36, 66-71]
Acetylcholine receptor ligands (also derived from intracutaneous nerve fibres)	Nicotinic and muscarinic acetylcholine receptors	Regulate keratinocyte proliferation, differentiation, migration, adhesion, and apoptosis; activate/attract neurotrophils; stimulate sweating; regulate sebocyte lipid production	[29-30, 72, 73]
Melatonin	MT-1, MT-2	Oxidative damage control, DNA damage repair	[74-80]
TRH	TRH-R	Promotes hair growth and keratinocyte mitochondrial function, stimulates hair follicle pigmentation; regulates keratin expression; stimulates epidermal TSH expression; regulates intracutaneous prolactin expression	[81, 82, 83, 84, 87, 88]
TSH	TSH-R	Stimulates keratinocyte mitochondrial activity and biogenesis; regulates keratin expression	[3, 27, 28, 84, 366]
Prolactin	PRL-R	Inhibits or promotes hair growth (species and gender dependent), stimulates sebum production; immunomodulatory; regulates keratin expression	[3, 87, 88-91]
PTH/PTHrp	PTH/PTHrp receptor	Regulates keratinocyte proliferation and differentiation; modulates hair growth (mice); regulate skin angiogenesis	[92-96]
Generated (mainly) outside the skin (transported via nerve fibres or blood vessels)
Substance P	NK1	Activates mast cells; pro-inflammatory; induces collapse of hair follicle immune privilege	[18, 97, 98]
CGRP	CGRP-R	Anti-inflammatory and tolerogenic; guardian of hair follicle immune privilege	[99, 100]
Dopamine	DR 1	Hair growth inhibitory	[101]
Thyroid hormones (T3, T4)	TRα, TRβ	Regulate keratin expression; prolong anagen; promote wound healing; stimulate keratinocyte mitochondrial activity and heat production; regulates epithelial stem cell functions	[84, 102-104, 195]

For additional relevant (neuro-) endocrine ligands and their receptors in human skin, see these papers and references cited therein: proenkephalin [105], dynorphin [106], galanin [107], somatostatin [108, 109, 110], oxytocin [111], neuropeptide Y [112, 113], serotonin [74, 114], leptin [22, 115], endovanilloids [49, 116, 117, 118, 119, 120], erythropoietin [121, 122, 123] and growth hormone (GH) [124].

ACTH, adrenocorticotropic hormone; AR, androgen receptor; CB1/2, cannabinoid receptors 1/2; CGRP/CGRP-R, calcitonin gene related protein/CGRP receptor; CRH, corticotropin-releasing hormone (synonym: CRF); CRHR1/R2, CRH receptors 1/2; DR1, dopamine receptor type 1; ER, oestrogen receptor; GR, glucocorticoid receptor; MC1R/2R, melanocortin receptors 1/2; α-MSH, α-melanocyte-stimulating hormone (synonym: melanotropin); MT1/2, melatonin receptors 1/2; NK1, neurokinin receptor 1; PPARγ, peroxisome proliferator activator receptor γ; PRLR, prolactin receptor; RAR/RORα/RXR, retinoid receptors; PTH, parathyroid hormone; TRs, thyroid hormone receptors; TRPV1/3/4, transient receptor potential ion channels of vanilloid type 1/3/4; VDR, vitamin D receptor.

Current evidence suggests that inflammation, oxidative and psycho-emotional stress, UV irradiation, microbiomal signals, certain nutritional and sensory stimuli and some drugs can influence intracutaneous hormone and neuromediator production, often in a similar though not necessarily identical manner to the regulation of central hormone production [3, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]. Thus, endocrine organs, the CNS, the peripheral nervous system and peripheral tissues such as the skin share and generate a comparable spectrum of hormonal and neuromediator ligands and express a very similar repertoire of cognate receptors which together facilitate intersystem communication, even though the subtype and preferred intracellular signal transduction pathways may differ (e.g. between the brain and the skin). Hormones such as prolactin and leptin also exert potent cytokine-like effects in addition to their classical endocrine activities [20, 21, 22, 23, 24], while steroid hormones, which regulate the function of multiple non-neuronal cells outside the nervous system, double as profound neuromodulators and neurotransmitters [25].

Key elements of the three major central endocrine signalling axes which regulate the body's overall metabolism, its response to environmental stressors and many aspects of specific organ function, i.e. the HPA axis, the HPT axis and the regulatory system of pituitary prolactin release, have been identified in other peripheral tissues together with negative and positive feedback regulatory loops. In conclusion, the traditional distinctions between hormones, neuropeptides, neurotransmitters and cytokines have ceased to be meaningful.

Skin as a (neuro-) endocrine organ

This explicitly includes human skin, which is now known to contain integral functional equivalents of the HPA axis (CRH → ACTH → cortisol) [3, 6, 8, 9, 26] and key elements of the HPT axis (TRH→TSH) [3, 27, 28]. In addition, human skin displays complete cholinergic and adrenergic signalling systems [29, 30, 31, 32, 33] with potentially important functions in the control of wound healing [34, 35, 36], epidermal immunity [37] and skin cancer [38]. Human skin also metabolizes numerous hormones and neuromediators, including glucocorticoids, androgens, oestrogens, neuropeptides such as substance P, and lipid neuromediators such as endocannabinoids [3, 8, 9, 10, 39].

Human skin and its pilosebaceous units and sweat glands also produce a bewildering range of steroid and peptide hormones, ranging from vitamin D, androgens, oestrogens and retinoids to erythropoietin, and classical neurohormones such as corticotrophin-releasing hormone (CRH), thyrotropin-releasing hormone (TRH), thyroid-stimulating hormone (TSH), adrenocorticotrophic hormone (ACTH), α-MSH, β-endorphin and other opioids, melatonin and prolactin, along with multiple other neuromediators; this is complemented by lipid mediators such as endocannabinoids and endovanilloids, which exert neuromediator and many other functions (see Tables 149.1 and 149.2 for details and references; see also [3, 8, 40, 41]). Intriguingly, human scalp hair follicles have been shown to be an important component of the neuroendocrine biology of human skin and show prominent expression/production of and/or sensitivity to a surprisingly wide range of neuromediators (Table 149.3), some of which are indicated in Figure 149.2.

Table 149.2 Examples of (neuro-) endocrine contributions to cutaneous pathogenesis.

Mediator	Condition	Action	References
Androgens +, Insulin –	Wound healing	Impairment	[55, 104, 158, 159, 160, 207-209]
T3/T4 –	Wound healing	Impairment	[104]
T3/T4 +	Telogen effluvium	Promotion	[50, 210, 327]
Oestrogens +	Melasma	Promotion	[211, 212]
	Chemotherapy-induced alopecia.	Promotion of hair regrowth?	[213]
DHT +	Androgenetic alopecia	Essential for pathogenesis	[207]
	Seborrhoea	Promotion	[50, 52]
	Acne vulgaris	Promotion	[214]
Prolactin +	Psoriasis	Aggravation (see text)	[215]
	Psoriatic arthritis		[20, 216-218]
	Seborrhoea	Promotion
	Systemic lupus erythematosus	Aggravation (see text)	[219, 220]
	Female pattern androgenetic alopecia	Aggravation	[221]
Substance P +	Atopic eczema	Triggering/aggravation via induction of neurogenic inflammation	[16]
	Psoriasis		[3, 97]
	Urticaria		[172]
	‘Stress’-induced telogen effluvium		[49]
	Alopecia areata		[169]
	Prurigo nodularis		[174]
	Chronic pruritus		[98, 222]
	Rosacea		[222, 223]
Cortisol +	Cushing syndrome	Induction	[211]
	Acne	Induction, aggravation
ACTH +	Hyperpigmentation	Induction	[211]
	Hypertrichosis
α-MSH +	Hyperpigmentation	Induction	[211]
	Melanoma	Potential suppression of anti-tumour immunity	[42]
		Autocrine secretion as survival factor for MM cells?	[224]
		Increased risk after Melanotan^® therapy?	[225-227]
	Psoriasis	Potential induction of tolerogenic dendritic cells and T-regs	[193]
	Chemotherapy-induced alopecia	Protective effect?	[44]
	Melanocytic/dysplastic naevi	Increased growth after Melanotan^® or α-MSH therapy	[228, 229]
α-MSH –	Alopecia areata	Insufficient maintenance of hair bulb immune privilege	[46, 189]
	Lichen planopilaris	Insufficient maintenance of bulge immune privilege	[190]
	Acne vulgaris	α-MSH analogue may be therapeutically beneficial	[230]
	Scleroderma	Insufficient α-MSH-mediated signalling and therapeutic effect of α-MSH postulated	[192]
MC1R variants	Melanoma	Increased risk	[231-234]
		Antagonizing MC1 signalling as antimelanoma strategy?	[233]
	Vitiligo	(association discussed)	[235, 236]
CRH +	Atopic eczema	Stress-related triggering/aggravation via induction of neurogenic inflammation (see text)	[9, 237]
	Alopecia areata		[238]
	Acne vulgaris	Promotion (see text)	[99, 239]
CGRP +	Atopic eczema	Increased IL-13-secretion (see text)	[99]
	AIDS	Inhibits HIV transmission from Langerhans to T cells
	UV-associated immunosuppression	Mediated by CGRP (mice)	[241]
	Allergic contact dermatitis	Induction of hapten-specific tolerance (mice)	[240, 241]
GH +	Acromegaly	Induction	[243]
	Seborrhoea	Promotion	[244, 245]
	Melanocytic naevi	Increased growth	[246]
Somatostatin	Merkel cell carcinoma	Growth inhibition by therapy with somatostatin analogue	[247]
	Alopecia areata	Important for hair follicle immune privilege?	[110]
β-endorphin	Atopic eczema	Increased serum level	[248, 249]
	Psoriasis		[250]
Bradykinin	Hereditary angio-oedema	Receptor antagonist may be therapeutically beneficial	[251]
β2-adrenergic receptor (BAR) –	Atopic eczema	Reduced signalling due to BAR point mutation	[69]
		Reduced catecholamine synthesis and increased catecholamine degradation in atopic epidermis	[68]
	Vitiligo, psoriasis	Insufficient BAR signalling (see text)
TRPVs	Rosacea	TRPVs overexpressed	[252]
		Can be stimulated by recognized rosacea trigger factors	[222]
	Pruritus	Stimulation can promote itch	[49, 118, 253, 254]

+, Increased; –, reduced. Abbreviations: see Table 149.1.

Table 149.3 When to suspect a hormonal basis for a skin disease: general signs and symptoms.

Sign/symptom	Underlying (neuro-) endocrine causes
Anaemia	Anaemia caused by insufficient renal production of erythropoietin (tumour anaemia, diabetic nephropathy and other renal diseases)
Body odour (unpleasant)	Acromegaly (due to enlarged apocrine glands)
Dryness	Thyroid dysfunction (mainly hypothyroidism)
Exophthalmus	Hyperthyroidism (Graves disease)
Extremities enlarged (notably fingers and toes, ‘spade-like’ hands)	Acromegaly
Facial features (overall change)	Coarse features: acromegaly (look also for other signs of acromegaly (see Tables 149.4 and 149.5) and bone and cartilage abnormalities (e.g. prognathism, frontal bossing, enlarged hands/feet)‘Moon facies’: hypercortisolism (Cushing syndrome)
Hyperhidrosis (see Chapter 94)	Hyperthyroidism; acromegaly (enlarged eccrine glands)
Hair, dry/brittle (see Chapter 89)	Hypothyroidism
Hair, loss or gain of (effluvium, alopecia, hirsutism, hypertrichosis) (see Chapter 89)	Virilizing tumour, adrenogenital syndrome, polycystic ovary syndrome Insufficient oestrogen serum level Hypo- or hyperthyroidism, hyperprolactinaemia ACTH-secreting tumour (e.g. pituitary tumour or small cell bronchial carcinoma)
Joints, swollen	Sometimes associated with acromegaly (note thickening of phalangeal joints!), hyper- or hypoparathyroidism, hypothyroidism
Libido, loss of/impotence	Hypopituitarism
Menorrhoea	Hyperprolactinaemia, hypogonadism
Pigmentary abnormalities (see Chapter 88)	Hypo-/hyperpigmentation: Addison disease, ACTH-secreting tumour, hypopituitarism Yellow tint of skin: hypercarotenaemia in association with hypothyroidism or diabetes Pallor with yellow tint: hypopituitarism
Pruritus (see Chapter 83)	Diabetes, thyroid dysfunction, anaemia due to insufficient EPO production (see later)
Psychological and neuropsychiatric disturbances (see Chapter 86)	Hypo- and hyperthyroidism, hypercortisolism
Skin texture thickened	Acromegaly
Skin thinning/atrophy(increased skin vulnerablility to minor trauma)	Cushing disease
Weight loss or gain	Thyroid dysfunction, Cushing syndrome
Wound healing impaired	Androgen excess or relative lack of oestrogens Diabetes, hypothyroidism, hypercortisolism

Human keratinocytes not only express functional adrenergic and cholinergic receptors but also synthesize and metabolize corresponding ligands (catecholamines, acetylcholine). Additionally, the skin displays a complex endocannabinoid and endovanilloid neuroendocrine signalling system, complete with intracutaneously expressed cannabinoid and vanilloid receptors, locally produced (typically lipid-based) ligands and a refined enzymatic machinery for synthesizing and degrading the latter. This machinery controls the intracutaneous level of these versatile neuromediators. The list of human skin functions known to be modulated by these systems is growing steadily: to name a few, it currently includes the regulation of keratinocyte and sebocyte proliferation, migration, apoptosis, differentiation, cytokine secretion and lipid production; mast cell differentiation from resident precursors and mast cell degranulation; control of epidermal and/or hair pigmentation; and the modulation of immune responses (see Tables 149.1 and 149.2). This list is bound to grow.

These multipurpose neuromediators are complemented by an array of intraepithelially generated classical (neuro-)hormones such as melatonin [75, 77], the pro-opiomelanocortin (POMC) products, α-MSH, ACTH and β-endorphin [46, 76, 125], as well as CRH [26, 45, 48, 126, 127], erythropoietin [121, 122, 123], neuropeptides released by sensory nerve fibres (mainly substance P, calcitonin gene related peptide (CGRP) and vasoactive intestinal peptide (VIP) [18, 128]) and cytokine-like hormones released by skin adipocytes (adipokines, e.g. leptin) [22], with important differences in the secretory profile of what is now called ‘dermal adipose tissue’ and adipocytes of the deeper subcutis [129, 130]. These neuromediators both exert long-distance effects (e.g. regulation of the hypothalamic controls of feeding behaviour) and regulate skin physiology and repair in a para- and autocrine manner via the stimulation of locally expressed specific receptors.

In addition, human skin and its appendages synthesize opioids (e.g. β-endorphin, a psychotropic, analgesic, and pigmentation-stimulatory POMC product) and enkephalins and express corresponding receptors. Apart from the long-recognized role of these neuromediators in the modulation of itch and pain (via the stimulation of cannabinoid, vanilloid, opioid or enkephalin receptors expressed by intracutaneous sensory nerve fibres and the corresponding neuons in dorsal root ganglia) they are now recognized as regulating numerous other aspects of skin biology. Examples of these are listed in Table 149.1.

Since Merkel cells engage in substantial neuroendocrine secretory activities (e.g. CGRP, VIP, neuropeptide Y, neurokinin A, galanin, substance P and somatostatin) [109, 113, 131, 132], their local functions within skin epithelium are likely to extend well beyond their long-recognized role as mechanosensory cells [132, 133, 134, 135].

This complex hormone and neuromediator-based dermatoendocrinological signalling mechanism intimately links specialized (neuro-)endocrine glands, the nervous system and peripheral tissue physiology, and is further complicated by the fact that the same hormones and neuromediators also regulate immune responses. Many of these substances are generated and secreted by immunocytes such as mast cells, T cells and macrophages [136, 137]. These immunocytes in turn, are now understood to alter peripheral tissue function in the skin and elsewhere, not only under conditions of infection, inflammation or tumour growth but also essentially ‘around the clock’, thereby contributing substantially to skin homeostasis [8, 13, 18, 138].

To illustrate this, skin mast cells are highly sensitive to neuroendocrine stimuli (e.g. by CRH, ACTH, endocannabinoids, substance P and catecholamines) and are often in intimate physical contact with sensory nerve fibres that release neuropeptides such as substance P or CGRP [16, 18, 64, 97, 127, 139, 140]. At least in mice, mast cells regulate T-cell function [141, 142], angiogenesis [143, 144], connective tissue turnover (including ‘mast cell-directed collagenolysis’) [145], wound healing [146] and hair growth; they also detoxify venoms [147, 148]. Thus, the neuroendocrine controls that these key protagonists of innate immunity are subjected to must have a profound impact on skin immune reponses and other aspects of skin function in health and disease. We can assume that every resident and transient cell found in human skin can respond to multiple different hormones and neuromediators, can generate, secrete and/or metabolize many of these, and can engage in very complex endocrinologically and neuroimmunologically relevant signalling interactions with its neighbours. Some of these signals can reach distant organs via the bloodstream or even the CNS by manipulating the firing sequence of action potentials generated by sensory skin nerve fibres.

Skin as a hormone target

Insight into the complexities of the skin's endocrine and neuroendocrine pathways can provide the practising dermatologist not only with a better understanding of how hormones such as glucocorticoids, retinoids and calcitriols exert their therapeutic action but also with valuable assistance in recognizing classical endocrine and neuroendocrine diseases that can affect the skin (see later).

The classical signalling scenarios consist in either stimulation of specific cell surface receptors (e.g. by peptide hormones, neuropeptides, neurotransmitters or lipid neuromediators), resulting in rapid cell responses or binding to nuclear hormone receptors, which produce slightly less immediate responses (as seen with all steroid hormones) [5, 149, 150]. While the effects of hormones and other mediators are commonly ascribed to their interaction (or lack of interaction) with cognate intracutaneously expressed receptors, it is unclear how exactly a given hormone or neuromediator induces any of the defined skin phenomena discussed later. Tables 149.1 and 149.2 list examples of important receptors for which there either is already persuasive clinical or preclinical evidence of a significant impact on human skin physiology or pathology, or for which studies with mutant mice have provided compelling evidence that signalling events mediated by these receptors regulate important mammalian skin functions in vivo.

When examining the skin as a hormone target, one needs to consider the many different means by which a given hormone or neuromediator can signal. Some steroid hormones (e.g. glucocorticoids, calcitriols and oestradiol) can exert very rapid and transient signalling effects via cell surface receptors as well as slower responses via classical interaction with hormone response elements in nuclear DNA. In addition, steroid hormones typically show dose-dependent ‘promiscuous’ effects, because they can also bind to receptors other than their chief signalling partner and can then influence the binding properties of these receptors to their main ligands [5, 25]. Such receptor promiscuity and cross-regulation has also been shown to occur with some peptide hormones and lipid-based neuromediators [25, 149, 150, 151].

Many of the biological effects of hormones depend on the extent to which a cell stimulated by a given hormone generates secondary mediators, such as the production of insulin-like growth factor 1 (IGF-1) by insulin and growth hormone (GH) (synonym: somatotropin (STH)), or whether the stimulation of a hair follicle with the potent androgen dihydrotestosterone (DTH) results predominantly in the production of hair growth inhibitory growth factors (e.g. TGFβ1, TGFβ2) or the hair growth promoting factor IGF-1 [50]. Thus, to understand the role of GH in human epidermal physiology [152] or of androgens in scalp hair follicle biology [153] it will be necessary to separate the direct from the indirect and perhaps clinically even more important growth factor-mediated effects of these hormones.

Another factor to be taken into account is the extent to which a given target tissue in the skin is capable of transforming (typically enzymatically) a prohormone into active metabolites [156]. This can make a significant difference to the biological outcome of hormone stimulation of a defined skin cell population or structure and has been extensively studied for intracutaneous steroid hormone metabolism. Examples include the intracutaneous conversion of the relatively weak androgen testosterone into the highly active androgen DTH by 5-α-reductase and of androgens into oestrogens (e.g. of testosterone to 17-β-oestradiol by aromatase) as well as vitamin D metabolism [10, 41, 154, 155]. Key examples for intracutaneous neurohormone metabolism are the conversion of POMC by prohormone convertases to either ACTH, α-MSH or β-endorphin [8, 10]. In addition, ligands produced in the skin (e.g. sex steroids, glucocorticoids, retinoids and β-endorphin) can pass through the blood–brain barrier to the CNS, where they may be further metabolized, producing complex neuropsychological and neuroendocrine responses with potential effects directed back at the skin.

Other ligands can exert additional, receptor-independent effects, such as the direct, intra- and extracellular scavenging of reactive oxygen species by melatonin [75], or the modulation of tyrosinase activity by α-MSH [156]. Finally, the specific biological activities exerted by most hormones or neuromediators in a given skin territory depend on how quickly they are degraded by local enzymes (e.g. rapid degradation of substance P by neutral endopeptidase and angiotensin-converting enzyme, or of endocannabinoids by fatty acid amide hydrolase (FAAH)), the activity of which may vary from one specific skin territory to another. All these signalling variations are again influenced by the simultaneous presence of other ligands, co-factors, decoy receptors, binding proteins, the overall cytokine signalling milieu (e.g. whether or not the skin area in question is inflamed) and numerous other variables.

Matters are further complicated by substantial gender and/or regional differences in the response of human skin and its appendages to a given hormone or neuromediator. While this has long been appreciated for androgens, as for example reflected in the ‘paradoxical’ response to DTH of beard versus temporofrontal scalp hair follicles [50, 153], it also applies to key hormones such as 17-β-oestradiol [53, 157] and prolactin [20]. Another example where sex steroid controlled gender differences have been demonstrated is cutaneous wound healing, which is promoted by oestrogens but inhibited by androgens [158, 159, 160, 161].

This level of (neuro-)endocrine signalling complexity explains why it is often extremely difficult to pinpoint mechanistically exactly how a given hormone or neuromediator has induced the human skin phenomena we observe in an individual patient. Conversely, it also explains why the same dose of the same agent used in hormonal dermatotherapy (e.g. in the management of acne or psoriasis) can produce such distinct clinical results in different individuals, despite belonging to the same gender and age group and having a comparable medical background. Obviously, dermatologists need to be aware of these hidden (neuro-)endocrine dimensions of their daily work.

Additional reasons why dermatoendocrinological considerations are inescapable in routine dermatological practice are listed below:

Key dermatological symptoms and signs such as pruritus, flushing, erythema, eczema and wealing, and complex skin parameters such as skin barrier function, wound healing, hair growth, pigmentation and skin immune status, are greatly influenced by the characteristics of this intracutaneous (neuro-)endocrine signalling mechanism.
The intraepithelial endocrine and neuroendocrine signalling milieu that is created under physiological circumstances in human epidermis and hair follicles is probably predominantly immunoinhibitory [3, 8, 9, 57], and this may well be a fundamental prerequisite for skin homeostasis: increasing evidence suggests that maintenance of this signalling milieu plays a key role in preventing excessive skin inflammation and itch that would otherwise result from the continuous onslaught of environmental stressors such as microbial pathogens, excessive colonization with skin microflora, UV irradiation, physical and chemical skin trauma, or metabolic, oxidative and psychoemotional stressors. Imbalances in this neuroendocrine signalling system may influence dermatoses where neurogenic skin inflammation plays a significant role and which are known to be triggered or aggravated by environmental stressors (e.g. psoriasis, atopic eczema, alopecia areata, urticaria, pruritus).
The skin's response to excessive or deficient systemic hormone levels is greatly influenced by underlying systemic and local factors, ranging from UV exposure, humidity, barrier function via inflammation and microbiological colonization of the skin to a patient's psychoemotional status, medication and concomitant chemical or physical skin trauma. All these factors impact on the endocrine and neuroendocrine signalling concert of human skin.
Emerging experimental evidence suggests that important stress-associated neuropeptides such as substance P may even modulate the composition of the skin microbiome [162].
Ageing of the human organism is associated with significant changes in systemic hormone levels, particularly in circulating androgens and oestrogens. These changes, such as the declining 17-β-oestradiol serum levels during and after the menopause, affect human skin on many more levels than were previously recognized, ranging from changes in overall skin architecture, skin immune responses and cutaneous microbiology via altered skin barrier function, cutaneous drug absorption and metabolism to hormone-dependent effects on hair growth, sebum production and wound healing [163, 164, 165, 166, 167, 168].

Neuroendocrine stress response systems in human skin and the brain–skin axis

The importance of hormones in dermatology is further underscored when considering the role of psychoemotional stress as a triggering or aggravating factor in common dermatoses such as psoriasis, atopic eczema, urticaria, nodular prurigo, lichen planopilaris or alopecia areata (see corresponding chapters, and Chapter 86). Rather than reliance on questionable psychoanalytical explanations, solid experimental and clinical data can now explain how major neuroendocrine stress mediators such as CRH, ACTH, prolactin and substance P may trigger or aggravate skin disease [9, 13, 17, 18, 49, 151, 169, 170, 171]. Stress-induced neurogenic skin inflammation represents the best-defined neuroendocrine explanation for how psychoemotional stress influences dermatoses such as psoriasis, atopic eczema or urticaria [13, 16, 18, 172, 173, 174, 175].

Pro-inflammatory activities of skin mast cells assume a ‘central switchboard’ role in neurogenic skin inflammation. Mast cells undergo enhanced degranulation after direct stimulation by increased serum and tissue levels of stress-associated mediators such as CRH, ACTH, nerve growth factor (NGF), substance P and NGF, all of which act as secretagogues for human skin mast cells. The latter are often found in close proximity to sensory nerve fibres and it has been shown that neuropeptide-releasing sensory neurons are stimulated by NGF to synthesize substance P for transport via sensory nerve fibres to the skin, where it is able to provoke or exacerbate skin inflammation [13, 49, 98]. Simultaneously, psychoemotional stress can up-regulate the intracutanous generation of stress-response hormones, such as CRH and ACTH [16, 18, 49, 169, 175, 176].

Certain stress-associated hormones are an important element of what has been termed ‘fetal programming’, a process during which maternal stress impacts on the offspring's stress responses in later life [177, 178]. For example, an inverse association between the maternal serum progesterone level and the risk of girls subsequently developing atopic eczema has been described. Since progesterone is thought to operate as an endocrine feto-maternal ‘stress sentinel’ and to promote fetal tolerance it is, thus, conceivable that a lowered maternal progesterone level may predispose the fetus to the development of atopic eczema [179].

These interactions along the brain–skin axis [13, 17, 98, 140, 176, 180, 181, 182] can recruit a cascade of secondary inflammatory events, thus conspiring to trigger or aggravate inflammatory, pruritic and/or hyperproliferative dermatoses. Conversely, most recent neuroimaging evidence suggests that chronic skin inflammation can exert profound retrograde pro-inflammatory effects on the human brain, for example in patients with psoriasis [183]. This may even negatively affect their cognitive performance [184]. If confirmed in follow-up studies, the presence of central neuroinflammation and cognitive impairment in association with chronic inflammatory dermatoses will add major new diagnostic and therapeutic challenges to the management of these patients.

Human skin and hair research models as discovery tools for general neuroendocrinology

It should be emphasized that a stringent (neuro-)endocrinological approach to the investigation of skin disease has already contributed to major translationally relevant progress in general endocrine and neuroendocrinological research, e.g. by organ-culturing intact human skin and scalp hair follicles and by studying human keratinocyte and sebocytes cell cultures.

Apart from the discovery that the skin and its pilosebaceous units rank among the most endocrinologically active organs of the human body and display regulatory neuroendocrine signalling loops that parallel in complexity those found in the central HPA and HPT axes [3, 6, 7, 8, 9, 39, 57], this line of research has identified novel neuroendocrine controls of pigmentation [15], e.g. the discovery of β-endorphin as melanotropin [125], and of TRH as a promoter not only of human hair pigmentation [82] but also of human epidermal re-epithelialization after wounding [161], suggesting that TRH is involved in the control of wound healing.

Human skin and hair research has helped to show that α-MSH, apart from its long-known key role as a pigmentation stimulatory melanocortin, exerts multiple anti-inflammatory activities [185] and acts as a powerful immune privilege guardian in human skin [46, 186, 187, 188, 189, 190]. Intriguing leads from mouse research suggest that α-MSH may also act as a potential antifibrotic neurohormone. This may be of relevance to the pathogenesis of systemic sclerosis [191, 192] and studies on human skin in situ are awaited to see whether α-MSH might promote the development of tolerogenic dendritic cells and/or regulatory T cells [193]. Preclinical evidence already suggests that α-MSH may exert local anti-allergic activity by inhibiting human basophils both in vitro and in situ [43] and that it may promote systems that limit UV-induced oxidative damage in human skin [194].

Prolactin and thyroid hormones have also been shown to act as important, previously unknown, stimulators of human hair follicle epithelial stem cells in situ [91, 103, 195]. Through the use of human skin and hair follicle organ culture assays, TRH, TSH, prolactin and the cannabinoid system have all recently surfaced as powerful novel neuroendocrine regulators of keratin gene and protein expression (e.g. TRH, TSH, prolactin and endocannabinoids potently regulate intraepidermal and intrafollicular expression of selected keratins, including stem cell and hair shaft associated keratins in situ; see [58, 87, 196, 197]). This has revealed an entire new level of keratin expression control.

The systematic dissection of vitamin D metabolism in human skin and keratinocytes [41, 198] has also identified an entire new line of secosteroids with intriguing therapeutic potential [39, 57]. The analysis of PPARγ-mediated signalling in human skin in health and disease has shown that these nuclear hormone receptors and their agonists operate as important ‘guardians’ of human epithelial (hair follicle) stem cells [59, 85, 199] and as molecular curbs on excessive skin inflammation [59, 200]. PPARγ stimulation may also help to reduce skin ageing induced by photoxidative damage [201].

Finally, recent research in healthy organ-cultured human epidermis and hair follicles has revealed that TRH and TSH act as previously unsuspected potent neuroendocrine stimulators of human mitochondrial activity and even mitochondrial biogenesis in situ [83, 84, 86]. This important and novel insight into the neuroendocrine control of mitochondria had been missed by mainstream mitochondrial and neuroendocrinology research, which has traditionally focused on the role of thyroid and steroid hormones as stimulators of mitochondrial function in tissues other than skin. Given the central role of mitochondrial dysfunction in ageing and in many degenerative diseases including maternally inherited mitochondrial DNA disorders [202, 203, 204, 205], this discovery opens up interesting new avenues for clinical research (see later).

These few examples demonstrate that the study of skin from a (neuro-)endocrinological perspective promises benefits well beyond the integument. They also show that human skin and pilosebaceous research models provide excellent discovery tools for exploring the full range of physiological activities in which neurohormones and other neuromediators play a part in human biology [3, 8, 87]. In this respect, important clues may be drawn from the study of evolutionarily much older vertebrate skin, namely that of frogs, which generates a striking variety and quantity of neurohormones and neuropeptides. An understanding of these may help to reveal evolutionarily conserved but as yet unappreciated functions common to both frog and human skin [161, 206].

(Neuro-)endocrine contributions to cutaneous pathogenesis

The example of stress-induced neurogenic skin inflammation (see earlier) has already illustrated how neuroendocrine and neuroimmunological mechanisms can contribute to the pathogenesis and course of stress-triggered or aggravated human skin diseases. Table 149.2 lists selected dermatoses for which a neuroendocrine contribution to disease pathogenesis has been postulated. For example, there is increasing evidence that the modulation of immune responses by a wide range of neurohormones, neuropeptides and neurotransmitters may contribute to the development and/or clinical course of psoriasis, atopic eczema, urticaria, pruritus, alopecia areata, systemic lupus erythematosus, systemic sclerosis, retarded wound healing and melanoma (see Table 149.3 for details and references).

However, conclusive proof that neuroendocrine mechanisms contribute fundamentally, rather than peripherally, to the primary pathogenesis of the most common human skin diseases such as atopic eczema and psoriasis, as opposed to playing a role in triggering or aggravating them, is still missing. Regrettably, a stringent neuroendocrine approach is rarely adopted when investigating human skin diseases. Mainstream neuroendocrinology research has been very slow to recognize and adopt human skin and its appendages as instructive research objects and experimental models.

One notable exception is the long line of research that has shed light on a significant β-adrenergic signalling defect in atopic eczema. Starting from Szentivanyi's β-adrenergic theory of atopy [255], this has arguably come very close to demonstrating primary relevance for the pathogenesis and neuropharmacological management of atopic diseases, including atopic eczema [68, 69, 256, 257, 258]. While the role of β-adrenergic receptors in haemangioma pathogenesis remains ill understood, the often impressive therapeutic response seen with β-blocker administration in haemangioma [259] certainly serves as an additional encouragement to undertake a systematic re-exploration of the role of catecholamines and their receptors in human skin physiology and pathology [30, 33, 66, 67, 6870, 71, 77260, 261, 262, 263, 264, 265]. It opens up the possibility that adrenergic receptor agonists and antagonists, which have already been used for decades in clinical medicine, may yet find new roles in the management of skin disease.

A second important exception is the melanocortin receptor type 1 (MC1R), the chief receptor for α-MSH, whose loss-of-function polymorphisms (found in red-haired individuals who fail to tan) is associated with a significantly increased risk of developing melanoma [231, 232, 233, 234, 266] (see Chapter 143). α-MSH also operates as a powerful immune privilege guardian in human skin [46, 186, 189]. A relative insufficiency of α-MSH/MC1R-mediated signalling may contribute to the collapse of immune privilege in the anagen hair bulb, a key element in the pathogenesis of alopecia areata [46, 189, 267]. Likewise, it has been speculated that insufficient α-MSH/MC1R-mediated signalling at the hair follicle bulge, the repository of follicle stem cells, may contribute to the pathogenesis of lichen planopilaris, facilitating the collapse of immune privilege at that site [190].

A third area where endocrine research has contributed to an understanding of skin disease pathogenesis is the study of endogenous or exogenously administered vitamin D and vitamin A derivatives in a wide range of dermatoses. It would not be surprising to find that abnormalities in intracutaneous calcitriol and retinoid synthesis and metabolism have an impact on an individual's susceptibility to skin diseases such as skin cancer, alopecia, retarded wound healing and acne, or on that individual's response to therapy with steroid hormones. The first intriguing examples of the potential importance of this new endocrine dimension in dermatology have already emerged [19, 41, 268, 269, 270, 271, 272, 273].

It is in this context of widely underappreciated progress in cutaneous (neuro-)endocrinology that traditional clinical dermatoendocrinology is still unfolding.

Basics of clinical dermatoendocrinology

How to evaluate a patient for a suspected (neuro-)endocrine disorder

Hormones may control or influence general body characteristics such as height, weight, body contour and posture, mood, agility, nervousness, hair phenotype, and also food and fluid intake. Being alert to changes in such general characteristics will therefore greatly help to identify a potential endocrinological dimension in patients presenting with a skin complaint. At the very least, when a patient presents with any of the lead signs or symptoms in Table 149.3 a hormonal basis for the dermatological problem should be considered and systematically confirmed or excluded.

Some characteristic skin signs provide invaluable indicators of specific endocrine diseases (Table 149.4). Additional ‘diagnostic pearls’ that will greatly help in identifying a potential underlying endocrine pathology are summarized in Table 149.5.

Table 149.4 Characteristic skin signs indicating specific endocrine diseases.

Sign	Figure	Associated endocrine disease/condition
Acanthosis nigricans (see Chapter 87)	Figure 149.3	Puberty, diabetes, other causes of insulin resistance including HAIR-AN syndrome of young black females (hyperandrogenism, insulin resistance, aconthosis nigricans; often associated with polycystic ovary syndrome, hirsutism and others signs of androgen excess), acromegaly
Acne (see Chapter 91)	Figure 149.4	Cushing syndrome, glucocorticoid therapy, ACTH-secreting tumours (e.g. small cell bronchial carcinoma, acromegaly)
Cutis verticis gyrata (see Chapter 107)	Figure 149.5	Acromegaly
Flushing (see Chapter 106)	Figure 149.6	Menopause-associated hot flushes Phaeochromocytoma, carcinoid
Galactorrhoea		Hyperprolactinaemia (e.g. due to prolactinoma, tumour-associated ectopic prolactin production or medication with neuroleptic drugs, oral contraceptives, tricyclic antidepressants)
Granuloma annulare (see Chapter 97)		Diabetes (see Chapter 64)
Gynaecomastia (usually, but not always, symmetrical)	Figures 149.7 and 149.8	Puberty, hypogonadism (e.g. Klinefelter syndrome), pituitary and gonadal tumours (e.g. prolactinoma), testis carcinoma), excessive endogenous oestrogen production (e.g. oestrogen-secreting tumours), insufficient oestrogen metabolism (e.g. liver cirrhosis), drugs (e.g. oestrogens, spironolactone, isoniazid, resumption of normal hypophyseal gonadotropin secretion after longer period of hunger, extreme diet, or major consumptive disease), hypopituitarism
Hyperpigmentation	Figure 149.9	Addison disease
Hypertrichosis (see Chapter 89)		Excess ACTH production (tumour, Cushing disease), hyperthyroidism (often in association with myxoedema in Graves disease)
Melasma (see Chapter 88)	Figure 149.10	Pregnancy, oral contraceptives, oestrogen-secreting tumour
‘Moon facies’		Cushing syndrome
Myxoedema, pretibial (see Chapter 59)	Figure 149.11	Hypo- or hyperthyroidism (notably in Graves disease)
Necrobiosis lipoidica (see Chapter 97)		Diabetes (search for additional cutaneous complications of diabetes in Box 149.1 and in Chapter 64)
Necrolytic migratory erythema (see Chapter 147)	Figure 149.12	Glucagonoma
Palmar erythema		Hyperthyroidism
Scleredema adultorum (Buschke) (see Chapter 59)		Diabetes
Stretch marks (striae distensae) (see Chapter 96)	Figure 149.13	Cushing disease, glucocorticoid therapy, ACTH-secreting tumours (e.g. small cell bronchial carcinoma); pregnancy, contraceptive therapy, puberty (growth spurt, namely in adipose individuals)

Adapted from Braverman 1998 [211], Du Vivier 2002 [274] and Luger and Böhm [275] © John Wiley.

Table 149.5 Additional ‘diagnostic pearls’ in clinical dermatoendocrinology.

Sign/symptom	Figure	Consider presence/role of…
Alopecia areata (see Chapter 89)		Thyroid autoimmune disease
Alopecia, rapidly progressing pattern balding (especially in female patients)		Virilizing tumour, adrenogenital syndrome; pregnancy or discontinuation of oral contraceptive; thyroid dysfunction, hyperprolactinaemia
Axillary hair, diminished growth		Panhypopituitarism
‘Buffalo hump’		Cushing syndrome
Digital clubbing (see Chapter 95)		Hyperthyroidism (thyroid acropachy), hyper- or hypoparathyroidism (periosteal formation of new bone)
Epidermoid cysts		Acromegaly
Eyelids, thickened/oedematous		Acromegaly
Gingival or oral hyperpigmentation	Figure 149.14	Addison disease (gingival hyperpigmentation: consider also hyperthyroidism and Cushing disease; dintinguish from ephelide-like hyperpigmentation of Peutz–Jeghers syndrome)
Hair, repigmentation of grey hair		ACTH-producting tumour (e.g. pituitary, or ectopic ACTH production by small cell bronchial carcinoma)
Macroglossia (tongue often also fissured)	Figure 149.15	Acromegaly; hypothyroidism (usually less pronounced than in acromegaly)
Mastitis, neonatal		Temporary stimulation of mammary gland by maternal hormones
Nails, thickened and hardened		Acromegaly
Sebaceous gland hypertrophy (see Chapter 93)		Acromegaly
Seborrhoea		Virilizing tumour, hyperprolactinaemia Parkinson disease (→ insufficient dopamine production, associated with hyperprolactinaemia)
Skin tags (acrochordons)		Acromegaly
Telogen effluvium (see Chapter 89)		Thyroid dysfunction, hyperprolactinaemia, virilizing tumour, adrenogenital syndrome; polycystic ovary syndrome; pregnancy or discontinuation of oral contraceptive
Ulcer, gangrene		Diabetes
Urticaria (see Chapter 42)		Chronic urticaria associated with thyroid hormone abnormalities (due to Hashimoto thyroiditis or Graves disease)
Vitiligo (see Chapter 88)	Figure 149.16	Can be associated with thyroid autoimmune disease and Addison disease

Adapted from Braverman 1998 [211], Du Vivier 2002 [274], Luger and Böhm [275], Jabbour 2003 [276] and Jabbour 2010 [277] © John Wiley.

Comparing a patient's current facial appearance with an older photograph, e.g. on a driver's licence or stored on a mobile phone, can offer important, rapidly collectable information about changes in overall facial features, as may occur from acromegaly or Cushing disease.

Given the increasing incidence and prevalence of diabetes in most societies, it is important to consider this as a factor. The cutaneous features of diabetes (Box 149.1) are discussed elsewhere (see Chapter 64).

Box 149.1 Cutaneous complications of diabetes (see Chapter 64)

Infections (notably, oral and genital candidosis, Candida intertrigo, recurrent folliculitis, abscess and erysipelas)
Ulcers and gangrene (as secondary skin changes due to diabetic microangiopathy and neuropathy)
Necrobiosis lipoidica
Diabetic rubeosis
Lipoatrophy/lipohypertrophy (including post-insulin injection)
Diabetic dermopathy (Binkley spots)
Diabetic sclerodactyly (diabetic stiff skin, diabetic cheiroarthropathy)
Diabetic scleredema
Diabetic bullae (bullosis diabeticorum)
Acanthosis nigricans (including its possible minor/disseminated variant, finger pebbles) [278]
Carotenaemia (‘aurantiasis cutis’, xanthochromia due to β-carotene accumulation in stratum corneum associated with diabetic hypercarotenaemia, especially over the knuckles, elbows, knees and in palmoplantar skin)
Acquired perforating dermatosis
Eruptive xanthomas
Generalized granuloma annulare
Vitiligo
Cutaneous adverse effects of antidiabetic agents (including phototoxic and photoallergic reactions)

There are also the cutaneous consequences of hypo- or hyperthyroidism to be considered. Finally, given that many tumours can be hormonally active, if still in doubt diagnostically, it is possible that the observed skin phenomena may have been caused by a hormone-secreting tumour, such as skin hyperpigmentation, acneform lesions and/or cushingoid features in ACTH-secreting small cell bronchial carcinoma [279], sudden attacks of hyperhidrosis (typically along with headache, tachycardia and hypertension) in catecholamine-secreting phaeochromocytoma, or necrolytic migratory erythema in glucagonoma syndrome [280, 281] (see Tables 149.4 and 149.5). (For details on paraneoplastic skin disease, see [277, 282] and Chapter 147.)

Endocrinological considerations in skin therapy

Once an endocrinological diagnosis has been made or is suspected, patients should be referred to an appropriate specialist, such as an endocrinologist or neurosurgeon, to undergo further diagnostic procedures and appropriate management. However, even replacement or suppressive therapy (as appropriate) may not always result in a rapid return of the skin to its premorbid state. The skin's response to corrective therapeutic intervention, for example with thyroxine, removal of a pituitary tumour or treatment of diabetes, can be very slow and protracted or may be impossible to achieve.

Some skin changes that occur in association with endocrine disease (e.g. necrobiosis lipoidica, lipoatrophy, striae distensae, ulcers, gangrene; see Tables 149.4 and 149.5) routinely result in irreversible skin damage, where progressive loss of function can only be halted if the underlying endocrine abnormality is aggressively sought and eliminated.

If a patient is known to have an endocrine disorder, whether that has come to light from its cutaneous manifestations or not, it should not be forgotten that the endocrine abnormality may also aggravate, increase susceptibility to or alter response to therapy of other skin disorders, including psoriasis, atopic eczema, alopecia areata and acne, and that it may affect wound healing (see Table 149.3). Accelerated skin ageing may also result from endocrine disease. The possibility that neuroendocrine abnormalities associated with psychoemotional stress may have triggered or aggravated a dermatosis via the induction of neurogenic skin inflammation (see earlier) should always be carefully considered and discussed with the patient.

Application of exogenous glucocorticosteroid hormone to human skin is likely to have a significant impact not only on the metabolism and synthesis of endogenous steroid hormones in human skin but also on intracutaneous neuroendocrine signalling axes, which affect the secretion of potent growth factors/cytokines or the signalling of receptors that are not classical targets of the administered therapeutic hormone. Glucocorticoid application may also synchronize or reset peripheral clock gene activity in human skin and its appendages. Given the increasing insight into the regulation of such diverse aspects of skin physiology as hair growth and pigmentation by changes in clock gene activity [283, 284, 285, 286, 287], this chronobiological dimension of glucocorticoid therapy could be more important than previously appreciated.

Whilst this field of dermatoendocrinology is still in its infancy and its practical implications for clinical disease management are still unclear, the dermatologist should be aware of the complex interplay between hormones and the skin. For example, it is known that cortisol administration reduces the intrafollicular expression of CRH in human scalp hair follicles ex vivo [26]. Although it remains to be formally demonstrated that this also happens in vivo, it is eminently conceivable that chronic glucocorticoid administration reduces the skin's CRH-dependent constitutive cortisol synthesis as well as that in the adrenal gland [9, 26]. This iatrogenic disruption of the brain–skin axis may contribute to the classical rebound phenomena typically seen after withdrawal of glucocorticoid therapy [3, 7, 9]. Moreover, all-trans-retinoic acid modulates the intrafollicular expression of key growth factors; that is, it up-regulates TGFβ1 and TGFβ2, thereby switching scalp hair follicles from anagen to catagen; this may explain why patients under retinoid therapy can experience a telogen effluvium [288]. Instead, the frequently observed retinoid-associated skin irritation results partly from sensory hypersensitivity induced by retinoid activation of a vanilloid receptor (TRPV1) [289].

Dermatologists should remember that abuse or misuse of hormonally active substances (e.g. anabolic hormones or glucocorticosteroids) may manifest in the skin; a patient's medication may also result in hormonally-mediated adverse cutaneous effects. These may be obvious when they result from systemic or high potency topical glucocorticosteroids or systemic retinoids but less so when due to anabolic agents or hormonal contraceptives. Adverse cutaneous effects may also manifest when a patient has taken novel neuroendocrine agents, such as the synthetic α-MSH analogue, Melanotan, which is increasingly (ab-)used for tanning purposes. This agent may also stimulate the appearance of multiple melanocytic naevi and/or dysplastic changes in existing naevi (see Table 149.3).

Other medications have less obvious (neuro-)endocrinological effects on the skin. For example, dopamine exerts direct hair growth-inhibitory properties on human scalp hair follicles [101]. This may explain why bromocriptine, the dopaminergic inhibitor of pituitary prolactin secretion used for treating prolactinoma patients, can cause effluvium in female patients [290]. Neuroleptics and other antipsychotic agents frequently cause hyperprolactinaemia [291]. Therefore, it is reasonable to consider whether skin abnormalities in which excessive prolactin signalling may play a part (see Table 149.3) might have been aggravated by such a medication, or whether standard therapy may have been less effective than expected in patients treated with these drugs. Where medically justifiable, it may be worth temporarily discontinuing or at least reducing such medication to observe whether this improves the dermatological complaint.

Long-term ciclosporin therapy not only frequently results in hypertrichosis, but rarely can also lead to reversible gynaecomastia associated with hyperprolactinaemia [292]. In murine skin, ciclosporin controls hair follicle stem cell activation and thus hair growth in a prolactin receptor-dependent manner [293], while it prolongs the duration of anagen in human scalp hair follicles [294]. It has been suggested, somewhat controversially, that the efficacy of ciclosporin therapy might be enhanced by co-administering bromocriptine, which lowers the serum prolactin level [295], while neuroleptics which cause hyperprolactinaemia may reduce the effectiveness of immunosuppression by ciclosporin [296].

Angiotensin-converting enzyme inhibitors are the leading cause of drug-induced angio-oedema and account for up to one third of cases of angio-oedema presenting to emergency departments [297, 298]. These agents inhibit the intracutaneous degradation of substance P, bradykinin and other pro-inflammatory, vasoactive and mast cell activating neuropeptides and as a result can aggravate neurogenic skin inflammation, vasodilatation and extravasation [16, 25, 49, 299, 300].

These considerations illustrate the value of a (neuro-)endocrinological perspective when approaching and managing patients with skin disease. Traditionally, however, it is the diagnostic benefits of such a perspective that secure the place of hormones in the dermatologist's mind.

Systematic review of clinical dermatoendocrinology

All the central endocrine signalling axes (see Figure 149.1) and (neuro-)endocrine systems described earlier can show abnormalities that result in either excessive or insufficient hormone levels. In the following sections, the skin signs or symptoms summarized in Tables 149.4 and 149.5 are briefly discussed. Further clinical information can be obtained from relevant review articles [98, 207, 276, 277, 282, 301, 302] and Braverman's classical monograph [211]. For clinical illustrations, follow the indicators listed in Tables 149.4 and 149.5.

As the diagnostic confirmation and therapeutic management of the endocrinological disease states listed later typically lie in the hands of endocrinologists, paediatricians, gynaecologists or neurosurgeons, investigation and management are not discussed here: see [5, 151] for details on diagnostic procedures and disease management.

Hypopituitarism

Inherited or acquired insufficiency of pituitary hormone production can result from many different causes. These range from congenital defects in POMC synthesis [303] or processing, via traumatic brain injury, ionizing irradiation, tumours (e.g. metastases, craniopharyngioma, adenomata, Langerhans cell histiocytosis) and infection (e.g. tuberculosis, syphilis) to sarcoidosis and postpartum pituitary necrosis (Sheehan syndrome) [211, 276]. As all of these may impair pituitary hormone production, the associated skin phenotype is dominated by the main neurohormones that are produced in insufficient amounts.

Given the usually non-hormone-selective nature of hypopituitarism, the levels of several pituitary neurohormones are reduced simultaneously, and the serum levels of other key hormones (e.g. thyroxine, cortisol) and growth factors (e.g. IGF-1) regulated by the former are typically reduced as well. Thus, it is often impossible to discern exactly which hormonal imbalance has caused the observed skin phenotype in a given patient. Once hypopituitarism is clinically suspected, it is diagnostically important to identify which pituitary hormone(s) is/are deficient [276, 304]. Classically, the absence of melanotropic neurohormones (e.g. ACTH, α-MSH) results in pale hypopigmented skin, not uncommonly with a yellowish tint. Instead, GH (STH) and/or prolactin deficiency gives rise to structural skin changes, e.g. overall skin thinning and reduced sebum and sweat secretion. Rarely, gynaecomastia can be a result of secondary hypopituitarism [305], as in the sarcoidosis patient shown in Figure 149.17.

In a small cohort of children and adolescents with isolated GH deficiency or multiple pituitary hormone deficiencies, elastin fibres were abnormal: besides being reduced in number, they were shorter and slimmer, suggesting that GH and other pituitary hormones regulate elastogenesis in human skin [306]. Reduced collagen synthesis has also been seen in patients with hypopituitarism [307]. In GH-deficient patients with Sheehan syndrome, skin capacitance and sebum content were found to be decreased compared with control subjects [308]. Patients affected by congenital POMC deficiency [303] are often red-haired and show a pale skin and early onset of obesity.

Hyperpituitarism

The major disease states resulting from hyperpituitarism (i.e. acromegaly, hyperprolactinaemia, and Cushing disease) each display very characteristic skin signs that aid diagnosis.

In acromegaly, a hormonally active adenoma of the anterior pituitary secretes excessive GH, leading to a secondary rise in IGF-1 serum and tissue levels. The clinical consequences depend upon whether this condition starts during childhood, adolescence or adulthood. It is characterized by excessive growth of acral cartilage and skin with prominent epidermal hyperplasia and dermal glycosaminoglycan accumulation (hence the name). External manifestations include progressive enlargement of earlobes and fingers (‘My gloves don't fit any more’), increasingly coarse facial features with a prominent chin (Figure 149.4) and large frontal skin folds (‘thinker's folds’), which may be associated with cutis verticis gyrata (Figure 149.5). Other characteristic changes include hypermelanosis, hypertrichosis, seborrhoea, thickened finger- and toenails, acanthosis nigricans, hyperhidrosis and macroglossia (Figure 149.15; Tables 149.4 and 149.5) [5, 309]. Decreased transepidermal water loss, reduced skin surface temperature and an increase in skin pH and elasticity have also been reported in patients with acromegaly [242, 310].

The treatment of acromegaly by neurosurgery, radiotherapy, with GH antagonists (e.g. pegvisomant), somatostatin analogues (e.g. octreotide) or dopaminergic agents (e.g. bromocriptine, cabergoline) may partially reverse the cutaneous phenomena induced by long-term exposure to excessive GH serum levels. Unfortunately, the opportunities to investigate how these surgical and medical therapies affect human skin have been largely overlooked as study of this might shed new light on the still fairly obscure role of GH receptor mediated signalling in human skin physiology. It has, however, recently been shown that dopamine actually inhibits human scalp hair growth in vitro [101] and may thus be a functional antagonist to GH in terms of hair growth control.

Conversely, GH therapy (e.g. in patients with Turner syndrome) can enhance the growth of melanocytic naevi, whose melanocytes appear to be activated as deduced from their increased HMB-45 and Ki-67 immunoreactivity [243, 244]. Indeed, in murine wound healing studies, treatment with GH-releasing hormone (GHRH) or a GH agonist increased the density of fibroblasts during the early stages of wound healing and accelerated re-epithelialization later on. Therefore, the undesired dermatological phenomena seen in acromegaly may well indicate novel strategies for promoting cutaneous wound healing [152]. It has also been proposed that replacement therapy with GH or the key growth factor stimulated by it (IGF-1) represents a potential anti-skin ageing strategy that is worth systematically exploring [166].

Although prolonged hyperprolactinaemia can have many causes, it most frequently results either from a prolactin-secreting micro- or macroadenoma of the anterior pituitary gland or from drugs (e.g. phenothiazine, dopamine antagonists, oestrogens, morphine derivates, H2-antagonists and neuroleptic agents) [5, 311].

As with GH, a sustained elevation of prolactin serum levels results in increased serum and tissue IGF-1 levels [20]. Thus, there is some limited overlap in the cutaneous phenomena seen in acromegaly and hyperprolactinaemia, namely seborrhoea, though less than expected if increased IGF-1 secretion were the primary mode of prolactin action on human skin. Instead, hyperprolactinaemia in women is typically associated with seborrhoea, acne, hirsutism and androgenetic alopecia, a combination for which the term SAHA syndrome has been coined. SAHA can also be associated with ovarian and adrenal dysfunction (see Chapter 90). Almost pathognomonically, hyperprolactinaemia can induce galactorrhoea in both men and women, usually along with gynaecomastia in men.

However, recent research suggests that the prominent effects on the pilosebaceous unit and the mammary gland seen in patients with hyperprolactinaemia represent only a small part of prolactin's full range of actions on human skin in health and disease. These range from the modulation of keratinocyte and sebocyte proliferation, differentiation and apoptosis to the regulation of cytokine secretion and keratin expression, including the promotion of keratin 15 expression, a key epithelial stem cell marker, indicating a wider role for prolactin in epithelial stem cell biology; moreover, prolactin actions may differ between the genders [3, 88, 89, 91, 312]. If it is confirmed that prolactin also modulates peripheral androgen metabolism [313] and plays a significant role in cutaneous autoimmunity [314], this only serves as additional motivation to explore fully the impact of excessive or insufficient prolactin production by the pituitary gland, human skin or its appendages [88, 90, 312] on skin physiology and function.

Most patients affected by Cushing disease suffer from an ACTH-producing tumour, either in the anterior pituitary or in an extrapituitary location where the tumour (most frequently small cell bronchial carcinoma) engages in ectopic ACTH production. The associated skin phenomena are dominated by the pigmentary effects of ACTH and the hypercortisolism induced by the chronically elevated serum cortisol level [275] (see later for description of cutaneous phenomena). The observation that a patient shows proximal repigmentation of previously grey/white hair and/or hyperpigmented palmar lines (in an individual of Caucasian ethnicity) should immediately indicate screening for an ACTH-secreting tumour or Addison disease (see later).

Adrenal hyperfunction

There are three forms of adrenal hyperfunction.

Hypercortisolism (Cushing syndrome) This is most frequently iatrogenic and results from systemic or occasionally extensive topical glucocorticoid therapy. Less common are ACTH-producing tumours (Cushing disease) and, very rarely, CRH-producing hypothalamic tumours [277] or autoimmune stimulation of adrenal ACTH receptors (MC2R) by autoantibodies (Carney syndrome) [315]; the latter is to be distinguished from the Carney complex, an exceptionally rare, dominantly inherited syndrome associated with multiple endocrine neoplasias, endocrine overactivity, and spotty skin hyperpigmentation that also shows ACTH-independent Cushing features resulting from a mutation in the PRKAR1A gene (see Chapter 150) [316, 317, 318].

The non-pigmentary skin manifestations of both Cushing disease and Cushing syndrome are thought to result primarily from excessive adrenal cortisol production (hypercortisolism), even though it is now well documented that ACTH also stimulates steroidogenesis in human skin and its appendages [26, 57]. Therefore, it remains to be explored whether some of the steroid hormone-dependent cutaneous phenomena in Cushing disease are actually due at least in part to excessive compensatory intracutaneous steroidogenesis.

The dermatological manifestations of hypercortisolism (including that due to Cushing disease) include facial and neck changes due to altered subcutaneous fat distribution (‘moon face’, ‘buffalo hump’) (see Tables 149.4 and 149.5), generalized hypertrichosis, acne and the consequences of increased collagen breakdown, namely skin fragility, stretch marks, vascular fragility with resulting purpura and impaired wound healing. Hypercortisolism can also induce diabetes, thus adding to the spectrum of diabetes-associated skin phenomena (see Box 149.1).

Hyperaldosteronism Several indications from the experimental literature (in the skin of mineralocorticoid receptor overexpressing mice) would lead one to expect that hyperaldosteronism (Conn syndrome) should also affect human skin, which expresses the aldosterone receptor [319, 320]. However, prominent skin signs or symptoms have not yet been described in individuals with hyperaldosteronism.

Congenital adrenal hyperplasia (CAH, formerly known as adrenogenital syndrome, MIM201910) is the third form of adrenal hyperfunction and is in the majority of cases due to an inherited deficiency of 21-hydroxylase due to mutations in the CYP21 gene. This means that the cortisol precursor 17-hydroxyprogesterone cannot be metabolized to cortisol but is shunted towards enhanced androgen production instead. This results in more or less pronounced virilization with associated skin signs including acne and hirsutism. ACTH levels are increased due to the body's attempts to compensate for insufficient cortisol production and this may result in hypermelanosis, especially of the buccal mucosa, lips and genitalia. In women, the clinical features of CAH may be difficult to distinguish from those of polycystic ovary syndrome, idiopathic hirsutism or hyperinsulinaemia and the possibility of underlying CAH in women presenting with these signs should not be overlooked: short stature may be a helpful clue [321].

Adrenal insufficiency (Addison disease)

Insufficient adrenal cortisol production is most commonly the consequence of autoimmune disease (autoimmune adrenalitis) or of suppression of ACTH secretion following long-term glucocorticoid therapy. Infectious diseases (tuberculosis, systemic fungal infection) are also important causes. For a full list of the many rarer causes endocrinology textbooks should be consulted [5]. For the practising dermatologist, it is important to remember that recovery of endogenous ACTH production after long-term suppression by exogenous glucocorticoids may be delayed and give rise to the same clinical picture.

Vitamin B₁₂ deficiency may rarely result in epidermal hyperpigmentation similar to Addison disease but is easily distinguishable by its haematological abnormalities (megaloblastic anaemia) [322, 323]. In early-onset obesity that is not otherwise explained, adrenal insufficiency should be considered as a potential cause [324], as this may indicate a rare loss-of-function mutation in POMC, the precursor hormone from which ACTH is produced enzymatically [303].

Apart from general symptoms (see Table 149.1) hypermelanosis due to compensatory pituitary (and possibly intracutaneous) oversecretion of ACTH or its enzymatic cleavage product, α-MSH, is a prominent component of Addison disease. This is often first and best visible as hyperpigmented palmar skin creases but also affects intertriginous skin, external genitalia, nail beds, oral mucosa and lips [325]. The biological reason for this apparently UV-independent peculiarly distributed but characteristic pigmentation phenotype is not yet clear; however, one may speculate that melanocytes in the preferentially affected skin regions are particularly sensitive to ACTH or α-MSH stimulation, perhaps as a result of higher levels of MC1R or MC2R expression.

If Addisonian pigmentation presents in association with vitiligo (see Chapter 88) or, rarely, alopecia areata (see Chapter 89) it is reasonable to assume that it is due to autoimmune adrenalitis; this possibility should then be excluded or confirmed by a specialist to guide subsequent therapy. It has been argued that patients presenting with generalized hyperpigmentation of unclear origin should have serum ACTH and mineralocorticoid function checked without delay for undiagnosed primary adrenal insufficiency [326].

Hyperandrogenism

While the testes (and to a minimal extent ovaries) are the primary site of testosterone synthesis, androgens are also produced in the adrenal glands. The main adrenal androgens are dehydroepiandrosterone (DHEA) and androstendione. Androgens are further metabolized in their target tissues. The biologically most important enzymatic processes involved are the conversion of testosterone to DHT by 5-α-reductase and the metabolic conversion of androgens to 17-β-oestradiol [50, 52]. All relevant enzyme activities and androgen receptors are present in human skin, most prominently in its pilosebaceous units. It is thought that the higher androgen content of male skin partly explains why it is thicker, more hairy and slightly more pigmented than female skin [275, 327]. This is in accordance with the skin phenotype seen in hyperandrogenism.

Apart from exogenous androgen use or abuse (e.g. anabolic steroids) and CAH, polycystic ovary syndrome with the production of androgens by ovarian theca cells is the most frequent cause of hyperandrogenism in women [328, 329, 330]. Acromegaly, Cushing disease, androgen-secreting (virilizing) tumours of the ovary or adrenal gland, and hyperprolactinaemia represent other less common causes. Amongst the latter, hyperprolactinaemia is probably the most common.

The cutaneous hallmarks of hyperandrogenism are seborrhoea, severe acne, hirsutism, telogen effluvium and female or male pattern balding [214, 331, 332]. Hyperandrogenism should be considered as a potential underlying cause of late-onet acne in women. Other signs of virilization (e.g. male habitus in women, deepening of the voice, clitoral hypertrophy) may make the clinical diagnosis of hyperandrogenism very straightforward.

A full endocrinological assessment should however be undertaken without delay. Polycystic ovary syndrome is often coupled with hyperprolactinaemia and/or increased luteinizing hormone levels [328, 329]. An important variant also shows insulin resistance, diabetes and acanthosis nigricans, but normal luteinizing hormone levels (HAIR-AN type: hyperandrogenism, insulin resistance, acanthosis nigricans; see Chapter 90).

Whether or not antiandrogenic therapy is indicated in patients with hyperandrogenism requires careful consideration, with the available pharmacological options differing greatly between countries [333]. Androgen suppression, e.g. via administration of a gonadotropin-releasing hormone antagonist, may also improve microvascular dilation in polycystic ovary syndrome, possibly via an endothelin receptor-dependent mechanism [51].

Hypoandrogenism

Whether it is caused by castration, tumours, antiandrogen therapy, genetically deficient androgen production (e.g. Klinefelter syndrome) or defective androgen receptor-mediated signalling, the dermatological picture of hypoandrogenism is mainly determined by the timing of androgen deficiency. If uncorrected hypoandrogenism occurs before puberty, it will manifest as eunuchoid habitus, with undeveloped male body habitus, minimal or absent male secondary hair and no acne during puberty; if it occurs after puberty, there is a slow progressive loss of male secondary sexual characteristics and androgenetic alopecia does not progress [5, 52, 150, 275].

As androgen deficiency is usually corrected in male patients, one dermatologically important point is that replacement therapy is not infrequently carried out too aggressively, so that features of hyperandrogenism such as seborrhoea, acne and rapidly progressive male pattern balding can develop.

One cause of infertility in hypoandrogenized males, typically with reduced beard and body hair growth, is a function-diminishing polymorphism of the androgen receptor [334].

Hyperoestrogenism

Besides oral contraceptives and excessive consumption of phyto-oestrogens (a possibility not to be ignored in the current climate of ‘health food’ frenzy), oestrogen-producing tumours are the most common and most serious causes of hyperoestrogenism. The classical skin signs are melasma, spider telangiectases and increased susceptibility to Candida vulvovaginitis. While girls can undergo precocious puberty, boys develop gynaecomastia. Much more rarely erythema nodosum, porphyria cutanea tarda, pemphigoid gestationis and systemic lupus erythematosus can be seen [211, 275, 276].

Hypo-oestrogenism

Hypo-oestrogenism as the natural consequence of the menopause or due to ovariectomy may result in female pattern balding. Sometimes this female variant of androgenetic alopecia may follow the male pattern or can also show a mixed phenotype, i.e. with features of both male and female pattern balding (see Chapter 89). The latter very common hair phenotype is associated with postmenopausal hypo-oestrogenism [54] and supports the hypothesis that, in contrast to androgenetic alopecia in males, a relative deficiency in the stimulation of scalp hair follicles by 17-β-oestradiol is a major factor in female pattern balding; this might well be as important an aetiological factor in female pattern balding as the hair growth inhibitory effects of DHT [53, 221, 335, 336, 337, 338, 339].

The significance of adequate oestrogen stimulation of female scalp hair follicles is further supported by observations in iatrogenic hypo-oestrogenism: aromatase inhibitors such as anastrazole and letrozole administered for management of breast cancer frequently cause a telogen effluvium [340] and sometimes may induce male pattern hair loss [338]. Reportedly, skin with higher aromatase expression (expected to be associated with elevated local oestrogen levels) also shows thicker elastic fibres than skin where aromatase expression is lower [40].

Given the recognized key role of 17-oestradiol as a promoter of wound healing [54, 159, 160, 161, 167], the observation that impaired wound healing can be associated with hypo-oestrogenism [158] and that 17-β-oestradiol may be capable of slowing or even partially reversing skin ageing [54, 55, 164, 167, 245], hypo-oestrogenism is a state that justifies consideration of hormone replacement, even by dermatologists. The potential benefits of this have to be carefully balanced against the long-debated, potentially serious adverse effects of oestrogen therapy [341].

In contrast to androgens, 17-β-oestradiol restricts sebocyte proliferation and sebum production [56]. Therefore, in addition to genital pruritus and burning mouth syndrome, seborrhoea is another dermatological manifestation of hypo-oestrogenism. If these occur concomitantly with osteoporosis, typical menopausal mood swings, paroxysmal hyperhidrosis and flushing, a clinical working diagnosis of hypo-oestrogenism is easily reached and can be confirmed by measuring serum hormone levels.

Phaeochromocytoma

Phaeochromocytoma is a potentially life-threating catecholamine-secreting endocrine neoplasm of the adrenal medulla. It can occur bilaterally or may be ectopic; in 10–15% of cases it is malignant. In comparison with its dominant clinical presentations (hypertension, paroxysmal hypertensive crises, headaches, tremor, palpitations, anxiety), its skin manifestations may not be obvious: hyperhidrosis, paroxysmal facial pallor, and sometimes aggravation of pre-existing Raynaud phenomenon [211, 276, 277, 342]. Dermatologists should be alert to the significance of such systemic signs in patients with these cutaneous features, particularly if they have neurofibromatosis type I, which conveys an increased risk of developing phaeochromocytoma [276, 342, 343].

Carcinoid

Carcinoid is a malignant neoplasm of specialized serotonin-producing epithelial cells of the small intestine and classically presents as paroxysmal flushing of the face and upper torso. These flushes, which may be triggered by psychoemotional stress, hot drinks or spicy food, often generate a sensation of rapidly ascending heat, which may be accompanied by wheezing and/or diarrhoea [277, 302, 342, 344]. It is assumed that not only the neurotransmitter serotonin but also neuropeptides such as substance P and bradykinin are involved in inducing these signs and symptoms.

However, as flushing reportedly occurs in less than 25% of patients with carcinoid [275], there are often no associated dermatological phenomena: the absence of flushing does not rule out the diagnosis. This must be kept in mind, as over the past 30 years the incidence of gastrointestinal carcinoid has been rising faster than that of any other cancer and the 5-year survival rate of patients with it is only 28.5% [345].

As an extreme rarity, human skin can also produce primary carcinoid tumours [346], while visceral neuroendocrine tumours may metastasize to the skin [347].

Glucagon and glucagonoma

Necrolytic migratory erythema (see Chapter 147) is the pathognomonic dermatological presentation of glucagonoma [280, 281, 348, 349], a neoplasm which typically arises from α-cells of the pancreas and secretes excessive amounts of glucagon. The associated cutaneous histopathological findings (intercellular epidermal oedema together with hydropic degeneration of epidermal keratinocytes) are shared with pellagra, zinc deficiency and acrodermatitis enteropathica, even though patients with glucagonoma syndrome have normal zinc serum levels [211].

Besides surgery, somatostatin analogues usually improve the skin eruption [350]. Little is known about how somatostatin or glucagon itself affects normal skin physiology or how excessive glucagon levels induce the pathognomonic clinical phenotype in the skin. Interestingly, some frog species generate substantial amounts of glucagon as well as many other neuropeptides in the skin [206]. Somatostatin may play a role in maintaining the physiological immune privilege of the hair follicle [110]. It has been postulated that it may not be glucagon itself but another as yet unidentified tumour-secreted peptide that is responsible for necrolytic migratory erythema [211]. Yet, it has been reported that necrolytic migratory erythema can be induced by glucagon therapy alone [351], suggesting that tumour-derived glucagon may well be capable of inducing this unique erythema phenotype.

Glucagon-like peptide-1 (GLP-1) is a gut hormone derived from proglucagon which has affinity for specific GLP-1 receptors present particularly in the pancreas and the brain. Amongst other actions, GLP-1 inhibits secretion of glucagon, enhances glucose-dependent insulin secretion and induces a feeling of satiety. Synthetic GLP-1 receptor agonists are now employed for treating type 2 diabetes. The observation that these drugs appear to be of some benefit for psoriasis in diabetic patients with insulin resistance before significant metabolic improvement can be detected [352, 353] provides interesting new clues to the possible role of GLP-1 in human skin, especially since psoriatic lesions can show increased expression of GLP-1 receptors [352, 354].

Polyendocrine disease

Multiple endocrine neoplasia (MEN-1 and MEN-2) and the autoimmune polyglandular syndromes (e.g. APS-1, due to a defective AIRE gene) can generate several of the skin lesions and symptoms described in this chapter [211, 355, 356, 357, 358, 359]. They are discussed in more detail in Chapter 147.

Diabetes

The dermatological dimensions of diabetes are described in Chapter 64 and summarized in Box 149.1.

Hyperthyroidism and hypothyroidism

Along with diabetes, hyper- and hypothyroidism are the endocrine disorders that most frequently cause skin abnormalities [210, 360]. If they occur together, pretibial myxoedema, weight loss, hyperhidrosis and an increased state of nervousness quickly indicate hyperthyroidism. By contrast, the combination of dry skin, telogen effluvium, brittle hair, pallor, weight gain, increased tiredness and decreased alertness is suggestive of hypothyroidism [211, 275]. Textbooks and reviews therefore traditionally deal with the dermatological aspects of hyper- and hypothyroidism in separate chapters. However, it may actually be more helpful to consider these conditions jointly, as we do here.

The dermatologist will identify the skin signs and symptoms (see Tables 149.3, 149.4 and 149.5) which point to the possibility of thyroid dysfunction; the dermatologist will then request routine thyroid function tests (TSH, free triiodothyronine (T3) and thyroxine (T4)) leaving the establishment of the finer details to the thyroid specialist.

Thyroid hormone receptors are abundantly expressed in human skin and hair follicles. Organ-cultured human skin and hair follicles respond directly to thyroid hormone stimulation in multiple ways with effects on keratinocyte proliferation and keratin expression, intracutaneous regulation of neurohormone production and hair follicle epithelial stem cell functions, stimulation of hair follicle pigmentation, keratinocyte energy metabolism and wound healing [28, 84, 102, 104, 195, 361, 362, 363]. Recently, it has been found that thyroxine strongly up-regulates expression of core clock genes in human skin [364].

Given increasing insight into the importance of peripheral clock activity for many different aspects of skin physiology ranging from cell cycle control and cell metabolism via pigmentation to hair growth control [283, 284, 285, 286, 287], this newly identified link between skin endocrinology and chronobiology only underscores the importance of hyper- or hypothyroidism from a dermatological perspective. Therefore it is not surprising that both excessive and insufficient serum levels of T3 or thyroxine T4 can generate a plethora of dermatological signs and symptoms. However, none of these mechanisms are yet fully understood.

A hyperthyroid state most frequently results from Graves disease, the most prevalent endocrinopathy after diabetes. It primarily affects women or occurs during the initial phase of autoimmune thyroiditis (Hashimoto type) [5, 150]. While the latter eventually results in hypothyroidism, its first dermatological presentation can be due to hyperthyroidism. Postpartum thyroiditis also manifests initially as hyperthyroidism, which is then followed by hypothyroidism – two more arguments for considering all thyroid disease associated skin and hair changes as one complex, at least for now. Excessive T4 medication or exposure to iodine (in particular in neonates or patients with impaired epidermal barrier function), hormone-secreting thyroid adenomas, multinodular goitre and, very rarely, a TSH-secreting pituitary adenoma are other important causes of hyperthyroidism. All these can induce the dermatological phenomena summarized earlier. An additional, thyroid automimmunity-related skin sign may be chronic urticaria, as it has been postulated to be inducible by IgE autoantibodies against thyroid peroxidase [365].

How stimulatory anti-TSH receptor autoantibodies are generated in Graves disease is still debated: one hypothesis holds that intradermal TSH receptors may provide the main inciting autoantigen [14, 366]. In any case, the stimulation of TSH receptors expressed by skin fibroblasts may induce the characteristic increased production and deposition of glycosaminoglycans responsible for pretibial and eyelid myxoedema as well as exophthalmos. Interestingly, excessive cutaneous glycosaminoglycans can induce both hypertrichosis and alopecia, possibly depending on whether the deposition of these extracellular matrix components act on relatively ‘quiescent’ telogen or maximally proliferating anagen hair follicles [215]. Likewise, both excessive and insufficient thyroid hormone serum levels can be associated with effluvium and alterations in hair shaft structure, elasticity or sheen [14, 102, 210]. The cutaneous microcirculation also appears to be affected in patients with Graves disease [367].

Hashimoto thyroiditis, the main cause of hypothyroidism apart from dietary iodine deficiency, can be associated with a few additional characteristic signs, namely lateral sparseness of the eyebrows (the sign of Hertoghe; key differential diagnoses include atopic eczema, trichotillomania and ulerythema ophryogenes) and a yellowish hue of the skin due to β-carotene accumulation (aurantiasis cutis) [368]. If these are accompanied by carpal tunnel syndrome, ptosis and brittle, slow-growing nails, this further indicates the presence of hypothyroidism [211, 276]. Apart from pretibial myxoedema (see Table 149.4) a more generalized myxoedema may also be present, often most prominently on the face, giving it a waxen appearance. With these characteristic signs, the diagnosis of hypothyroidism should not be challenging.

Iatrogenic hypothyroidism used to be largely restricted to patients following thyroid surgery, thyroid radiation or thyrostatic drug therapy. More recently, however, another iatrogenic form of hypothyroidism must also be considered: bexarotene, a second line therapy for cutaneous T-cell lymphoma, can induce significant hypothyroidism [369, 370], accompanied by the corresponding skin signs and symptoms (see Chapter 19).

Hyperparathyroidism

Most frequently, the typical hypercalcaemia that is associated with hyperparathyroidism causes only mild pruritus and/or skin dryness or no dermatological signs and symptoms at all. It is therefore easily overlooked by dermatologists.

Primary hyperparathyroidism most commonly results from an autonomous parathormone-secreting adenoma; secondary hyperparathyroidism may arise from vitamin D deficiency or from disturbances of the calcium-phosphate balance, especially in patients with chronic kidney disease. This predisposes to calciphylaxis (calcific uraemic arteriolopathy), though not all patients with this frequently fatal condition have either abnormal parathormone levels or high calcium-phosphate products [5, 120, 121]. Calciphylaxis is characterized by widespread vascular calcification and thrombotic occlusion of cutaneous blood vessels with resultant extensive tissue necrosis. Calciphylaxis is a frequent complication of end-stage renal insufficiency [371, 372]. Given the very high mortality of calciphylaxis, its development requires rapid diagnostic action so as to ascertain its origin and treatability [373].

Rarely, hyperparathyroidism may result in metastatic calcification in the dermis and subcutis: this does not specifically affect blood vessels and, if treated promptly, is reversible. It again is commoner in patients with renal failure and typically presents as firm papules, nodules or plaques, particularly around large joints or flexural sites. Unlike calciphylaxis it does not lead to tissue necrosis [374, 375]. These conditions are described in more detail in Chapter 61.

Primary hyperparathyroidism can also occur as a component of the MEN-1 syndrome (see earlier) [376].

Interestingly, both autosomal recessive congenital ichthyosis and epidermolytic ichthyosis in children were recently found to be associated with significantly increased parathyroid hormone (PTH) serum levels as well as with a higher radiological rickets score compared with common ichthyosis [377]. This raises the intriguing question whether some chronic dermatoses, namely those characterized by abnormalities of keratinization, are associated with increased intracutaneous production of PTH/PTH-related peptide (PTHrp). As associated endocrinological abnormalities (i.e. hyperparathyroidism and hypercalcaemia) could have major clinical consequences for distant organs such as the skeletal and gastrointestinal systems, this possibility deserves systematic investigation.

Hypoparathyroidism

Hypoparathyroidsm used to arise predominantly as a surgical complication following thyroidectomy. It is clinically most obvious from the characteristic resultant tetanic muscle cramps. However, hypoparathyroidism can also be seen in polyglandular autoimmune endocrinopathies or rare genetic defects, where its onset and progression can be far more insidious. In patients with pseudohypoparathyroidism the target tissues show greatly decreased sensitivity to stimulation with PTH.

In both cases, hair and nail growth disorders and skin dryness are the most frequently associated dermatological phenomena. ‘Dry, rough skin; coarse, brittle hair; and lustreless distally split nails’ [378] are frequently encountered skin signs in patients with idiopathic hypoparathyroidism. In a small Indian cohort of such patients, loss of axillary and pubic hair, coarsening of body hair, brittle and ridged nails and xerosis cutis were the most common skin signs observed [379]. Another study involved patients exposed to ionizing radiation following the nuclear reactor accident in Chernobyl who had had to undergo thyroidectomy and who developed permanent hypoparathyroidism as a surgical complication. Apart from frequently recurring paraesthesiae and joint pains, hair loss and progressively dry skin were the chief dermatological abnormalities [380]. A study in Brazilian women found hypoparathyroidism to be associated with a significantly reduced hair shaft content of calcium and phosphate, while these minerals were increased in patients with hyperparathyroidism or hyperthyroidism [381].

This hair phenotype corresponds well with the knowledge that both PTH/PTHrp agonists and antagonists profoundly modulate rodent hair follicle cycling in vivo as well as the hair follicle response to chemotherapy-induced damage [92, 93, 94, 382]. It has been proposed that this may in part be related to the effects of PTH/PTHrp-mediated signalling on angiogenesis in murine skin [95]. While the fine details of these murine studies do not concern us here, what really matters from a clinical dermatoendocrinological perspective are three concepts:

The interactions between PTH/PTHrp with its receptor constitute a functionally important, intracutaneous hormonal signalling system, the epithelial–mesenchymal interactions of which have a profound impact on rodent hair follicle growth and skin vasculature.
That patients with hypoparathyroidism also show hair growth abnormalities strongly suggests that this signalling system also operates in human skin, is clinically important and might be targeted therapeutically (e.g. in the management of telogen effluvium and chemotherapy-induced alopecia) [92, 383].
Any hormonal signalling system that modulates both the activities of an entire mini-organ (here, the hair follicle) and a process as complex as angiogenesis is almost guaranteed to have effects on multiple other aspects of skin physiology and pathology that await discovery. The finding that PTHrp regulates proliferation and differentiation in a human keratinocyte cell line in vitro [384] supports this notion.

Thus, hypoparathyroidism illustrates the principle that the skin phenotype seen in patients with a defined hormone deficiency, if interpreted in conjunction with the experimental literature, can highlight clinically relevant novel research frontiers in investigative dermatoendocrinology.

Future perspectives

For now, recognizing the skin signs and symptoms that indicate underlying endocrine abnormalities or conditions remains the core endocrinological challenge for the practising dermatologist. However, owing to the discovery that human skin is a major endocrine organ and metabolizes numerous hormones and neuromediators, the future scope and practice of dermatoendocrinology will change. With increasing insight into previously unsuspected (neuro-)endocrine dimensions of common skin diseases (see Table 149.2) and of standard therapeutic agents used to treat them, clinical dermatology cannot escape or ignore these important recent developments.

A few examples for future perspectives of dermatoendocrinology underscore this point:

All routinely administered hormone-based agents used for treating skin disease (glucocorticoids, retinoids, vitamin D derivatives) are likely to have a major impact on the physiological (neuro-)endocrine signalling milieu of a given patient's skin. A better understanding of these interactions should help to reduce adverse effects and enhance therapeutic efficacy.
In all endocrine disorders that affect the skin, there are likely to be alterations in the ways that relevant (neuro-)hormones and neuromediators are expressed or metabolized in the skin; and receptor-mediated signalling must be expected to be altered as a secondary consequence of systemic endocrine abnormalities.
It remains a matter of speculation to what extent secondary intracutaneous (neuro-)endocrine abnormalities contribute to discernible skin phenotypes associated with classical endocrine disease (see Tables 149.4 and 149.5 and Box 149.1). This deserves further systematic study.
Better understanding of how the extremely complex signalling interactions between the individual (neuro-)hormones and neuromediators present in the skin affect its functioning in health and disease should help to develop new strategies for improving skin function and retarding or even partially reversing skin ageing [54, 166].
Understanding in greater detail which specific (neuro-)endocrine abnormalities associated with inflamed, hyperproliferating and/or fibrosing skin diseases significantly contribute to the triggering, aggravation or pathogenesis of common dermatoses will identify novel therapeutic targets and corresponding management strategies.
The ‘epidemic’ of allergic diseases throughout Western civilization [385] calls for the development of innovative and effective anti-allergic strategies. Neuroendocrine strategies are particularly promising in this context. Examples include studies in human skin and hair follicle organ culture which have shown that: (i) endogenous and exogenous agonists of cannabinoid receptor 1 have a potent suppressive effect on mast cell degranulation and the maturation of functional mast cells from resident progenitor cells in human skin and airway mucosa [64, 139], while CRH promotes both [3, 127, 169]; (ii) SP stimulates human skin mast cell degranulation and neurogenic inflammation [13, 18, 97]; and (iii) the α-MSH-derived peptide, K(D)PT, inhibits human mast cell degranulation in situ [187]. Thus, in addition to neuropeptides, CB1 agonists and CRH and SP receptor (NK1) antagonists are promising candidates for anti-allergy therapeutics.
Antagonizing CRH [127, 237] or substance P [98, 143, 169, 386], or stimulating CGRP receptors [99], holds promise for the future management of inflammatory skin diseases such as atopic eczema by counteracting SP- and/or CRH-driven neurogenic skin inflammation. Although pharmacological antagonists of substance P or CRH have been shown to abrogate or significantly mitigate neurogenic skin inflammation and its sequelae (e.g. hair growth inhibition, itch, eczema) in animal experiments, this remains to be fully explored in human skin.

Finally, it has long been suspected that the modern penchant for sun exposure, despite its much publicised effects on skin ageing and skin cancer risk, has a habit-forming, if not addictive, neuroendocrine component. A recent animal study supports this notion: UV-seeking addictive behaviour in mice was influenced greatly by the UV-induced intracutaneous synthesis of the endogenous opioid, β-endorphin, leading to elevated plasma levels even after low-dose UV irradiation in vivo [387]. This is exciting because it shows that endorphins synthesized peripherally in mammalian skin can, in principle, reach both the circulation and the brain, and can cause complex behavioural changes. Interestingly, an increase of enkephalin plasma levels has already been reported in patients undergoing UVA phototherapy [388]. Therefore, it has become at least conceivable that future dermatological therapy will be capable of modulating this brain–skin axis in ways that may help dermatological patients whose skin disease has a major psychological component [389].

In addition to its paramount importance for the identification of underlying endocrine disorders, it is for these reasons that dermatoendocrinology lies at the very heart of modern dermatological practice.