Children with comorbid attention-deficit-hyperactivity disorder and tic disorder Evidence for additive inhibitory deficits within the motor system.код для вставкиСкачать
BRIEF COMMUNICATIONS Children With Comorbid Attention-DeficitHyperactivity Disorder and Tic Disorder: Evidence for Additive Inhibitory Deficits Within the Motor System Gunther H. Moll, MD,1 Hartmut Heinrich, PhD,1 Götz-Erik Trott, MD,2 Sigrun Wirth, MD,2 Nathalie Bock, MD,1 and Aribert Rothenberger, MD1 For children with attention-deficit-hyperactivity disorder (ADHD) or tic disorder (TD), we recently reported deficient inhibitory mechanisms within the motor system by using transcranial magnetic stimulation. These deficits— stated as reduced intracortical inhibition in ADHD and shortened cortical silent period in TD—could be seen as neurophysiological correlates of motor hyperactivity and tics, respectively. To investigate neurophysiological aspects of comorbidity, we measured motor system excitability for the first time also in children with combined ADHD and TD. The findings of a reduced intracortical inhibition as well as a shortened cortical silent period in these comorbid children provide evidence for additive effects at the level of motor system excitability. Ann Neurol 2001;49:393–396 In attention-deficit-hyperactivity disorder (ADHD), motoric hyperactivity is usually the striking abnormality to parents and physicians, typically seen in restlessness, fidgeting, and generally unnecessary gross body movements.1 In tic disorder (TD), fluctuating motor and phonic tics can be characterized as involuntary, sudden, abrupt, repetitive movements, gestures, or utterances that may be seen as poorly modulated fragments of normal sensorimotor behavior.2 As a possible neurophysiological correlate of these hypermotoric symptoms in children with ADHD or TD, we recently tested the hypothesis of deficient inhibitory motor control throughout the sensorimotor circuit by investigating motor system excitability using the technique of focal transcranial magnetic stimulation (TMS) in a single- and paired-stimulus paradigm.3,4 Whereas children with ADHD showed a reduced intracortical inhibition,5 children with TD had a shortened cortical silent period compared to healthy children.6 Because these two hypermotoric disorders coexist in up to 50% in clinical samples,7 we investigated neurophysiological correlates of motor system excitability in a comorbid group (ADHD ⫹ TD) relative to the two single-disorder groups (ADHD only or TD only) to test whether a distinct dysfunctional pattern for this common comorbidity exists. Patients and Methods Patients The study was performed on 16 children with ADHD (ADHD only), 16 children with TD (chronic motor tic disorder/Tourette’s disorder; TD only), 16 children with comorbid ADHD and TD (ADHD ⫹ TD), and 16 healthy children (controls). The demographic and clinical data are summarized in Table 1. All children (60 boys, 4 girls) were right-handed and of at least normal intelligence (IQ ⬎ 80). The four groups did not differ significantly in mean age and sex. All disordered children had to fulfill the diagnostic criteria for ADHD and/or chronic motor tic disorder/Tourette’s disorder according to the APA’s DSM-IV. None of the ADHD only, TD only, or ADHD ⫹ TD children fulfilled the criteria for any other psychiatric disorder as stated in a structured clinical interview (DISC). In all of the children with an ADHD diagnosis, the score of the abbreviated (10 item) Conners scale was ⱖ15.8 In addition, none of the ADHD only children had any actual tic symptoms nor any history of tics. For the TD-only group, a Conners score ⱖ15 was used as the exclusionary criterion. Children with an ADHD diagnosis either had never taken any psychostimulant medication or had been drug-free for at least 48 hours before TMS investigation (the only drug used was methylphenidate in a standard formulation). Some of the children with a TD diagnosis received a neuroleptic medication (Table 1). The healthy children were drug-free and devoid of any child psychiatric disorder. The subjects of all four groups lacked gross neurological or other organic disorders. After a complete description of the study to the children and their parents, assent was obtained from the children and written informed consent from their parents. The study was conducted according to the declaration of Helsinki and was approved by the local ethics committee. TMS From the 1Child and Adolescent Psychiatry Department, University of Göttingen, Göttingen; and 2Practice of Child and Adolescent Psychiatry, Aschaffenburg, Germany Received Sep 1, 2000, and in revised form Oct 13. Accepted for publication Oct 18, 2000. Address correspondence to Prof Rothenberger, von-Siebold-Str. 5, D-37075 Göttingen, Germany. E-mail: email@example.com Focal TMS was applied to the hand area of the left motor cortex (figure-eight magnetic coil with a diameter of one wing of 70 mm, Magstim 200 HP magnetic stimulator; Magstim, Whiteland, United Kingdom), and surface electromyography was recorded from the right abductor digiti minimi muscle. The exact equipment and protocol are reported in detail by Ziemann et al.4 Resting and active motor threshold were expressed as a percentage of the maximum stimulator output to elicit a mo- © 2001 Wiley-Liss, Inc. 393 Table 1. Demographic and Clinical Data Age (yr; mo) Mean (⫾SD) Range Sex Conners (10 items) Mean (⫾SD) Range TD medication Controls (n ⫽ 16) ADHD only (n ⫽ 16) TD only (n ⫽ 16) ADHD ⫹ TD (n ⫽ 16) 12;3 (⫾1;8) 9;5–15;5 14 Boys, 2 Girls 12;0 (⫾1;6) 9;7–14;8 15 Boys, 1 Girl 12;8 (⫾1;10) 9;0–15;0 15 Boys, 1 Girl 12;5 (⫾2;1) 9;6–15;11 16 Boys 4.3 (⫾3.1) 1–12 — 20.5 (⫾3.4) 15–28 — 6.3 (⫾4.1) 0–14 Tiapride (n ⫽ 5) 18.9 (⫾5.1) 15–30 Tiapride (n ⫽ 6), tiapride/pimozide (n ⫽ 1) tor evoked potential. The duration of the cortical silent period (in milliseconds) was investigated at a stimulus intensity of 40% above active motor threshold (single magnetic stimulation). Intracortical excitability was determined in a conditioning test paired-stimulus paradigm and measured as changes in the amplitude of the suprathreshold test response by the conditioning subthreshold shock (expressed as a percentage of the unconditioned mean across all inhibitory (2–5 msec) and facilitatory (7–20 msec) interstimulus intervals tested). Because activation of the relaxed target muscle may reduce intracortical inhibition and intracortical facilitation,9 for artefact control, all sweeps were visually inspected offline. Sweeps showing voluntary activation of the target muscle were rejected and excluded from further processing. Statistical Analysis Unique TMS variables were subjected to a 2 ⫻ 2 factorial analysis of variance with ADHD (with ADHD vs. without ADHD) and TD (with TD vs. without TD) as factors using SPSS for Windows v.10. Possible significant interactions were planned to be examined with post-hoc Tukey tests. For all statistical procedures, significance was assumed at p ⬍ 0.05. Results No significant effects were obtained for passive and active motor thresholds. Concerning the cortical silent period, a main effect of TD was found, indicating that children with a TD diagnosis (TD only or ADHD ⫹ TD) had a shorter cortical silent period than children without TD (controls or ADHD only; Table 2 and Fig). Post-hoc analysis (t test) revealed no significant cortical silent period difference between medicated (139.7 ⫾ 37.8 msec) and unmedicated (134.4 ⫾ 44.3 msec) children with TD [t(1,30) ⫽ 0.36, n.s.]. Analysis of intracortical inhibition revealed a main effect of ADHD; i.e., children with an ADHD diagnosis (ADHD only or ADHD ⫹ TD) had a reduced intracortical inhibition compared to children without ADHD (controls or TD only; see Table 2 and Fig). For intracortical facilitation, no significant effects were obtained. Discussion Morphometric and neuroimaging studies suggest that ADHD and TD as well involve neuropathology of the sensorimotor circuit, responsible for control of sensorimotor behavior.10 To study aspects of comorbidity of these two hypermotoric disorders (i.e., additive vs. nonadditive effects), we applied TMS to assess inhibitory mechanisms within the motor system in children with ADHD, children with TD, and children with comorbid ADHD ⫹ TD in comparison to healthy children by means of a 2 ⫻ 2 factorial design. The finding Table 2. Summary of TMS Results Resting motor threshold (%) Active motor threshold (%) Cortical silent period (msec) Intracortical inhibition (%) Intracortical facilitation (%) 394 F Values and Significance Controls (n ⫽ 16) Mean (⫾SD) ADHD only (n ⫽ 16) Mean (⫾SD) TD only (n ⫽ 16) Mean (⫾SD) ADHD ⫹ TD (n ⫽ 16) Mean (⫾SD) Main Effect ADHD Main Effect TD Interaction ADHD ⫹ TD 57.7 (⫾13.1) 35.0 (⫾6.4) 163.3 (⫾26.7) 69.8 (⫾20.6) 135.5 (⫾24.8) 56.7 (⫾11.3) 33.4 (⫾3.7) 172.1 (⫾25.3) 87.4 (⫾19.4) 124.6 (⫾17.8) 54.9 (⫾8.6) 35.9 (⫾8.1) 138.0 (⫾40.9) 70.7 (⫾26.2) 133.9 (⫾29.3) 53.7 (⫾10.6) 33.1 (⫾5.5) 137.3 (⫾39.8) 88.1 (⫾16.5) 125.3 (⫾18.9) F(1,63) ⫽ 0.16, n.s. F(1,63) ⫽ 2.04, n.s. F(1,63) ⫽ 0.23, n.s. F(1,63) ⫽ 11.1, p ⫽ 0.002 F(1,63) ⫽ 2.82, n.s. F(1,63) ⫽ 1.11, n.s. F(1,63) ⫽ 0.04, n.s. F(1,63) ⫽ 11.8, p ⫽ 0.001 F(1,63) ⫽ 0.02, n.s. F(1,63) ⫽ 0.01, n.s. F(1,63) n.s. F(1,63) n.s. F(1,63) n.s. F(1,63) n.s. F(1,63) n.s. Annals of Neurology Vol 49 No 3 March 2001 ⫽ 0.00, ⫽ 0.17, ⫽ 0.31, ⫽ 0.00, ⫽ 0.04, Fig. Main effects of tic disorder (TD) and attention-deficithyperactivity disorder (ADHD) on cortical silent period (top) and intracortical inhibition (bottom). For each group, mean value ⫾ standard error is depicted. Open boxes represent children without tic disorder; solid boxes represent children with tic disorder. of normal motor thresholds in all psychiatric groups investigated gives no evidence for hyperexcitability at the neuronal membrane level of cortical neurons within the motor system11 for children with ADHD, children with TD, or comorbid children. Concerning the cortical silent period, children with a TD diagnosis had a shorter cortical silent period than children without TD (see Table 2 and Fig). This finding confirms our previously reported TMS data for children with TD, in which a significantly shorter cortical silent period was found in children with TD compared to healthy children. This result was seen as a neurophysiological correlate of deficient motor inhibition within the sensorimotor circuit in children with TD, probably at the level of the basal ganglia, and might be at least partly due to dopaminergic hyperinnervation of the striatum or a supersensitivity of striatal D2 recep- tors.2,6 The normal duration of the cortical silent period in children with an ADHD only diagnosis does not support the hypothesis of generally deficient inhibitory mechanisms within their sensorimotor circuit.5 Concerning intracortical inhibition, children with an ADHD diagnosis had a reduced intracortical inhibition compared to children without ADHD (see Table 2 and Fig). This result also confirms our previously reported TMS data for children with ADHD only, in which a significantly decreased intracortical inhibition was found, providing evidence for inhibitory deficits in ADHD within the motor cortex.5 Owing to the method used, this inhibitory deficit within the motor cortex might be only part of a general deficit in behavioral inhibition.12 Whether these discriminative TMS findings on motor system excitability in children with ADHD or TD might be related to their different neuromotor functioning at the behavioral level, eg, the time to complete motor movements,13 remains to be tested. In children with combined ADHD ⫹ TD, a significantly reduced intracortical inhibition as well as a significantly shortened cortical silent period were found, suggesting an additive effect concerning motor system excitability. This may be in line with neuropsychological and behavioral data, especially on neuromotor functioning, indicating the most severe impairment in children with ADHD ⫹ TD.13,14 On the other hand, neuropathological data at the level of the basal ganglia15 and neurophysiological parameters reflecting task-dependent cognitive information processing do not fully support the notion of additive deficits and dysfunctions, respectively.16 –19 Thus, it might be assumed that purely additive neurobiological effects in ADHD ⫹ TD comorbidity are restricted to abnormalities within the motor system (as measured by TMS in this study), but this may not hold true for all aspects of cognitive information processing. In conclusion, a distinct dysfunctional pattern of deficient inhibitory motor control can be postulated as a neurobiological correlate of hypermotoric symptoms in children with ADHD, TD, or ADHD ⫹ TD. These findings should be verified with different drug treatments for both kinds of inhibitory deficits on motor control, i.e., methylphenidate in ADHD and probably neuroleptics in TD.20 The authors thank Susanne Mock and Eva Nobbe. References 1. Barkley RA. Attention-deficit-hyperactivity disorder, 2nd edition. New York: Guilford, 1998. 2. Leckman JF, Peterson BS, Anderson GM, et al. Pathogenesis of Tourette’s Syndrome. J Child Psychol Psychiatr 1997;38:119 – 142. Brief Communication: Moll et al: Motor System Excitability in ADHD and Tic Disorder 395 3. Kujirai T, Caramia MD, Rothwell JC, et al. Corticocortical inhibition in human mortor cortex. J Physiol (Lond) 1993;471: 501–519. 4. Ziemann U, Paulus W, Rothenberger A. Decreased motor inhibition in Tourette syndrome: evidence from transcranial magnetic stimulation. Am J Psychiatr 1997;154:1277–1284. 5. Moll GH, Heinrich H, Trott GE, et al. Deficient intracortical inhibition in drug-naive children with attention-deficit hyperactivity disorder is enhanced by methylphenidate. Neurosci Lett 2000;284:121–125. 6. Moll GH, Wischer S, Heinrich H, et al. Deficient motor control in children with tic disorder: evidence from transcranial magnetic stimulation. Neurosci Lett 1999;272:37– 40. 7. Spencer T, Biederman J, Wilens T. Attention-deficit/hyperactivity disorder and comorbidity. Pediatr Clin North Am 1999;46:915– 127. 8. Goyette CH, Conners CK, Ulrich RF. Normative data on revised Conners parent and teacher rating scales. J Abnorm Child Psychol 1978;6:221–236. 9. Ridding MC, Taylor JL, Rothwell JC. The effect of voluntary contraction on cortico-cortical inhibition in human motor cortex. J Physiol (Lond) 1995;487:541–548. 10. Sheppard DM, Bradshaw JL, Purcell R, Pantelis C. Tourette’s and comorbid syndromes: obsessive compulsive and attention deficit hyperactivity disorder. A common etiology? Clin Psychol Rev 1999;19:531–552. 11. Ziemann U, Rothwell JC, Ridding MC. Interaction between intracortical inhibition and facilitation in human motor cortex. J Physiol (Lond) 1996;496:873– 881. 12. Barkley RA. Behavioral inhibition, sustained attention, and executive functions: constructing a unifying theory of ADHD. Psychol Bull 1997;121:65–94. 13. Schuerholz LJ, Cutting L, Mazzocco MM, et al. Neuromotor functioning in children with Tourette syndrome with and without attention deficit hyperactivity disorder. J Child Neurol 1997;12:438 – 442. 14. Harris EL, Schuerholz LJ, Singer HS, et al. Executive function in children with Tourette syndrome and/or attention deficit hyperactivity disorder. J Int Neuropsychol Soc 1995;1:511–516. 15. Castellanos FX, Giedd JN, Hamburger SD, et al. Brain morphometry in Tourette’s syndrome: the influence of comorbid attention-deficit/hyperactivity disorder. Neurology 1996;47: 1581–1583. 16. Rothenberger A, Banaschewski T, Heinrich H, et al. Comorbidity in ADHD-children: effects of coexisting conduct disorder or tic disorder on event-related brain potentials in an auditory selective-attention task. Eur Arch Psychiatr Clin Neurosci 2000;250:101–110. 17. Schuerholz LJ, Baumgardner TL, Singer HS, et al. Neuropsychological status of children with Tourette’s syndrome with and without attention deficit hyperactivity disorder. Neurology 1996;46:958 –965. 18. Yordanova J, Dumais-Huber C, Rothenberger A. Coexistence of tics and hyperactivity in children: no additive effect at the psychophysiological level. Int J Psychophysiol 1996;21:121–133. 19. Yordanova J, Dumais-Huber C, Rothenberger A, Woerner W. Frontocortical activity in children with comorbidity of tic disorder and attention-deficit hyperactivity disorder. Biol Psychiatr 1997;41:585–594. 20. Gadow KD, Sverd J, Sprafkin J, et al. Long-term methylphenidate therapy in children with comorbid attention-deficit hyperactivity disorder and chronic multiple tic disorder. Arch Gen Psychiatr 1999;56:330 –336. 396 Annals of Neurology Vol 49 No 3 March 2001 Neurological and Neuropathologic Heterogeneity in Two Brothers with Cobalamin C Deficiency James M. Powers, MD,1 David S. Rosenblatt, MD,2 Robert E. Schmidt, MD, PhD,3 Anne H. Cross, MD,4 Joseph T. Black, MD,4 Ann B. Moser, BS,5 Hugo W. Moser, MD,5 and Daniel J. Morgan, BS6 Two adult brothers, one documented to have methylmalonic acidemia with homocystinuria, or cobalamin C deficiency, after autopsy, displayed severe but divergent neurological presentations. One exhibited a myelopathy and the other chronic endocrine problems (Schmidt’s syndrome) followed by a neuropsychiatric and dementing disorder owing to cerebral perivascular demyelination. The recognition of cobalamin C deficiency has practical implications because it is one of the few inherited diseases of central white matter that is treatable. Ann Neurol 2001;49:396 – 400 Neurological deficits due to vitamin B12, or cobalamin (Cbl), deficiency are highly variable and have been seen most commonly in the setting of pernicious anemia, in which a myelopathy and peripheral neuropathy predominate. Neuropsychiatric symptoms are seldom mentioned, and brain involvement is said to occur only after the spinal cord lesion.1 Likewise, neuropathology textbooks emphasize the myelopathy almost to the exclusion of other central nervous system (CNS) lesions.2 There also are rare inherited disorders of transport or intracellular processing that result in Cbl deficiency;3 the latter include methylmalonic acidemia with homocystinuria (cblC; McKusick 277400), in which there is an early intracellular defect leading to a failure of synthesis of both Cbl cofactors, adenosylcobalamin From the 1University of Rochester Medical Center, Departments of Pathology and Neurology, Rochester, NY; 2McGill University Health Centre, Division of Medical Genetics, Montreal, Canada; 3 Departments of Pathology and Immunology and 4Washington University School of Medicine, Departments of Neurology and Neurosurgery, St. Louis, MO; 5Johns Hopkins Medical Institutes, Kennedy-Krieger Institute, Baltimore, MD; and 6University of Rochester Medical Center, Department of Pathology, Rochester, NY. Received Aug 18, 2000, and in revised form Oct 23. Accepted for publication Oct 25, 2000. Address correspondence to Dr Powers, University of Rochester Medical Center, Department of Pathology, 601 Elmwood Avenue, Rochester, NY 14642. E-mail: firstname.lastname@example.org (AdoCbl) and methylcobalamin (MeCbl). Two adult brothers, one documented to have cblC after his death, demonstrated a striking neurological and neuropathologic heterogeneity. Neither displayed subnormal vitamin B12 levels or the expected hematological abnormalities. This report broadens the spectrum of cblC and reminds us that neuropsychiatric deficits due to cerebral white matter lesions can dominate in diseases of Cbl deficiency. Patients and Methods Patient 1 In good health until age 32, Patient 1 presented with numbness of his extremities. Within 3 months, leg weakness, incontinence, and Lhermitte’s sign developed. Five months after presentation, neurological examination revealed bitemporal disk pallor; weakness and diminished sensation to temperature, pinprick, touch, and vibration of lower extremities; flexor plantar responses; and waddling gait. Analysis of the cerebrospinal fluid (CSF) revealed a protein level of 91 mg/100 ml, IgG elevation, IgG synthesis of 7.0 mg/24 hour, no oligoclonal bands, and negative results on the VDRL test. Human immunodeficiency virus (HIV) test results were unavailable. Mean corpuscular volume, erythrocyte sedimentation rate (ESR), and results of antinuclear antibodies (ANAs) studies, serum protein electrophoresis, very long–chain fatty acids studies, and thyroid tests were normal. Mild leukopenia and a slight excess of hypersegmented polymorphonuclear cells (retrospectively) developed only during his final admission. The vitamin B12 level was 415 mg/dl (normal 100 –700). Nerve conduction in the posterior tibial and sural nerves was mildly slowed. Brainstem auditory evoked responses were normal. Visual evoked responses were 98 m/sec oculus dexter and 105 m/sec oculus sinister. Magnetic resonance imaging (MRI) revealed a confluent periventricular white matter process about both trigone regions and the posterior body of the lateral ventricles. Five months after presentation, prednisone was administered without benefit. By 6 months after onset, the patient displayed a scissoring gait, by 9 months he needed a wheelchair, and by 16 months he was quadriparetic. Two years after onset he had dysphagia; paraplegia; upper extremity weakness; diminished light touch, proprioception, and vibratory sense below the neck; absent pinprick response below the waist; and urinary retention. The patient developed multiple lung and urinary tract infections, expiring 26 months after onset. There were “no obvious cognitive problems” at time of death. but had no recall at 5 minutes, was unable to spell, and had difficulty initiating speech. There was right optic disk pallor. Motor function, sensation, and coordination were normal. CSF analysis revealed a protein level of 61 mg/100 ml, a normal CSF IgG index, and no oligoclonal bands. The thyroxine level was elevated (18 mg/dl), and the cortisone level was normal (9.9 mg/dl). The ESR and levels of ceruloplasmin, vitamin E, vitamin B12, ANAs, arylsulfatase A, and very long–chain fatty acids were normal. Folate levels were not determined. HIV and VDRL test results were negative. An electroencephalogram revealed 3.5-Hz background. Proton density and T2-weighted MRI demonstrated periventricular nonenhancing increased signal abnormalities that were confluent in the occipital lobes and patchy in frontal white matter. T1-weighted images were normal (Fig 1). Brain biopsy revealed some perivascular demyelinative foci and a few perivascular lymphocytes without specific diagnostic features. Within 1 year, he developed dysarthria, difficulty walking, and inability to feed himself. At that time he had flexor plantar reflexes and rigidity. Deep venous thromboses with pulmonary embolism caused his death at age 45. Family History The brothers were of Italian-French-Irish descent without evidence of consanguinity. Their father had Grave’s disease, their mother hypothyroidism. A younger brother, age 39, had insulin-dependent diabetes mellitus but no neurological symptoms. The younger brother had normal serum vitamin Fig 1. Magnetic resonance imaging study of Patient 2. This T2-weighted (2400/90) axial image at the level of the atria shows abnormal increased signal in the periventricular occipital white matter bilaterally. Patient 2 The older brother of Patient 1, Patient 2 was a college professor with hypoadrenalism and hypothyroidism, diagnosed as Schmidt’s syndrome, since age 16. Medications included cortisol, synthroid, and beclomethasone. At 44, he developed social withdrawal and difficulty doing calculations. Over 4 months he deteriorated such that he was unable to interact normally and became lost going to familiar locations. Results of a computed tomography scan of the head at that time were normal. Seven months after presentation he was alert Brief Communication: Powers et al: cblC Deficiency and PGA-II 397 B12, folate, methylmalonic acid, methylcitric acid, homocysteine, and cystathionine levels and normal urinary methylmalonic acid and homocysteine levels. Methods In Patient 1 the postmortem examination was confined to the CNS. Sections from CNS were stained with Luxol fast blue–periodic acid–Schiff (LFB-PAS) and modified Bielschowsky techniques. In addition, plastic-embedded sections of sural nerve from Patient 2 were examined. In view of the neuropathologic findings suggestive of a methylation defect and because of the family history, cultured fibroblasts from Patient 2 were analyzed for a possible inherited metabolic disorder.3 Results The sural nerve of Patient 2 exhibited a loss of large and, especially, small myelinated axons with ongoing axonal degeneration. A dorsal root ganglion of Patient 2 contained multifocal clusters of lymphocytes, as did the testis. There were no distinctive cytosomes, demyelination, “onion bulbs,” or inflammation. In addition to pulmonary emboli, systemic findings in Patient 2 consisted of (1) no demonstrable adrenal tissues at autopsy (presumably severe primary atrophy), (2) Hashimoto’s thyroiditis, (3) testicular atrophy, Sertoli cell– only phenotype, and (4) severe macrovesicular hepatic Fig 2. Patient 2. Perivascular demyelination of white matter with periodic acid–Schiff–positive macrophages (arrowhead) can be seen. Luxol fast blue–periodic acid–Schiff, ⫻125. Fig 3. Patient 1. The thoracic cord is atrophic, with loss of myelinated fibers, particularly in the posterior and posterolateral funiculi, and vacuolated myelin adjacent to the anterior horns. Modified Bielschowsky’s stain, ⫻4.25 before 41% reduction. steatosis. The gastric mucosa was unremarkable on gross examination. Microscopic examination was omitted. The CNS of both brothers was unremarkable on gross examination except for the spinal cord of Patient 1, in which there were ill-defined gelatinous areas. The cerebrum in both cases demonstrated perivascular demyelination (mainly of arterioles and venules) with occasional vacuoles, oligodendrocytic loss, PAS- and LFB-positive macrophages, reactive astrocytes, and relative sparing of axons with occasional axonal spheroids (Fig 2). Patient 1 also displayed a severe vacuolar myelopathy typical of subacute combined degeneration seen in Cbl deficiency,4 resulting in atrophy of the thoracic cord owing to marked and asymmetric loss of myelinated fibers in the posterior, lateral, and anterior columns (Fig 3). Fibroblast analysis revealed impaired synthesis of both AdoCbl and MeCbl (Tables 1 and 2). Propionate fixation was not increased when fibroblasts from Patient 2 were fused with two cblC cell lines (Table 3), establishing the diagnosis of cblC. Discussion Subacute combined degeneration of the spinal cord and its accompanying clinical myelopathy is dominant in Cbl deficiency of pernicious anemia, and neuropsychiatric symptoms and supraspinal lesions are seldom publicized.1–5 However, the classic study of pernicious anemia by Adams and Kubik describes and illustrates demyelinative lesions in deep cerebral white matter, which were responsible for neuropsychiatric manifestations.6 In addition to a deficiency in Castle’s intrinsic factor needed for the absorption of the external factor, Cbl, 398 Annals of Neurology Vol 49 No 3 March 2001 Table 1. Propionate and Methyl-THF Uptake [14C] Propionate (nmol/mg protein/18 h) [14C] Methyl-THF (pmol/mg protein/ 18 h) Cell Line ⫺OHCbl ⫹OHCbl ⫺OHCbl ⫹OHCbl Patient 2 Control subjects 0.76, 0.46a 10.8 ⫾ 3.7b 3.7, 3.1 10.9 ⫾ 3.5 59, 49 225 ⫾ 165 201, 185 305 ⫾ 125 a Mean of triplicate values for two separate experiments. Based on 12 determinations (3 different control subjects). OHCbl ⫽ hydroxycobalamin; THF ⫽ tetrahydrofolate. b Table 2. Cobalamin Uptake and Distribution Cbla Distribution (%) Cell Line Cbla Uptake (pg/106 cells) AqCbl CNCbl AdoCbl MeCbl Other Patient 2 Control subjects 2.7 4.60b ⫾ 2.01 9.6 8.4 ⫾ 3.9 69.8 11.3 ⫾ 6.9 4.5 15.3 ⫾ 4.2 6.7 58 ⫾ 6.7 9.4 7.2 ⫾ 2.7 a 57 [ Co] CNCbl 25 pg/ml, 4 days incubation. Based on 12 determinations (3 different control subjects). AqCbl ⫽ aquacobalamin; AdoCbl ⫽ adenosylcobalamin; Cbl ⫽ cobalamin; amin; MeCbl ⫽ methylcobalamin. b Table 3. Complementation Analysis Propionatea Uptake (nmol/mg protein/18 h) Fusion with Cells from Patient 2 Cell Line Self-Fusion ⫺ PEG ⫹ PEG Patient 2 cblC cblC cblF cblD cblD mut 0.50 0.20 0.10 1.10 0.95 0.60 0.28 0.33 0.25 1.30 0.87 0.69 0.44 0.28 0.30 3.20 2.60 1.30 1.0 a 14 [ C] propionate. mut ⫽ mut0 or mut⫺, the vitamin B12–nonresponsive type of methylmalonic acidemia; PEG ⫽ polyethylene glycol. in pernicious anemia,7 deficiencies of Cbl also can be inherited and due to abnormalities in its transport or intracellular processing. The latter have been divided into complementation groups cblA to cblH.3,8 Our patients fall into the cblC complementation group, the most common, which causes combined methylmalonic acidemia and homocystinuria. Neither of the highly effective biochemical screening tests for methylmalonic acidemia and homocystinuria was performed in our patients. It is noteworthy that serum Cbl and folate levels are usually normal in cblC because there is no abnor- Co ⫽ radioactive cobalt; CNCbl ⫽ cyanocobal- 57 mality in the absorption of either vitamin. Clinical heterogeneity in cblC has been previously reported.9 The major cblC phenotype is that of an infant who demonstrates feeding difficulties, hypotonia, seizures, developmental delay, decreased visual acuity, nystagmus, and pigmentary retinopathy. Hematologic manifestations usually consist of thrombocytopenia, macrocytic anemia, megaloblastosis, and hypersegmented neutrophils. Perivascular (arteriole and precapillary) demyelination throughout the centrum semiovale, often associated with fibrinoid degeneration of the vascular wall, was reported in 1 child with cblC.10 No spinal symptoms or lesions were noted. The later-onset phenotype has a more benign clinical course, characterized by dementia, myelopathy, and seizures; neuropathologic data have not been reported. Chronic progressive and relapsing-remitting courses may occur, and some patients respond to hydroxycobalamin treatment.11–14 One of 2 sisters developed a fatal myelopathy at 13 years of age, while the other presented with reversible dementia at 10 years of age.13 An abnormality in folate metabolism due to 5, 10-methylenetetrahydrofolate reductase deficiency15 also may result in subacute combined degeneration and cerebral perivascular demyelination or clinical combinations of dementia and myelopathy.16,17 The autopsy findings of cerebral perivascular demyelination and subacute combined degeneration in our patients suggested a methylation defect, but biochemical studies of cultured skin fibroblasts were necessary to make a specific diagnosis. Brief Communication: Powers et al: cblC Deficiency and PGA-II 399 The neuroradiological similarity and neuropathologic identity of the cerebral white matter lesions in both brothers provide strong evidence that both had cblC, even though biochemical confirmation could only be obtained in Patient 2. We cannot relate the endocrinologic problems of Patient 2 to cblC.3 Our patients’ family exhibited signs and symptoms of polyglandular autoimmune syndrome (PGA) type II (Schmidt’s syndrome). “Pernicious anemia” is rarely (0.5%) noted in PGA type II but is more commonly associated with PGA type I.18 To our knowledge, myelopathy or encephalopathy in PGA–“pernicious anemia” has not been reported. The chromosomal localizations of both cblC and PGA type II are unknown; consequently, we cannot comment on the genetic implications of these two rare diseases in one family. Patient 1 suffered only from cblC, Patient 2 from both cblC and PGA type II, and the third brother only from PGA type II. The precise molecular defect and cellular pathogenesis of cblC are unknown, although the activity of several enzymes presumably involved in the reduction of the central cobalt of Cbl is decreased. The end result is the failure to retain Cbl within cells and to synthesize the active cofactors, AdoCbl and MeCbl. AdoCbl catalyzes the conversion of methylmalonyl CoA to succinyl CoA, and MeCbl catalyzes the methylation of homocysteine to methionine. The pathogenesis of the myelin lesion in Cbl deficiency is also disputable. Our cblC patients provide three practical lessons. First, in contrast to most inherited myelin diseases, cblC is treatable. Second, endogenous defects in Cbl can occur without low serum vitamin B12 levels and hematologic abnormalities. Third, the neurologic spectrum of Cbl deficiency includes optic or neuropsychiatric presentations, which may occur without the myelopathy. Hence, cblC should be considered in enigmatic optic neuropathy, neuropsychiatric disease, or subacute myelopathy, even with normal serum vitamin B12 or folate levels. Laboratory studies should include serum or urine screening for homocysteine and methylmalonic acid. We thank DeWitte T. Cross, III, MD, for his assistance with the radiological studies and Mrs Tina Blazey for her usual outstanding secretarial assistance. We also recognize the importance of the United Leukodystrophy Foundation, because at one of its annual scientific sessions the primary author was reminded (learned?) of the association of perivascular demyelination and methylation disorders by Robert Surtees. Without that encounter, the diagnosis of our patients probably would still be unknown. References 1. Adams RD, Victor M. Principles of neurology, 6th ed. New York: McGraw-Hill, 2000. 400 Annals of Neurology Vol 49 No 3 March 2001 2. Harper C, Butterworth R. Nutritional and metabolic disorders. In: Graham DI, Lantos PL, eds. Greenfield’s neuropathology, 6th ed. London: Arnold, 1997:621– 624. 3. Fenton WA, Rosenberg LE. Inherited disorders of cobalamin transport and metabolism. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic and molecular bases of inherited disease, 7th ed. New York: McGraw-Hill, 1995:3129 –3149. 4. Pant SS, Asbury AK, Richardson Jr EP. The myelopathy of pernicious anemia: a neuropathological reappraisal. 1968;44:8 – 36. 5. Savage DG, Lindenbaum J. Neurological complications of acquired cobalamin deficiency: clinical aspects. In: Wickramasinghe SN, ed. Baillière’s clinical haematology: international practice and research. Megaloblastic anaemia. London: Bailliére Tindall, 1995;8:657– 678. 6. Adams RD, Kubik CS. Subacute degeneration of the brain in pernicious anemia. N Engl J Med 1944;231:1–9. 7. Cooper BA, Castle WB. Sequential mechanisms in the enhanced absorption of vitamin B12 by intrinsic factor in the rat. J Clin Invest 1960;39:199 –214. 8. Watkins D, Matiaszuk N, Rosenblatt DS. Complementation studies in the cblA class of inborn error of cobalamin metabolism: evidence for interallelic complementation and for a new complementation class (cblH). J Med Genet 2000;37:510 – 513. 9. Rosenblatt DS, Aspler AL, Shevell MI, et al. Clinical heterogeneity and prognosis in combined methylmalonic aciduria and homocystinuria (cblC). J Inherit Metab Dis 1997;20:528 –538. 10. Dayan AD, Ramsey RB. An inborn error of vitamin B12 metabolism associated with cellular deficiency of coenzyme forms of the vitamin: pathological and neurochemical findings in one case. J Neurol Sci 1974;23:117–128. 11. Shinnar S, Singer HS. Cobalamin C mutation (methylmalonic aciduria and homocystinuria) in adolescence: a treatable cause of dementia and myelopathy. N Engl J Med 1984;311:451– 454. 12. Gold R, Bogdahn U, Kappos L, Toyka K. Hereditary defect of cobalamin metabolism (homocystinuria and methylmalonic aciduria) of juvenile onset. J Neurol Neurosurg Psychiatry 1996; 60:107–108. 13. Augoustides-Savvopoulou P, Mylonas I, Sewell AC, Rosenblatt DS. Reversible dementia in an adolescent with cblC disease: clinical heterogeneity within the same family. J Inherit Metab Dis 1999;22:756 –758. 14. Enns GM, Barkovich AJ, Rosenblatt DS, et al. Progressive neurological deterioration and MRI changes in cblC methylmalonic acidaemia treated with hydroxycobalamin. J Inherit Metab Dis 1999;22:599 – 607. 15. Rosenblatt DS. Inherited disorders of folate transport and metabolism. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic and molecular bases of inherited disease, 7th ed. New York: McGraw-Hill, 1995:3111–3128. 16. Clayton PT, Smith I, Harding B, et al. Subacute combined degeneration of the cord, dementia and parkinsonism due to an inborn error of folate metabolism. J Neurol Neurosurg Psychiatry 1986;49:920 –927. 17. Haworth JC, Dilling LA, Surtees RAH, et al. Symptomatic and asymptomatic methylenetetrahydrofolate reductase deficiency in two adult brothers. Am J Med Genet 1993;45:572–576. 18. Neufeld M, Maclaren NK, Blizzard RM. Two types of autoimmune Addison’s disease associated with different polyglandular autoimmune (PGA) syndromes. Medicine 1981;60:355– 362. Irreversible Brain Creatine Deficiency with Elevated Serum and Urine Creatine: A Creatine Transporter Defect? Kim M. Cecil, PhD,1 Gajja S. Salomons, PhD,3 William S. Ball, Jr., MD,1 Brenda Wong, MD,2 Gail Chuck,2 Nanda M. Verhoeven, PhD,3 Cornelis Jakobs, PhD,3 and Ton J. DeGrauw, MD,2 Recent reports highlight the utility of in vivo magnetic resonance spectroscopy (MRS) techniques to recognize creatine deficiency syndromes affecting the central nervous system (CNS). Reported cases demonstrate partial reversibility of neurologic symptoms upon restoration of CNS creatine levels with the administration of oral creatine. We describe a patient with a brain creatine deficiency syndrome detected by proton MRS that differs from published reports. Metabolic screening revealed elevated creatine in the serum and urine, with normal levels of guanidino acetic acid. Unlike the case with other reported creatine deficiency syndromes, treatment with oral creatine monohydrate demonstrated no observable increase in brain creatine with proton MRS and no improvement in clinical symptoms. In this study, we report a novel brain creatine deficiency syndrome most likely representing a creatine transporter defect. Ann Neurol 2001;49:401– 404 Creatine (␣-methyl-guanidinoacetic acid) plays an important role in energy metabolism. In humans, creatine is synthesized in the liver, kidney, and pancreas. Creatine is ultimately transported via the blood to the muscles, heart, and nervous system, which are rich in creatine kinase. Creatine kinase is an essential enzyme to catalyze phosphorylation of creatine to provide a highenergy phosphate buffer system. Previous reports describe an inborn error in metabolism representing a severe deficiency of hepatic guanidinoacetate methyltransferase activity [S-adenosylmethionine: guanidinoacetate N-methyltransferase (EC From the 1Divisions of Radiology, and 2Neurology, Children’s Hospital Medical Center and the University of Cincinnati, Cincinnati, OH; and 3Metabolic Unit of the Department of Clinical Chemistry, Free University Hospital, Amsterdam, The Netherlands. Received Jul 17, 2000, and in revised form Nov 3. Accepted for publication Nov 6, 2000. Address correspondence to Dr Cecil, Imaging Research Center, Children’s Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, OH 45229. E-mail: email@example.com 184.108.40.206; GAMT)].1–3 This condition results in an accumulation of guanidinoacetic acid (GAA) and a deficiency of creatine. Neurologic symptoms associated with creatine deficiency have some variability but present overall as extrapyramidal movement disorders, developmental regression, behavioral problems, and intractable epilepsy. Patients with the GAMT defect, treated with oral creatine, demonstrate improvement of clinical symptoms and biochemical abnormalities. Residual deficits may arise from neurotoxic GAA accumulation or an imbalance of brain creatine and highenergy phosphates.4 – 6 A technique useful in recognizing creatine deficiencies and other metabolic abnormalities is proton magnetic resonance spectroscopy (MRS). As a complementary technique with magnetic resonance imaging (MRI), the use of proton MRS in the clinical MR examination provides a noninvasive method for assessing unique metabolic information for focal and diffuse disorders in vivo. Creatine signal was absent in multple regions of the brain of a 6-year-old boy examined with in vivo proton MRS. Metabolic testing revealed elevated serum and urine creatine levels but normal plasma and urine GAA levels. In this study, we report a novel brain creatine deficiency syndrome hypothesized to be a creatine transporter defect. Case Study A white male was diagnosed at 7 months of age with mild developmental delay and central hypotonia. Prenatal and perinatal histories were unremarkable. MRI at 8 months demonstrated prominent extraaxial spaces. At age 2 years, he was admitted to the hospital in partial status epilepticus. He recovered completely and was discharged in good condition. An electroencephalograph (EEG) showed multifocal epileptiform discharges. The MRI performed revealed a small focus of hyperintensity in the right posterior periventricular white matter on T2-weighted images. A repeat EEG at age 5 years was unchanged, with no further seizures reported. A repeat MRI continued to demonstrate a stable region of T2-weighted hyperintense signal in the right posterior periventricular white matter. Physical examination at 6 years of age showed severe delay in speech and language function. He did not follow commands and spoke only a few single words. Gross and fine motor functions were normal, but he was mildly hypotonic. He had a short attention span; however, he did establish good eye contact. Head circumference increased from the 75th percentile to the 95th percentile, prompting MRI and proton MRS. He presented with no other neurological or general physical abnormalities. The small focus of hyperintense signal in the right posterior periventricular white matter remained unchanged compared to the previous MRI examinations. Cecil et al: Creatine Deficiency 401 Fig 1. (A) T2-weighted image and (B) initial short (PROBE-PRESS, TE 35 msec, TR 2 sec) and long (PROBEPRESS, TE 272 msec, TR 2 sec) echo spectroscopy of 6-year-old male patient. Note the absence of the creatine and phosphocreatine at 3.0 ppm. The small resonance at 3.0 ppm most likely represents ␥-aminobutyric acid insofar as the signal is higher on the short echo scan. Fig 2. Normal short (PROBE-PRESS, TE 35 msec, TR 2 sec) and long (PROBEPRESS, TE 272 msec, TR 2 sec) echo spectroscopy of 7-year-old male patient. Proton MRS was acquired on the patient using PRESS single voxel sequences [short echo time (TE) 35 msec and long TE 288 msec, repetition time (TR) 2 sec] on 8 cc volumes in the frontal white matter and basal ganglia. Spectral reconstruction demonstrated an almost complete loss of creatine and phosphocreatine signal (Fig 1, compare Fig 2). N-acetyl aspartate and choline levels were within normal limits for age. Myoinositol levels appeared slightly diminished on comparison with normal control data, with no lactate signal detected in the spectra. The small peak at 3.0 ppm, where creatine normally appears, was most likely due to a small concentration of ␥-aminobutyric acid (GABA). Further off-line processing with custom soft- 402 Annals of Neurology Vol 49 No 3 March 2001 ware was performed to determine metabolite concentrations using methods similar to that previously described in the literature.7 The laboratory findings pertaining to creatine and related metabolites are found in the Table. Creatinine in blood and amino acid levels in blood and urine were normal. Other metabolic testing showed serum lactate, pyruvate, ammonia, electrolytes, glucose, and liver function tests were normal. Carnitine and acylcarnitine profiles were also normal. A cytogenetic investigation of the chromosomes revealed a normal karyotype (46,XY). With the initial MRS result indicating a creatine deficiency, the patient was put on oral creatine monohydrate 2,500 mg three times/day (340 mg/kg/day). Table. Laboratory Results Prior to Oral Creatine Substitution Creatine Serum (mol/liter) Urine (mmol/liter) Creatinine Blood (mol/liter) Creatine kinase (U/liter) Guanidinoacetate Plasma (mol/liter) Urine (mmol/mol creatinine) Patient Normal 75 2.2 15–44 ⬍0.35 35.4 100 1.05 74 17.7–61.9 75–215 0.65–1.44 10.3–98.8 A follow-up proton MRS examination performed 4 months later demonstrated a similar absence of creatine. However, creatine in the urine increased tenfold, and elevated cerebrospinal fluid (CSF) levels were measured [61.7 mol/liter (normal 38 ⫾ 4.8 mol/liter)]. The patient’s clinical symptoms did not improve after 3 months. Oral substitution of creatine was discontinued. Discussion Initial reports of creatine deficiency revealed by proton MRS of the brain led clinicians to uncover a synthesis defect in hepatic guanidinoacetate methyltransferase activity.4,8 Three patients with confirmed GAMT defects demonstrated developmental delay, intractable epilepsy, and an extrapyramidal movement disorder.1,2,4 Van der Knaap et al.5 reported GAMT deficiency in two males without specific neurological findings and a presentation much milder than that of the patients with the original GAMT defect. Treatment of these patients with oral creatine substitution resulted in higher concentrations of creatine in the brain as observed by proton MRS. The patients showed marked clinical and neurologic improvement. Laboratory results in our patient demonstrate that the substrates and products involved in creatine synthesis and excretion (glycine, ornithine, GAA, and creatinine) are at normal levels. Supplementation with oral creatine results in a tenfold increase in urine creatine concentrations. In an isolated defect in creatine synthesis, treatment with creatine should result in an increased and measurable concentration of creatine in the brain. Normal adults, who take creatine supplements, increase their creatine concentration in the CNS as measured by MRS.9 Therefore, we exclude GAMT deficiency as the source of our patient’s brain creatine deficiency. Although the main synthesis of creatine proceeds in liver, it has been reported that rat brain cells are capable of creatine synthesis.10,11 However, this work has been demonstrated only in animal cell cultures. The data from our patient indicates that, if de novo brain creatine synthesis is occuring, it is not detectable by proton MRS. We suspected a defect in the creatine transporter of our patient. Gregor et al12 assigned the human creatine transporter (CT1) gene (SLC6A8) to chromosome Xq28. Sora et al.13 demonstrated the presence of a creatine transporter in human brainstem and spinal cord. The highest levels of creatine transporter mRNA expression are in human skeletal muscle, kidney, and heart, with lower levels in brain, small and large intestine, epididymis, testis, vas deferens, seminal vesicles, prostate, and adrenal.14,15 Creatine uptake into tissues is an active process and has been shown to depend on sodium and/or chloride channels.16,17 Creatine transport in the brain is hypothesized to be an astroglial rather than a neuronal function.16,18 Within the brain, there is a heterogenous, regional distribution of creatine transporter transcripts.15 In situ hybridization showed high creatine transporter levels in cerebellum and hippocampus.19 Several neurological, neuromuscular, and hearing disorders are mapped to Xq28, but the association of the disorder with the creatine transporter is not known.18 An autosomal, testis-specific form of the human creatine transporter (CT2) has been assigned to chromosome 16p11.220 but may represent a pseudogene.21 At high serum concentrations, creatine uptake in some tissues could represent passive diffusion across the plasma membrane or alternatively transport via other transporters. The patient’s abilities to include supplemental dietary creatine into the blood stream and to maintain relatively normal muscle function indicate some mechanism of transport or compensation. Two sisters were recently described with a brain creatine deficiency syndrome presumed to be a creatine tranporter defect in the brain.22 Again, proton MRS of the brain detected the absence of creatine. Normal creatine and GAA concentrations in the sister’s blood indicated normal creatine synthesis. Mutation detection in the SLC6A8 gene was not reported. The improvement in clinical symptoms and the increase in CNS creatine concentration after creatine supplementation indicate either an X-linked mosaicism with a less severe mutation in the transporter or a creatine deficiency syndrome different from that of our patient. Preliminary sequence analysis of amplified cDNA of SLC6A8 skin fibroblasts in our patient indicates a hemizygous nonsense mutation. In the GAMT deficiency cases, a mechanism of injury could be inferred from neurotoxic GAA levels. However, in the cases reported by Bianchi et al.22 and our patient, a mechanism for injury must arise from the deficiency of creatine. Unknown is whether another neurotoxin could be responsible for the neurological symptoms. Our case further demonstrates a utility of MRS in diagnosing and treating children with neurodevelopmental problems. This patient represents another defect in creatine metabolism of the CNS revealed by Cecil et al: Creatine Deficiency 403 MRS. This case also shows the importance of creatine metabolism for neurological development. Creatine supplementation may play a role in improving symptoms of neuromuscular disorders.23 Beal and colleagues report neuroprotective effects of oral supplementation with creatine in Huntington’s disease, amyotrophic lateral sclerosis, and other neurodegenerative diseases.24 –27 Creatine may attenuate 1-methyl-4-phenyl1,2,3,6-tetrahydropyridine (MPTP) neurotoxicity and possibly reverse mitochondrial dysfunctions.25 Further investigations into the metabolism of creatine and pathophysiology of creatine disorders are necessary. References 1. Schulze A, Hess T, Wevers R, et al. Creatine deficiency syndrome caused by guanidinoacetate methyltransferase deficiency: diagnostic tools for a new inborn error of metabolism [see comments]. J Pediatr 1997;131:626 – 631. 2. Ganesan V, Johnson A, Connelly A, et al. Guanidinoacetate methyltransferase deficiency: new clinical features. Pediatr Neurol 1997;17:155–157. 3. Stockler S, Isbrandt D, Hanefeld F, et al. Guanidinoacetate methyltransferase deficiency: the first inborn error of creatine metabolism in man. Am J Hum Genet 1996;58:914 –922. 4. Stockler S, Holzbach U, Hanefeld F, et al. Creatine deficiency in the brain: a new, treatable inborn error of metabolism. Pediatr Res 1994;36:409 – 413. 5. van der Knaap MS, Verhoeven NM, Maaswinkel-Mooij P, et al. Mental retardation and behavioral problems as presenting signs of a creatine synthesis defect. Ann Neurol 2000;47:540 – 543. 6. Hirayasu Y, Morimoto K, Otsuki S. Increase of methylguanidine and guanidinoacetic acid in the brain of amygdala-kindled rats. Epilepsia 1991;32:761–766. 7. Kreis R, Ernst T, Ross BD. Absolute concentrations of water and metabolites in human brain. II Metabolite concentrations. J Magn Reson Ser B 1993;102:9 –19. 8. Stockler S, Hanefeld F. Guanidinoacetate methyltransferase deficiency: a newly recognized inborn error of creatine biosynthesis. Wien Klin Wochenschr 1997;109:86 – 88. 9. Dechent P, Pouwels PJ, Wilken B, et al. Increase of total creatine in human brain after oral supplementation of creatinemonohydrate. Am J Physiol 1999;277:R698 –R704. 10. Defalco AJ, Davies RK. The synthesis of creatine by the brain of the intact rat. J Neurochem 1961;7:308 –312. 11. Dringen R, Verleysdonk S, Hamprecht B, et al. Metabolism of glycine in primary astroglial cells: synthesis of creatine, serine, and glutathione. J Neurochem 1998;70:835– 840. 12. Gregor P, Nash SR, Caron MG, et al. Assignment of the creatine transporter gene (SLC6A8) to human chromosome Xq28 telomeric to G6PD. Genomics 1995;25:332,333. 13. Sora I, Richman J, Santoro G, et al. The cloning and expression of a human creatine transporter. Biochem Biophys Res Commun 1994;204:419 – 427. 14. Nash SR, Giros B, Kingsmore SF, et al. Cloning, pharmacological characterization, and genomic localization of the human creatine transporter. Recept Channels 1994;2:165–174. 15. Wyss M, Kaddurah-Daouk R. Creatine and creatinine metabolism. Physiol Rev 2000;80:1107–1213. 16. Moller A, Hamprecht B. Creatine transport in cultured cells of rat and mouse brain. J Neurochem 1989;52:544 –550. 404 Annals of Neurology Vol 49 No 3 March 2001 17. Guimbal C, Kilimann MW. A Na(⫹)-dependent creatine transporter in rabbit brain, muscle, heart, and kidney. cDNA cloning and functional expression. J Biol Chem 1993;268:8418 – 8421. 18. Hiel H, Happe HK, Warr WB, Morley BJ. Regional distribution of a creatine transporter in rat auditory brainstem: an insitu hybridization study. Hear Res 1996;98:29 –37. 19. Schloss P, Mayser W, Betz H. The putative rat choline transporter CHOT1 transports creatine and is highly expressed in neural and muscle-rich tissues. Biochem Biophys Res Commun 1994;198:637– 645. 20. Iyer GS, Krahe R, Goodwin LA, et al. Identification of a testisexpressed creatine transporter gene at 16p11.2 and confirmation of the X-linked locus to Xq28. Genomics 1996;34:143– 146. 21. Eichler EE, Lu F, Shen Y, et al. Duplication of a gene-rich cluster between 16p11.1 and Xq28: a novel pericentromericdirected mechanism for paralogous genome evolution. Hum Mol Genet 1996;5:899 –912. 22. Bianchi MC, Tosetti M, Fornai F, et al. Reversible brain creatine deficiency in two sisters with normal blood creatine level. Ann Neurol 2000;47:511–513. 23. Guerrero-Ontiveros ML, Wallimann T. Creatine supplementation in health and disease. Effects of chronic creatine ingestion in vivo: down-regulation of the expression of creatine transporter isoforms in skeletal muscle. Mol Cell Biochem 1998; 184:427– 437. 24. Klivenyi P, Ferrante RJ, Matthews RT, et al. Neuroprotective effects of creatine in a transgenic animal model of amyotrophic lateral sclerosis. Nature Med 1999;5:347–350. 25. Matthews RT, Ferrante RJ, Klivenyi P, et al. Creatine and cyclocreatine attenuate MPTP neurotoxicity. Exp Neurol 1999; 157:142–149. 26. Ferrante RJ, Andreassen OA, Jenkins BG, et al. Neuroprotective effects of creatine in a transgenic mouse model of Huntington’s disease. J Neurosci 2000;20:4389 – 4397. 27. Malcon C, Kaddurah-Daouk R, Beal MF. Neuroprotective effects of creatine administration against NMDA and malonate toxicity. Brain Res 2000;860:195–198. Reduction of Menkes mRNA and Copper in Leukocytes of Patients with Primary Adult-Onset Dystonia Niels Kruse, PhD,1 Daniela Berg, MD,1 Michael J. Francis, DPhil,2 Markus Naumann, MD,1 Wolf-Dieter Rausch, PhD,3 K. Reiners, MD,1 Peter Rieckmann, MD,1 Andreas Weishaupt, PhD,1 and Georg Becker, MD1 Studies on postmortem tissue of patients with primary adult-onset dystonia revealed a significant increase in copper levels and a reduction of copper transporting Menkes protein of the lentiform nuclei. Here we demonstrate that patients with idiopathic adult-onset cervical dystonia (n ⴝ 14) have reduced Menkes mRNA copies and lower copper levels in leukocytes compared to controls (n ⴝ 17; U test, p < 0.05). Changes were less distinct in patients with blepharospasm. Therefore, disturbances of copper metabolism in focal dystonia may not be restricted to the basal ganglia. intracellular copper level. This change in copper level may cause a dysfunction in the basal ganglia loop at the receptor level.5–7 In this study we evaluated whether disturbances of copper metabolism are restricted to the basal ganglia or whether they can also be detected in blood cells from patients with idiopathic adult-onset dystonia. Patients and Methods Twenty-four patients with primary adult-onset focal dystonia were included in this study. Fourteen patients suffered from spasmodic torticollis (mean age 51.7 ⫾ 10.6 years, 10 females and 4 males) and 10 from blepharospasm (mean age 71.9 ⫾ 7.7 years, 2 males and 8 females; Table). The mean duration of the disease was 10.9 ⫾ 8.1 years. The severity of dystonia in patients with spasmodic torticollis was graduated according to the Tsui score (mean Tsui score: 13.0 ⫾ 3.2). All patients with dystonia were treated with botulinum toxin and responded well to the injections. For comparison, 17 healthy adults (mean age 42.4 ⫾ 9.5 years, 10 females and 7 males) were included in this study. Informed consent was obtained from all individuals participating in this study, which was approved by the local ethics committee. From patients and controls, peripheral blood mononuclear cells were purified by Lymphoprep gradient centrifugation (Nycomed, Oslo, Norway) according to the manufacturer’s instruction and stored at – 80°C until use. Ann Neurol 2001;49:405– 408 Measurement of Menkes mRNA in Leukocytes Evidence from recent studies suggests abnormalities in copper metabolism of patients with idiopathic adultonset dystonia.1,2 These examinations were stimulated by transcranial sonography (TCS) findings demonstrating abnormal signals at the lentiform nuclei.3,4 Subsequent neurochemical analyses of postmortem tissue of patients with primary adult-onset dystonia revealed a significant increase in copper levels of the lentiform nuclei but not of the thalamus or caudate nucleus.1 One reason for the elevated tissue copper levels may be a change in the copper-transporting Menkes protein, which has been found to be reduced in the lentiform nuclei but not in the thalamus and caudate nucleus.2 A reduction in Menkes protein expression results in a reduced efflux of copper out of the cell and an increased From the 1Department of Neurology, Bayerische JuliusMaximilians-Universität, Würzburg, Germany; 2Human Genetics/ Clinical Medicine, Wellcome Trust Center for Human Genetics, Oxford, United Kingdom; and 3Institute of Medical Chemistry, University of Veterinary Medicine, Vienna, Austria. Received Aug 4, 2000, and in revised form Nov 10. Accepted for publication Nov 11, 2000. Address correspondence to Dr Becker, Neurologische Klinik und Poliklinik des Klinikums der Universität Würzburg, JosefSchneider-Str. 11, 97080 Würzburg, Germany. E-mail: firstname.lastname@example.org PREPARATION OF TOTAL RNA AND REVERSE TRANSCRIPTION. Total RNA was prepared from 5 ⫻ 106 lympho- cytes using the Qiagen RN easy kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer’s instructions. Purified RNA was eluted with 33 l DEPC-treated water. For reverse transcription 3 l oligo(dT) (500 g/ml; Pharmacia Biotech, Freiburg, Germany) were added, incubated for 10 minutes at 70°C, and chilled on ice. After mixing with 12 l First Strand Buffer and 6 l 0.1M dithiothreitol (DTT; both from Life Technologies, Karlsruhe, Germany), 3 l 10 mM dNTPs (Pharmacia Biotech), and 3 l Superscript II reverse transcriptase (200 U/l; Life Technologies), the assays were incubated for 50 minutes at 42– 45°C. Reverse trancriptase was denatured by incubation for 10 minutes at 95°C. QUANTIFICATION OF ␤-ACTIN AND MENKES mRNA BY REAL-TIME POLYMERASE CHAIN REACTION. Polymerase chain reaction (PCR) for ␤-actin mRNA quantification was performed as previously described.8 For quantification of Menkes mRNA expression, the following oligonucleotides were used: ATP7A S: 5⬘-CCC GGT TAC CAA TGA GGA CTT-3⬘; ATP7A AS: 5⬘-CAT CAA CTT CGT CTT CCA TGA TTA TT-3⬘; ATP7A probe: 5⬘-FAM-ATG CCC AGA TCT CAA GTG CTC TTA ATG CTC A-TAMRA-3⬘. Equivalents of 108 copies of ␤-actin mRNA were used as templates for semiquantitative PCR of Menkes mRNA. Serial dilutions of standard Menkes cDNA were prepared in 10 mM Tris, pH 8.0, and amplified in parallel with unknown cDNAs. Each determination was performed in triplicate. Reaction conditions were as described previously.8 All amplifi- Brief Communication: Kruse et al: Menkes Protein in Dystonia 405 Table. Characteristics of Patients and Controls, Including Copper and Menkes mRNA Levels in Leukocytes Focal Dystonia Total group Number of subjects Mean age (years) Sex (females/males) Menkes mRNA copies/108 ␤-actin mRNAs of leukocytesa Copper level of leukocytes (ng/mg protein)a Serum copper (mg/dl) 24 59.2 ⫾ 13.9 18/6 123.0 (7; 310)b 6.3 (5.7; 7.6)b 111 (98; 123) Torticollis 14 51.7 ⫾ 10.6 10/4 70.5 (6; 162)b 6.3 (5.6; 6.8)b 111 (89; 125) Blepharospasm Controls 10 71.9 ⫾ 7.7 8/2 287 (208; 343) 17 42.4 ⫾ 9.5 10/7 289 (119; 456) 6.9 (6; 8) 7.8 (7; 9.5) 111 (107; 121) 113 (107; 121) a Median values and 25th and 75th quartiles. Intergroup comparison using the U test revealed a significant difference compared to the control group ( p ⬍ 0.05). b cation reactions were performed on an ABI PRISM 7700 Sequence Detection System. Determination of Copper Content in Leukocytes Leukocytes (1.8 ⫻ 107) were suspended in 200 ml phosphate-buffered saline and 1% Triton X-100 containing protease inhibitors. Insoluble material was pelleted by centrifugation, and the protein concentration supernatant was determined by Comassie Protein Assay Reagent (Pierce, Rockford, IL). Leukocytes suspensions with defined protein content (2–5 mg/ml) were directly injected into graphite pyrotubes. Following a drying program at 120°C, samples were ashed at 600°C and measured in duplicate by atomic absorption at 2,700°C using a polarized Zeeman atomic absorption spectrophotometer (Hitachi Z 8100). Commercial standards (Merck Titrisol) were diluted and used in a concentration range of 0 – 8 g/ml. Measurements of copper and Menkes mRNA were made blindly with respect to the clinical characteristics of the subjects. Fig 1. Menkes mRNA copies/108 ␤-actin mRNA of leukocytes from patients with focal dystonia and controls. Differences in Menkes mRNA copies of patients with focal dystonia (spasmodic torticollis and blepharospasm) and controls were significant (U test, p ⫽ 0.04). In analyzing mRNA levels of both subgroups of focal dystonia, differences were significant for patients with torticollis (U test, p ⫽ 0.02) compared to controls but not for patients with blepharospasm (U test, p ⬎ 0.05). 406 Annals of Neurology Vol 49 No 3 March 2001 Results Analyses of leukocyte mRNA revealed a significant decrease of Menkes mRNA in patients with spasmodic torticollis [70.5 (25th and 27th percentiles, 6; 162)] compared to controls [289 (25th and 75th percentiles; 119; 456); U test, p ⫽ 0.02]. Menkes mRNA levels of patients with blepharospasm [287.0 (208; 343)] were similar to those of controls, reflecting differences in the severity of these dystonic disorders (Table, Fig 1). In addition, patients with a more severe spasmodic torticollis (Tsui score ⬎ 13; 9 of the 14 patients) had lower Menkes mRNA levels [median 22 (2; 92) Menkes mRNA copies/108 ␤-actin mRNA copies] than less disabled patients [Tsui score ⱕ 13; 5 of the 14 patients; median 162 (123– 473) Menkes mRNA copies/ 108 ␤-actin mRNA copies; U test, p ⫽ 0.04]. No correlation was identified between age and number of mRNA copies (Spearman’s rank correlation, r ⫽ – 0.07, p ⫽ 0.91), sex and mRNA copies (U test, p ⫽ Fig 2. Leukocytes copper levels of patients with focal dystonia and healthy controls. Copper levels of patients in both groups with focal dystonia were significantly reduced compared to controls (U test, p ⫽ 0.005). Patients with spasmodic torticollis had lower copper levels than those with blepharospasm, and only for patients with torticollis were differences significant compared to controls (U test, p ⫽ 0.002). 0.39), or the duration of the disease and number of mRNA copie (Spearman’s rank correlation, r ⫽ – 0.47, p ⫽ 0.63). As with Menkes mRNA, the copper concentrations of leukocytes in patients with spasmodic torticollis were lower [6.3 (5.6; 6.8) ng/mg protein] than in patients with blepharospasm [6.9 (6; 8) ng/mg protein] or controls [7.8 (7.0; 9.5) ng/mg protein] paralleling changes of Menkes mRNAs (Table, Fig 2). No correlation between Menkes mRNA levels and copper concentration in leukocytes could be detected (Spearman’s rank correlation, r ⫽ – 0.03, p ⫽ 0.87). Leukocyte copper levels were not related to age (Spearman’s rank correlation, r ⫽ – 0.23, p ⫽ 0.14) or sex of patients and controls (U test, p ⫽ 0.67). Serum copper levels in patients and controls are outlined in the Table. No differences between patients and controls and no correlations with other parameters were obtained. Discussion Our data revealed reduced Menkes mRNA levels in leukocytes of patients with primary adult-onset cervical dystonia compared to controls. Menkes mRNA levels were lower in more severely affected patients. In addition, the copper concentration of leukocytes was lower in patients with focal dystonia. Reductions of copper and particularly mRNA levels were less distinct in patients with bleparospasm, which might reflect differences in the pathogenesis or severity of both disorders. These data support previous findings of a reduced Menkes protein content in the lentiform nuclei of patients with primary dystonia and indicate that a disturbance in copper metabolism in dystonia is not restricted to the brain. However, we cannot exclude that Menkes protein synthesis might be influenced by posttranscriptional gene-regulating mechanisms. Menkes mRNA levels presented in this study represent steadystate mRNA levels, so it is not clear whether they are the result of decreased levels of gene expression or an increased rate of mRNA degradation. Menkes protein is a membrane-bound protein thought to be involved in both the sequestration of copper to essential copper requiring enzymes and the efflux of copper from the cell.9,10 In humans, expression has been detected in all tissues tested except the liver. Menkes protein transports copper into the lumen of the Golgi complex. In case of intracellular copper levels, Menkes protein effluxes copper out of the cell. A reduction in the levels of Menkes protein results in an increase in intracellular copper levels. Menkes protein reduction will mainly affect the intestinum and the brain, notably the lentiform nuclei; these areas exhibit the highest copper turnover of Menkes proteinexpressing cells.11 Although Menkes protein reduction causes accumulation of copper in some tissues,12,13 body copper content decreases as a result of the reduced transportation of copper across the intestinal cells.14 This “mosaic pattern” of cellular copper concentration typical for patients with Menkes protein deficiency was also observed in patients with primary dystonia: A recent postmortem study revealed an increased copper level in the lentiform nuclei of patients with idiopathic dystonia,1 whereas copper levels of leukocytes were found to be reduced in this study. Changes in tissue copper levels may have a substantial influence on the activity of the basal ganglia loop neurons; copper is an endogenous modulator of synaptic function and a potent inhibitor of several receptor types, such as glutamate, opiate, and GABAA receptors, all of which are highly expressed in the basal ganglia.5–7 Therefore, one may postulate a relation between Menkes protein depletion, increased copper tissue level in the lentiform nuclei, changes in the neuronal activity of the basal ganglia loop, and subsequently dystonic movements. In analyzing our data, one may question whether Menkes mRNA depletion is the central pathogenetic process in dystonia because of the broad overlap of Menkes mRNA copies and copper levels in patients and controls. In addition, Menkes disease, the disorder associated with Menkes protein depletion, is phenotypically very different from dystonia,14 although one report mentioned extrapyramidal symptoms in milder forms of Menkes disease.15 Therefore, the reduction in Menkes mRNA might reflect the consequence of an as yet unidentified primary pathology. The authors thank S. Hellmig for her excellent technical support. We gratefully acknowledge Prof. K. V. Toyka for his continuous support and most helpful comments on the manuscript. References 1. Becker G, Berg D, Rausch WD, et al. Increased tissue copper and manganese content in the lentiform nucleus in primary adult-onset dystonia. Ann Neurol 1999;46:260 –263. 2. Berg D, Weishaupt A, Francis MJ, et al. Changes of coppertransporting proteins and ceruloplasmin in the lentiform nuclei in primary adult-onset dystonia. Ann Neurol 2000;47:827– 830. 3. Naumann M, Becker G, Toyka KV, et al. Lenticular nucleus lesion in idiopathic dystonia detected by transcranial sonography. Neurology 1996;47:1284 –1290. 4. Becker G, Naumann M, Scheubeck M, et al. Comparison of transcranial sonography, magnetic resonance imaging, and single photon emission computed tomography findings in idiopathic spasmodic torticollis. Mov Disord 1997;12:79 – 88. 5. Sharonova IN, Vorobjev VS, Haas HL. High-affinity copper block of GABA(A) receptor-mediated currents in acutely isolated cerebellar Purkinje cells of the rat. Eur J Neurosci 1998; 10:522–528. 6. Weiser T, Wienrich M. The effects of copper ions on glutamate Brief Communication: Kruse et al: Menkes Protein in Dystonia 407 7. 8. 9. 10. 11. 12. 13. 14. 15. receptors in cultured rat cortical neurons. Brain Res 1996;742: 211–218. Sadee W, Pfeiffer A, Herz A. Opiate receptor: multiple effects of metal ions. J Neurochem 1982;39:659 – 667. Kruse N, Pette M, Toyka KV, Rieckmann P. Quantification of cytokine mRNA expression by RT PCR in samples of previously frozen blood. J Immunol Methods 1997;210:195–203. La Fontaine SL, Firth SD, Camakaris J, et al. Correction of the copper transport defect of Menkes patient fibroblasts by expression of the Menkes and Wilson ATPases. J Biol Chem 1998; 273:31375–31380. Francis MJ, Jones EE, Levy ER, et al. A Golgi localization signal identified in the Menkes recombinant protein. Hum Mol Genet 1998;7:1245–1252. Turnlund JR. Human whole-body copper metabolism. Am J Clin Nutr 1998;67(5 Suppl):960S–964S. Danks DM, Cartwright E, Stevens BJ, Townley RRW. Menkes’ kinky hair disease: further definition of the defect in copper transport. Science 1973;179:1140 –1142. Kodama H, Abe T, Takama M, et al. Histochemical localization of copper in the intestine and kidney of macular mice: light and electron microscopic study. J Histochem Cytochem 1993;41:1529 –1535. Menkes JH. Kinky hair disease: twenty five years later. Brain Dev 1988;10:77–79. Haas RH, Robinson A, Evans K, et al. An X-linked disease of the nervous system with disordered copper metabolism and features differing from Menkes disease. Neurology 1981;31:852– 859. Anticardiolipin Antibodies Are Not a Useful Screening Tool in a Nonselected Large Group of Patients with Multiple Sclerosis Jaume Sastre-Garriga, MD,1 Juan Carlos Reverter, MD,2 Josep Font, MD, PhD,3 Mar Tintoré, MD,1 Gerard Espinosa, MD,3 and Xavier Montalban, MD, PhD1 Recent works claiming that primary antiphospholipid syndrome (PAPS) cannot be clinically distinguished from multiple sclerosis (MS) recommend that MS patients be screened for anticardiolipin antibodies (ACA). In this study 296 randomly selected patients with MS and clinically isolated syndromes and 51 healthy controls were analyzed; ACA, anti-␤2-glycoprotein I, or antiprothrombin was found in 6 patients. No predominance of any kind of clinical manifestations and no cardinal manifestations of PAPS were found in these patients. ACA tests should be performed only when a suspicion of PAPS is raised and atypical clinical presentation for MS is found. Ann Neurol 2001;49:408 – 411 408 Annals of Neurology Vol 49 No 4 April 2001 Antiphospholipid antibodies (aPL) are associated with venous and arterial thrombosis and recurrent spontaneous abortions or fetal losses.1,2 They may occur with no other manifestation of systemic disease, a scenario known as primary antiphospolipid syndrome (PAPS) or Hughes syndrome.3 Because IgG anticardiolipin antibodies (ACA) are also present in healthy populations, positivity in a person with no signs or symptoms of PAPS and no associated connective tissue disorder will lack significance. Recently, ␤2-glycoprotein I (␤2GPI) has been shown to be the real antigen of ACA.4 – 6 Although several attempts have been made to correlate ACA with specific neurological complaints, such as migraine, epilepsy, myelitis, optic neuritis, or vascular dementia, most of these associations have never been clearly established. Four recent studies have examined the presence of ACA, stressing the significance in patients with multiple sclerosis (MS).7–10 These studies claim that PAPS must be considered as an alternative diagnosis to MS, and even that MS cannot be differentiated from PAPS according only to clinical and magnetic resonance imaging (MRI) findings, and a recommendation to test ACA in MS populations routinely is, therefore, made. To address the question of whether it is necessary to test MS patients for ACA, we performed tests for ACA, anti-␤2-glycoprotein I antibodies (a␤2-GPI), and antiprothrombin antibodies (aPT) in a large, unselected population of MS patients. We also reviewed the medical records of ACA- or a␤2-GPI-positive patients to determine whether a specific clinical subset of patients was linked to this serological pattern. Materials and Methods Patients and Controls Samples from 251 patients with MS [109 relapsing-remitting (RRMS), 67 secondary progressive (SPMS), and 75 transitional and primary progressive (PPMS)] from a cohort of 1,400 MS patients assisted at our outpatient MS unit and from 45 patients with clinically isolated syndromes (CIS) of the central nervous system suggestive of demyelinating disease were randomly selected (with a random-number generator) from our serum bank. Demographic features of these patients were reviewed. The MS patient population was From the 1Unitat de Neuroimmunologia Clı́nica, H. Universitari Vall d’Hebron, 2Hemotherapy and Hemostasis Department, Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Hospital Clı́nic, and 3Systemic Autoimmune Diseases Unit, IDIBAPS, Hospital Clı́nic, Barcelona, Spain. Received Jul 25, 2000, and in revised form Oct 30. Accepted for publication Nov 22, 2000. Address correspondence to Xavier Montalban, Unitat de Neuroimmunologia Clı́nica, H. Universitari Vall d’Hebron, Escola d’Infermeria, 5a planta, Passeig de la Vall d’Hebron 119 –129, 08035 Barcelona, Spain. E-mail: email@example.com Table 1. Neurological and PAPS-Related Manifestations in ACA and/or a␤2-GPI-Positive Patients Disease Duration (yr) Patient Number (age in yr/sex) ACA a␤2-GPI Clinical Course 1 (23/female) IgM⫹⫹⫹ IgM⫹⫹ RR 3 1.5 2 (58/male) IgM⫹ — RR 6 1 3 (33/female) 4 (48/female) IgM⫹ IgG⫹ — — RR SP 6 21 5 (40/male) 6 (50/male) IgG⫹ IgM⫹ IgG⫹⫹ IgM⫹⫹⫹ SP PP 5 5 EDSS Location of Symptomatic Lesion in Relapses Spinal cord/brainstem Optic nerve Spinal cord Brainstem Supratentorial Optic nerve Brainstem Supratentorial Brainstem Myelopathy (no relapses) 1.5 7.5 — 4 PAPS-Related and Other Symptomsa MS Therapy — IFN␤ — IFN␤ Migraine — IFN␤ — — — IVIG/AZA IFN␤1b vs placebo RR ⫽ relapsing-remitting; SP ⫽ secondary progressive; PP ⫽ primary progressive; EDSS ⫽ Expanded Disability Status Scale; PAPS ⫽ primary antiphospholipid syndrome; IFN␤ ⫽ interferon-beta; IVIG ⫽ intravenous immunoglobulin; AZA ⫽ azathioprine; ACA ⫽ anticardiolipin antibody; a␤2-GPI ⫽ anti-␤2/glycoprotein I antibody. a Thrombotic events and/or recurrent miscarriages. composed of 95 men and 156 women, and in the CIS group there were 22 men and 23 women. The mean age of both MS and CIS groups was 38.7 years [standard deviation (SD) 11.7 years]. Fifty-one serum samples from presumed healthy controls were picked from the blood bank of our hospital; there were 32 men and 19 women with a mean age of 41.4 years (SD 11.5 years). ACA, a␤2-GPI, and aPT Tests ACA were measured using standardized ELISA.11 Results were expressed in IgG and IgM units (GPL and MPL) and reported as negative (⬍15 units), low-positive (16 –25 units, ⫹), moderate positive (26 – 40 units, ⫹⫹), and high positive (⬎40 units, ⫹⫹⫹). Detection of a␤2-GPI was done by a standardized enzyme-linked immunosorbent assay (ELISA) previously described12; aPT were measured using an ELISA kit. Optical density at 492 nm (OD492) was measured for both aPT and a␤2-GPI assays. OD492 values higher than 5 SD above the mean of negative controls were considered positive, with low positive between 5 and 7 SD (⫹), moderate between 7 and 9 SD (⫹⫹), and high when above 9 SD (⫹⫹⫹). Results Laboratory Tests ACA or a␤2-GPI were found in 6 patients (Table 1). No patients were positive for aPT. No controls had positive titers for any of the antibodies tested. Clinical Findings in ACA- and/or a␤2-GPI-Positive Patients Table 1 summarizes neurological and PAPS-related manifestations in the 6 positive patients. There was no predominance of any type of clinical manifestation in relapses of patients with RRMS or SPMS. None of the patients presented thrombotic events, recurrent miscarriages, or livedo reticularis. No patient presented valvulopathy, pulmonary hypertension, or chorea, which are clinical manifestations thought to be related to PAPS. Table 2 summarizes paraclinical studies performed in the 6 positive patients. In all patients, visual, somatosensory, and brainstem auditory evoked potentials Table 2. Paraclinical Studies in ACA And/or a␤2-GPIab-Positive Patients Patient Number Altered Evoked Potentialsa MRI Suggestive of MS OCB aPTT ratio Syphilis Reaginic Serology ANA Platelet Count/mm3 1 2 3 4 5 SSEP SSEP and VEP None VEP SSEP, VEP, and BAEP SSEP and BAEP Yes Yes Yes Yes Yes ⫹ Np Np ⫺ Np N N N N N ⫹ ⫺ Np ⫺ ⫺ 1/40 — Np — — N N N N N Yes Np N ⫺ 1/80 N 6 ⫹ ⫽ Positive; ⫺ ⫽ negative; N ⫽ normal; Np ⫽ not performed; SSEP ⫽ somatosensory evoked potentials; VEP ⫽ visual evoked potentials; BAEP ⫽ brainstem auditory evoked potentials; MRI ⫽ magnetic resonance imaging; MS ⫽ multiple sclerosis; OCB ⫽ oligoclonal bands; aPTT ⫽ activated partial thromboplastin time; ANA ⫽ antinuclear antibodies. a All three kinds of evoked potentials were performed in all patients. Brief Communication: Sastre-Garriga et al: MS and Antiphospholipid Antibodies 409 (VEP, SSEP, and BAEP, respectively) were performed. All patients had MRI scans that fulfilled Paty criteria for MS.13 Two patients had antinuclear antibodies (ANA) in low titers (1/40 for patient 1, 1/80 for patient 6). Patient 1 had false-positive reaginic syphilis serology and oligoclonal bands (OCB). Discussion Our work shows that there is no higher prevalence of ACA than expected (based on our own control population) in the largest unselected MS patients population tested to date. A recent communication by Cuadrado et al7 supports the idea that it is not possible to distinguish the neurological manifestations of PAPS from those of MS, and so it is mandatory to test all MS patients for ACA routinely. This the authors concluded after examining 27 patients referred to their lupus clinic because of symptoms suggesting underlying connective tissue disease, MRI findings uncommon for MS, atypical evolution of MS, or ACA positivity when a diagnosis of MS had been made by a neurologist. We agree that in the face of such a clinical setting it is difficult to be satisfied with a diagnosis of MS; however, this should not be applied to all MS patients. From the work of Cuadrado et al, it might be expected that every patient diagnosed with MS would be under suspicion of being a symptomatic PAPS patient (because, as the authors conclude, it is impossible to distinguish between the two entities). As a consequence, every patient diagnosed with MS must be tested for ACA and, when positive, treated. Nevertheless, in their study, patients were well-grouped, and the suspicion of PAPS was correct. In 1998, Karussis et al10 screened a population of 70 classic and 100 “nonclassic” MS patients, labeling as “nonclassic” those with features unusual for MS. They found a strikingly significant proportion of ACApositive patients in the nonclassic MS group. Most of these positive patients in the nonclassic MS group had a similar clinical manifestations pattern, such as progressive myelopathy (which is really not surprising, insofar as patients were selected for this reason), spinocerebellar syndrome, or neuromyelitis optica. They concluded that this clinical subset of patients has ACA positivity as a defining feature and recommended performing this test in patients with MS showing such clinical features. On the other hand, D’Olhaberriague et al14 found no differences in ACA positivity between MS patients and patients with other neurological diseases. Tourbah et al15 found a 6.2% prevalence of ACA among 161 MS patients. Cordoliani et al16 found an 8% prevalence among 62 consecutive MS patients (other than the present study, this is the only study testing for ACA, a␤2-GPI, and aPT). Although with some controversy,17 there is wide 410 Annals of Neurology Vol 49 No 4 April 2001 agreement for therapeutic abstinence in asymptomatic patients with ACA antibodies.18 At this point we need not remind the reader that MS is an exclusion diagnosis or that patients with other neurological conditions may fulfill criteria for MS19 (for instance, a patient with two lacunar infarcts). Also, clinical criteria for MS were designed, as stated by their authors, as a guideline for research protocols. PAPS has been said to mimic MS.20 After careful reading of these case reports, it becomes clear that neurologists would never have taken some of the cases as suspected MS. Otherwise, clinical criteria for MS are useful when clinical suspicion is strong enough, and they should not be used to screen populations with relapsing neurological disturbances. Therefore, we must be cautious in making such a diagnosis when atypical features are present, and we must always seek other possible diagnoses, including PAPS. Now that a new set of diagnostic criteria for MS is under development, it might well be useful to include the clinical and paraclinical manifestations of PAPS (venous and arterial thrombosis and recurrent spontaneous abortions or fetal losses, low platelet counts, positive reaginic serologies, and high activated partial thromboplastin time ratios) as red flags that should prompt us to perform ACA and preferably a␤2-GPI tests. Our results show that it is not useful to test a large unselected MS population for ACA antibodies. The manifestations of PAPS are related more to a␤2GPI than to ACA.12 Positivity for a␤2-GPI was found in 3 patients only. In patient 6 (PPMS) SSEP and BAEP were pathologic, VEP were normal, and the MRI showed medullary and supratentorial lesions suggestive of MS; ANA were positive with a titer of 1/80, and OCB were not detected. The question therefore arises of whether this patient has actual symptomatic ACA with neurological manifestations or whether the patient has a PPMS plus asymptomatic ACA. Patient 1 has RRMS with 5 relapses and MRI highly suggestive of MS; a false-positive reaginic syphilis test was detected, and ANA were positive (titer 1/40); OCB were detected. No other relevant findings suggesting PAPS were encountered. In this case we are prone to think that we are dealing with RRMS with asymptomatic ACA. Patient 5 (SPMS) received intravenous immunoglobulin treatment for 6 days before blood was collected for serology, and the results are thus invalidated for the purpose of this study. We found no relationship between positivity for ACA antibodies and MS clinical profiles. In summary, our work, performed on the largest unselected MS population to date, supports the insight that ACA and, preferably, a␤2-GPI tests should be performed only when a suspicion of PAPS is raised and atypical clinical presentation for MS is ascertained. Screening for these antibodies in MS patients with no clinical or laboratory manifestations of PAPS is not useful. Furthermore, after careful analysis of the positive patients in our cohort, we believe that this positivity, when found in isolation, could be taken as incidental and so should not be treated. 17. The authors are indebted to Mr. J. Graells for help with the English. 19. 18. 20. References 1. Hughes GRV, Harris EN, Charavi AE. The anticardiolipin syndrome. J Rheumatol 1986;13:486 – 489. 2. Merkel PA, Chang Y, Pierangeli SS, et al. The prevalence and clinical associations of anticardiolipin antibodies in a large inception cohort of patients with connective tissue diseases. Am J Med 1996;101:576 –583. 3. Hughes GRV. Thrombosis, abortion, cerebral disease and the lupus anticoagulant. Br Med J 1983;287:1088 –1089. 4. McNeil HP, Simpson RJ, Chesterman CN, Krilis SA. Antiphospholipid antibodies are directed against a complex antigen that includes a lipid-binding inhibitor of coagulation: beta 2-glycoprotein I (apolipoprotein H). Proc Natl Acad Sci USA 1990;87:4120 – 4124. 5. Tsusumi A, Matsuura E, Ichikawa K, et al. Antibodies to beta 2-glicoprotein I and clinical manifestations in patients with systemic lupus erythematosus. Arthritis Rheum 1996;39:1466 – 1474. 6. Inanç M, Donohoe S, Ravijaran CT, et al. Anti-beta2glycoprotein I, antiprothrombin and anticardiolipin antibodies in a longitudinal study of patients with systemic lupus erythematosus and the antiphospholipid syndrome. Br J Rheumatol 1998;37:1089 –1094. 7. Cuadrado MJ, Khamashta MA, Ballesteros A, et al. Can neurological manifestations of Hughes (antiphospholipid) syndrome be distinguished from multiple sclerosis? Analysis of 27 patients and review of the literature. Medicine 2000;79:57– 68. 8. Ijdo JW, Conti-Kelly AM, Greco P, et al. Anti-phospholipid antibodies in patients with multiple sclerosis and MS-like illnesses: MS or APS? Lupus 1999;8:109 –115. 9. Fukuzawa T, Morikawa F, Mukai M, et al. Anticardiolipin antibodies in Japanese patients with multiple sclerosis. Acta Neurol Scand 1993;88:184 –189. 10. Karussis D, Leker RR, Ashkenazi A, Abramski O. A subgroup of multiple sclerosis patients with anticardiolipin antibodies and unusual clinical manifestations: do they represent a new nosological entity? Ann Neurol 1998;44:629 – 634. 11. Cervera R, Font J, López-Soto A. Isotype distribution of anticardiolipin antibodies in systemic lupus erythematosus: prospective analysis of a series of 100 patients. Ann Rheum Dis 1990; 49:109 –113. 12. Teixidó M, Font J, Reverter JC, et al. Anti-␤2glycoprotein I antibodies: a useful marker for the antiphospholipid syndrome. Br J Rheumatol 1997;36:113–116. 13. Paty DW, Oger JJ, Kastrukoff LF, et al. MRI in the diagnosis of MS: a prospective study with comparison of clinical evaluation, evoked potentials, oligoclonal banding, and CT. Neurology 1988;38:180 –185. 14. D’Olhaberriague L, Levine SR, Salowich-Palm L, et al. Specificity, isotype and titer distribution of anticardiolipin antibodies in CNS diseases. Neurology 1998;51:1376 –1380. 15. Tourbah A, Clapin A, Gout O, et al. Systemic autoimmune features and multiple sclerosis: a 5-year follow-up study. Arch Neurol 1998;55:517–521. 16. Cordoliani MA, Michon-Pasturel U, Rerat K, et al. Multiple sclerosis and antiphospholipid antibodies: study of 62 consecutive patients. Rev Med Intern 1998;19:635– 639. Martini A, Ravelli A. The clinical significance of antiphospholipid antibodies. Ann Med 1997;29:159 –163. Ordi-Ros J, Perez-Peman P, Monasterio J. Clinical and therapeutic aspects associated to phospholipid binding antibodies (lupus anticoagulant and anticardiolipin antibodies). Haemostasis 1994;24:165–174. Poser CM, Paty DW, Scheinberg L, et al. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol 1983;13:227–231. Scott TF, Hess D, Brillman J. Antiphospholipid antibody syndrome mimicking multiple sclerosis clinically and by magnetic resonance imaging. Arch Intern Med 1994;154:917–920. Low Cerebral Blood Flow Velocity and Risk of White Matter Hyperintensities Christophe Tzourio, MD, PhD,1 Claude Lévy, MD,2 Carole Dufouil, PhD,1 Pierre-Jean Touboul, MD,3 Pierre Ducimetière, PhD,4 and Annick Alpérovitch, MD, MSc1 Cerebral blood flow velocity (CBF-V) measured by transcranial doppler was assessed in 628 elderly individuals who had cerebral magnetic resonance imaging performed as part of a population-based study on vascular aging. Cerebral white matter hyperintensities (WMHs) were associated with low CBF-V, such as the adjusted odds ratios of severe WMHs from highest (referent) to lowest quartile of mean CBF-V were 1.0, 1.7, 3.7, and 4.3 ( p ⴝ 0.001). Further, CBF-V was found to be a stronger risk factor for WMHs than high blood pressure. These findings suggest that the assessment of CBF-V might be a powerful tool in future studies on WMHs. Ann Neurol 2001;49:411– 414 White matter bral magnetic sociated with disturbances,4 hyperintensities (WMHs) seen on cereresonance imaging (MRI) have been ascognitive decline and dementia,1–3 gait depression,5 and risk of future stroke.6 From the 1INSERM (National Institute for Health and Medical Research) U360, Hôpital Pitié-Salpêtrière, Paris; 2Department of Radiology Institut hospitalier Jacques Cartier, Massy; 3Department of Neurology Hôpital Lariboisière, Paris; and 4INSERM U258, Hôpital Paul Brousse, Villejuif, France. Received Jul 1, 2000, and in revised form Nov 11. Accepted for publication Nov 11, 2000. Address correspondence to Dr Tzourio, INSERM U360, Hôpital de la Salpêtrière, 75651 Paris cedex 13, France. E-mail: firstname.lastname@example.org Brief Communication: Tzourio et al: Cerebral Blood Velocity and White Matter Hyperintensities 411 They have been shown to be related to vascular risk factors—in particular hypertension—and are generally interpreted as a marker of cerebral ischemia due to diffuse small vessel disease.7,8 However, despite their importance, little is known about the etiology of WMHs. It has been suggested that changes in cerebral blood flow might play a role in the occurrence of WMHs.7,8 Transcranial doppler is an established, non-invasive method of assessing cerebral hemodynamics through the measurement of cerebral blood flow velocities (CBF-V). The aim of the study presented here was to analyze the relationship between WMHs and CBF-V in a large sample of elderly individuals from the general population. Methods Subjects The EVA (Epidemiology of Vascular Aging) study has already been described previously.9 The sample included 834 participants (484 [58%] women, mean age [SD] of 68.9 years [2.9], age range, 63 to 75 years) who participated in the 4-year follow-up of the EVA study and had a cerebral MRI. MRI Cerebral MRI was performed using a 1.0 T scanner (Siemens, Munich, Germany). WMHs were read on the T2 weighted acquisition, including a fast multislice double echo T2-weighted axial acquisition (repetition time [TR] ⫽3500 msec, echo time [TE] ⫽ 140 msec) and a proton density axial acquisition (TR ⫽ 3500 msec, TE ⫽ 85 msec) with 26 slices 5 mm thick. The images were rated visually with respect to the presence of WMHs using a modified version of the scale of Sheltens et al.10 It provided an overall WMHs grade ranging from A (no lesion) to D. All the ratings were done by a trained medical doctor who was blind to any clinical data or diagnoses. After the rater had read all the MRIs, 100 MRIs were randomly selected and reread by the same rater. The intra-reader kappa coefficient obtained for the overall grading was 0.76. Because only 12 subjects had no WMH (grade A), it was decided to pool grade A and B. In the rest of the paper the three grades will be denominated mild (grade A and B), moderate (grade C), and severe (grade D). Transcranial Doppler Transcranial doppler was accepted by all participants. Exams were performed by four sonographers with a 2-MHz pulsed wave probe (Transflo, Atys Medical, France). The study was performed in a quiet room, with participants lying in a supine position without any visual or auditory stimulation. Middle cerebral artery was insonated on both sides and the mean of both CBF-V values was used in the analyses. CBF-V was missing in 206 participants (24.7%) because of the absence of bilateral temporal window. Participants with missing CBF-V were older (69.6 [2.9] vs 68.7 [2.9]; p ⬍ 0.001), more frequently women (76.7% vs 51.9%; p ⫽ 0.006), and had more severe WMHs (23.3% vs 15.0%; p ⫽ 0.006) than the rest of the population. There were no other significant differences between both groups concerning body mass index, systolic or diastolic blood pressure, and hypertension. One of the sonographers did not report correctly mean velocity and these values were not taken into account in the analyses. Common carotid artery intima-media thickness was measured with a 7.5-MHz transducer (SSD650, Aloka, Tokyo, Japan), as previously described.11 Statistical Analysis The association between WMHs and mean CBF-V was expressed in terms of odds ratio (OR) estimated by quartile of CBF-V. Multiple logistic regression analysis was used to evaluate the association after adjustment on potential confounders by a backward elimination procedure. The 95% confi- Table 1. Clinical Characteristics of Participants, According to WMHs Grade Grade of WMHs Women (%) Age, years (SD) Body mass index, kg/m2 (SD) Systolic BP, mmHg (SD) Diastolic BP, mmHg (SD) Hypertensionb (%) Tobacco smokingc (%) Diabetes (%) Intima-media thicknessd, mm (SD) Mild (n ⫽ 251) Moderate (n ⫽ 283) Severe (n ⫽ 94) pa 49.8 68.3 (3.0) 25.9 (3.6) 133.3 (18.0) 76.6 (10.2) 34.7 47.4 5.3 6.9 (1.1) 54.1 68.9 (2.7) 25.6 (3.8) 136.0 (18.0) 77.0 (11.0) 34.3 41.7 8.2 7.0 (1.2) 51.1 69.2 (3.1) 25.8 (3.3) 138.1 (19.5) 79.1 (10.5) 48.9 46.8 7.5 7.2 (1.1) 0.61 0.010 0.56 0.016 0.08 0.027 0.38 0.41 0.042 p for trend, obtained by regression analysis for quantitative variables and with Mantel-Haenszel 2 statistic for qualitative variables. b Defined as a systolic BP ⱖ 160 mmHg or a diastolic BP ⱖ 95 mmHg or being under antihypertensive treatment. c Present or ex-smokers. d Common carotid artery intima-media thickness measured longitudinally. WMHs ⫽ white matter hyperintensities; BP ⫽ blood pressure. a 412 Annals of Neurology Vol 49 No 3 March 2001 Table 2. Cerebral Blood Flow Velocities According to Grade of WMHs Grade of WMHs Transcranial doppler variables Mild (n ⫽ 251) Moderate (n ⫽ 283) Severe (n ⫽ 94) pa Systolic velocity, m/s (SD) Diastolic velocity, m/s (SD) Mean velocity, m/s (SD) Resistance index (SD) Pulsatility index (SD) 69.9 (11.5) 30.7 (8.2) 46.7 (10.4) 0.56 (0.10) 0.86 (0.23) 68.7 (11.8) 28.5 (7.8) 44.1 (9.6) 0.59 (0.10) 0.91 (0.23) 64.2 (10.8) 27.1 (8.0) 41.4 (10.4) 0.58 (0.10) 0.94 (0.30) ⬍0.001 ⬍0.001 ⬍0.001 0.023 0.017 a p for linear trend. WMHs ⫽ white matter hyperintensities. dence interval (CI) are given for each OR, and all p values are 2-tailed. Statistical analyses were performed using the SAS 8.0 software package (SAS Inc. Cary, NC). Results In Table 1 clinical characteristics of the participants are indicated by grade of WMHs. There was a strong positive association between age and grade of WMHs. Frequency of hypertension, levels of systolic and diastolic blood pressure, and common carotid artery intimamedia thickness also increased with increasing grade of WMHs (Table I). Table 2 shows the relationship between grade of WMHs and cerebral blood flow velocities and indexes. There was a highly significant negative association between velocities and grade of WMHs: the higher the grade the lower were the velocities. There was also a significant trend of increasing resistance and pulsatility indexes with WMHs grade. With regard to potential confounders, age, hypertension, diabetes, body mass index, and hematocrit were significantly associated with lower velocities and were therefore included in multivariate models (data not shown). ORs of moderate and severe WMHs by quartiles of Table 3. Risk of Moderate and Severe WMHs According to Quartiles of CBF-V Risk of Moderate WMHs Risk of Severe WMHs Quartiles of mean CBF-Va ORb (95% CI) p ORb (95% CI) p 0c 1 2 3 1 1.4 (0.8–2.3) 1.6 (0.9–2.7) 2.3 (1.3–4.1) 1 1.7 (0.7–4.2) 3.7 (1.6–8.8) 4.3 (1.8–10.6) — 0.22 0.11 0.006 — 0.22 0.003 0.001 a Decreasing values of mean CBF-V stratified by quartiles; range of velocities for each quartile are as follows: ⬎51 m/s; 45.5 to 51 m/s; 38.5 to 45 m/s; ⬍38.5 m/s. b Adjusted for age, sex, BMI, hypertension, diabetes, hematocrit, intima-media thickness. c Reference category. WMHs ⫽ white matter hyperintensities; CBF-V ⫽ cerebral blood flow velocity; OR ⫽ odds ratio; BMI ⫽ body mass index. mean velocity were estimated in multiple logistic regression models adjusting for potential confounders (Table 3). The risks of moderate and severe WMHs increased continuously with decreasing mean velocity. The adjusted ORs of severe WMHs from highest (referent) to lowest quartile of mean CBF-V were 1.0, 1.7, 3.7, and 4.3 ( p ⫽ 0.001). We then compared the effect of age, systolic blood pressure, and mean CBF-V on the risk of WMHs by comparing the ORs of severe WMHs for the highest versus the lowest quartile for each variable. For each variable, the ORs estimated in separate models were all significant: OR ⫽ 2.4 (95% CI, 1.4 – 4.2 ; p ⫽ 0.003) for age, 2.1 (95% CI, 1.1– 4.1; p ⫽ 0.023) for systolic blood pressure, and OR ⫽ 4.3 (95% CI, 1.8 –10.6; p ⫽ 0.001) for mean CBF-V. When age, systolic blood pressure, and mean CBF-V were included in the same model, the only OR remaining significant was for mean CBF-V: OR ⫽ 4.7 (95% CI, 1.7–12.6; p ⫽ 0.002). Age and systolic blood pressure were no more associated with the risk of severe WMHs. Discussion In this population-based study we found that WMHs were strongly associated with CBF-V. The adjusted ORs of severe WMHs from highest (referent) to lowest quartile of mean CBF-V were 1.0, 1.7, 3.7, and 4.3 ( p ⫽ 0.001). This relationship was independent of age and hypertension and seems to explain in part the association between these variables and WMHs. Previous cross-sectional studies have shown that patients with WMHs had a decreased cerebral blood flow12 or an impaired cerebral autoregulation.13,14 However, because of the complex methods used (single photon emission computed tomography or positron emission tomography scan) these studies were done on small samples of selected patients. The CBF-V is an indirect estimate of actual blood flow,15 and the equivalence between both variables, although suggested by some studies,15–17 have been challenged by others.18,19 It has been shown that, when interindividual variability of CBF-V was reduced, changes in CBF-V reflect ac- Brief Communication: Tzourio et al: Cerebral Blood Velocity and White Matter Hyperintensities 413 curately changes in cerebral blood flow.15,18 It can therefore be assumed that in a large, population-based group of individuals with ultrasonic examinations performed in the same conditions, CBF-V is a good surrogate of cerebral blood flow and that the differences of CBF-V observed between grades of WMHs actually reflect differences of cerebral blood flow.15 The relationship observed between CBF-V and WMHs is only an association, the direction of which remains speculative. The most plausible explanation is that the hypertension-related small vessel disease causes a decrease in CBF-V and cerebral blood flow, which would contribute to the occurrence of WMHs. Other explanations may however be considered. It could be suggested that the differences of CBF-V between grades of WMHs were actually partly due to cognitive decline. The reduced CBF-V in patients with severe WMHs would be related to a reduced metabolic demand in subjects having a cognitive deterioration.20 However, when cognitive measures like the MiniMental State Examination, trail making test, or Wechsler test were included in the models the relationship between CBF-V and WMHs was unchanged (data not shown). Further, when participants with severe cognitive decline were excluded,9 the adjusted OR of severe WMHs for the quartile of lowest mean velocity was 3.7 ( p ⫽ 0.005), very similar to what was observed in the full sample. These data therefore suggest that the relationship between CBF-V and WMHs is not related to cognitive decline, although this could be definitively established only in longitudinal studies. Apart from pathophysiological considerations, our study has shown that CBF-V is a stronger risk factor for WMHs than age and high blood pressure, which are the most important known risk factors for WMHs. These results suggest that transcranial doppler may be a powerful tool to help to identify groups of patients exposed to the risk of WMHs and to the associated risk of cognitive decline,1–3 and future stroke.6 They also raise the possibility that therapeutic interventions aiming at reducing the risk of WMHs in hypertensives or other high-risk groups of individuals could be selected on their effect on CBF-V. This will be best explored in longitudinal studies and therapeutic trials. The EVA study was carried out under an agreement between the INSERM and the Merck, Sharp and Dhome-Chibret Laboratories (West-Point, PA). It is also supported by the Eisai Company (France). The authors are grateful to Dr Pierre Démolis for his helpful comments during the analysis of the data. We thank the ultrasound physicians, Drs J.M. Fève, C. Leroux, I. Ruelland, and C. Magne, who performed the examinations, and Dr V. Besançon for help with the interpretation of the MRIs. 414 Annals of Neurology Vol 49 No 4 April 2001 References 1. Longstreth WT, Manolio TA, Arnold A, et al. Clinical correlates of white matter findings on cranial magnetic resonance imaging of 3301 elderly people: the cardiovascular health study. Stroke 1996;27:1274 –1282. 2. Steingart A, Hachinski V, Lau C, et al. Cognitive and neurologic findings in subjects with diffuse white matter lucencies on computed tomographic scan (leuko-araiosis). Arch Neurol 1987;44:32–35. 3. de Groot JC, deLeeuw FE, Oudkerk M, et al. Cerebral white matter lesions and cognitive function: The Rotterdam Scan Study. Ann Neurol 2000;47:145–151. 4. van Zagten M, Lodder J, Kessels F. Gait disorder and parkinsonian signs in patients with stroke related to small deep infarcts and white matter lesions. Mov Disord 1998;13:89 –95. 5. Hickie I, Scott E, Wilhelm K, Brodaty H. Subcortical hyperintensities on magnetic resonance imaging in patients with severe depression—a longitudinal evaluation. Biol Psychiatry 1997;42:367–374. 6. Kobayashi S, Okada K, Koide H, et al. Subcortical silent brain infarction as a risk factor for clinical stroke. Stroke 1997;28: 1932–1939. 7. Roman GC. From UBOs to Binswanger’s disease: impact of magnetic resonance imaging on vascular dementia research. Stroke 1996;27:1269 –1273. 8. Pantoni L, Garcia JH. Pathogenesis of leukoaraiosis: a review. Stroke 1997;28:652– 659. 9. Tzourio C, Dufouil C, Ducimetiere P, Alpérovitch A. Cognitive decline in individuals with high blood pressure: a longitudinal study in the elderly. Neurology 1999;53:1948 –1952. 10. Scheltens P, Barkhof F, Leys D, et al. A semiquantative rating scale for the assessment of signal hyperintensities on magnetic resonance imaging. J Neurol Sci 1993;114:7–12. 11. Bonithon-Kopp C, Touboul PJ, Berr C, et al. Relation of intima-media thickness to artherosclerotic plaques in carotid arteries. The Vascular Aging (EVA) Study. Arterioscler Thromb Vasc Biol 1996;16:310 –316. 12. Hatazawa J, Shimosegawa E, Satoh T, et al. Subcortical hypoperfusion associated with asymptomatic white matter lesions on magnetic resonance imaging. Stroke 1997;28:1944 –1947. 13. Isaka Y, Okamoto M, Ashida K, Imaizumi M. Decreased cerebrovascular dilatory capacity in subjects with asymptomatic periventricular hyperintensities. Stroke 1994;25:375–381. 14. Matsushita K, Kuriyama Y, Nagatsuka K, et al. Periventricular white matter lucency and cerebral blood flow autoregulation in hypertensive patients. Hypertension 1994;23:565–568. 15. Sorteberg W. Cerebral artery blood velocity and cerebral blood flow. In: Newell DW, Aaslid R, eds. Transcranial doppler. New York: Raven Press 1992:57– 66. 16. Sorteberg W, Lindegaard KF, Rootwelt K, et al. Blood velocity and regional blood flow in defined cerebral artery systems. Acta Neurochir 1989;97:47–52. 17. Sugimori H, Ibayashi S, Fujii K, et al. Can transcranial Doppler really detect reduced cerebral perfusion states? Stroke 1995; 2053–2060. 18. Bishop CC, Powell S, Rutt D, Browse NL. Transcranial Doppler measurement of middle cerebral artery blood flow velocity: a validation study. Stroke 1986;17:913–915. 19. Kontos HA. Validity of cerebral arterial blood flow calculations from velocity measurements. Stroke 1989;20:1–3. 20. Franceschi M, Alberoni M, Bressi S, et al. Correlations between cognitive impairment, middle cerebral artery flow velocity and cortical glucose metabolism in the early phase of Alzheimer’s disease. Dementia 1995;6:32–38.