Pain 96 (2002) 9–12

www.elsevier.com/locate/pain
Research papers

Intractable postherpetic itch and cutaneous

deafferentation after facial shingles q
Anne Louise Oaklander a,*, Steven P. Cohen b,1, Shubha V.Y. Raju a,c,2
a
Neuropathic Pain Study Group, Departments of Anesthesiology, Neurology, Neuropathology, Massachusetts General Hospital, Harvard Medical School,
Boston, MA, USA
b
Department of Anesthesiology and Critical Care, Massachusetts General Hospital, Boston, MA, USA
c
M.S. Ramaiah Medical College, Bangalore, India
Received 25 May 2001; accepted 6 August 2001

Abstract
Some patients develop chronic itch from neurological injuries, and shingles may be a common cause. Neuropathic itch can lead to self-
injury from scratching desensate skin. A 39-year-old woman experienced severe postherpetic itch, but no postherpetic neuralgia, after
ophthalmic zoster. Within 1 year, she had painlessly scratched through her frontal skull into her brain. Sensory testing and skin biopsies were
performed on itchy and normal scalp to generate preliminary hypotheses about mechanisms of neuropathic itch. Quantitation of epidermal
neurites in PGP9.5-immunolabeled skin biopsies demonstrated loss of 96% of epidermal innervation in the itchy area. Quantitative sensory
testing indicated severe damage to most sensory modalities except itch. These data indicate that in this patient, severe postherpetic itch was
associated with loss of peripheral sensory neurons. Possible mechanisms include electrical hyperactivity of hypo-afferented central itch-
specific neurons, selective preservation of peripheral itch-fibers from neighboring unaffected dermatomes, and/or imbalance between
excitation and inhibition of second-order sensory neurons. q 2002 International Association for the Study of Pain. Published by Elsevier
Science B.V. All rights reserved.
Keywords: Nervous system/physiopathology; Human; Pruritus/physiopathology; Herpes zoster; Neuralgia; Postherpetic neuralgia

1. Introduction
Itch was defined in the 17th century as an unpleasant
sensation producing a desire to scratch, a definition not
improved on since (Haffenreffer, 1660). Cutaneous inflam-
mation and several systemic illnesses are the most common
causes of itch, but it can also be caused by neurological
abnormalities (reviewed in Bernhard, 1994). “Notalgia
paresthetica” has been reported from nerve-root compres-
sion (Raison-Peyron et al., 1999). Central itch has been
reported in multiple sclerosis (Yamamoto et al., 1981),
brain tumor (Andreev and Petkov, 1975), stroke (Shapiro
q
This manuscript is dedicated to the memory of Professor Patrick D.
Wall whose own work on itch provided a foundation for this study, and
whose interest in this manuscript was inspirational.
* Corresponding author. Massachusetts General Hospital, 55 Fruit Street/
Clinics 3, Boston, MA 02114, USA. Tel.: 11-617-726-4656; fax: 11-617-
726-5845.
E-mail address: aoaklander@partners.org (A.L. Oaklander).
1
Present address: Anesthesia Service, Walter Reed Army Medical
Center, 6825 16th Street NW, Washington, DC 20307-5001, USA.
2
Present address: 891 Massachusetts Ave., Apt #2, Cambridge, MA
02139, USA.
0304-3959/02/$20.00 q 2002 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved.
PII: S 0304-395 9(01)00400-6
and Braun, 1987), and brain abscess (Sullivan and Drake,
1984). Although only briefly mentioned in the literature
(Liddell, 1974; Tada et al., 1991; Darsow et al., 1996), the
most common cause of focal neuropathic itch today is prob-
ably prior shingles. Shingles (herpes zoster) is an often
painful rash caused by reactivation of latent varicella-zoster
virus in sensory ganglia (Straus, 1993). It affects 10–15% of
Americans, usually the elderly or immunosuppressed. It
destroys peripheral sensory neurons, and leaves some
patients with chronic postherpetic neuralgia (PHN) pain
(Straus, 1993). Postherpetic itch (PHI) can also occur in
the region of prior shingles, especially after facial shingles
(Oaklander and Bowsher, 2001). Although it is stated that
itch cannot occur simultaneously with pain, PHI can occur
with or without PHN pain at the same site. We have used
Quantitative Sensory Testing and punch skin biopsies from
a patient with severe PHI to generate hypotheses about
potential mechanisms of neuropathic itch.
2. Patient
A 39-year-old woman developed typical shingles on her
10 A.L. Oaklander et al. / Pain 96 (2002) 9–12
Fig. 1. A CT scan of the patient’s head shows a 4 £ 6 cm right frontal skull
defect, hypo-density in the underlying brain, and extra-axial fluid. No other
lesions were present. The defect was self-induced by scratching an area
affected by postherpetic itch after shingles. The scratching was painless.
right forehead. Her history revealed asymptomatic hepatitis
C and HIV infections, previous drug abuse controlled by
methadone, personality disorder and depression, but no
history of obsessive-compulsive disorder, and no other
neurological complaints or diagnoses. Acyclovir was
given, the shingles lesions resolved, and itch without pain
ensued. She began to scratch her right forehead, using
fingernails only. Her scratching was interpreted as trichotil-
lomania, but several serotonin-specific re-uptake inhibitors,
as well as other medications (e.g., gabapentin, diphenhydra-
mine) did not relieve the itch.
She came to the Emergency Department 10 months post-
shingles with fever, weight loss, headache, nausea, and foul
scalp drainage. Mental status and motor exams were normal.
Her CD4 1 cell count was 218/mL 3. Cerebral cortex was
herniating through a skull defect (Fig. 1). A split-thickness
skin graft was placed, destroyed by scratching (Fig. 2),
replaced, and re-lacerated. Cultures from her wound, dura,
epi- and sub-dural fluid, and brain grew mixed flora includ-
ing Staph. aureus and Strep. species. Pathological examina-
tion of skull fragments showed osteomyelitis. Ultimately,
scalp was translocated over the skull defect to close it.
Phenytoin and nortriptyline were ineffective against the
itch. Intra-nasal lidocaine gel, Lidoderm patches applied
to the right forehead, or lidocaine block above the right
eye gave temporary relief. Section of the supraorbital
nerve helped for a few weeks only. A helmet that impeded
scratching during sleep or while inattentive was very useful.
Left-hand weakness, the onset of seizures, and emotional
lability were attributed to frontal brain injury. A shingles
recurrence 15 months after the first, in the same dermatome,
caused new spontaneous, touch-evoked, and lancinating
pain diagnosed as new postherpetic neuralgia. The pain
did not stop the itching and scratching. Eighteen months
after the first shingles, she rated her pain severity at 8 on
a 0–10 scale of ascending severity, and itch severity at 10/
10. There was improvement; at 21 months the PHN severity
was 2/10, and PHI severity was 7/10. A Lidoderm patch was
helpful with both pain and itch.
2.1. Sensory testing
Stimuli applied to the patient’s arm produced normal
sensation. The itchy area on the right forehead had loss of
sensation to pin and inability to perceive mechanical stimuli
including 283 mN of force applied with a von Frey mono-
filament. When thermal testing was performed by a techni-
cian, warmth on the left forehead was detected at 35.28C,
but was not perceived on the right forehead. The threshold
for detecting cooling was 27.88C on the left forehead and
12.48C on the right. The threshold for perceiving heat pain
was 44.08C on the left, and 36.38C on the right (although this
is less than the reported threshold for warmth detection).
Cold pain thresholds were not recorded. Intradermal injec-
tion of histamine on her arm produced normal flare and itch.
Histamine injected into the itchy forehead reportedly
produced new itch sensations, but the presence of a flare
could not be confirmed because of hair stubble.
2.2. Measuring cutaneous innervation
After consent to an IRB-approved protocol was obtained,
a 3-mm diameter punch was removed from anesthetized
itchy skin 4 cm anterolateral to the wound margin. Because
unilateral shingles can cause bilateral nerve damage
(Oaklander et al., 1998; Watson et al., 2000) the control
biopsy was taken from left parieto-occipital scalp in the
C2 dermatome. The biopsies were removed prior to section
of the supraorbital nerve. The biopsies were fixed,
sectioned, immunolabeled against PGP9.5, a pan-axonal
marker, and epidermal neurite density measured using stan-
dard methods (Oaklander et al., 1998).
The biopsy from normal scalp had 588 intra-epidermal
neurites/mm 2 skin surface area (Fig. 3A). The biopsy from
PHI-affected skin was almost devoid of somatic innervation
(Fig. 3B). The density of 22 intra-epidermal neurites/mm 2
was 4% of the control value. Autonomic innervation of
glands, hair follicles, and blood vessels appeared normal
(not shown).
3. Discussion
Pain and itch are closely related. Thin, unmyelinated,
 A.L. Oaklander et al. / Pain 96 (2002) 9–12 11

shown dramatic loss of innervation (Oaklander et al.,

1998). One study of patients with notalgia paresthetica
(that grouped patients with itch and other neuropathic symp-
toms) reported increased dermal innervation (Springall et
al., 1991). Our results do not explain why some patients
experience PHI in addition to, or instead of, PHN.
We offer three hypothetical mechanisms to explain how
this patient’s severe itch co-existed with profound damage
to her cutaneous sensation and innervation. Shingles causes
the entire peripheral sensory neuron to degenerate (Head
and Campbell, 1900) so PHI patients are unlikely to have
normal innervation deeper in the skin. Yet any hypothesis
must account for this patient’s clear-cut temporary relief
from itch after local anesthetics.
Analogous to phantom-limb pain, the itch could be gener-
ated by central itch neurons (Andrew and Craig, 2001).
firing excessively when deprived of afferent input (phantom
itch). This is possible even though modulating peripheral
sensory neurons by scratching or local anesthetics was help-
ful. Patients with neuropathic itch from entirely central
causes (e.g., brain tumor) are driven to scratch, feel relief

Fig. 2. Quantitative sensory testing and an intradermal injection of hista-

mine were followed by punch skin biopsies. When the patient’s forehead
was bared for procedures, she was unable to resist scratching. Multiple
medications were ineffective at relieving her itch. Her hands were often
bandaged or tied to the bed in an effort to prevent further scratching.

sensory axons transmit both sensations, and many stimuli

can produce either sensation. Itch-specific C-fibers with
large innervation territories mediate chemical and inflam-
matory itch (Schmelz et al., 1997). Histamine-specific
second-order itch neurons have recently been identified in
the dorsal horn (Andrew and Craig, 2001).
Less is know about mechanical itch (prickle) from punc-
tate stimuli such as wool fibers (Garnsworthy et al., 1988). It
may be triggered by minute stimuli that excite single dorsal-
horn neurons but are not large or strong enough to also
excite inhibitory interneurons (Greaves and Wall, 1996).
Normally, these inhibitory interneurons are also excited
by primary afferents, and synapse onto second-order
sensory neurons to temporally and spatially limit their elec-
trical responses. Scratching may relieve itch by expanding
the area of mechanical stimulation enough to recruit inhibi-
tory interneurons in addition to directly exciting second-
order projection neurons (Brull et al., 1999). Fig. 3. PGP 9.5 immunolabeling of nerve fibers in punch skin biopsies. The
epidermis is near the top and the superficial dermis is below. Representative
In our patient, itch was associated with loss of sensory
labeled vertical sections from (A) a normal area of the patient’s scalp
neurons. PHI is thus likely to have different mechanisms showing axons extending up into the epidermis, and (B) previously shin-
than inflammatory itch. Skin biopsies from patients with gles-affected, itchy skin in which no epidermal or dermal neurites are
neuropathic pain syndromes, including PHN, have also visible.
12 A.L. Oaklander et al. / Pain 96 (2002) 9–12
from it, and can scratch unconsciously during sleep (as did
our patient) (Andreev and Petkov, 1975). The return of itch
after transection of her supraorbital nerve may have
involved central mechanisms.
An alternative is suggested by the mechanism for
mechanical itch mentioned above. Preservation of a few
scattered sensory neurons after shingles may have allowed
activation of occasional second-order neurons without
concomitant recruitment of inhibitory interneurons. Thus,
loss of most sensory neurons in the skin may create an
electrophysiological situation similar to when fully inner-
vated skin is stimulated with a minute punctate stimulus.
This hypothesis readily explains why scratching and local
anesthetics gave relief.
It is also possible that the few remaining cutaneous
neurons may have been itch-fibers from adjacent unaffected
dermatomes, since these have unusually large innervation
territories (up to 8.5 cm diameter) that extend beyond tradi-
tional dermatomes (Schmelz et al., 1997). These might be
sensitized by an abnormal environment that included neuro-
transmitters from nearby autonomic effector neurons or
products of inflammation. This hypothesis is consistent
with the patient’s report that histamine produced additional
itch in the PHI-affected area.
Although self-injury is usually attributed to psychiatric
causes, some patients have underlying sensori-neural disor-
ders. This patient was concerned about her scratching and
sought medical treatment for it, but without the deterrent of
pain sensation, had difficulty overcoming her drive to
scratch. Other factors contributed to her self-injury, includ-
ing osteomyelitis that softened her skull. Her history of
personality disorder and drug abuse suggest impulsive
tendencies. Additionally, frontal-lobe brain injuries have
been associated with loss of impulse control (Hahm et al.,
2001). Regardless of the reasons for her scratching, her itch
was caused by shingles. Hopefully, better recognition and
understanding of postherpetic itch will give better treatment
options.
Acknowledgements
We gratefully acknowledge the technical assistance of Yu
Yang and the clinical contributions of Robert Cluff, Carol
Garner, J. Peter Rubin, and Ronald P. Silverman. Martin
Schmelz, Bob LaMotte, and Oliver Sacks provided helpful
discussion. This study was supported by the Paul Beeson
Physician Faculty Scholarship in Aging Research, the
Beatrice and Roy Backus Foundation, and the Ruth and
Maurice Freeman Fund for Pain Research. Preliminary
versions were presented in abstract form. We thank the
subject of this study.
References
Andreev VC, Petkov I. Skin manifestations associated with tumours of the
brain. Br J Dermatol 1975;92:675–678.
Andrew D, Craig AD. Spinothalamic lamina I neurons selectively sensitive
to histamine: a central neural pathway for itch. Nat Neurosci
2001;4:72–77.
Bernhard JD. Neurogenic pruritus and strange skin sensations. In: Bernhard
JD, editor. Itch: mechanisms and management of pruritis, New York:
McGraw-Hill, 1994. pp. 185–201.
Brull SJ, et al. Attenuation of experimental pruritus and mechanically
evoked dysesthesiae in an area of cutaneous allodynia. Somatosens
Mot Res 1999;16:299–303.
Darsow U, et al. Pruritus circumscriptus sine materia: a sequel of postzos-
teric neuralgia. Acta Derm Venereol (Stockholm) 1996;76:45–47.
Garnsworthy RK, et al. Identification of the physical stimulus and the
neural basis of fabric- evoked prickle. J Neurophysiol 1988;59:1083–
1097.
Greaves MW, Wall PD. Pathophysiology of itching. Lancet 1996;348:938–
940.
Haffenreffer S. Nosodochium in quo cutis…affectus omnes…et cognos-
cendi et curandi fidelissime traduntur. Ulm: Balthasar Kühnen, 1660.
Hahm DS, et al. A compulsive collecting behavior following an A-com
aneurysmal rupture. Neurology 2001;56:398–400.
Head H, Campbell AW. The pathology of herpes zoster and its bearing on
sensory localisation. Brain 1900;23:353–529.
Liddell K. Post-herpetic pruritus. Br Med J 1974;4:165.
Oaklander AL, et al. Unilateral postherpetic neuralgia is associated with
bilateral sensory neuron damage. Ann Neurol 1998;44:789–795.
Oaklander AL, Bowsher D. Post-herpetic itch (PHI), a common neuro-
pathic complication after shingles (herpes zoster). J Pain 2001;2S1:18.
Raison-Peyron N, et al. Notalgia paresthetica: clinical, physiopathological
and therapeutic aspects. A study of 12 cases. J Eur Acad Dermatol
Venereol 1999;12:215–221.
Schmelz M, et al. Specific C-receptors for itch in human skin. J Neurosci
1997;17:8003–8008.
Shapiro PE, Braun CW. Unilateral pruritus after a stroke. Arch Dermatol
1987;123:1527–1530.
Springall DR, et al. Symptoms of notalgia paresthetica may be explained by
increased dermal innervation. J Invest Dermatol 1991;97:555–561.
Straus SE. Shingles: Sorrows, salves, and solutions. J Am Med Assoc
1993;269:1836–1839.
Sullivan MJ, Drake MEJ. Unilateral pruritus and Nocardia brain abscess.
Neurology 1984;34:828–829.
Tada J, et al. Trigeminal trophic syndrome-a report of three patients. J
Dermatol 1991;18(613):615.
Watson CPN, et al. Trigeminal postherpetic neuralgia postmortem: clini-
cally unilateral, pathologically bilateral. In: Devor M, Rowbotham MC,
Weisenfeld-Hallin Z, et al., editors. Proceedings of the 9th World
Congress on Pain, Vol. 16. Seattle, WA: IASP Press Seattle, 2000.
pp. 733–739.
Yamamoto M, et al. Paroxysmal itching in multiple sclerosis: a report of
three cases. J Neurol Neurosurg Psychiatry 1981;44:19–22.