~~.-~J ~.;i;,~....i·i ;., ' FLSF\11:R The Functional Neuroanatomy of Tourette's Syndrome: An FDG-PET Study. II: Relationships between Regional Cerebral Metabolism and Associated Behavioral and Cognitive Features of the Illness Allen R. Braun, M.D., Christopher Randolph, Ph.D., Brigitte Stoetter, M.D., Erich Mohr, Ph.D., Christine Cox, Ph.D., Katalin Vladar, M.D., Ro1/ Sexton, B.S., Richard E. Carson, Ph.D., Peter Hcrscovitclz, M.D., and Thomas N. ·chase, M.D. We :malyzed F-18 Jluoro-deoxygliwisc PET scans earned stria/um), paralimbic (parahippocampal gyrus), or out in 18 drug-free patients ,l'ith Tourette's S!fndrome srnsorimotor regions (supplementary motor area, lateral (TS1 in order to evaluate relationships between cerebral premotor or Rolandic cortices), in which metabolism had, metabolism and complex cognifrue and behauioral _features in some cases more robustly, distinguished these TS commonly associated with th is disorder. These features patients from controls (Braun et al., 1993). These results (obsessions and compulsions, impulsiuity, coprola/ia, suggest that a subset of regions in which metabolic self-injurious behavior, echophenomcna, depression, and activity appears to be associated with the diagnosis of TS measures of attentional and u1s1wspatial dysfunction! per sc, may be explicitly associated with the emergence of wert.' associated with significant increases in metaboli( cumplex behavioral and cognitive features of the illness. acti:1ity in the orbitofrontal cortices. Similar increases. Th is is most conspicuous in the orbitofrontal cortices, and although less robust, were obsen 1cd in the puta111e11 and, 11 i~ consistent with the observation that these features in t1,e case of attentional and l'isuospatial mcasu res, 111 resemble tlze elements of a behavioral syndrome typically the inferior portions of the insula. On the other hand. seen in patients with lesions of the orbitofrontal cortex. behavioral and cognitiI 1e features ,1 1crc not associated ,(li//1 {Neuropsychopharmacology 13:151-168, 1995] metabolic rates in other subcortica/ (midbraln, z,entra/ Kl\ WORDS: Tourette's (Tourette) 51/ndrome; Positron Fr.im the VSS, YSLB. NlDCD (Al\fl 1, the L rn. \:ll\.DS \ BS, li\.l 1 and the PET department (REC, ['1-l). l\.c1tional Institutes of Ht·dlth rnussim1 tomography; FOG; obsessive-compulsive Bethesda, Maryland; the depc1rtments of \Jeurolug\ and Psvchiatn disorder; Impulsivity; Self-injurious behavior; (CR:. Northwestern Universitv School of \1edicine, Chicago, Illino1s Depression; Coprolalia; Echophenomena; Attention; the Institute uf Mental Health Research (EM I. Universit\· nf Otta\\ a \'isuospatial; Orbitofrontal cortex; Insula; Basal ganglia Canada; the Kennedy Krieger Institutt' (CCI Baltimore. \1arvland and the CBDB, NIMH Neuroscience· C,,nlt'r at C,t Eli,abeths 1K\ RS), Washington. DC Sinct' tht' disorder was initially described (Gilles de la The 1993 part I Braun t't al. arti<"k a~'P<'M<'d in tht· Y( 4) 1ssut> ot \,·,, rop,1clwpharmacolog.~ as pages 277 2Y I (sec· u,mplell' refrrencc·I and fourettt' 1885), it has been clear that the spectrum of as of Octob<'r 1995, the part Ill manuscript is in preparation wmptoms in Tourette's syndrome (TS) includes com­ Address correspondence to Dr Allen R. Braun. I\JID( D. :'.:IH. Huild ple, neurobehavioral and cognitive disturbances in ad­ ing 10, Room 'iD-38, Bethesda MD 208Ll2 RPceived June 20. !YY4; re\ iscd \1,mh I )Yli~. d<c<'pkd \l<1r, h, dition to simple vocal and motor tics. These symptoms 0 19Y , include obsessions and compulsions, impulsivity, self- NlLR()J'½Y(H()f'H\KMA(()[()(,\ I·'"-- 1,11 ,,, © }(195 An1erlcan (_ "ullegt' ut '\Jl'lH( )P'-,\ l hur1h,ll lll,h [ dog\ l'uhlished b\· Et,,,,·icr Sncncc· In, 0893-133Xi95/$9.50 l'f,, '\, \ 111 SSDI 0893-133X(95)00052-F 152 A.R. Braun et a.I. NEUROPSYCHOPHARMACOLOGY 1995-VOL. 13, NO. 2 injurious behavior, echophenomena (echolalia, palila­ PET Scans lia, echopraxia), attentional disturbances that may be PET scans were performed on the NeuroPET (Brooks associated with hyperactivity, complex, socially disrup­ et al. 1980), a seven-slice positron tomograph with an tive tics (such as coprolalia), and mood lability (Com­ in-plane resolution of 6 to 7 mm and an axial resolution ings and Comings 1988; Riddle et al., 1988; Towbin, of 11 to 12 mm FWHM, according to a protocol described 1988). In addition, patients may experience anxiety, previously (Braun et al. 1993). Five mCi of FDG were sleep disturbances, learning disabilities, and second­ injected intravenously over a period of 1 minute, and ary behavioral problems such as social isolation (Glaze 10-minute scans were initiated 30 to 45 minutes after et al., 1983; Hagin and Kugler, 1988; Silver, 1988). infusion. Between each scan, the position of the pa­ The anatomical substrate of motor symptoms in TS tient's head was shifted slightly within the gantry to is poorly understood. The pathophysiology of the as­ increase the anatomical sampling by obtaining inter­ sociated behavioral features is even more obscure. The leaved image sets. A calculated attenuation correction initial results of our FDG-PET study (Braun et al. 1993) was performed using a visually fitted outline of the suggested that the TS brain may be characterized by scalp. The scanner and gamma well counter were cross­ lower metabolic rates in the inferior, limbic-related calibrated using a phantom containing a solution of regions of the cortex and striatum, and in the midbrain, I F-18] fluoride. Local cerebral metabolic rates for glu­ and higher metabolic rates in the superior, sensorimo­ cose were calculated according to the method of Brooks tor cortices. This may represent a metabolic pattern as­ (1982). Patients were asked to remain still during the sociated with the diagnosis of TS per se. Which of these course of the procedure. Although they were not ex­ regions might participate in the generation of the be­ plicitly instructed to suppress symptoms, vocal or mo­ havioral and cognitive symptoms is not clear. However, tor tics were rarely observed during the period of FDG our patient group was heterogeneous, possessing a uptake. wide enough dynamic range of behavioral and cogni­ tive features to make it possible to address this question. We therefore used clinical, neuropsychological and Image Analysis FDG-PET data to identify patterns of regional activity Eight image planes were selected for analysis on the that might be associated with specific neurobehavior~l basis of recognizable cerebral landmarks identified in and cognitive symptoms in TS. The identification of standard anatomical atlases (Damasio and Damasio functional neuroanatomical patterns that correspond 1989; Duvernoy 1991; Matsui and Hirano 1989). Image to the various phenotypic expressions of this unusually planes were parallel to the canthomeatal line and ex­ heterogeneous disorder might suggest further avenues tended from the level of the cerebellum, temporal pole, of research that could potentially lead to successful and gyms rectus inferiorly to the superior frontoparie­ treatment. tal convexities. In our earlier report, two separate anal­ vses, each based on a different technical strategy, were performed, and these provided essentially convergent METHODS results. For this reason, only regional values derived using one of these methods were utilized in the pres­ Subjects ent study. Irregular regions of interest were identified Eighteen TS patients, 16 males and 2 females age 33 ± in the cortex using thresholding and edge-detection 7 years (mean ± SD; range 23-49 years) consented tu techniques; circular regions of interest (ROis) were ap­ participate in this study after providing informed con­ plied to subcortical structures and cortical regions in sent. Sixteen patients were right-handed, and two, one \Vhich anatomical detail was less precise, and data male and one female, were left-handed. reduction was performed in these instances as previ­ The subjects included the same cohort of 16 patients ously described (Braun et al. 1993). analyzed in our previous report as well as two patients In our initial report, the results of both independent from whom blood samples were not taken. These sub­ analyses showed the highest degree of convergence jects were included in the present analysis because onlv when metabolic rates were normalized, controlling for normalized metabolic rates were evaluated. intersubject differences in CMRglu and thereby reduc­ The diagnosis of TS was confirmed by history and ing group variances. For this reason, normalized values physical examination that excluded concurrent medi­ were utilized in the present analysis. Regional meta­ cal or other neurological or psychiatric illnesses and con­ bolic rates were divided by global metabolic rates; global formed to DSM-IV criteria. All patients were drug free rates were estimated by averaging all regional grey mat­ at the time of scanning, medications having been with­ ter metabolic rates, weighting these for the area en­ drawn at least 2 weeks prior to the study (Braun et al. closed within each ROI. For the analyses summarized Jq93). in Tables 3-6, normalized regional values were further NEUROPSYCHOPHARMACOLO(,Y IYlJ'i-VllL. I,, NO. 2 TS, Behavioral-Cognitive Features and PET 153 reduced by averaging metabolic rates in functionally 6. Depression (significantly depressed mood without related ROis (e.g., medial orbital cortex contained in psychotic features lasting at least 2 weeks, which planes 2 and 3), again weighting these for the area en­ may have represented a dysphoric response to neu­ closed within each region. roleptic medications). Although the criteria for depression and obsessive­ compul: ive behaviors do not strictly conform to DSM Clinical Evaluation [V criteria, they identify several behavioral features, such as dysphoric response to medications, obsessive Within 1 week of PET scanning, all patients underwent or compulsive symptoms that do not cause distress, that a formal neurological evaluation that included a physi­ are frequently seen in TS patients. Attentional para­ cal examination, structured historical interview, and re­ meters were not evaluated in this portion of the study view of medical records.