Please upgrade to the latest version of Flash Player.


Relief of post-herpetic neuralgia by surgical removal of painful skin

Karinn L. Petersen, Frank L. Riceb, Fred Suess, Marlene Berroa, Michael C. Rowbotham

Department of Neurology, UCSF Pain Clinical Research Center, 1701 DivisaderoStreet, Suite 480, University of California, San Francisco, CA 941} 5, USA Department of Anesthesia, UCSF Pain Clinical Research Center, University of California, San Francisco, CA, USA 'Center for Veuropharmacology and Neuroscience, Albany Medical College, Albany, NY, USA California Pacific Medical Center, San Francisco, CA, USA

We present a case of longstanding PHN treated by skin excision of the area of greatest pain (11.3 X 26.0 cm2). The operation reduced pain, nnated tactile allodynia, and facilitated greatly reduced medication use over a 1-year follow-up period. Fourteen punch biopsies and 10 jps of skin (each 10 mm long) from the excised painful PHN skin were qualitatively assessed by double-label immunofluorescence using bodies against protein-gene-product 9.5 (PGP9.5), 200 kDa neurofilament protein (NF), calcitonin gene-related peptide (CGRP) and lilloid receptor-! (VR-1). Comared with a punch biopsy from mirror image skin, the pattern of cutaneous innervation in PHN skin was isistently and substantially different. The results may explain the anatomical basis of the capsaicin-response test and have implications for understanding of clinical mechanisms underlying PHN pain. © 2002 International Association for the Study of Pain. Published by vier Science B.V. All rights reserved.


Surgical removal of painful skin as a treatment for post-berpetic neuralgia (PHN) was first mentioned in 1900 (Cowers, 1900) and was the subject of several case series isports published between 1949 and 1976 (Browder and ifeVeer, 1949; Abbott and Martin ,1951; Weidmann, 1976, review, see Loeser, 2001). Pre-operative sensory testing K microscopic evaluation of excised skin was not described. Follow-up was variable and in several cases pain recurred srithin a year. In one case permanent 'cure' followed removal ?f additional skin just outside the original excision. Without a jnethod for predicting benefit and a success-rgte under 33%, Han resection procedures were abandoned. We report a case longstanding PHN treated by skin excision, including (iouble-labeling immunofluorescence assessment of cuta-aeous innervation in excised and contralateral skin.

Case History

A 65-year-old male with PHN since 1992 underwent surgical skin resection on the back below the shoulder costal; nerve blocks spanning three levels and various oral medications without lasting pain relief. Very brief trials of topical capsaicin (0.025 and 0.075%) in 1993 produced intense burning pain. On initial evaluation at UCSF in 1993, daily pain was reported as 6 on a 0-10 numerical pain scale. Nortriptyline and methadone were the most successful of the many medi¬cations tried subsequently, but neither reduced pain by more than 50%. In 1995, he underwent extensive sensory testing and skin biopsies as a research subject (Rowbotham et al., 1996). At that time, daily pain was rated as 38 on a 0-100 mm pain visual analog scale (VAS) and allodynia to brush was 8.5 on a 0-10 numerical allodynia scale. In the area of great¬est pain, there was a modest loss of thermal sensory function (heat pain deficit of 0.9°C, overall thermal sensory deficit of 1.0°C). For comparison, the study group means were a heat pain deficit of 3.0°C and an overall thermal sensory deficit of 3.1°C. Skin biopsy evaluation using immunofluorescence with PGP 9.5 showed his cutaneous innervation to be rela¬tively intact. In the patient's PHN skin there were 19 fibers/ mm and in contralateral skin there were 27 fibers/mm. Study group means were seven fibers in PHN skin and 23 fibers in mirror-image skin. He participated in a clinical trial of lidocaine patches and responded well to the use of two patches per day in a post-study compassionate use protocol. The later addition of gabapentin further reduced his pain. For more than a year pre-surgery, a stable daily regimen of gabapentin 1600 mg/day, methadone 20 mg/day. nortriptyline 40 ing/ day and two lidocaine patches/day kept his pain within toler­able limits, although he was never pain free.

Approval from the UCSF Committee on Human Research was received for the pre-operative sensory testing and the skin biopsy analyses. On the day prior to surgery, he rated his average daily PHN pain intensify' for the prior week as 30 on a 0-100 mm VAS. On examination, the area of allodynia was 251 cm2 (Fig. 1A). Allodynia produced by three foam brush strokes within the most painful area was rated as 76 by the patient on a 0-100 allodynia VAS. As illustrated in Fig. !A, the 'capsaicin response test" (Petersen et al.. 2000) was performed using capsaicin cream 0.075% (Clay-Park Labs Inc., NY, USA) applied to a 9-crrr location within the most painful area. The intended application time was 60 min, but the capsaicin was removed after 35 min because of intolerable worsen­ing of symptoms. By 35 min. pain in the capsaicin applica­tion site had increased from 0/100 to 40/100, the area of allodynia enlarged from 251 to 382 cm2, allodynia severity in a neighboring site increased to 90/100 and overall PHN pain intensity increased from 17/100 to 81/100. After capsaicin removal, the patient rated the net effect of capsaicin stimulation on his PHN pain as +43 mm on a modified 100-mni VAS scale marked in the middle (0 mm) as 'no effect* and anchored at the left end as —50 mm = ' relieved pain", and at the right end as -+-50 mm = ' worsened pain'. According to the criteria used in our prior study (Petersen et al.. 2000), the patient was a 'capsaicin responder".

Surgery was performed on 10/27/00. Under general anesthesia, the 11.3 X 26.0 cm2 piece of skin corresponding to the area of pain and allodynia marked on the pre-surgical examination was excised through skin and fatty tissue to muscle fascia. Superior and inferior flaps were elevated off the muscle fascia and advanced and rotated into position under mild tension and the layers were closed. Immediately-after removal of the skin, 14 3-mm punch biopsies and ten strips (10x3 mm") were collected for analysis from the excised skin (Fig. IB). For comparison, a 3-mm punch biopsy was taken from the unaffected contralateral (mirror-image) skin. The patient would not consent to biopsies from a distant normal site. Telephone contact 2 days post-operation revealed immediate reduction in pain, loss of painful skin sensitivity and cessation of lidocaine patch use. He was eval­uated in the PCRC at frequent intervals over a 53-week period post-surgery. At the visits, pain ratings were obtained, a short clinical exam was performed and medication changes discussed. The results are summarized in Table 1. Post-operatively, the patient immediately ceased lidocaine patch use. At the follow-ups from 12 through 20 weeks, he was nearly symptom free, without deep pain, shooting pain, or allodynia, while eliminating nortriptyline and greatly reducing neurontin and methadone use. At the 32-week and 1 year follow-up visits, he described some deep and shooting pain located on the anterior chest, but remained free of allodynia and maintained reduced medication use. Overall he considered himself 50% improved by the surgery.

Skin biopsy analysis methods

;Skin biopsies were fixed overnight in 4% paraformalde-i in 0.1 M phosphate buffered saline at pH 7.4 and main-ned at 4CC. Serial sections 14-(jun thick and perpendicular > the skin surface were cut by cryostat and thawed onto ne alum gelatin coated slides. The strips of skin were sented so that the sections were parallel to their long axis. Dnsecutive sections were mounted alternating across a of ten slides. The first and sixth slide of sections ( each biopsy were incubated for immunofluorescence anti-protein gene product 9.5 (PGP), a cytoplasmic me presumed to be present in all axons and nerve tings (UltraClone Ltd: rabbit polyclonal, 1:1000 dilution), intervening sections were processed for double-label nunofiuorescence in the following combinations: (1) iti-calcitonin gene-related peptide (GGRP) (Chemicon c: rabbit polyclonal, 1:1006) and anti-PGP; (2) anti-) kDa neurofilament protein (NF) (Chemicon Inc.: rabbit Olyclonal, 1:1000) and anti-PGP; (3) anti-CGRP (Infiniti d:sheeppojycjonal, 1:800)and anti-NF; (4)anti-vanilloid ceptor 1 (VR-1) (Quo et al., 1999; guinea pig polyclonal, •1500) and anti-PGP; and (5) anti-VR-1 and anti-CGRP. condary antibodies for rabbit, sheep and guinea pig were all raised in donkey and were conjugated to either Cy3 or Cy2 (Jackson laimunoLaboratories Inc.) for red and green fluorescence, respectively. Details of double-labeling procedures are described in Pare el til. (2001). High resolution digital images were captured on an Olympus Provis epifluorescence microscope equipped with a SONY DKC-ST5 digital camera interfaced with Northern Eclipse digital imaging software (Empix Imaging Inc., see Pare et a!.. 2001). Lower resolution (800 pixel/inch2) photomontages were created and the loca¬tions of labeled structures were mapped with the aid of X. Y and Z axis stepping motors interfaced to Neurolucida :m N euro Explorer software (MicroBrightiield Inc.).

Skin biopsy results

Photomicrographs of skin biopsies are shown in and 3. A schematic summary of the differences between lucida and contralateral and PHN skin is presented in Fig. 4. Although sly one biopsy of contralateral skin was available, the osity, morphology and immunochemistry of cutaneous nervation of this specimen was substantially different [;lroin all 24 of the samples taken from the excised PHN skin. Dermal papillae were large and highly irregular in PHN kin but were uniformly small and narrow in the contralat-|"*ral skin biopsy (Fig. 2A, C-E, G, H). The epidermis in skin was reduced and more irregular, especially ove the dermal papillae, than in contralateral skin. The jSpiderrnis of PHN skin, but not contralateral skin, contained ptnti-PGP labeled clusters of presumed dendritic cells. These pvere primarily located in lamina basalis between the dermal apillae, especially where the epidermal innervation was ticularly low (Fig. 2E-H). As revealed by anti-PGP labeling, PHN skin had a much ower overall density of innervation than contralateral skin Fig. 2), although the innervation density per unit length was j-jjot uniform in either. The variability encountered in each of |the 1-cm strips suggested that the 3-mm biopsies were qtiately representative of the overall skin innervation. he contralateral epidermis was supplied by three general ources of innervation: (1) axons located in the dermal ppapillae (#2,3 in Fig. 2A-D); (2) axons from small nerves Un the upper dermis but not in contact with the basement nembrane (#4,5); and (3) axons in intimate contact with the p-basement membrane (#6,7). Endings supplied by the latter pwo sources directly enter the epidermis between the dermal apillae (Figs. 2C, D and 3B). In contrast, nearly all of the I epidermal innervation in PHN skin was derived from axons |%ntering the dermal papillae (Figs. 2G, H and 3K, L). In |«ontralateral epidermis, endings usually had a simple, radial nentation and rarely terminated near stratum lucidum. Some terminated abruptly as short T-shaped branches oriented parallel to the skin surface. Endings in the PHN epidermis formed more tortu­ous, highly branched endings that could have long oblique trajectories and often terminated very superficially, near or at stratum lucidum (Fig. 3E, F, K, L).

The immunochemical composition of the innervation in PHN skin differed from that in the contralateral biopsy. Less than 20% of PGP positive endings in the contralateral epidermis (Fig. 31) co-labeled for anti-CGRP, while nearly all of the PGP-positive endings in the PHN epidermis labeled with anti-CGRP (Fig. 3L). Anti-VR-1 labeled profiles were rarely seen in the dermal papillae of the contralateral skin and were not detected in the epidermis (Fig. 3Ct H). Innervation with detectable anti-VR-1 labeling was present in PHN skin (Fig. 3C, G, H, K). At least one and often as many as three VR-1 positive axons could be detected in most of the enlarged dermal papillae of the PHN skin (Fig. 3G, K) and VR-1 positive endings with long oblique and superficial trajectories were present in the epidermis (Fig. 3F, K, M). Many of these endings also co-labeled with anti-CGRP (Fig. 3M). Interestingly, over half of the VR-1 positive endings in the PHN skin had little or no PGP immunoreactivity (Fig. 3K, M).

Finally, PHN skin had less thicker-caliber innervation than the contralateral skin sample. First, hair follicles had fewer of the lanceolate endings (Fig. 3 A, D) that are derived from large-caliber, myelinated axons in other species (e.g. Rice et al., 1997; Rice and Rasmusson, 2000). Second, PHN skin lacked relatively thick-caliber innervation that, in contralateral skin, intensely labeled with anti-NF and termi­nated as coiled endings in dermal papillae (Fig. 3J). Contral­ateral skin lacked detectable anti-NF labeling on thin-caliber axons in the dermis or on endings in the epidermis (Fig. 3J). In PHN skin, NF labeling was present on thin-caliber axons and epidermal endings and often co-expressed anti-CGRP immunoreactivity.

5. Conclusions

In the present patient, skin excision for PHN pain and allodynia reduced pain, eliminated tactile allodynia and allowed a substantial decrease in medication use that has been maintained over a 1-year follow-up period. In prior studies, we have provided evidence to support the hypoth¬esis that PHN associated neural dysfunction spans a spectrum. At one end is 'deafferentation' with marked sensory loss and central reorganization, and at the other end 'irri¬table nociceptors' with minimal deafferentation and allo¬dynia maintained by abnormal nociceptor input to chronically sensitized CNS targets (Fields et al., 1998; Rowbotham et al., 1998). 'Irritable nociceptors' refers to preserved, and possibly sensitized, primary afferent noci¬ceptors that remain connected to their peripheral cutaneous and central targets. The abnormal function and possible spontaneous activity in these nociceptors may provide sufficient ongoing input to maintain the CNS in a state of chronic  sensitization.  Selectively  stimulating  cutaneous nociceptors by applying topical 0.075% capsaicin to a small area of PHN skin {the  'capsaicin response test'): increased pain  and enlarged  the  area  of allodynia in : many of the PHN patients studied previously, analogous to experimentally  produced  secondary hyperalgesia in healthy volunteers (Petersen et al., 2000).

The pain relief following surgical skin excision experi­enced by this patient suggests that afferents originating in or very near the skin contributed to ongoing pain and allody­nia. Pre-operatively, there were several clues that the mechanisms of his pain were consistent with the 'irritable nociceptor' hypothesis. First, the results of his prior research study participation showed relatively modest thermal sensory loss and reduction in cutaneous innervation when compared to study group means. Second, allodynia to gentle touch was a major part of his symptom complex. Third, daily use of lidocaine patches produced a stable reduction in PHN pain and aliodynia. Fourth, the marked worsening of pain and allodynia during brief focal application of 0.075% capsaicin cream suggested the presence of functioning, and possibly sensitized, cutaneous afferents expressing the VR-1 receptor.

The skin excision provided the unique opportunity to nalyze the cutaneous innervation of a quantity of PHN that is far greater than has heretofore been possible, nor studies of PHN skin have been limited to PGP 9.5 nalysis of one or two 3-mm punch biopsies of the affected srea per patient. In contrast, the large piece of skin excised Mowed us to analyze 14 punch biopsies and 10 strips of t (each 10-mm long) from the painful side. Furthermore, omprehensive double-labeling immunofiuorescence eval-ation of human skin affected by neuropathic pain has not en previously reported. Although the patient would only sent to a single contralateral punch biopsy and would not Blow biopsy of distant normal skin, the morphology and nervation in the contralateral biopsy was unlike that of I the PHN skin samples. A prior study by Oaklander and oworkers (Oaklander et al., 1998) provided evidence that ntralateral skin may have reduced cutaneous innervation ompared to distant normal skin. If so, the difference in nervation density between contralateral skin and PHN an described here could underestimate the differences etween PHN and normal skin. The descriptions in the present study are qualitative observations without the use Of quantitative morphological techniques in part described ' Coggeshall (1999), because there is only a single punch biopsy of contralateral skin compared to 24 samples of PHN kin.

The qualitative analysis of the skin biopsies is relevant to nechanistic theories of PHN pain and neural dysfunction, he results confirm and expand upon the findings of prior 1 Studies by indicating that many specific subsets of cutaneous pnnervation in PHN skin are reduced, including the larger Caliber afferents innervating hair follicles and the dermal apillae as well as the smaller caliber fibers that normally -innervate the epidermis without first passing through the Idermal papillae. The small caliber afferents labeling for line VR-1 receptor appeared in increasing number in PHN |*kin. The VR-1 receptor is the site of action of capsaicin and been implicated in mediating polymodal nociception Fominaga et al, 1998; Caterina et al., 2000). What is nost   interesting   is   the   correspondence   between   this itient's worsened pain and allodynia during brief focal apsaicin application and an increase in innervation expres-ng the VR-1 receptor. Furthermore, much of this innerva-on apparently lacks or has low levels of PGP 9.5, which as been the most reliable marker for immunolabeling cuta-eous   innervation.   Other  changes   in  PHN   skin   were elected; the origins of which cannot yet be fully explained, early all of the epidermal innervation that was labeled by nti-PGP in our study was also immunoreactive for CGRP, |which has also been implicated in nociception (Urban et al., |i995). Although PHN skin lacked the coarse NF positive [Innervation to the dermal papillae. NF labeling was detected ion numerous thin-caliber axons in the PHN epidermis, a £ pattern that was not present in contralateral skin. Since  labeling is commonly expressed on myelinated axons |(Rice et al., 1998), the NF positive endings in the PHN epidermis may also be abnormal innervation. PGP labeled dendritic cells were also observed only in PHN skin. This finding has been previously recognized in animals under­going experimental deafferentation (Rice et al., 1993). The pattern of reduction in some types of innervation and increase (or altered morphology) in others may be a source of abnormal spontaneous activity and altered responses to sensory stimuli, thus contributing to the enduring neuro­pathic pain of PHN.

In summary, surgical skin excision is an irreversible ther­apy that requires a prolonged healing time and leaves a lengthy scar. We cannot recommend skin excision for others with PHN until pre-operative testing can reliably predict a positive long-term outcome. Despite 1-year follow-up, there is no guarantee that pain and allodynia will not eventually return. The intensive study of the large amount of resected PHN skin in this patient has allowed identification of several changes in cutaneous innervation. The normal sources of this innervation, the reasons for the alterations following herpes zoster and the potential contribution to PHN symp­toms warrant further investigation, and could lead to more effective treatments for PHN.

Supported by NTNDS grants K24 NS02164, RO-1 NS39521, RO-1 NS34692 and a grant from the VZV Research Foundation Inc.

Abbott K.H, Martin BC. Surgical treatment of post-herpetic neuralgia by subdermal denervation. Neurology 1951;l:275-282.
Browder J, deVeer J A. Herpes zoster: a surgical procedure for the treatment of post-herpetic neuralgia. Ann Surg 1949;) 30:622-636.
Caterina MJ, Leffler A. Malmberg AB. Martin WJ, Trafton J, Petersen-Zeitz KR, Koltzenburg M, Basbaum Al, Julius D. Impaired nociception and pain sensation in mice lacking the capsaicin receptor. Science 2000:288:306-313.
Coggeshall RE. Assaying structural changes after nerve damage, an essay on quantitative morphology. Pain !999;Suppl 6:S21-S25.
Fields HL, Rowbotham M. Baron R. Post-herpetic neuralgia: irritable noci-ceptors and deafferentation. NeurobioS Dis !998:5:209-227.
Gowers WR. A manual of diseases of the nervous system, Philadelphia, PA: BlakistonCo, 1900.
Guo A, Vulchanova L, Wang J, Li X, Elde R. Immunocytochemical loca­lization of the vanilloid receptor 1 (VR1): relationship to neuropeptides, the P2X3 purinergic receptor and IB4 binding sites. Eur J Neurosci 1999; 11:946-958.
Locser J. Surgery for post-herpetic neuralgia. In: Watson, Gershon, editors. Her pes zoster and post-herpetic neuralgia. Elsevier, Amsterdam, 2nd ed. 2001. p. 255-264.
Oaklander AL, Romans K, Horasek S, Stocks A, Hauer P, Meyer RA. Unilateral post-herpetic neuralgia is associated with bilateral sensory neuron damage. Ann Neurol 1998:44:789-795.
Pare M, Eide R, Mazurkiewtcz JE, Rice FL. The meissner corpuscle revised: a multi-afferented mechanoreccptor with nociceptor immuno-chemical properties. J Neurosci 2001 ;21:7236-7246.
Petersen KL. Fields H, Brennum J, Sandroni P, Rowbotham MC. Capsaicin evoked pain and allodynia in post-herpetic neuralgia. Pain 2(XX);88: 125-133.
Rice FL, Kinnman E. Aldskogius H, Johansson O, Arvkisson I. The inner-vation of the mystaciai pad of the rat a.s revealed by PGP 9.5 immuno-fluoreseenee. J Comp Neural 1993:337:366-385.
Rice FL, Fundin BT, Arvidsson J. Aldskogius H, Johansson O. A compre­hensive immunofluorescence and lectin binding of vibrissai follicle sinus complex innervalion in the mystaciai pad of the rat. J Comp Neurol 1997:385:149-184.
Rice FL, Albers KM, Davis BM. Silos-Santiago I. Wilkinson GA, LeMaster AM. Emfors P, Smeyne RJ, Aldskogius H, Phillips HS, Barbacid M, DeChiara TM, Yancopoulos GD, Dunne CE, Fundin BT. Differential effects of various neurotrophin and trk receptor deletions on the unmye-iinaled innervation of the epidermis in the whisker pad of the mouse. Dev Biol 1998:198:57-8!.
Rice FL. Rasinusson DD. Innervation of the digit on the forepaw of I
raccoon J Comp Neurol 2000;417:467-490. Rowbotham MC. Yosipovitch G, Connolly MK. Finlay D, Forde G. i
HL. Cutaneous innervation density in the ailodynic form of post-ner]
tic neuralgia. Neurobiol 0is 1996:3:205-214. Rowbotham MC, Petersen KL, Fields HL. Is post-herpetic neuralgia t
than one disorder? Pain Forum i998;7:231-237. Tominaga M, Caterina MJ. Malmberg AB. Rosen TA, Gilbert H, Skim
K, Raumann BE. Basbaum Al, Julius D. The cloned capsaicin rece
integrates multiple pain-producing stimuli. Neuron 1998:21:531-5' Urban L, Thompson SW. Fox AJ, Jeftinija S, Dray A. Peptidergic afferett
physiological aspects. Prog Brain Res 1995:104:255-269. Weidinann MJ. Surgical relief of post-herpetic pain by excision of sea
skin. Med J Aust 1976:1:472-473.


  Face Lift
  Nose Surgery
  Eyelid Surgery
  Chin Augmentation
  Neck Lift
  Breast Augmentation
  Breast Reduction
  Breast Lift
  Breast Reduction for Men
  Ultrasonic Liposuction
  Tummy Tuck
  BodyTite Lipo
  Pectoral Implants for Men
  Endermologie Cellulite Treatment
  Skin Cancer Treatment
  Skin Rejuvenating Peel
  Jane Iredale® Cosmetics
  Latisse (TM)
  Kinerase® Products

Top Doctor Plastic Surgery San Francisco, CA 94109

Copyright © 2007 Aesthetic Enhancement Clinic All Rights Reserved.