Nerve fibre and sensory end organ density in the epidermis and papillary dermis of the human hand

Transcription

1 British Journal of Plastic Surgery (2005) 58, Nerve fibre and sensory end organ density in the epidermis and papillary dermis of the human hand E.J. Kelly a,b,c, G. Terenghi d, A. Hazari d, M. Wiberg a,c,d, * a Departments of Anatomy and Hand and Plastic Surgery, Umeå University, Umeå, Sweden b Department of Plastic Surgery, Wythenshawe Hospital, Manchester, UK c Department of Plastic, Hand and Reconstructive Surgery, St James s University Hospital, Leeds, UK d Plastic and Reconstructive Surgery Research, University of Manchester, Manchester, UK Received 13 July 2004; accepted 9 December 2004 KEYWORDS Hand; Skin; Nerve fibre density; Protein gene product 9.5 Summary Quantification of sensory recovery after peripheral nerve surgery is difficult and no accurate techniques are available at present. Quantification of reinnervated skin has been used experimentally, and in some clinical studies, but the lack of knowledge about the normal sensory distribution has been a problem. The purpose of this study was, therefore, to map the density of sensory end organs, nerve fibres and free nerve endings in the glabrous skin of the human hand. Skin biopsies were taken from patients undergoing acute and elective hand surgery. Nerve fibres were stained in the epidermis and papillary dermis and quantified in five sites on the palm of the hand, using protein gene product 9.5 immunoreactivity a panneuronal marker. The finger tip skin was found to have more than twice the nerve fibre density in the papillary dermis than the skin of the palm, and the number of Meissner corpuscles in the finger tip was also higher than in the palm. We found a reduction in innervation density with increasing age in the dermis, however, that was not the case for the epidermis. The innervation of the epidermis showed high interindividual variability and unlike the papillary dermis did not display any pattern of distribution in the hand. Q 2005 The British Association of Plastic Surgeons. Published by Elsevier Ltd. All rights reserved. Introduction The sensory target organs and the free nerve endings in the glabrous skin of the finger and toes * Corresponding author. Address: Department of Hand and Plastic Surgery, University Hospital, S Umeå, Sweden. Tel.: C ; fax: C address: mikael.wiberg@handsurg.umu.se (M. Wiberg). have been studied in humans, 1 5 but there is not yet any detailed knowledge of the quantity or pattern of distribution of sensory target organs and free nerve endings in the glabrous skin of the human hand. The quality of sensory outcome following nerve repair is difficult to objectively quantify. There is not any specific method available, that alone can quantify functional sensibility in the hand and efforts to establish reliability in the evaluation of S /$ - see front matter Q 2005 The British Association of Plastic Surgeons. Published by Elsevier Ltd. All rights reserved. doi: /j.bjps

2 Nerve fibre density in the hand 775 hand sensibility continue. 6 Sensory outcome is an even more difficult assessment in laboratory animals in whom all the currently evolving neurobiological advances in nerve repair are being tested. There has been recent interest in correlating sensory target organ histology with sensory outcome following nerve repair. 7,8,21 More knowledge about the anatomy of the sensory end organs and the free nerve endings in the human hand may help in developing an outcome assessment following peripheral nerve surgery. Our aim was to study and quantify the neuronal structures in the skin on the palmar aspect of the normal human hand. Patients and methods Patients In accordance with local ethics committee approval in Umeå, Sweden and South Manchester, UK, patients undergoing elective and emergency hand surgery volunteered to donate one or more 8!4mm 2 biopsy of glabrous skin. The skin was chosen from an area where the patient reported no sensory deficit and where there were no signs of inflammation. Sixty biopsies in all were studied, 30 biopsies represented 21 volunteers over 60 years and 30 biopsies represented 24 volunteers less than 40 years. The group over 60 had a mean age of 68.3G4.1 and the group less than 40 had a mean age of 30.3G8.4. There were three females in the over 60 group and eight females in the less than 40 group. Five key areas were selected to represent the entire hand. These are shown in Fig. 1. Each site was represented by 12 biopsies. Figure 1 The five sites on the hand where the innervation in the skin was quantified. Skin Following harvest, the skin biopsies were fixed in Zamboni s solution for 12 h at 4 8C, and then rinsed three times over h in a phosphate buffered solution containing sodium azide and 15% sucrose and then stored at 4 8C. The skin samples were frozen in OCT and 15 mm sagittal sections were cut using a cryostat and every sixth section was collected onto a Vectabonde (Vector UK) coated glass slide. The sections were allowed to dry at 37 8C overnight, and were then permeabilised in 0.2% Triton X-100 for 1 min. Following three washes in PBS, the sections were incubated in normal goat serum (diluted 1:100) at room temperature for 20 min. The sections were then incubated with polyclonal antibody to protein gene product 9.5 a pan neuronal marker (PGP 9.5) (diluted 1:8000) for 24 h at 4 8C, and then placed twice in Tween for 30 s, washed three times in PBS and then incubated in biotynylated goat anti-rabbit sera (diluted 1:100) for 1 h at room temperature. The sections were then again placed twice in Tween, (2!30 s), washed three times in PBS and then incubated in avidin biotin complex solution (diluted 1:1000) for 1 h at room temperature. After final immersions in Tween (2!30 s), followed by washes in PBS, (3! 3 min), the sections were incubated in a sodium acetate buffer, (ph 6.0), (2!5 min), and developed in diaminobenzidine until the staining was clearly visible. The developing reaction was stopped by immersion for 10 min in sodium acetate, and then the sections were dehydrated through graded alcohols and cleared in xylene prior to mounting in DPX. The immunohistochemical staining of PGP 9.5 produces a dark blue black reaction product, which contrasts well with the pale nonstained background. The morphology of the innervation of the skin was studied under direct light microscopy using 10!, 20! and 40! objective lenses. The innervation of the papillary dermis for each specimen was quantified by randomly selecting six representative fields at!20 magnification. The field captured by the digital camera always included the most superficial part of the granular layer of the epidermis. Quantification was carried out as follows using Image-Pro w Plus 4.5 (Media Cybernetics). Each selected field was converted into an 8-bit greyscale digital image before automatic thresholding to separate the stained nerve fibres from the background. The threshold was maintained within a narrow range in order to reduce any bias, and to avoid dark background staining being counted. Following final manual editing of

3 776 any structures other than nerve fibres, the area of staining was quantified and expressed as a percentage of the area of papillary dermis in the field. The mean value was calculated from the measurements of six fields. In every section processed, the total Meissner corpuscle count was manually recorded along with the corresponding length of epidermis so that the Meissner count per mm could be calculated for each biopsy. The number of intra-epidermal nerve fibres was also manually counted in every section processed and the number per mm was calculated for each biopsy. Statistical analysis This was carried out using SigmaState (Version 2.0) software. All data was subjected to normality and equal variance tests. All comparisons were made between groups using the student s t-test. Results Morphology E.J. Kelly et al. Clearly stained nerve fibres in the papillary dermis were seen to run parallel to the epidermis just below the tips of the dermal papillae. This collection of fibres is conventionally known as the subpapillary plexus. From this plexus branches or terminal fibres left obliquely to enter the epidermis either directly, or to end at the epidermal dermal junction, or to enter a dermal papilla. The branches that ended at the epidermal dermal junction merged with Merkel cell neurite complexes which are disc like sensory end organs which stained densely like the nerve fibres. Some dermal papillae were found to be occupied by one to three Meissner corpuscles, but mostly by just one. The Meissner corpuscle is a low threshold mechanoreceptor and its rich neural component allows visualisation of its architecture. Between two and seven fibres were Figure 2 (a) IEF, intra-epidermal nerve fibre; MC, Meissner corpuscle; NF, nerve fibre in papillary dermis. (b) DP, dermal papilla; MK, Merkel cell neurite complex, at the epidermal dermal junction. (c) Represents a higher power view and (d) displays an intra-epidermal nerve fibre cluster (NFC) never previously described (measurement barz10 mm).

4 Nerve fibre density in the hand 777 seen entering these papillae. They measured approximately 80 mm long and 45 mm wide. We observed that in the older subjects the Meissner corpuscles were smaller. The other dermal papilla were either empty of any neuronal structures or filled with a rich network of nerve fibres. Nerve fibres in the epidermis were visible in the deepest three layers. They appeared in short segments, which reflects the tortuous route they take from the stratum germinativum to stratum lucidum. Some fibres had a corksrew like appearance in the more superficial layers. The fibres in the epidermis were thinner than those in the dermis and were either smooth or beaded. In some specimen, we noted a cluster of fibres closely bunched together Fig. 2(d). Quantification The percentage area of staining in the papillary dermis is shown in Table 1 (over 60 years group) and Table 2 (less than 40 years group). The innervation density was significantly less in the older group of subjects in three of the five sites measured, the proximal and distal palm, p!0.001 and the finger tip, pz The percentage of staining was significantly greater (p!0.001) in the fingers than in the palm for both age groups, while the innervation density in the proximal and distal palm showed no difference. The Meissner corpuscle count per mm is depicted in Tables 3 and 4. There was a statistically significant difference between the palm and finger tip for the younger age group (pz0.004), but not for the older group (pz0.031). The number of intra-epidermal nerve fibres per mm of epidermis is shown in Table 5 (over 60 years group) and Table 6 (under 40 years group). They was no statistically significant difference between both groups nor was there any statistical difference within the groups between the key areas studied. Discussion Skin innervation The method of visualising sensory endorgans and free nerve endings in the epidermis and dermis of the human hand, demonstrated in the present study a typical anatomical pattern that corresponded to previous qualitative investigations. 1 5,9 Jackson and Thompson detected PGP9.5 in 1981, 10 and Thompson et al. in 1983 showed that it was a cytoplasmic neuron-specific protein widely distributed in vertebrate brains and cells of the human diffuse neuroendocrine system. 11 Using immunohistochemistry a number of authors have used specific antiserum to PGP9.5 to effectively visualise all the types of nerve fibres in human and mammalian skin There have been many attempts at visualising and quantifying nerves and sensory end organs in human skin, including the glabrous type, but earlier attempts have been with silver stains and methylene blue techniques which are both difficult and time consuming. 1,3,5 The more recent studies have used immunohistochemistry with antisera to PGP9.5, and other neuropeptides which are specific to different fibre types, 9 and it is now widely accepted that antiserum to PGP9.5 will reliably stain most if not all fibres in skin. In the present study, the innervation density of the papillary dermis was measured and in spite of the interindividual variability, a certain pattern could be demonstrated. In both age groups the percentage of staining in the finger tips was over twice that found in the palm. The reduction in innervation density with age has not previously been demonstrated, although, Ridley described Meissner corpuscle degeneration with age. 3 Meissner corpuscles have also been described as reducing in size and number with age in glabrous skin. 15 Any reduction in number was not seen in the present study, but the reduction in their size could Table 1 The percentage area of innervation density in the papillary dermis in the five key areas in the over 60 years of age group Palm proximal Palm distal Finger proximal Finger middle Finger tip

5 778 E.J. Kelly et al. Table 2 Percentage area of innervation density in the papillary dermis for those subjects less than 40 years of age Palm proximal Palm distal Finger proximal Finger middle Finger Tip Table 3 Meissner corpuscle count per mm of epidermis in the older group Palm proximal Finger tip Table 4 Meissner corpuscle count per mm of epidermis in the younger group Palm proximal Finger tip Table 5 Manual intra-epidermal count per mm epidermis in older group Palm proximal Palm distal Finger proximal Finger middle Finger tip Table 6 Manual intra-epidermal count per mm epidermis in younger group Palm proximal Palm distal Finger proximal Finger middle Finger Tip well have accounted for the reduction in percentage area stained in the papillary dermis. A significant increase in the density of Meissner corpuscles in the finger tip compared to the palm was also observed which most possibly accounts for the better two-point discrimination at the finger tip. The Merkel cell neurite complex, a densely innervated disc like structure, was seen in the papillary dermis, demonstrated to convey information about prolonged skin deformation and, therefore, also relevant in appreciation of two point discrimination. 16 In contrast to the dermal innervation, the intraepidermal fibre counts varied a lot in both age groups, and there was no obvious pattern of fibre distribution. The reason for this is unclear but may

6 Nerve fibre density in the hand 779 be that the presence of these fibres is related to occupation and manual activity, and we know that none of the older group was engaged in manual labour and we have no knowledge of the occupation in the younger group but they were more likely to be involved in manual labour as the majority of volunteers in this group were having surgery for traumatic injuries to their hands. Analysis of epidermal nerve fibre distribution in the thigh and distal part of the leg, using PGP9.5 as a marker was performed previously 14 in order to assess the spatial distribution of involvement in peripheral nerve disease and the response to neurotrophic and other restorative therapies. In this study, no significant decrement with age was found which was also the case in a study using the older technique of silver staining. 3 Clinical implications Return of sensory nerve fibres to the skin following sciatic nerve injury using immunohistochemical methods has been reported in mice and rats. 17,18 A comparison of functional recovery was made with nerve terminal morphology in the rat after sciatic nerve crush and reappearance of nociception coincided with nerve terminal reinnervation. 18 A number of investigators have addressed the correlation between sensory recovery measured clinically and skin innervation of the human hand, 7,8,19 21 but these have either been too small to draw a worthwhile conclusion or have failed to show any relationship. 7,19 21 However, Fu Chang Wei showed statistically significant correlation between moving two point discrimination and the number of regenerated Meissner corpuscles in the transplanted toe wrap-around flaps in children undergoing a debulking procedure of the flap at secondary surgery. 8 The conclusion from our study is that quantitative analysis of skin innervation is possible in the human hand, but because of interindividual variability a comparison with the patients contralateral hand is necessary particularly if it is to be used in sensory outcome evaluation. If this type of measurement is applied, in combination with clinical and neurophysiological techniques, it would be a useful complement to existing methods of sensory outcome evaluation improving the accuracy of evaluation and quantification of sensory recovery. Acknowledgements The authors would like to thank staff at Umea University Hospital and Wythenshawe Hospital for their help in collecting and storing the biopsies. References 1. Arthur RP, Shelley WB. The innervation of human epidermis. J Invest Dermatol 1959;32: Miller MR, Ralston III HJ, Kasahara M. The pattern of cutaneous innervation of the human hand. Am J Anat 1958; 102: Ridley A. Silver staining of nerve endings in human digital glabrous skin. J Anat 1969;104: Cauna N. Fine morphological characteristics and microtopography of the free nerve endings of the human digital skin. Anat Rec 1980;198: Novotny GEK, Gommert-Novotny E. Intraepidermal nerves in human digital skin. Cell Tissue Res 1988;254: Novak CB, Mackinnon SE, Williams JI, Kelly L. Development of a new measure of fine sensory function. Plast Reconstr Surg 1993;92: Dellon AL, Munger BL. Correlation of histology and sensibility after nerve repair. J Hand Surg [Am] 1983;8: Wei FC, Carver N, Lee YH, Chuang DC, Cheng SL. Sensory recovery and Meissner corpuscle number after toe-to-hand transplantation. Plast Reconstr Surg 2000;105: Nolano M, Provitera V, Crisci C, Stancanelli A, Wendelschafer-Crabb G, Kennedy WR, et al. Quantification of myelinated endings and mechanoreceptors in human digital skin. Ann Neurol 2003;54: Jackson P, Thompson RJ. The demonstration of new human brain-specific proteins by high-resolution two-dimensional polyacrylamide gel electrophoresis. JNeurolSci1981;49: Thompson RJ, Doran JF, Jackson P, Dhillon AP, Rode J. PGP 9.5 a new marker for vertebrate neurons and neuroendocrine cells. Brain Res 1983;278(14): Kennedy WR, Wendelschafer-Crabb G. The innervation of human epidermis. Neurol Sci 1993;115: Rice FL, Rasmusson DD. Innervation of the digit on the forepaw of the raccoon. Comp Neurol 2000;417(21): McArthur JC, Stocks EA, Hauer P, Cornblath DR, Griffin JW. Epidermal nerve fiber density: normative reference range and diagnostic efficiency. Arch Neurol 1998;55: Bolton CF, Winkelmann RK, Dyck PJ. A quantitative study of Meissner s corpuscles in man. Neurology 1966;16: Nolano M, Provitera V, Crisci C, Stancanelli A, Wendelschafer-Crabb G, Kennedy WR, et al. Quantification of myelinated endings and mechanoreceptors in human digital skin. Ann Neurol 2003;54: Hsieh ST, Chiang HY, Lin WM. Pathology of nerve terminal degeneration in the skin. J Neuropathol Exp Neurol 2000;59: Verdu E, Navarro X. Comparison of immunohistochemical and functional reinnervation of skin and muscle after peripheral nerve injury. Exp Neurol 1997;146: Ramieri G, Stella M, Calcagni M, Cellino G, Panzica GC. An immunohistochemical study on cutaneous sensory receptors after chronic median nerve compression in man. Acta Anat (Basel) 1995;152: Jabaley ME, Burns JE, Orcutt BS, Bryant M. Comparison of histologic and functional recovery after peripheral nerve repair. J Hand Surg [Am] 1976;1: Wiberg M, Hazari A, Ljungberg C, et al. Sensory recovery after hand reimplantation: a clinical, morphological, and neurophysiological study in humans. Scand J Plast Reconstr Surg Hand Surg 2003;37: