Pain 98 (2002) 331–334 www.elsevier.com/locate/pain Clinical note Dissociation between cutaneous and deep sensibility in central post-stroke pain (CPSP) Angela Mailis a,b,*, Gary J. Bennett c,1 a Comprehensive Pain Program, Toronto Western Hospital, Toronto, Ontario, Canada Toronto Western Research Institute, Toronto Western Hospital, Toronto, Ontario, Canada c Department of Anesthesia and Faculty of Dentistry, Anesthesia Research Unit, McGill University, Montreal, Quebec, Canada b Received 5 December 2001; accepted 29 April 2002 Abstract We present three cases of central post-stroke pain after right hemorrhagic or ischemic stroke associated with severe impairment of cutaneous sensibility but preservation of stimulus-evoked pain from periosteum. This is the first such report of dissociation of cutaneousfrom deep-tissue sensibility loss. q 2002 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. Keywords: Algometry; Central pain syndrome; Periosteum; Post-stroke pain syndrome; Pressure pain 1. Introduction Central post-stroke pain (CPSP) is a chronic pain syndrome that sometimes follows lesions in the spinothalamic pathway and its associated thalamo-cortical projections. CPSP patients generally have clear somatosensory deficits and a paradoxical ongoing (‘spontaneous’) pain that is felt within the area of deficit. In addition, cutaneous hyperalgesia, allodynia and/or hyperpathia are frequently, but not invariably, present (Boivie et al., 1989; Boivie and Leijon, 1991; Boivie, 1999). We present three cases of CPSP with ongoing pain and profound loss of cutaneous pain sensibility, but preservation of stimulus-evoked pain from deep tissues (periosteum). 2. Methods Patients were asked to indicate their pain regions on standardized body drawings. Cutaneous sensibility to touch, pinprick and cold were examined using a brush, a pinwheel and a cold roller, respectively, while vibration sense was tested by a 248 Hz tuning fork. These stimuli are well above the sensory detection threshold and are invariably sensed by normal subjects. * Corresponding author. Tel.: 11-416-603-5380; fax: 11-416-603-5725. E-mail addresses: angela.mailis@uhn.on.ca (A. Mailis), gary.bennett@mcgill.ca (G.J. Bennett). 1 Tel: 11-514-398-3432; fax: 11-514-398-8241. Pain pressure thresholds (PPTs) over bony prominences were measured by a pressure algometer (Pain Diagnostics and Therapeutics Inc., Great Neck, New York) with a rubber tip of 1 cm diameter; pressure was applied at a rate of 1 kg/s. Seven potential test sites in each upper extremity and eight in each lower extremity were predetermined (Figs. 1–3); not all sites were tested in all patients. The sum of all measured values from the symptomatic extremity divided by the sum of values from the asymptomatic extremity provides the pressure pain threshold ratio (PPTR). As expected, the PPTRs of normal subjects show no left–right differences. Thus, the normal PPTR is about 1.0 (mean 0.998; SD ¼ 4.3; N ¼ 20); Mailis et al., in preparation). 3. Results 3.1. Case report A A 68-year-old vasculopathic man suffered a right hemispheric stroke with left-sided hemiplegia as a complication of carotid endarterectomy 2 years prior to his appearance in our clinic for left-sided pain. His radiology report noted a right parietal lobe infarct in the distribution of the middle cerebral artery documented via computer-aided tomography (CAT) scan. The patient reported that his pain began within 2–3 days after the stroke and that it involved the left forearm/hand and leg/foot. It was intermittent, disappearing at night, and worsened by activity, movement, cold weather or 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(02)00156-2 332 A. Mailis, G.J. Bennett / Pain 98 (2002) 331–334 and clasp knife phenomenon in both left arm and leg, well preserved motor strength (at least 4/5), dystonic posturing of the upper extremity with extension of metacarpo-phalangeal joints and the small joints of the fingers, left homonymous hemianopia, and hyperactive left-sided reflexes. The Babinski sign was present on the left. He walked with the typical circumduction of the left leg while the arm was kept flexed at the elbow. There was definite discoloration of the left knee, while the left hand and foot were colder (compared to the uninvolved side) by 1.5 and 1.38C, respectively, as measured by a thermistor probe. There was dense hemianesthesia to pinprick, touch and cold in the left torso, arm, and leg. Pinprick in the left V2-3 facial distribution evoked a slight sensation of dull pressure. There was minor perception of joint movement in the left big toe, but not in the ankle, knee, or any joint in the upper extremity. Position sense was absent in both left arm and leg. Vibration sense was minimally present over the left acromio-clavicular joint and left forehead, but absent on the rest of the left side. Fig. 1. Case Report A: right parietal infarct, middle cerebral artery territory. Notice slightly increased pain pressure threshold ratios in the left symptomatic arm and slightly decreased ratios in the left leg, despite dense cutaneous left hemianesthesia. For this and the following figures, the symptomatic side is bracketed ([L]), and the averages for the locations tested in the upper and lower limbs are shown in parentheses. PPTRs: pressure pain threshold ratios. stress. On the McGill Pain Questionnaire, the patient described the pain as “pounding, stabbing, cramping, aching and sharp”. The patient indicated specifically that the pain was “deep“ and “not on the skin“ and denied touch intolerance in the past or during the examination. He also complained of numbness on the left face, arm, and leg. His wife reported occasional discoloration and coldness of the left hand and foot, but the patient did not perceive this, as he had no appreciation of temperature. During the interview, the pain was rated as 6/10 on a verbal numeric scale (0 ¼ no pain and 10 ¼ worst imaginable pain). The patient reported that the pain could go as high as 10/10 during activities. On examination, he presented with significant spasticity Fig. 2. Case Report B: right hemorrhagic stroke involving the basal ganglia and the corona radiata. Pain pressure thresholds ratios were only slightly elevated in both left leg and arm, despite significant loss of cutaneous sensibility. A. Mailis, G.J. Bennett / Pain 98 (2002) 331–334 333 patient reported that the pain in his left leg began within 2 months after his stroke. Examination revealed the following abnormalities on the left side: very severe arm and leg spasticity with fixed hand and foot equinovarus deformity; hyperactive reflexes in both extremities, no functional use of arm or leg; moderate to severe loss of sensibility to touch, pin prick and cold in the face, arm, trunk and leg; and loss of vibration sense in the arm and leg. He was unable to use the McGill Pain Questionnaire to characterize or rate the pain, but he indicated that it was constant, moderate to severe, fluctuating, and involved the whole left lower extremity from the groin to the foot. Despite his significant sensory loss, pain pressure thresholds were easily detectable. PPTRs were 1.10 in the upper extremity and 1.23 in the lower extremity (i.e. both symptomatic limbs demonstrated only slightly elevated pressure thresholds to deep pain) (Fig. 2). 3.3. Case report C Fig. 3. Case Report C: large right frontal and parietal region hemorrhage. Pain pressure threshold ratios were only slightly elevated in the left arm and slightly reduced in the left leg (i.e. the patient was actually more sensitive to pressure in the anesthetic lower extremity when compared to the normal contralateral limb) despite severe loss of cutaneous sensitivity. Pain pressure thresholds were easily detectable. PPTRs were 1.38 in the left upper extremity and 0.94 in the left lower extremity (i.e. the symptomatic arm demonstrated slightly elevated pressure thresholds, while the symptomatic leg demonstrated almost normal pain pressure thresholds) (Fig. 1). A 60-year-old hypertensive man suffered a right intracerebral hemorrhage 7 years prior to referral. The radiology report on his CAT scan taken at the time of admission to the hospital showed a large intracerebral hematoma involving the right frontal and parietal region, extending into the lateral ventricle, with blood in the fourth ventricle and a midline shift. A right frontoparietal craniotomy with evacuation of the hematoma was performed. Unfortunately the patient was left with dense left hemiplegia. The patient reported that he became aware of hypersensitivity and pain in both the left upper and lower extremity within 1–2 days after the stroke. The hypersensitivity seemed to decrease with time. The leg pain was worse than the arm pain, but both were described on the McGill Pain Questionnaire as severe, constant, “throbbing, shooting, sharp, cramping and aching”. During the interview, the pain was rated as 5–6/10. On examination, the patient had dense left hemiplegia associated with significant spasticity and hyperreflexia. His left hand was tightly clenched. On the entire left side, he displayed dense hemianesthesia to pinprick, cold, and light touch, while vibration sense was reduced in both affected arm and leg. Despite the severe loss of cutaneous sensitivity, pain pressure thresholds were only slightly elevated in the left arm (1.14) and slightly reduced in the left leg (0.90) (Fig. 3). 4. Discussion 3.2. Case report B A 67-year-old hypertensive man suffered a right hemispheric stroke with left-sided hemiplegia 3 years prior to referral to our clinic. His radiology report described a large area of hemorrhage involving the right basal ganglia and the corona radiata documented via CAT scan. The Painful levels of pressure applied over a bony prominence potentially activate the nociceptors that innervate the underlying skin, subcutaneous tissues, periosteum and bone. However, evidence indicates that it is largely, and probably entirely, activation of periosteal nociceptors that determine the pressure pain threshold over a bony prominence. 334 A. Mailis, G.J. Bennett / Pain 98 (2002) 331–334 Anesthetizing the skin with EMLA cream has been shown to have little or no effect on the PPT obtained over a bony prominence (Kosek et al., 1995, 1999). There is little reason to believe that activation of the nociceptors that innervate the subcutaneous tissues lying between the skin and periosteum, contribute to the PPTs. First, there is very little such tissue over a bony prominence. Second, topical application of EMLA cream reduces or eliminates pain to a depth of a millimeter or so (Bjerring and Arendt-Nielsen, 1990; Wahlgren and Quiding, 2000). Thus, the EMLA cream studies also indicate that stimulation of subcutaneous tissues immediately below the skin is unlikely to contribute significantly to the PPT. In animals, at least, nociceptive afferents penetrate deeply into bone via the Haversian canals (Mach et al., 2000). However, it seems highly unlikely that the pressures used to obtain the PPTs are great enough to influence the environment within the bone. Human periosteum is richly innervated by unmyelinated peptidergic sensory afferents (Gronblad et al., 1984; Wotjys et al., 1990). Landau and Bishop (1953) showed that stimulation of their own periosteal innervation (selectively activated with electric shock delivered via a subcutaneous needle electrode, or activated directly via mechanical stimulation) evoked a pain sensation like that produced by pressure over bone. In summary, it seems nearly certain that the PPTs obtained over a bony prominence are determined by the activation of periosteal nociceptors. The three patients described here were typical cases of the CPSP syndrome. All had pain localized within the region of hypoesthesia. All had very severe deficits in their ability to detect pain (pinwheel stimulation) and cold, as well as clear deficits in touch and proprioception. Deficits in cutaneous pain and thermal sensibility are nearly universal in CPSP patients; deficits in haptic and proprioceptive sensibility are variable (Boivie et al., 1989; Boivie and Leijon, 1991; Boivie, 1999). The three patients presented here are remarkable for having dense hyposensitivity to cutaneous pain, but wellpreserved sensitivity to pain evoked from periosteum. To our knowledge, this is the first report of a dissociated loss of cutaneous- vs. deep-tissue pain sensibility in CPSP patients. Hypoesthesia is not uncommon in stroke patients who do not develop the CPSP syndrome. We are unaware of any reports of a dissociated loss of cutaneous- vs. deep-tissue pain sensibility in such pain-free stroke patients. We have no hypothesis to explain the dissociated loss of cutaneous- vs. deep-tissue pain sensibility described here. The spinothalamic tract has long been considered to be the most important projection pathway for pain from both cutaneous and deep tissues. This belief is founded on the profound analgesia that develops below and contralateral to a lesion of the anterolateral quadrant of the spinal cord. It is now known that information about pain may ascend via several pathways in the spinal white matter. For example, there is particularly convincing evidence for an additional pathway for visceral pain traveling in the midline dorsal column of the human spinal cord (Willis et al., 1999; Nauta et al., 2000). However, we have no evidence showing that input from periosteal nociceptors has a special pathway to the brain. Input from periosteal nociceptors onto spinal cord neurons has been little studied. Evidence from cats indicates that inputs from nociceptors innervating vertebral periosteum converge on the same spinal cord neurons that receive input from nearby skin and muscle (Gillette et al., 1993). We are aware of only one report suggesting that there might be separate thalamic or cortical representations for cutaneous- vs. deep-tissue pain sensibility. A recent fMRI study comparing activation patterns of pain evoked by noxious heat applied to the skin over the sternum versus pain evoked by esophageal distension revealed very similar patterns, except for the location of activity within the insular cortex (Strigo et al., 2001). References Bjerring P, Arendt-Nielsen L. Depth and duration of skin analgesia to needle insertion after topical application of EMLA cream. Br J Anaesth 1990;64:173–177. Boivie J. Central pain. In: Wall PD, Melzack R, editors. Textbook of pain, 4th ed.. Edinburgh: Churchill Livingstone, 1999. pp. 879–914. Boivie J, Leijon G. Clinical findings in patients with central post-stroke pain. In: Casey KL, editor. Pain and central nervous system disease; the central pain syndromes, New York, NY: Raven Press, 1991. pp. 65–75. Boivie J, Leijon G, Johansson I. Central post-stroke pain – a study of the mechanisms through analyses of the sensory abnormalities. Pain 1989;37:173–185. Gillette RG, Kramis RC, Roberts WJ. Characterization of spinal somatosensory neurons having receptive fields in lumbar tissues of cats. Pain 1993;54:85–98. Gronblad M, Liesi P, Korkala O, Karaharju E, Polak J. Innervation of human bone periosteum by peptidergic nerves. Anat Rec 1984;209:297–299. Kosek E, Ekholm J, Hansson P. Increased pressure pain sensibility in fibromyalgia patients is located deep to the skin but not restricted to muscle tissue. Pain 1995;63:335–339. Kosek E, Ekholm J, Hansson P. Pressure pain thresholds in different tissues in one body region. The influence of skin sensitivity in pressure algometry. Scand J Rehabil Med 1999;31:89–93. Landau W, Bishop GG. Pain from dermal, periosteal and fascial endings and inflammation. Arch Neurol Psychiatry 1954;69:490–504. Mach DB, Rogers SD, Kotz CM, Clohisy DR, Mantyh PW. Sensory nerves that innervate bone are involved in conveying skeletal pain. Soc Neurosci Abstr 2000;26:1960. Nauta HJ, Soukup VM, Fabian RH, Lin JT, Grady JJ, Williams CG, Campbell GA, Westlund KN, Willis WD. Punctate midline myelotomy for the relief of visceral cancer pain. J Neurosurg 2000;92(2 suppl):125–130. Strigo IA, Bushnell CB, Boivin M, Duncan GH. Cerebral activity during painful stimulation of skin and viscera. Soc Neurosci Abstr 2001;27:Program No. 507.6. Wahlgren CF, Quiding H. Depth of cutaneous analgesia after application of a eutectic mixture of the local anesthetics lidocaine and prilocaine (EMLA cream). J Am Acad Dermatol 2000;42:584–588. Willis WD, Al-Chaer ED, Quast MJ, Westlund KN. A visceral pain pathway in the dorsal column of the spinal cord.. Proc Natl Acad Sci USA 1999;96:7675–7679. Wotjys EM, Beaman DN, Glover RA, Janda D. Innervation of the human knee joint by substance-P fibers. Arthroscopy 1990;6:254–263.