From the Departments of Dermatology, University of Münster, Münster (Drs Ständer, Steinhoff, Metze, and Luger), Anesthesiology and Intensive Care Medicine, Faculty of Clinical Medicine, Mannheim (Dr Schmelz), and Social Medicine, Occupational and Environmental Dermatology, University of Heidelberg, Heidelberg (Dr Weisshaar), Germany. The authors have no relevant financial interest in this article.
Itching is defined as an unpleasant cutaneous sensation leading to the desire to scratch. It serves as a physiological self-protective mechanism as do other cutaneous sensations like pain, touch, vibration, cold, and heat to help defend the skin against harmful external agents. Pruritus can be evoked in the skin directly by mechanical and thermal stimuli or indirectly through chemical mediators. It may also be generated in the central nervous system independently of peripheral stimulation. Single-nerve-fiber recordings have shown that histamine-evoked itch is transmitted by selective slow-conducting subpopulations of unmyelinated C-polymodal neurons. Recent experimental studies using improved methods have demonstrated which of the suspected chemical itch mediators such as histamine, neuropeptides, prostaglandins, serotonin, acetylcholine, or bradykinin act pruritogenically on C-fibers. Moreover, investigations have revealed new receptor systems such as vanilloid, opioid, and cannabinoid receptors on cutaneous sensory nerve fibers that may modulate itch and thereby represent targets for antipruritic therapy. This review focuses on the peripheral generation of itch, including neurotransmitters, neuropeptides, and inflammatory mediators.
Itching is defined as an unpleasant cutaneous sensation leading to the desire to scratch. It serves as a physiological self-protective mechanism as do other cutaneous sensations like pain, touch, vibration, cold, and heat to help defend the skin against harmful external agents, including parasites and plants. Pruritus is a major symptom of skin diseases and various systemic diseases.1,2 Acute itch that occurs along with urticaria or as an adverse effect of many drugs, including chloroquine, can easily be treated.
However, the chronic, severe, and often generalized pruritus that may develop in cholestatic and renal diseases or in atopic dermatitis (AD) is more difficult to treat.1,3- 5 Chronic rubbing, scratching, or pinching leads to secondary skin lesions such as erosions, excoriations, crusts, hyperpigmentation or hypopigmentation, lichenification, prurigo, and scars and causes the release of inflammatory mediators that potentially induce or aggravate pruritic sensations resulting in an itch-scratch cycle. Where conventional therapies often fail to alleviate these more severe types of chronic pruritus, current progress in understanding the neurophysiology of pruritus has contributed to the development of new antipruritic therapeutic strategies.
Pruritus can be directly evoked in the skin by chemical mediators and by physical and thermal stimuli. In addition, it can arise in the diseased peripheral nerve or even in the central nervous system.5- 7 This review summarizes chemical mediators that elicit itch by excitation of cutaneous nerves (Table 1).
The skin is equipped with a dense network of highly specialized afferent sensory and efferent autonomic nerve branches that can be found in all skin layers.8 Autonomic (sympathetic) nerves represent only a minority of cutaneous fibers innervating eccrine, apocrine, and sebaceous glands, hair follicles, hair erector muscles, and blood vessels. On activation, they release acetylcholine (ACh) and norepinephrine, which activate the target cells via muscarinergic and catecholamine receptors. Afferent sensory nerve fibers transmit noxious information and cold via thinly myelinated Ad-fibers (conduction velocity, 10-30 m/s). Burning pain, heat, and itch are transmitted through slow-conducting unmyelinated C-fibers (conduction velocity, 0.5-2 m/s). Single-fiber recordings in humans have shown that the sensation of histamine-induced itch is transmitted by a very slow-conducting (0.5 m/s) subpopulation of C-neurons (pruriceptors) that also respond to other pruritic substances such as prostaglandin E2.9,10 However, it is doubtful that this particular type of pruriceptor is responsible for all "flavors" of the itch sensation. For example, mucanain, a protease of the fine spicules of cowhage (Mucuna pruriens), will elicit the sensation of pure itch, while endothelin induces itch with a burning component.11
Pruritic information is transmitted via the dorsal root ganglion to the spinal cord. At the spinal level, spinothalamic projection neurons are selectively excited by histamine and thus probably participate in the transmission of pruriceptive information in a dedicated neuronal pathway.12,13 In addition, positron emission tomography studies in humans show what is possibly activation of the primary sensory cortex after cutaneous histamine application.14,15 Motor-associated areas were activated, which correlates with the desire to scratch. In contrast to pain, itch does not provoke a spinal reflex. The scratching movement is governed by a center in the distal medulla close to the bottom of the fourth ventricle being under control of midbrain structures.16 The observation that rubbing, scratching, and pressing temporarily relieve the sensation of itch has yet not been explained, but recent studies have contributed to understanding this phenomenon (see, eg, Ikoma et al17 in this issue of the ARCHIVES).
Free nerve endings of cutaneous sensory C-nerve fibers serve as pruriceptors located in the papillary dermis and epidermis.18 A number of substances have been found to produce experimental pruritus, including amines, prostaglandins, and neuropeptides.9 Pruritogenic agents may bind specifically to itch receptors on the surface of chemosensitive nerve endings and thereby induce firing of the axons. The existence of specific binding sites on cutaneous nerves is underlined by the observation that experimental pruritus induced by application of histamine can be blocked by systemically administered H1 antagonists.19,20 In recent years, a large variety of receptors on sensory nerve fibers have been described. Moreover, many ligands of these receptors are constitutively expressed in skin cells such as keratinocytes, endothelial cells, and epithelial cells. This suggests that various sensory nerve fiber receptors interacting with their mediators from different skin cells are involved in cutaneous elicitation of itch.
Among the known itch mediators, histamine has been the most thoroughly studied pruritogen. About 70 years ago, Lewis et al21 reported that intradermal injections of histamine provoked redness, wheal, and flare (the so-called triple response of neurogenic inflammation) accompanied by pruritus. Accordingly, histamine is used for most experimental studies to induce an intense itching.9 However, recent studies have shown that histamine is not the major mediator in elicitation of pruritus in AD,4 for example, which partly explains the weak therapeutic efficacy of antihistamines.
Histamine is stored in mast cells and keratinocytes while H1 and H2 receptors are present on sensory nerve fibers.22 Consequently, histamine-induced itch may be evoked by release from mast cells or keratinocytes. Along with neuronal depolarization, neuropeptides such as substance P (SP) are released from nerve fibers. This is evidenced by the fact that in the presence of histamine H1 and H2 receptor antagonists, histamine has an inhibitory effect on the electrically evoked release of SP. Current studies strongly support the hypothesis that histamine also regulates SP release via prejunctional histamine H3 receptors located on peripheral endings of sensory nerves.23 Furthermore, many of the known itch mediators are potent histamine liberators, and thereby mast cell liberators, and may lead to the induction of pruritus.
Acetylcholine is a major neurotransmitter of the autonomic nervous system binding to muscarinic (types M1-M5) and nicotinic receptors. In experimental studies, intracutaneously injected ACh induces mainly pain over itching.9 Recent electrophysiologic and neurochemical studies with skin preparations from muscarinic receptor–mutant mice indicated that activation of cutaneous M2 receptors on nerve fibers can reduce the sensitivity of peripheral nociceptors.24,25 Also nicotine was able to weakly excite C-nociceptors and to induce a mild sensitization to heat stimulation.26
However, several muscarinic agonists such as carbachol and bethanechol produce itch in an animal model through activation of cutaneous M3 receptors.27 Furthermore, it was shown that neurotransmitters of the autonomic system may influence the sensory system. Nicotine induced a bimodal increase of calcitonin gene-related peptide (CGRP) release, while muscarine significantly decreased the basal CGRP release.26
In summary, these findings suggest that the autonomic system may be involved in cutaneous itch elicitation. Since ACh is synthesized and also released by other skin cells (eg, keratinocytes),28 it is feasible that non-neuronally released ACh plays a role in modulating peripheral nociception. Interestingly, patients with AD develop pruritus when injected with ACh in lesional skin, while healthy controls report a burning pain, which supports the idea that nicotinergic receptors and ACh are involved in the pathophysiology of AD pruritus.29- 32
Bradykinin belongs to the group of kinins that bind to bradykinin receptors (types B1 and B2). Injected into the skin, bradykinin induces pain9 rather than pruritus, mediated by the B2-receptor,33 although it is reported to induce mast cell degranulation for the release of histamine.34 Furthermore, bradykinin sensitizes nerve fibers for other chemical stimuli (reviewed in Lang et al35). In experimental studies, administration of bradykinin enhanced subsequent histamine responses.35,36 Stimulation with bradykinin also significantly augments the release of SP, CGRP, and prostaglandin E2,37 while serotonin and prostaglandin E2 have no effect on neuropeptide release but facilitate the bradykinin-evoked neuropeptide release.
It has been known for many years that serotonin can excite nociceptive C-fibers.38 When serotonin is injected intradermally or applied by iontophoresis in humans, it induces itch but is less potent than histamine.20,39 The widespread distribution of serotonin 3 receptors in the peripheral and central nervous system indicates that this receptor type may have a role in various disease states, which has sparked the investigation of serotonin 3 receptor antagonists in other applications such as the treatment of itching. However, in experimental and controlled clinical trials, serotonin 3 receptor antagonists did not show significant antipruritic effects in renal itch.20,40- 42
Endothelin is a peptide with many biological properties produced by endothelial cells. Endothelin 1, 2, and 3 induce neurogenic inflammation associated with burning pruritus. Itch may be induced by stimulation of polymodal nociceptor fibers and the release of nitric oxide independently from histamine.11,43
The vanilloid receptor subtype 1 (VR1)/transient receptor potential (TRP) V1 is a nonselective heat-activated cation channel that binds vanilloids and was originally found to be activated by the naturally occurring alkaloid capsaicin, which is not constitutively endogenously expressed in humans.44 The endogenous cannabinoid anandamide and an environment of increased temperature (> 42°C) and/or increased number of protons (pH <5.9), can also activate VR1.44,45
The recently confirmed expression of VR1 on cutaneous sensory nerve fibers in humans46 supports the notion that vanilloids and their receptors contribute to the induction and modulation of nociceptive cutaneous sensations such as pain and pruritus. On sensory nerve fibers, vanilloid receptor activation leads to depolarization and release of secretory granules containing neuropeptides such as SP or CGRP. Regardless of neuropeptide release, the action potentials transmit the information to the spinal cord resulting in initial burning pain or pruritus.47 Repeated application of capsaicin for several days induces desensitization of nerve fibers, inhibition of neuropeptide accumulation, and suppression of painful and pruritic sensations. Such application is therefore used as a treatment for pruritic or painful diseases.48
Other members of the vanilloid heat-activated receptor family such as vanilloid receptorlike protein (VRL) 1/TRP V2, TRP V3, and VRL-2/TRP V4 are identified, but their implication in the pathophysiology of pruritus is unknown.
A role of proteinases such as trypsin, chymotrypsin, and papain as pruritogenic agents has long been proposed49 but has not received much attention in recent years. However, new findings in this field have renewed the interest in proteinases. For example, stratum corneum chymotryptic enzyme was found to be increasingly expressed in itchy inflammatory skin diseases.50 Moreover, the pruritogenic effect and triple response of neurogenic inflammation induced by proteinases can be blocked by antihistamines,51 indicating an interaction of proteinases and mast cells in pruritus.
Recently, it has been demonstrated that mast cell tryptase mediates some of its cellular effects by activating a proteinase-activated receptor 2 (PAR-2).52 It was clearly shown that the concentration of tryptase is enhanced in the lesional skin of patients with AD compared with nonlesional skin from these patients and the skin of normal controls.53 In contrast, no significant difference was found for the concentration of histamine in these groups, which indicates a specific role for mast cell tryptase in patients with AD. Accordingly, the receptor for tryptase, PAR-2, was markedly enhanced on primary afferent nerve fibers in the lesional skin of patients with AD, suggesting that the pruritic effects of tryptase in AD may be at least in part mediated by PAR-2.53 Together, these results suggest that PAR-2 may be a receptor on sensory nerve fibers involved in induction of pruritus during inflammatory skin diseases.
Prostaglandins. Prostaglandins have an important synergistic function in itching and are believed to potentiate histamine-induced itch independently on the liberation of endogenous histamine.54,55 Pretreatment of human skin with prostaglandin E1 significantly lowers the threshold to itching evoked by both histamine and papain.56 Prostaglandin E2 causes a dose-dependent intense local vasodilation, weak pruritus, and pain in normal skin.57 Prostaglandins may thus potentiate pruritus in inflammatory skin disease, presumably by a nonspecific effect on nerve endings.
Interleukins. Three interleukins (ILs), IL-2, IL-4, and IL-6, are known to play a role in the elicitation of itch. Interleukin 2 was claimed to be a potent inducer of pruritus because high doses of recombinant IL-2 applied for cancer therapy frequently provoke redness and cutaneous itching.58 In fact, IL-2 elicited itch on experimental skin-prick testing59 by activation of a subpopulation of cutaneous C-fibers that are chemosensitive to histamine, bradykinin, and capsaicin.60 Bradykinin thereby seems to enhance the effect of IL-2–induced pruritus on sensory nerves.61
In a transgenic mouse line that overexpresses epidermal IL-4, mice spontaneously developed a pruritic inflammatory skin disease resembling human AD, which suggests that IL-4 is involved in the pathogenesis of inflammation and pruritus.62 Interleukin 6 and its receptor are expressed in nerve and Schwann cells,63 and IL-6–like immunoreactivity was increased in nerve fibers of patients with positive epicutaneous patch tests and prurigo nodularis,64 which suggests a role for this cytokine in pruritus.
Neurotrophins and Nerve Growth Factor. Neurotrophins and their receptors play an important role in cutaneous nerve development and reconstruction after injury. They are released by non-neuronal cells, and after binding to specific receptors on the peripheral nerve endings they are transported along the axon on the cell bodies in the dorsal root ganglia where they regulate expression of a variety of proteins involved in neuronal growth and sensitivity. A direct correlation between nerve growth factor (NGF) and pruritus has not been reported. However, a recent study showed that NGF is overexpressed in prurigo nodularis, and NGF and its receptors were hypothesized to contribute to the neurohyperplasia of the disease (Figure 1).65 Furthermore, patients with AD had significant increases in plasma levels of NGF and SP compared with controls.66 Prolonged treatment of rats with moderate doses of NGF is sufficient to stimulate neuropeptide synthesis in primary afferent neurons without causing long-lasting changes in the thermal nociceptive threshold.67
Increased and thickened dermal nerve fibers (arrows) in prurigo nodularis. The epidermis (E) shows irregular hyperplasia, and a dense inflammatory infiltrate (asterisk) can be seen in the dermis (immunohistochemical staining with anti-S100 protein antibody, original magnification ×200).
Recent observations indicate that neurotrophin-4 may be involved in inflammatory and itch responses of patients with AD. Neurotrophin-4 is a keratinocyte-derived agent that is highly expressed under inflammatory conditions. Its expression was found to be significantly increased in lesional skin of patients with AD and in prurigo lesions of AD skin.68 Interestingly, neurotrophin-4 production can be induced by interferon gamma, which itself is known to have a beneficial effect on pruritus.
Opioids. Opioid peptides such as β-endorphin, enkephalins, and endomorphins play a major role in the nociceptive pathway in the central nervous system by interacting with opioid receptors (µ-, δ-, κ-, and orphan receptors). Based on clinical observations that systemically administrated morphins may induce pruritus, it seems likely that endogenous and exogenous opioids modify the peripheral and central sensations of pruritus.69 Antagonizing endogenous opioids suppresses localized and systemic pruritus, which suggests a major role of central opioid receptors in the perception of pruritus.70- 72 Interestingly, opioids have been shown to act on capsaicin-sensitive nerve fibers to inhibit the release of inflammatory neuropeptides such as SP, neurokinin A, and CGRP,73,74 which suggests that opioids diminish sensitivity of peripheral nerve endings. Recent studies have revealed the presence of opiate receptors on cutaneous sensory nerve fibers,46,75 suggesting that peripheral opioid receptors may be involved in nociceptive mechanisms. This is supported by a previous observation that opioid receptor antagonists may significantly diminish experimentally evoked histamine-induced itch of the skin.39
Cannabinoids and Cannabinoid Receptors. Endogenous as well as synthetic cannabinoids are known for their analgesic potency. Recently, the cannabinoid receptor agonist HU210 was shown to attenuate histamine-induced excitation of nerve fibers and thus reduce itch and axon reflex erythema.76 These findings suggest that cannabinoid receptors are present on cutaneous nerve fibers. In fact, recently cannabinoid receptors CB1 and CB2 (Figure 2) were found to be present on sensory nerve fibers in human skin (S.S. and Roman Rukwied, PhD, unpublished observation, 2003).
Immunofluorescence staining for cannabinoid 2 (CB2) receptor (A; fluoroscein isothiocyanate conjugated stain) and the axonal marker neurofilament (B; Texas red) in a cutaneous nerve of normal skin. Axons (arrows) and the perineurium (arrowheads) of a large dermal nerve fiber stain for CB2 receptor. C, Yellow staining (arrow) in an overlay of A and B indicates colocalization. (Original magnification ×600 for all photographs.)
Cold Receptors. Two cold receptors have been cloned to date, the cold- and menthol-activated channel TRPM8 (CMR1) and ANKTM1 (ankyrinlike protein with transmembrane domains), a cold-activated channel with a lower activation temperature than TRPM8.77,78 The ANKTM1 receptor is found in a subset of nociceptive sensory neurons where it is coexpressed with VR1 but not TRPM8. It has been shown that lowering the skin temperature reduced the intensity of experimentally induced itch.79 A similar effect was achieved with menthol, although the skin temperature was not decreased.79 It was concluded that these findings suggest a central inhibitory effect of cold-sensitive A-delta fiber activation on itch. The role of both receptors in peripheral inhibition or modification of pruritus awaits further exploration.
Pruritus, whealing, and axon-reflex erythema due to histamine release appear in human skin after intradermal injection of vasoactive intestinal polypeptide, neurotensin, and secretin. Vasoactive intestinal polypeptide is more potent than the other neuropeptides and has approximately the same pruritogenic potency as SP.30 The 2 adjacent basic amino-acid residues and the amide substitution of the terminal C group of vasoactive intestinal polypeptide, in addition to its strong net basic charge, may explain its potency as a histamine releaser.80 In addition, somatostatin has been reported to stimulate histamine release of human skin mast cells.81 Other neuropeptides such as angiotensin, carnosine, melanocytic stimulating hormone, neuropeptide Y, and neurophysins were not reported to have a major pruritogenic potency.
The tachykinins SP, neurokinin A, and neurokinin B bind with different affinities to 3 neurokinin receptors (NKR 1-3) that are present on mast cells, fibroblasts, keratinocytes, Merkel cells, endothelial cells, and Langerhans cells. Intradermally injected SP releases histamine via binding to NKRs on mast cells and thereby acts as a pruritogen.82 However, an animal study demonstrated that mast cells are one but not the key factor in SP-induced itching.83 Intradermal SP increases the amount of cutaneous nitric oxide, possibly through the action on epidermal NKR-1 receptors, and enhances SP-induced itch-associated responses.84
Interestingly, there is no evidence for SP-induced activation of nociceptors under physiologic conditions.85 However, SP is an important sensory neuropeptide in the skin constitutively expressed in nerve fibers. Increased SP-immunoreactive nerve fibers have been observed in certain inflammatory skin diseases such as psoriasis, AD, and prurigo nodularis.86- 88 Furthermore, a connection between neuropeptides and stress has been shown in animal studies.89 Acute immobilization stress triggers skin mast cell degranulation via SP. However, neurokinin A and B seem to play a minor role in pruritus generation. In an experimental study, intradermally injected neurokinin A elicited itch in normal but not in inflamed skin. Neurokinin B did not elicit itch in either normal or inflamed skin.82
Intradermal injection of corticotropin-releasing hormone or its analogue, urocortin, induces skin mast cell degranulation and thereby histamine liberation.90 Moreover, acute immobilization stress triggers skin mast cell degranulation via corticotropin-releasing hormone in addition to other neuropeptides.89 However, corticotropin-releasing hormone is not known as a major pruritogen.
One of the most prominent sensory neuropeptides in skin nerves is CGRP, which is colocalized with SP. The corresponding CGRP receptors have been divided into 2 major classes: CGRP-1 and CGRP-2. Recently, CGRP-1 was found on arteriolar smooth muscle, venular and capillary endothelium, hair follicles, and sweat glands in the skin.91 Further studies to characterize CGRP-2 and expression of CGRP receptors on mast cells or nerve fibers are pending.
Interestingly, an inhibitory effect between CGRP and SP seems to exist. When CGRP was injected 10 minutes before an SP injection, a significant prolongation of itch latency was observed.92 However, increased CGRP contents were found in nerve fibers in pruritic diseases (eg, in lesional skin of AD and nummular eczema88 and in prurigo nodularis93).
Bombesin, a tetradecapeptide originally isolated from frog skin, induces dose-related long-lasting scratching when administered intracerebroventricularly to rats.94,95 Pruritus induced by bombesin is not antagonized by antihistamines or opioid antagonists (naloxon) but by opioid analgesics of the benzomorphan class, which suggests an association with the opioid system.94
The role of leukotrienes in the pathogenesis of pruritus is speculative, although there is some evidence about their relevance. Intradermally injected leukotriene B4 provokes scratching in mice.96 Additionally, a correlation between nocturnal itch and high urinary leukotriene B4 levels was found, suggesting that leukotrienes may contribute to severe pruritus at night in AD.97 Furthermore, the pruritic potency of SP may be partly explained by its ability to induce an arachidonate cascade producing prostaglandins and leukotrienes.98 Recent studies have revealed an antipruritic potency of the leukotriene receptor antagonists zafirlukast and zileuton.99- 101
Intraneuronal chemical deposits may influence nociceptive sensations or even directly induce them. A high incidence of severe pruritus has been observed after therapeutic administration of hydroxyethyl starch. A formation of intracytoplasmic storage vacuoles in cutaneous macrophages may be demonstrated in all infused patients.102,103 Patients with pruritus uniformly showed an additional deposition of hydroxyethyl starch in small peripheral nerves (Figure 3). The accumulation of hydroxyethyl starch in various cell elements of peripheral nerves may directly influence neuronal functions and result in generalized itch. Consequently, neural devacuolization paralleled clinical improvement of the symptoms.102,103 Furthermore, persisting and burning itch is related to intramuscular administration of polyvinylpyrrolidone, a high-molecular-weight substance used as a slow-release agent. Likewise, this adverse effect has been assumed to be induced by a storage of polyvinylpyrrolidone in perineural and Schwann cells.104,105
Intraneuronal deposits of hydroxyethyl starch (HES). Monoclonal HES-specific antibody subjected to a postembedding immunogold technique and counterstained with uranyl acetate–lead citrate. Vacuoles (V) in the cytoplasm of a Schwann cell (C, core; A, axon) are immunoreactive (arrow) to anti-HES antibody (original magnification ×11 500).
Although renewed interest and research in pruritus has provided a deeper insight into its mechanisms in recent years, the basis of itch is not yet fully understood. Many chemical mediator candidates have been explored for their pruritic potency. Furthermore, various receptors on cutaneous sensory nerve fibers have been identified, which suggests direct elicitation of itch on pruriceptors in addition to histamine receptors, which were for years believed to be the only direct itch inducer. However, our knowledge of itch neurophysiology is still fragmentary; we are as yet unable to explain the different clinical manifestations of itch. Nevertheless, the increase in knowledge in itch mechanisms should encourage us to develop new therapeutic regimens to treat the severe and distressing symptom of pruritus.
Corresponding author and reprints: Sonja Ständer, MD, Department of Dermatology, University of Münster, Von-Esmarchstrasse 58, D-48149 Münster, Germany (e-mail: firstname.lastname@example.org).
Accepted for publication July 28, 2003.
Thank you for submitting a comment on this article. It will be reviewed by JAMA Dermatology editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Register and get free email Table of Contents alerts, saved searches, PowerPoint downloads, CME quizzes, and more
Subscribe for full-text access to content from 1998 forward and a host of useful features
Activate your current subscription (AMA members and current subscribers)
Purchase Online Access to this article for 24 hours
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 91
Customize your page view by dragging & repositioning the boxes below.
and access these and other features:
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.