- Email: [email protected]
Absent innervation of skin and sweat glands in congenital insensitivity to pain with anhidrosis
Clinical Neurophysiology 111 (2000) 1596±1601
www.elsevier.com/locate/clinph
Absent innervation of skin and sweat glands in congenital insensitivity to pain with anhidrosis M. Nolano a,*, C. Crisci a, L. Santoro b, F. Barbieri b, R. Casale c, W.R. Kennedy d, G. Wendelschafer-Crabb d, V. Provitera a, N. Di Lorenzo b, G. Caruso b b
a Salvatore Maugeri Foundation, I.R.C.C.S., Center of Telese Terme (BN), Italy Department of Neurological Sciences, University of Naples, Federico II, Naples, Italy c Salvatore Maugeri Foundation, I.R.C.C.S., Center of Montescano (PV), Italy d Department of Neurology, University of Minnesota, Minneapolis, MN, USA
Accepted 8 May 2000
Abstract Objectives: A case of a 10-year-old girl with congenital insensitivity to pain with anhidrosis (CIPA) is reported. Methods and results: Parents referred several hyperpyretic episodes without sweating occurring since birth, and insensitivity to pain, noticed when the child was 2 years old. Her body had many bruises and scars, bone fractures and signs of self-mutilation. Neurological examination was normal except for insensitivity to pain. Her IQ was 52. Electrical and tactile sensory nerve conduction velocities were normal. The patient was unable to detect thermal stimuli. Histamine injection evoked a wheal but not a ¯are; pilocarpine by iontophoresis did not induce sweat. Microneurography showed neural activity from A-beta sensory ®bers while nociceptive and skin sympathetic C ®ber nerve activity was absent. No small myelinated ®bers and very rare unmyelinated ®bers were found in the sural nerve. Immunohistochemistry showed a lack of nerve ®bers in the epidermis and only few hypotrophic and uninnervated sweat glands in the dermis. Conclusions: The lack of innervation of the skin (C and A-delta ®bers) appears to be the morphological basis of insensitivity to pain and anhidrosis, and is consistent with the loss of unmyelinated and small myelinated ®bers in the sural nerve biopsy. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Skin; Sweat glands; Congenital insensitivity to pain; Anhidrosis; Cutaneous innervation
1. Introduction Congenital insensitivity to pain with anhidrosis (CIPA) or hereditary sensory neuropathy type IV (Dyck et al., 1983) is a very rare and severe syndrome. The condition is characterized by mental retardation, congenital analgesia leading to selfmutilation, multiple scars and fractures, and anhidrosis with repeated episodes of fever, especially in hot weather (Rosemberg et al., 1994). Death from hyperpyrexia occurs within the ®rst 3 years of life in about 20% of patients (Rosemberg et al., 1994). Ultrastructural and morphometric studies of sensory nerves show a loss of unmyelinated and small myelinated ®bers (Goebel et al., 1980). Skin and sweat glands have been described as `normal' (Swanson, 1963; Brown and Podosin, 1966; Pinsky and Di George, 1966; Mitaka et al., 1985; * Corresponding author. Department of Neurology, Salvatore Maugeri Foundation, I.R.C.C.S., Via Bagni Vecchi, 82037 Telese Terme (BN), Italy. Tel.: 139-824-909-111; fax: 139-824-909-614. E-mail address: [email protected] (M. Nolano).
Rosemberg et al., 1994); only one case of lack of innervation of eccrine sweat glands has been reported (Langer et al., 1981). We studied the clinical, electrophysiologic and morphologic features of a case of CIPA. The immunohistochemical study of the skin, in keeping with the electrophysiologic results and with the ultrastructural and morphometric studies of the sural nerve, revealed a lack of innervation of epidermis and sweat glands.
2. History and clinical examination Our patient was a 10-year-old girl who suffered asphyxia at birth and in whom a marked hypotonia was noticed. The parents were not consanguineous; a sister had died at 6 days from asphyxia. A 6-year-old sister had no symptoms or signs of central or peripheral nervous system involvement. At the age of 20 months, our patient was found to have microcephaly, mental retardation, and low weight and height; chronic diarrhea was diagnosed as due to intolerance of arti-
1388-2457/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S13 88-2457(00)0035 1-5
CLINPH 99752
M. Nolano et al. / Clinical Neurophysiology 111 (2000) 1596±1601
®cial milk. When the child was 2 years old, the mother noticed that she seemed unaware of occasional burns and cuts; once she had accidentally put a needle in her thigh without any evident pain. She had many infections and bone fractures due to asymptomatic lesions to the limbs, particularly the feet and legs. Her ®ngers bore many signs of self-mutilation, secondary to bites (Fig. 1). In addition, there had been several episodes of hyperpyrexia (up to 408C) in summer without sweating. The clinical examination of the skeletal musculature showed normal tone, normal strength and no muscle wasting. Tendon re¯exes were normal in the upper and lower limbs. The main clinical signs were insensitivity to pain with impaired temperature sense and inability to discriminate heat from cold; tactile sensation was normal. The skin was warm and dry, with no sweating. The patient was shy but cooperative during all the tests. Her language was poor but understandable. The psychometric evaluation showed a low IQ of 52. A computerized tomographic scan of the encephalus was normal. The parents of our young patient gave informed consent for all the procedures undertaken.
1597
orthodromic sensory conduction studies along the median, ulnar, peroneal and tibial nerves were performed. Sensory potentials evoked by mechanical stimulation were recorded along the median nerve. Tactile stimulation was applied on the distal phalanx of digit III using a vibrator triggered by a square-wave generator. Somatosensory (SSEPs), visual (VEPs) and brainstem (BAEPs) evoked potentials were investigated. SSEPs were elicited by electrical stimulation of median and tibial nerve and surface electrodes were used for recording. 3.1.2. Microneurography Using the technique devised by Valbo and Hagbarth (1968), the left peroneal nerve was impaled through the intact skin by means of a tungsten ®lament and the potentials were acquired using a 1±5 mm recording tip.
3. Methods
3.1.3. Sympathetic skin response The evoked potentials were recorded simultaneously from the hands and feet using surface electrodes. For the electrical arousal stimulation, random single 1 ms stimuli of 5±20 mA intensity were delivered at the volar wrist.
3.1. Electrophysiological tests
3.2. Autonomic tests
3.1.1. Nerve conduction study Conventional motor nerve conduction and near-nerve
3.2.1. Cardiovascular re¯exes R-R variation during rest and deep breathing, Valsalva
Fig. 1. Picture of the hands showing signs of self-mutilation of the ®ngertips.
1598
M. Nolano et al. / Clinical Neurophysiology 111 (2000) 1596±1601
ratio, 30/15 ratio, tilt test, and blood pressure response to handgrip and to standing were performed. 3.2.2. Sudomotor function Sweating was elicited by iontophoresis with direct current (2.0 mA) of a 1% aqueous solution of pilocarpine nitrate for 5 min. A count of sweat drops per cm 2 was made using a silastic mold (Kennedy et al., 1984). The vasomotor re¯ex in the skin was evaluated with intradermal injection of 1/1000 histamine (0.1 cc). 3.3. Morphological study 3.3.1. Sural nerve biopsy A 3 cm sural nerve sample excised in toto without anesthesia was ®xed in GTA, post-®xed in osmium tetroxide, embedded in Epon and stained with Toluidin Blue. Semithin sections were analyzed with a computerized system of image-analysis (VIDAS-Zeiss) for morphometric evaluation. Ultrathin sections were stained with uranyl acetate and lead cytrate and examined with a Philips EM 301 electron microscope. 3.3.2. Skin biopsy Samples of skin (3 mm punch biopsies) were taken from the thigh, back and calf (no samples were taken from the ®ngers because they were too damaged), ®xed in Zamboni solution and stored overnight in 0.1 M phosphate buffer saline with 20% sucrose added and then processed by immunohistochemical methods. Cryostat sections of 100 mm were stained using antisera to protein gene product 9.5 (PGP 9.5), calcitonin gene-related protein (CGRP), substance P (SP), vasomotor intestinal peptide (VIP), collagen IV (Col IV), S100, myelin basic protein (MBP) and secondary antibodies labeled with cyanine 3.18, cyanine 5.18 and ¯uorescein. Using a laser scanning confocal microscope (MRC-1000, Biorad, Boston, MA), a zseries of 2 mm optical sections was collected and then projected into a single in-focus digitized image of the 100 mm section. 4. Results 4.1. Electrophysiological ®ndings Conventional motor and orthodromic sensory nerve conduction by electrical and mechanical stimuli were normal. Similarly, somatosensory, visual and brainstem evoked potentials were normal. The microneurographic study of the left peroneal nerve showed intraneural activity from cutaneous fascicles only from A-beta afferent units connected to low threshold cutaneous mechanoreceptors. No activity from either sensory or skin sympathetic C ®bers was recorded. Moreover, intraneural electrical stimulation (about 3 mV strength) that produces unbearable pain in normals did not evoke any
painful sensation in our patient, but merely evoked tactile sensations that were localized to coherent cutaneous provinces. In a pure motor bundle, intraneural activity from mechanoreceptor units, evoked by the stretching of the anterior tibialis muscle, and bursts of muscular sympathetic activity were recorded. The sympathetic skin response was absent. 4.2. Autonomic test ®ndings Cardiovascular re¯exes were normal compared to the results obtained from a cohort of age-matched controls. The intradermal injection of histamine on the volar forearm produced a wheal but no axon ¯are and failed to evoke itch sensation. Pilocarpine by iontophoresis failed to produce any sweating. 4.3. Morphological ®ndings 4.3.1. Sural nerve biopsy The morphometric study of the sural nerve showed a unimodal histogram with a marked reduction of the total number of myelinated ®bers and a complete loss of myelinated ®bers ,4 mm (Fig. 2). Electron microscopy revealed an almost complete lack of unmyelinated ®bers. 4.3.2. Skin biopsy Confocal study of skin biopsies demonstrated a complete loss of cutaneous nerve ®bers in all the skin specimens; only in the thigh were very rare nerve ®bers found in the dermis (Fig. 3C). These showed some varicosities that were not observed in control subjects (Fig. 3A,B). In the same site specimen only two hypotrophic and uninnervated sweat glands were identi®ed (Fig. 3D). 5. Discussion The typical clinical features of CIPA are well known (Goebel et al., 1980; Rafel et al., 1980; Rosemberg et al., 1994); only the rarity of this condition can lead to misdiagnosis. In our case, CIPA was diagnosed when the patient was 10 years old. She came to our observation after several hospitalizations in the orthopedic ward for fractures and osteomyelitis. The impairment of pain and temperature sensation and the absence of sweating suggested a small ®ber neuropathy. Because of the patient's mental retardation, it was dif®cult to perform reliable quantitative sensory testing and to assess the subject's ability to perceive cold, warm, and painful stimuli. However, our diagnosis was con®rmed by the absence of SSR and by the lack of sweating after pilocarpine by iontophoresis. Cardiovascular re¯exes were normal in our patient in agreement with previous ®ndings in CIPA (Hilz et al., 1999). In addition, microneurography revealed intraneural activity from A-beta sensory units using mechanical stimuli (brush and von Frey stimulators), but no skin C ®ber activity. Moreover, sensory
M. Nolano et al. / Clinical Neurophysiology 111 (2000) 1596±1601
potentials evoked along the median and ulnar nerves by supramaximal electrical shocks and mechanical stimulation of the ®ngertips of digits III and V were normal, a ®nding compatible with a normal density of Meissner's and Pacinian corpuscles (Caruso et al., 1993, 1994), and with a good state of preservation of large diameter nerve ®bers. In fact, the ultrastructural and morphometric study of the sural nerve showed a selective loss of unmyelinated and small myelinated ®bers (Goebel et al., 1980; Itoh et al., 1986). In agreement with the clinical, electrophysiological and neuropathological features, the immunohistochemical study of the skin revealed a complete absence of epidermal ®bers (all unmyelinated ®bers) and an almost complete absence of dermal ®bers to blood vessels and erector pilorum muscles (mostly unmyelinated ®bers and small myelinated ®bers). While in normal subjects sweat glands present very dense innervation (Kennedy et al., 1994), only uninnervated and markedly hypotrophic sweat glands were found in our patient (Fig. 3D). The pathophysiology of CIPA is still unknown although a recent genetic study of 3 subjects affected by CIPA demonstrated a mutation in the TRKA/NGF receptor gene (Indo et al., 1996). The only autopsy study available, which was performed on a 12-year-old boy who succumbed during a bout of hyperpyrexia (Swanson et al., 1965), showed an absence of Lissauer tract and loss of small myelinated ®bers in the dorsal roots and in the spinal tract of the trigeminal
1599
nerve, and loss of small ganglion cells of the dorsal spinal ganglia. In this case, a genetically determined defect in differentiation and migration of neural crest elements early in embryogenesis was postulated (Swanson et al., 1965). The absence of sympathetic skin activity against the well preserved cardiovascular re¯exes and muscle sympathetic activity could be explained by the presence in sympathetic ganglia of structurally and functionally distinct populations of cells and by their time-dependence in differentiation from the neural crest, which was observed in an animal model (Hamburger and Levi-Montalcini, 1949). The congenital lack of small neurons in the spinal cord reported by Swanson as well as the frequently described loss of small myelinated ®bers and unmyelinated ®bers in sural nerve suggests, in CIPA, a lack of innervation of the target organs, the skin and its annexes. Nonetheless, most conventional histologic studies failed to demonstrate skin and sweat gland alterations (Swanson, 1963; Brown and Podosin, 1966; Pinsky and Di George, 1966; Mitaka et al., 1985; Rosemberg et al., 1994). In the present study we obtained clear evidence of a lack of skin and sweat gland innervation by means of immunohistochemical techniques. Recently, nerve ®bers in epidermis (Wang et al., 1990; Kennedy and Wendelschafer-Crabb, 1993; Kennedy et al., 1996) and in dermis around vessels and sweat glands (Kennedy et al., 1994) have been extensively studied using the pan-axonal marker PGP 9.5 (Thompson and
Fig. 2. Distribution of the diameters of myelinated ®bers in the sural nerve of the patient (dark columns) compared to the mean values of a group of agematched controls (white columns). In the patient: endoneurial area, 0.5 mm 2; total number of myelinated ®bers, 1424 (.7 mm, 72%; .11 mm, 4.7%); total number of unmyelinated ®bers, 500. In normal subjects: endoneurial area, 0.6±1.0 mm 2; total number of myelinated ®bers, 6200±8500; density, 10 333± 14 166; number of unmyelinated ®bers, 50 000±60 000.
1600
M. Nolano et al. / Clinical Neurophysiology 111 (2000) 1596±1601
Fig. 3. Laser scanning confocal microscope images of skin from the thigh of a 14-year-old healthy subject (A,B) and of our patient (C,D) immunostained for PGP 9.5 with Cy-3 ¯uorophore (shown in green) and for collagen IV with Cy-5 ¯uorophore (shown in red). Note the severe loss of cutaneous nerve ®bers in the epidermis, dermis (C) and around the few hypotrophic sweat glands in the patient (D) compared with a normal distribution of dermal and epidermal ®bers (A) and normal-size and innervation of sweat glands (B). E, epidermis; ENFs, epidermal nerve ®bers; BM, basement membrane.
Day, 1988). A reduction of epidermal nerve ®ber density has been demonstrated in sensory neuropathy (McCarthy et al., 1995), diabetic neuropathy (Kennedy et al., 1996) and in small ®ber neuropathy (Holland et al., 1998; Mendell et al., 1998). The loss of epidermal nerve ®bers observed in diabetic neuropathy correlates with hypoalgesia to heat and mechanical stimuli (pinprick), and thermal threshold abnormalities in affected patients (Nolano et al., 1996). Topical application of capsaicin in humans produces a degeneration of epidermal ®bers with hypoalgesia to heat and mechanical stimulations (Nolano et al., 1999). These reports suggest a primary role of epidermal nerve ®bers in conveying pain sensation. Therefore, we assume that the lack of epidermal nerve ®bers and the absence of nerve ®bers around sweat glands are the morphological basis of analgesia and anhidrosis in CIPA. The results from our immunohistochemical study of the skin are in keeping with a report of a loss of SP ®bers in a patella biopsy from a patient with CIPA (Derwin et al., 1994), although SP ®bers are only a small amount of cutaneous ®bers. Moreover, this pathologic picture would account for the frequent bone fractures because the nociceptive ®bers may play a trophic role in the skeletal system (Hill and Elde, 1991).
In conclusion, this is the ®rst study that shows in CIPA a lack of unmyelinated and small myelinated ®bers in the skin, providing a morphological basis for analgesia and anhidrosis. Consequently, skin biopsy should be considered a primary diagnostic approach in patients with suspected CIPA, instead of the more invasive sural nerve biopsy. References Brown JW, Podosin R. A syndrome of the neural crest. Arch Neurol 1966;15:294±301. Caruso G, Nilsson J, Crisci C, Nolano M, Massini R, Lullo F. Sensory nerve ®ndings by tactile stimulation of median and ulnar nerves in healthy subjects of different ages. Electroenceph clin Neurophysiol 1993;89:392±398. Caruso G, Nolano M, Lullo F, Crisci C, Nilsson J, Massini R. Median nerve sensory responses evoked by tactile stimulation of the ®nger proximal and distal phalanx in normal subjects. Muscle Nerve 1994;17:269±275. Derwin KA, Glover RA, Wojtys EM. Nociceptive role of substance-P in the knee joint of a patient with congenital insensitivity to pain. J Pediatr Orthop 1994;14:258±262. Dyck PJ, Mellinger JF, Reagan TJ, Horowitz SJ, McDonald JW, Litchy WJ, Daube JR, Fealey RD, Go VL, Kao PC, Brimijoin WS, Lambert EH. Not indifference to pain but varieties of hereditary sensory and autonomic neuropathy. Brain 1983;106:373±390. Goebel HH, Veit S, Dyck PJ. Con®rmation of virtual unmyelinated ®ber
M. Nolano et al. / Clinical Neurophysiology 111 (2000) 1596±1601 absence in hereditary sensory neuropathy type IV. J Neuropathol Exp Neurol 1980;39:670±675. Hamburger V, Levi-Montalcini R. Proliferation, differentiation and degeneration in the spinal ganglion of the chick embryo under normal and experimental condition. J Exp Zool 1949;111:457±500. Hill EL, Elde R. Distribution of CGRP-, VIP-, DbH-, SP-, and NPY-immunoreactive nerves in the periosteoum of the rat. Cell Tissue Res 1991;264:469±480. Hilz MJ, Stemper B, Axelrod FB. Sympathetic skin response differentiates hereditary sensory autonomic neuropathies III and IV. Neurology 1999;52:1652±1657. Holland NR, Crawford TO, Hauer P, Cornblath DR, Grif®n JW, McArthur JC. Small-®ber sensory neuropathies: clinical course and neuropathology of idiopathic cases. Ann Neurol 1998;44:47±59. Indo Y, Tsuruta M, Hayashida Y, Karim MA, Ohta K, Kawano T, Mitsubuchi H, Tonoki H, Awaya Y, Matsuda I. Mutations in the TRKA/NGF receptor gene in patients with congenital insensitivity to pain with anhidrosis. Nat Genet 1996;13:485±488. Itoh Y, Yagishita S, Nakajima S, Nakano T, Kawada H. Congenital insensitivity to pain with anhidrosis: morphological and morphometric studies on the skin and peripheral nerves. Neuropediatrics 1986;17:103±110. Kennedy WR, Wendelschafer-Crabb G. The innervation of human epidermis. J Neurol Sci 1993;115:184±190. Kennedy WR, Sakuta M, Sutherland D, Goetz F. Quantitation of the sweating de®ciency in diabetes mellitus. Ann Neurol 1984;15:482±488. Kennedy WR, Wendelschafer-Crabb G, Brelje T. Innervation and vasculature of human sweat glands. An immunohistochemistry-laser scanning confocal ¯uorescence microscopy study. J Neurosci 1994;14:6825±6833. Kennedy WR, Wendelschafer-Crabb G, Johnson T. Quantitation of epidermal nerves in diabetic neuropathy. Neurology 1996;47:1042±1048. Langer J, Goebel HH, Veit S. Eccrine sweat glands are not innervated in hereditary sensory neuropathy type IV. An electron microscopic study. Acta Neuropathol (Berlin) 1981;54:199±202. McCarthy BG, Hsieh ST, Stocks MA, Hauer P, Macko C, Cornblath DR,
1601
Grif®n JW, McArthur JC. Cutaneous innervation in sensory neuropathies: evaluation by skin biopsy. Neurology 1995;45:1848±1855. Mendell JR, Periquet I, Kissel JT, et al. Distal painful axonal idiopathic neuropathy: criteria for diagnosis and distinction from other sensory neuropathies (abstract). Neurology 1998;50:343. Mitaka C, Tsunoda Y, Hikawa Y, Sakahira K, Matsumoto I. Anesthetic management of congenital insensitivity to pain with anhidrosis. Anesthesiology 1985;63:328±329. Nolano M, Wendelschafer-Crabb G, Kennedy WR. Correlation of sensation with epidermal innervation in diabetic neuropathy. Presented at the 43rd AAEM Annual Scienti®c Meeting, October 3±5, 1996, Minneapolis, MN. Nolano M, Simone DA, Wendelschafer-Crabb G, Kennedy WR. Topical capsaicin in humans: parallel loss of epidermal nerve ®bers and pain sensation. Pain 1999;81:135±145. Pinsky L, Di George AM. Congenital familial sensory neuropathy with anhidrosis. J Pediatr 1966;68:1±13. Rafel E, Alberca R, Bautista J, Navarrete M, Lazo J. Congenital insensitivity to pain with anhidrosis. Muscle Nerve 1980;3:216±220. Rosemberg S, Nagahashi Marie SK, Kliemann S. Congenital insensitivity to pain with anhidrosis (hereditary sensory and autonomic neuropathy type IV). Pediatr Neurol 1994;11:50±56. Swanson AG. Congenital insensitivity to pain with anhidrosis. A unique syndrome in two male siblings. Arch Neurol 1963;8:83±90. Swanson AG, Buchan GC, Alvord EC. Anatomical changes in congenital insensitivity to pain. Arch Neurol 1965;12:12±18. Thompson RJ, Day INH. Protein gene product (PGP) 9.5: a new neuronal and neuroendocrine marker. In: Marangos PJ, Campbell X, Cohen RM, editors. Neuronal and glial proteins: structure, function and clinical application, Neurobiological research, 2. Millbrae, CA: California Academic Press, 1988. pp. 209±228. Ê B, Hagbarth KE. Activity from skin mechanoreceptors recorded Valbo A percutaneously in awake human subjects. Exp Neurol 1968;21:270±289. Wang L, Hilliges M, Jernberg T, Wiegleb-EdstroÈm D, Johansson O. Protein gene product 9.5 ± immunoreactive nerve ®bres and cells in human skin. Cell Tissue Res 1990;261:25±33.
www.elsevier.com/locate/clinph
Absent innervation of skin and sweat glands in congenital insensitivity to pain with anhidrosis M. Nolano a,*, C. Crisci a, L. Santoro b, F. Barbieri b, R. Casale c, W.R. Kennedy d, G. Wendelschafer-Crabb d, V. Provitera a, N. Di Lorenzo b, G. Caruso b b
a Salvatore Maugeri Foundation, I.R.C.C.S., Center of Telese Terme (BN), Italy Department of Neurological Sciences, University of Naples, Federico II, Naples, Italy c Salvatore Maugeri Foundation, I.R.C.C.S., Center of Montescano (PV), Italy d Department of Neurology, University of Minnesota, Minneapolis, MN, USA
Accepted 8 May 2000
Abstract Objectives: A case of a 10-year-old girl with congenital insensitivity to pain with anhidrosis (CIPA) is reported. Methods and results: Parents referred several hyperpyretic episodes without sweating occurring since birth, and insensitivity to pain, noticed when the child was 2 years old. Her body had many bruises and scars, bone fractures and signs of self-mutilation. Neurological examination was normal except for insensitivity to pain. Her IQ was 52. Electrical and tactile sensory nerve conduction velocities were normal. The patient was unable to detect thermal stimuli. Histamine injection evoked a wheal but not a ¯are; pilocarpine by iontophoresis did not induce sweat. Microneurography showed neural activity from A-beta sensory ®bers while nociceptive and skin sympathetic C ®ber nerve activity was absent. No small myelinated ®bers and very rare unmyelinated ®bers were found in the sural nerve. Immunohistochemistry showed a lack of nerve ®bers in the epidermis and only few hypotrophic and uninnervated sweat glands in the dermis. Conclusions: The lack of innervation of the skin (C and A-delta ®bers) appears to be the morphological basis of insensitivity to pain and anhidrosis, and is consistent with the loss of unmyelinated and small myelinated ®bers in the sural nerve biopsy. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Skin; Sweat glands; Congenital insensitivity to pain; Anhidrosis; Cutaneous innervation
1. Introduction Congenital insensitivity to pain with anhidrosis (CIPA) or hereditary sensory neuropathy type IV (Dyck et al., 1983) is a very rare and severe syndrome. The condition is characterized by mental retardation, congenital analgesia leading to selfmutilation, multiple scars and fractures, and anhidrosis with repeated episodes of fever, especially in hot weather (Rosemberg et al., 1994). Death from hyperpyrexia occurs within the ®rst 3 years of life in about 20% of patients (Rosemberg et al., 1994). Ultrastructural and morphometric studies of sensory nerves show a loss of unmyelinated and small myelinated ®bers (Goebel et al., 1980). Skin and sweat glands have been described as `normal' (Swanson, 1963; Brown and Podosin, 1966; Pinsky and Di George, 1966; Mitaka et al., 1985; * Corresponding author. Department of Neurology, Salvatore Maugeri Foundation, I.R.C.C.S., Via Bagni Vecchi, 82037 Telese Terme (BN), Italy. Tel.: 139-824-909-111; fax: 139-824-909-614. E-mail address: [email protected] (M. Nolano).
Rosemberg et al., 1994); only one case of lack of innervation of eccrine sweat glands has been reported (Langer et al., 1981). We studied the clinical, electrophysiologic and morphologic features of a case of CIPA. The immunohistochemical study of the skin, in keeping with the electrophysiologic results and with the ultrastructural and morphometric studies of the sural nerve, revealed a lack of innervation of epidermis and sweat glands.
2. History and clinical examination Our patient was a 10-year-old girl who suffered asphyxia at birth and in whom a marked hypotonia was noticed. The parents were not consanguineous; a sister had died at 6 days from asphyxia. A 6-year-old sister had no symptoms or signs of central or peripheral nervous system involvement. At the age of 20 months, our patient was found to have microcephaly, mental retardation, and low weight and height; chronic diarrhea was diagnosed as due to intolerance of arti-
1388-2457/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S13 88-2457(00)0035 1-5
CLINPH 99752
M. Nolano et al. / Clinical Neurophysiology 111 (2000) 1596±1601
®cial milk. When the child was 2 years old, the mother noticed that she seemed unaware of occasional burns and cuts; once she had accidentally put a needle in her thigh without any evident pain. She had many infections and bone fractures due to asymptomatic lesions to the limbs, particularly the feet and legs. Her ®ngers bore many signs of self-mutilation, secondary to bites (Fig. 1). In addition, there had been several episodes of hyperpyrexia (up to 408C) in summer without sweating. The clinical examination of the skeletal musculature showed normal tone, normal strength and no muscle wasting. Tendon re¯exes were normal in the upper and lower limbs. The main clinical signs were insensitivity to pain with impaired temperature sense and inability to discriminate heat from cold; tactile sensation was normal. The skin was warm and dry, with no sweating. The patient was shy but cooperative during all the tests. Her language was poor but understandable. The psychometric evaluation showed a low IQ of 52. A computerized tomographic scan of the encephalus was normal. The parents of our young patient gave informed consent for all the procedures undertaken.
1597
orthodromic sensory conduction studies along the median, ulnar, peroneal and tibial nerves were performed. Sensory potentials evoked by mechanical stimulation were recorded along the median nerve. Tactile stimulation was applied on the distal phalanx of digit III using a vibrator triggered by a square-wave generator. Somatosensory (SSEPs), visual (VEPs) and brainstem (BAEPs) evoked potentials were investigated. SSEPs were elicited by electrical stimulation of median and tibial nerve and surface electrodes were used for recording. 3.1.2. Microneurography Using the technique devised by Valbo and Hagbarth (1968), the left peroneal nerve was impaled through the intact skin by means of a tungsten ®lament and the potentials were acquired using a 1±5 mm recording tip.
3. Methods
3.1.3. Sympathetic skin response The evoked potentials were recorded simultaneously from the hands and feet using surface electrodes. For the electrical arousal stimulation, random single 1 ms stimuli of 5±20 mA intensity were delivered at the volar wrist.
3.1. Electrophysiological tests
3.2. Autonomic tests
3.1.1. Nerve conduction study Conventional motor nerve conduction and near-nerve
3.2.1. Cardiovascular re¯exes R-R variation during rest and deep breathing, Valsalva
Fig. 1. Picture of the hands showing signs of self-mutilation of the ®ngertips.
1598
M. Nolano et al. / Clinical Neurophysiology 111 (2000) 1596±1601
ratio, 30/15 ratio, tilt test, and blood pressure response to handgrip and to standing were performed. 3.2.2. Sudomotor function Sweating was elicited by iontophoresis with direct current (2.0 mA) of a 1% aqueous solution of pilocarpine nitrate for 5 min. A count of sweat drops per cm 2 was made using a silastic mold (Kennedy et al., 1984). The vasomotor re¯ex in the skin was evaluated with intradermal injection of 1/1000 histamine (0.1 cc). 3.3. Morphological study 3.3.1. Sural nerve biopsy A 3 cm sural nerve sample excised in toto without anesthesia was ®xed in GTA, post-®xed in osmium tetroxide, embedded in Epon and stained with Toluidin Blue. Semithin sections were analyzed with a computerized system of image-analysis (VIDAS-Zeiss) for morphometric evaluation. Ultrathin sections were stained with uranyl acetate and lead cytrate and examined with a Philips EM 301 electron microscope. 3.3.2. Skin biopsy Samples of skin (3 mm punch biopsies) were taken from the thigh, back and calf (no samples were taken from the ®ngers because they were too damaged), ®xed in Zamboni solution and stored overnight in 0.1 M phosphate buffer saline with 20% sucrose added and then processed by immunohistochemical methods. Cryostat sections of 100 mm were stained using antisera to protein gene product 9.5 (PGP 9.5), calcitonin gene-related protein (CGRP), substance P (SP), vasomotor intestinal peptide (VIP), collagen IV (Col IV), S100, myelin basic protein (MBP) and secondary antibodies labeled with cyanine 3.18, cyanine 5.18 and ¯uorescein. Using a laser scanning confocal microscope (MRC-1000, Biorad, Boston, MA), a zseries of 2 mm optical sections was collected and then projected into a single in-focus digitized image of the 100 mm section. 4. Results 4.1. Electrophysiological ®ndings Conventional motor and orthodromic sensory nerve conduction by electrical and mechanical stimuli were normal. Similarly, somatosensory, visual and brainstem evoked potentials were normal. The microneurographic study of the left peroneal nerve showed intraneural activity from cutaneous fascicles only from A-beta afferent units connected to low threshold cutaneous mechanoreceptors. No activity from either sensory or skin sympathetic C ®bers was recorded. Moreover, intraneural electrical stimulation (about 3 mV strength) that produces unbearable pain in normals did not evoke any
painful sensation in our patient, but merely evoked tactile sensations that were localized to coherent cutaneous provinces. In a pure motor bundle, intraneural activity from mechanoreceptor units, evoked by the stretching of the anterior tibialis muscle, and bursts of muscular sympathetic activity were recorded. The sympathetic skin response was absent. 4.2. Autonomic test ®ndings Cardiovascular re¯exes were normal compared to the results obtained from a cohort of age-matched controls. The intradermal injection of histamine on the volar forearm produced a wheal but no axon ¯are and failed to evoke itch sensation. Pilocarpine by iontophoresis failed to produce any sweating. 4.3. Morphological ®ndings 4.3.1. Sural nerve biopsy The morphometric study of the sural nerve showed a unimodal histogram with a marked reduction of the total number of myelinated ®bers and a complete loss of myelinated ®bers ,4 mm (Fig. 2). Electron microscopy revealed an almost complete lack of unmyelinated ®bers. 4.3.2. Skin biopsy Confocal study of skin biopsies demonstrated a complete loss of cutaneous nerve ®bers in all the skin specimens; only in the thigh were very rare nerve ®bers found in the dermis (Fig. 3C). These showed some varicosities that were not observed in control subjects (Fig. 3A,B). In the same site specimen only two hypotrophic and uninnervated sweat glands were identi®ed (Fig. 3D). 5. Discussion The typical clinical features of CIPA are well known (Goebel et al., 1980; Rafel et al., 1980; Rosemberg et al., 1994); only the rarity of this condition can lead to misdiagnosis. In our case, CIPA was diagnosed when the patient was 10 years old. She came to our observation after several hospitalizations in the orthopedic ward for fractures and osteomyelitis. The impairment of pain and temperature sensation and the absence of sweating suggested a small ®ber neuropathy. Because of the patient's mental retardation, it was dif®cult to perform reliable quantitative sensory testing and to assess the subject's ability to perceive cold, warm, and painful stimuli. However, our diagnosis was con®rmed by the absence of SSR and by the lack of sweating after pilocarpine by iontophoresis. Cardiovascular re¯exes were normal in our patient in agreement with previous ®ndings in CIPA (Hilz et al., 1999). In addition, microneurography revealed intraneural activity from A-beta sensory units using mechanical stimuli (brush and von Frey stimulators), but no skin C ®ber activity. Moreover, sensory
M. Nolano et al. / Clinical Neurophysiology 111 (2000) 1596±1601
potentials evoked along the median and ulnar nerves by supramaximal electrical shocks and mechanical stimulation of the ®ngertips of digits III and V were normal, a ®nding compatible with a normal density of Meissner's and Pacinian corpuscles (Caruso et al., 1993, 1994), and with a good state of preservation of large diameter nerve ®bers. In fact, the ultrastructural and morphometric study of the sural nerve showed a selective loss of unmyelinated and small myelinated ®bers (Goebel et al., 1980; Itoh et al., 1986). In agreement with the clinical, electrophysiological and neuropathological features, the immunohistochemical study of the skin revealed a complete absence of epidermal ®bers (all unmyelinated ®bers) and an almost complete absence of dermal ®bers to blood vessels and erector pilorum muscles (mostly unmyelinated ®bers and small myelinated ®bers). While in normal subjects sweat glands present very dense innervation (Kennedy et al., 1994), only uninnervated and markedly hypotrophic sweat glands were found in our patient (Fig. 3D). The pathophysiology of CIPA is still unknown although a recent genetic study of 3 subjects affected by CIPA demonstrated a mutation in the TRKA/NGF receptor gene (Indo et al., 1996). The only autopsy study available, which was performed on a 12-year-old boy who succumbed during a bout of hyperpyrexia (Swanson et al., 1965), showed an absence of Lissauer tract and loss of small myelinated ®bers in the dorsal roots and in the spinal tract of the trigeminal
1599
nerve, and loss of small ganglion cells of the dorsal spinal ganglia. In this case, a genetically determined defect in differentiation and migration of neural crest elements early in embryogenesis was postulated (Swanson et al., 1965). The absence of sympathetic skin activity against the well preserved cardiovascular re¯exes and muscle sympathetic activity could be explained by the presence in sympathetic ganglia of structurally and functionally distinct populations of cells and by their time-dependence in differentiation from the neural crest, which was observed in an animal model (Hamburger and Levi-Montalcini, 1949). The congenital lack of small neurons in the spinal cord reported by Swanson as well as the frequently described loss of small myelinated ®bers and unmyelinated ®bers in sural nerve suggests, in CIPA, a lack of innervation of the target organs, the skin and its annexes. Nonetheless, most conventional histologic studies failed to demonstrate skin and sweat gland alterations (Swanson, 1963; Brown and Podosin, 1966; Pinsky and Di George, 1966; Mitaka et al., 1985; Rosemberg et al., 1994). In the present study we obtained clear evidence of a lack of skin and sweat gland innervation by means of immunohistochemical techniques. Recently, nerve ®bers in epidermis (Wang et al., 1990; Kennedy and Wendelschafer-Crabb, 1993; Kennedy et al., 1996) and in dermis around vessels and sweat glands (Kennedy et al., 1994) have been extensively studied using the pan-axonal marker PGP 9.5 (Thompson and
Fig. 2. Distribution of the diameters of myelinated ®bers in the sural nerve of the patient (dark columns) compared to the mean values of a group of agematched controls (white columns). In the patient: endoneurial area, 0.5 mm 2; total number of myelinated ®bers, 1424 (.7 mm, 72%; .11 mm, 4.7%); total number of unmyelinated ®bers, 500. In normal subjects: endoneurial area, 0.6±1.0 mm 2; total number of myelinated ®bers, 6200±8500; density, 10 333± 14 166; number of unmyelinated ®bers, 50 000±60 000.
1600
M. Nolano et al. / Clinical Neurophysiology 111 (2000) 1596±1601
Fig. 3. Laser scanning confocal microscope images of skin from the thigh of a 14-year-old healthy subject (A,B) and of our patient (C,D) immunostained for PGP 9.5 with Cy-3 ¯uorophore (shown in green) and for collagen IV with Cy-5 ¯uorophore (shown in red). Note the severe loss of cutaneous nerve ®bers in the epidermis, dermis (C) and around the few hypotrophic sweat glands in the patient (D) compared with a normal distribution of dermal and epidermal ®bers (A) and normal-size and innervation of sweat glands (B). E, epidermis; ENFs, epidermal nerve ®bers; BM, basement membrane.
Day, 1988). A reduction of epidermal nerve ®ber density has been demonstrated in sensory neuropathy (McCarthy et al., 1995), diabetic neuropathy (Kennedy et al., 1996) and in small ®ber neuropathy (Holland et al., 1998; Mendell et al., 1998). The loss of epidermal nerve ®bers observed in diabetic neuropathy correlates with hypoalgesia to heat and mechanical stimuli (pinprick), and thermal threshold abnormalities in affected patients (Nolano et al., 1996). Topical application of capsaicin in humans produces a degeneration of epidermal ®bers with hypoalgesia to heat and mechanical stimulations (Nolano et al., 1999). These reports suggest a primary role of epidermal nerve ®bers in conveying pain sensation. Therefore, we assume that the lack of epidermal nerve ®bers and the absence of nerve ®bers around sweat glands are the morphological basis of analgesia and anhidrosis in CIPA. The results from our immunohistochemical study of the skin are in keeping with a report of a loss of SP ®bers in a patella biopsy from a patient with CIPA (Derwin et al., 1994), although SP ®bers are only a small amount of cutaneous ®bers. Moreover, this pathologic picture would account for the frequent bone fractures because the nociceptive ®bers may play a trophic role in the skeletal system (Hill and Elde, 1991).
In conclusion, this is the ®rst study that shows in CIPA a lack of unmyelinated and small myelinated ®bers in the skin, providing a morphological basis for analgesia and anhidrosis. Consequently, skin biopsy should be considered a primary diagnostic approach in patients with suspected CIPA, instead of the more invasive sural nerve biopsy. References Brown JW, Podosin R. A syndrome of the neural crest. Arch Neurol 1966;15:294±301. Caruso G, Nilsson J, Crisci C, Nolano M, Massini R, Lullo F. Sensory nerve ®ndings by tactile stimulation of median and ulnar nerves in healthy subjects of different ages. Electroenceph clin Neurophysiol 1993;89:392±398. Caruso G, Nolano M, Lullo F, Crisci C, Nilsson J, Massini R. Median nerve sensory responses evoked by tactile stimulation of the ®nger proximal and distal phalanx in normal subjects. Muscle Nerve 1994;17:269±275. Derwin KA, Glover RA, Wojtys EM. Nociceptive role of substance-P in the knee joint of a patient with congenital insensitivity to pain. J Pediatr Orthop 1994;14:258±262. Dyck PJ, Mellinger JF, Reagan TJ, Horowitz SJ, McDonald JW, Litchy WJ, Daube JR, Fealey RD, Go VL, Kao PC, Brimijoin WS, Lambert EH. Not indifference to pain but varieties of hereditary sensory and autonomic neuropathy. Brain 1983;106:373±390. Goebel HH, Veit S, Dyck PJ. Con®rmation of virtual unmyelinated ®ber
M. Nolano et al. / Clinical Neurophysiology 111 (2000) 1596±1601 absence in hereditary sensory neuropathy type IV. J Neuropathol Exp Neurol 1980;39:670±675. Hamburger V, Levi-Montalcini R. Proliferation, differentiation and degeneration in the spinal ganglion of the chick embryo under normal and experimental condition. J Exp Zool 1949;111:457±500. Hill EL, Elde R. Distribution of CGRP-, VIP-, DbH-, SP-, and NPY-immunoreactive nerves in the periosteoum of the rat. Cell Tissue Res 1991;264:469±480. Hilz MJ, Stemper B, Axelrod FB. Sympathetic skin response differentiates hereditary sensory autonomic neuropathies III and IV. Neurology 1999;52:1652±1657. Holland NR, Crawford TO, Hauer P, Cornblath DR, Grif®n JW, McArthur JC. Small-®ber sensory neuropathies: clinical course and neuropathology of idiopathic cases. Ann Neurol 1998;44:47±59. Indo Y, Tsuruta M, Hayashida Y, Karim MA, Ohta K, Kawano T, Mitsubuchi H, Tonoki H, Awaya Y, Matsuda I. Mutations in the TRKA/NGF receptor gene in patients with congenital insensitivity to pain with anhidrosis. Nat Genet 1996;13:485±488. Itoh Y, Yagishita S, Nakajima S, Nakano T, Kawada H. Congenital insensitivity to pain with anhidrosis: morphological and morphometric studies on the skin and peripheral nerves. Neuropediatrics 1986;17:103±110. Kennedy WR, Wendelschafer-Crabb G. The innervation of human epidermis. J Neurol Sci 1993;115:184±190. Kennedy WR, Sakuta M, Sutherland D, Goetz F. Quantitation of the sweating de®ciency in diabetes mellitus. Ann Neurol 1984;15:482±488. Kennedy WR, Wendelschafer-Crabb G, Brelje T. Innervation and vasculature of human sweat glands. An immunohistochemistry-laser scanning confocal ¯uorescence microscopy study. J Neurosci 1994;14:6825±6833. Kennedy WR, Wendelschafer-Crabb G, Johnson T. Quantitation of epidermal nerves in diabetic neuropathy. Neurology 1996;47:1042±1048. Langer J, Goebel HH, Veit S. Eccrine sweat glands are not innervated in hereditary sensory neuropathy type IV. An electron microscopic study. Acta Neuropathol (Berlin) 1981;54:199±202. McCarthy BG, Hsieh ST, Stocks MA, Hauer P, Macko C, Cornblath DR,
1601
Grif®n JW, McArthur JC. Cutaneous innervation in sensory neuropathies: evaluation by skin biopsy. Neurology 1995;45:1848±1855. Mendell JR, Periquet I, Kissel JT, et al. Distal painful axonal idiopathic neuropathy: criteria for diagnosis and distinction from other sensory neuropathies (abstract). Neurology 1998;50:343. Mitaka C, Tsunoda Y, Hikawa Y, Sakahira K, Matsumoto I. Anesthetic management of congenital insensitivity to pain with anhidrosis. Anesthesiology 1985;63:328±329. Nolano M, Wendelschafer-Crabb G, Kennedy WR. Correlation of sensation with epidermal innervation in diabetic neuropathy. Presented at the 43rd AAEM Annual Scienti®c Meeting, October 3±5, 1996, Minneapolis, MN. Nolano M, Simone DA, Wendelschafer-Crabb G, Kennedy WR. Topical capsaicin in humans: parallel loss of epidermal nerve ®bers and pain sensation. Pain 1999;81:135±145. Pinsky L, Di George AM. Congenital familial sensory neuropathy with anhidrosis. J Pediatr 1966;68:1±13. Rafel E, Alberca R, Bautista J, Navarrete M, Lazo J. Congenital insensitivity to pain with anhidrosis. Muscle Nerve 1980;3:216±220. Rosemberg S, Nagahashi Marie SK, Kliemann S. Congenital insensitivity to pain with anhidrosis (hereditary sensory and autonomic neuropathy type IV). Pediatr Neurol 1994;11:50±56. Swanson AG. Congenital insensitivity to pain with anhidrosis. A unique syndrome in two male siblings. Arch Neurol 1963;8:83±90. Swanson AG, Buchan GC, Alvord EC. Anatomical changes in congenital insensitivity to pain. Arch Neurol 1965;12:12±18. Thompson RJ, Day INH. Protein gene product (PGP) 9.5: a new neuronal and neuroendocrine marker. In: Marangos PJ, Campbell X, Cohen RM, editors. Neuronal and glial proteins: structure, function and clinical application, Neurobiological research, 2. Millbrae, CA: California Academic Press, 1988. pp. 209±228. Ê B, Hagbarth KE. Activity from skin mechanoreceptors recorded Valbo A percutaneously in awake human subjects. Exp Neurol 1968;21:270±289. Wang L, Hilliges M, Jernberg T, Wiegleb-EdstroÈm D, Johansson O. Protein gene product 9.5 ± immunoreactive nerve ®bres and cells in human skin. Cell Tissue Res 1990;261:25±33.