Motor symptoms

TS, the most common cause of tics, is manifested by a broad spectrum of motor and behavioral disturbances (Table 16.3). This clinical heterogeneity often causes diagnostic difficulties and presents a major challenge in genetic linkage studies. To aid in the diagnosis of TS, the Tourette Syndrome Classification Study Group (1993) formulated the following criteria for definite TS: (1) Both multiple motor and one or more phonic tics have to be present at some time during the illness, although not necessarily concurrently. (2) Tics must occur many times a day, nearly every day, or intermittently throughout a period of more than 1 year. (3) The anatomic location, number, frequency, type, complexity, or severity of tics must change over time. (4) Onset must be before age 21. (5) Involuntary movements and noises cannot be explained by other medical conditions. (6) Motor and/or phonic tics must be directly witnessed by a reliable examiner at some point during the illness or must be recorded by videotape or cinematography. Probable TS type 1 meets all the criteria except for number 3 and/or number 4, and probable TS type 2 meets all the criteria except for number 1; it includes either a single motor tic with phonic tics or multiple motor tics with possible phonic tics. In contrast to the criteria outlined by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) (1994), the Tourette Syndrome Classification Study Group criteria do not include a statement about “impairment.” There is considerable controversy about the DSM-IV criteria, which require that “marked distress or significant impairment in social, occupational or other important areas of functioning” be present. Therefore, patients with mild tics that do not produce an impairment would not satisfy the diagnostic criteria for TS, according to DMS-IV. This criterion will be deleted from the fifth edition of the DSM. Kurlan (1997) suggested another set of diagnostic criteria for genetic studies and introduced the term Tourette disorder for patients who have “functional impairment.” This, however, does not take into account the marked fluctuation in symptoms and severity; some patients may be relatively asymptomatic at one time and clearly functionally impaired at another time. The TSA International Genetic Collaboration developed the Diagnostic Confidence Index, which consists of 26 confidence factors, with weightings given to each of them and a total maximum score of 100. The most highly weighted diagnostic confidence factors include history of coprolalia, complex motor or vocal tics, a waxing and waning course, echophenomenon, premonitory sensations, an orchestrated sequence, and age at onset. The Diagnostic Confidence Index was found to be a useful instrument in assessing the lifetime likelihood of TS (Robertson et al., 1999). Several instruments, some based on ratings of videotapes, have been developed to measure and quantitate tics, but they all have some limitations (Goetz and Kompoliti, 2001; Goetz et al., 2001). The most widely used instrument to assess tics is the Yale Global Tic Severity Scale (YGTSS), which consists of two broad domains: Total Tic Severity (with two sub-domains: Motor and Phonic Tics) and Impairment. Within each category there are five dimensions, score 0–5: number of tics, frequency, intensity, complexity, and interference. The total tic score ranges from 0 to 50; the usual ranges in most studies is 15 to 30. A health-related quality of life scale (HR-QOL) has been developed and validated for internal consistency, test–retest reliability and against other clinical scales (Cavanna et al., 2008). Utilization of the HR-QOL scale showed that comorbidities such as ADHD and OCD rather than tic severity are more predictive of the long-term outcome.

Table 16.3 Demographic and clinical features in a large database of patients with TS (Tourette International Consortium) (Freeman et al., 2009)

The clinical criteria are designed to assist in accurate diagnosis, in genetic linkage studies and in differentiating TS from other tic disorders (Jankovic, 1993b) (Table 16.2). There is a body of evidence to support the notion that many, if not all, patients with other forms of idiopathic tic disorders represent one end of the spectrum in a continuum of TS (Kurlan et al., 1988). The most common and mildest of the idiopathic tic disorders is the transient tic disorder (TTD) of childhood. This disorder is essentially identical to TS except the symptoms last less than 1 year and therefore the diagnosis can be made only in retrospect. Transient tic disorder has been estimated to occur in up to 24% of schoolchildren (Shapiro et al., 1988). Chronic multiple tic disorder is also similar to TS, but the patients have either only motor or, less commonly, only phonic tics lasting at least 1 year. Chronic single tic disorder is the same as chronic multiple tic disorder, but the patients have only a single motor or phonic tic. This separation into transient tic disorder, chronic multiple tic disorder, and chronic single tic disorder seems artificial because all can occur in the same family and probably represent a variable expression of the same genetic defect (Kurlan et al., 1988).

Although the diagnostic criteria require that the onset is present before the age of 21, in 96% of patients the disorder is manifested by age 11 (Robertson, 1989). In 36–48% of patients, the initial symptom is eye blinking, followed by tics involving the face and head. Blink rate in TS is about double of that of normal, age-matched controls (Tulen et al., 1999). During the course of the disease, nearly all patients exhibit tics involving the face or head, two-thirds have tics in the arms, and half have tics involving the trunk or legs. According to one study, the average age at onset of tics is 5.6 years, and the tics usually become most severe at age 10; by 18 years of age, half of the patients are tic-free (Leckman et al., 1998). In a study of 58 adults who had been diagnosed with TS during childhood, Goetz and colleagues (1992) found that tics persisted in all patients but were moderate or severe in only 24%, although 60% had moderate or severe tics during the course of the disease. Tic severity during childhood had no predictive value for the future course, but patients with mild tics during the preadult period had mild tics during adulthood. In a longitudinal study that involved structured interviews of 976 children aged 1–10 years, 776 of whom were reassessed 8, 10, and 15 years later, tics and ADHD symptoms were associated with OCD symptoms in late adolescence and early adulthood (Peterson et al., 2001). Furthermore, ADHD was associated with lower IQ and lower social status, whereas OCD was associated with higher IQ. These findings are similar to those of another study designed to address the long-term prognosis of children with TS as they reach adulthood (Bloch et al., 2006). In this study 46 children with TS underwent a structured interview at a mean age of 11.4 years, and again at 19.0 years. The mean worst-ever tic severity score was 31.6 out of a possible 50 on the YGTSS, and occurred at a mean age of 10.6 years. By the time of the second interview, mean YGTSS score decreased to 10. This first prospective longitudinal study also showed that only 22% continued to experience mild or greater tic symptoms (YGTSS scores, ≥10) at follow-up, while nearly one-third were in complete remission of tic symptoms at follow-up. In contrast to the study by Goetz et al. (1992), the severity of childhood tics was predictive of increased tic severity at follow-up. The peak OCD severity occurred 2 years after peak tic severity. Interestingly, a 10-point increase in baseline IQ increased the risk of OCD symptoms at follow-up by 2.8-fold. The authors point out that the later average onset of OCD symptoms indicates the importance of counseling parents about the possibility of OCD development in children recently diagnosed with tics. Although the long-term prognosis for TS is generally favorable for most patients, a minority of cases may have persistent, severe tic symptoms which may be resistant to medications (Eapen et al., 2002). In another study, the investigators reviewed videotapes of 31 patients, with an average age of 24.2 ± 3.5 years, approximately 12 years after their initial video and found that 90% of the adults still had tics, even though they often considered themselves tic-free (Pappert et al., 2003). There was, however, a significant improvement in tic disability and tic severity. When 40 children and 31 adults with TS were compared no difference in tic phenomenology or severity was found, but children were more frequently managed without medications, and sedation was more common in adults but weight gain was more common in children (Cubo et al., 2008).

Although the vast majority of tics in adults represent recurrences of childhood-onset tics, rare patients may have their first tic occurrence during adulthood (Chouinard and Ford, 2000). In adults with new-onset tics, it is important to search for secondary causes, such as infection, trauma, stroke (Kwak and Jankovic, 2002; Gomis et al., 2008), multiple sclerosis (Nociti et al., 2009), cocaine use, neuroleptic exposure, and peripheral injury (Video 16.14) (Chouinard and Ford, 2000; Jankovic, 2001b; Jankovic and Mejia, 2006). One study of eight patients with adult-onset tics (three of whom had childhood-onset OCD and three of whom had a family history of tics and OCD) found that in comparison to the patients with more typical childhood-onset tics, the former group had more severe symptoms, greater social morbidity, and less favorable response to medications (Eapen et al., 2002). Poor motor control, which can lead to poor penmanship and, at times, almost illegible handwriting, can contribute to the academic difficulties faced by many patients with TS.

Tics, although rarely disabling, can be quite troublesome for TS patients because they cause embarrassment, interfere with social interactions, and at times can be quite painful or uncomfortable. Rarely, cervical tics may be so forceful and violent, the so-called “whiplash tics,” that they may cause secondary neurologic deficits, such as cervical artery dissection (Norris et al., 2000), and noncompressive (Isaacs et al., 2010) or compressive cervical myelopathy (Krauss and Jankovic, 1996) (Video 16.2). The truncal bending tics, which resemble intermittent, repetitive camptocormia, may cause secondary degenerative changes in the thoracic spine (Azher and Jankovic, 2005) (Video 16.5). These disabling tics and other severe symptoms of TS draw attention to the subset of patients with TS so severe that they may be life-threatening, hence labeled as “malignant” TS (Table 16.4). Of 332 TS patients evaluated at Baylor College of Medicine Movement Disorders Clinic during a 3-year period, 17 (5.1%) met criteria for malignant TS, defined as ≥2 emergency room (ER) visits or ≥1 hospitalizations for TS symptoms or its associated behavioral comorbidities (Cheung et al., 2007). The patients exhibited tic-related injuries, self-injurious behavior (SIB), uncontrollable violence and temper, and suicidal ideation/attempts. Compared to patients with nonmalignant TS, those with malignant TS were significantly more likely to have a personal history of obsessive-compulsive behavior/disorder (OCB/OCD), complex phonic tics, coprolalia, copropraxia, SIB, mood disorder, suicidal ideation, and poor response to medications. Severe or malignant TS, associated with SIB and other disabling features, has been also reported in families, including consanguineous kindreds (Motlagh et al., 2008).

Table 16.4 Clinical features of severe (“malignant”) TS

Vocalizations have been reported as the initial symptom in 12–37% of patients, throat clearing being the most common (Robertson, 1989). Phonic tics can be quite troublesome for patients and those around them. In addition to involuntary noises, some patients have speech dysfluencies that resemble developmental stuttering, and up to half of all patients with developmental stuttering have been thought to have undiagnosed TS (Abwender et al., 1998). Coprolalia, perhaps the most recognizable and certainly one of the most distressing symptoms of TS, is actually present in only half of patients (Videos 16.8 and 16.15). When describing the distress caused by his severe coprolalia, one of our patients remarked that immediately after shouting an obscenity, he reaches out with his hand in an attempt to “catch the word and bring it back before others can hear it.” Coprolalia appears to be markedly influenced by cultural background. Although in one retrospective analysis of 112 children with TS, only 8% exhibited coprolalia (Goldenberg et al., 1994), the true prevalence of coprolalia in TS children and adults is about 50% in the US population, when mental coprolalia (without actual utterance) is included. In a study of 597 individuals with TS from seven countries, coprolalia occurred at some point in the course of the disease in 19.3% of males and 14.6% of females, and copropraxia in 5.9% of males and 4.9% of females (Freeman et al., 2009). Coprolalia has been reported to occur in only 26% of Danish patients and 4% of Japanese patients (Robertson, 1989). Copropraxia has been found in about 20% of patients, echolalia in 30%, echopraxia in 25%, and palilalia in 15%. Although coprolalia is a characteristic feature of TS and, based on functional MRI studies, attributed to abnormal activation particularly in the left middle frontal gyrus and right precentral gyrus, and possibly the caudate nucleus, cingulate gyrus, cuneus, left angular gyrus, left inferior parietal gyrus, and occipital gyri (Gates et al., 2004), this language abnormality is not universally present or specific for TS.

Except for tics, the neurologic examination in patients with TS is usually normal. In one case-control study, TS patients were found to have a shorter duration of saccades, but the saccades were performed with a greater mean velocity than in normal controls, and the TS patients had fewer correct antisaccade responses, suggesting a mild oculomotor disturbance in TS (Farber et al., 1999). Although the ability to inhibit reflexive saccades is normal, TS patients make more timing errors, indicating an inability to appropriately inhibit or delay planned motor programs (LeVasseur et al., 2001).

Behavioral symptoms

Patients with TS generally have normal intelligence and may even perform better and faster than age-matched controls on certain tasks that require grammar skills that depend on procedural, rather than declarative memory (Walenski et al., 2007). While TS patients have no cognitive deficits, they often exhibit a variety of behavioral symptoms, particularly ADHD and OCD (Figs 16.1 and 16.2) (Gaze et al., 2006). In the Tourette International Consortium (TIC) database, which includes information on 3500 patients with TS evaluated by neurologists or psychiatrists, only 12% had tics only, without other comorbidities (Freeman et al., 2000). Kurlan and colleagues (2002) interviewed 1596 children, aged 9–17, in schools in Rochester and Monroe Counties, New York, and identified tics in 339 children (21%) after 60–150 minutes of observation. They found the following behavioral problems more frequently (P < 0.05) in children with tics than in those without tics: OCD, ADHD, separation anxiety, overanxious disorder, simple phobia, social phobia, agoraphobia, mania, major depression, and oppositional defiant behavior. Also, children with tics were younger (mean age: 12.5 vs. 13.3 years) and were more likely to require special education services (27% vs. 19.8%).

Figure 16.1 An overlap of disorders typically coexisting in patients with TS.

Redrawn from Jankovic J: Tourette’s syndrome. N Engl J Med 2001;345:1184–1192.

Figure 16.2 Typical progression of symptoms associated with TS.

Redrawn from Jankovic J: Tourette’s syndrome. N Engl J Med 2001;345:1184–1192.

A thorough discussion of the pathogenesis of comorbid disorders and their relationship to TS is beyond the scope of this review, and the reader is referred to some recent reviews on these topics (Tannock, 1998; Stein, 2002). The diagnosis of ADHD and OCD is based on clinical history; there are no laboratory or other tests that reliably diagnose these neurobehavioral disorders (Dulcan and AACAP Works Group on Quality Issues, 1997; Goldman et al., 1998; Swanson et al., 1998). These comorbid behavioral conditions often interfere with learning and with academic and work performance (Singer et al., 1995b). In contrast to tics, ADHD and obsessional symptom severity are significantly associated with impaired social and emotional adjustment (Carter et al., 2000). The clinician should be skilled not only in the recognition and treatment of ADHD but also in documenting the ADHD-related deficits (Richard et al., 1998). Such documentation is essential for the parents and educators in order to provide the optimal educational setting for the affected individual.

Since nearly all studies on the frequency of associated features have been based on a population of TS patients who have been referred to physicians, usually specialists, there is a certain selection bias; therefore, accurate figures on the prevalence of these behavioral disorders in TS patients are not available. It has been estimated, however, that 3–6% of the school-aged population suffers from ADHD (Goldman et al., 1998), and probably a majority of patients with TS have had symptoms of ADHD, OCD, or both at some time during the course of their illness (Coffey and Park, 1997). In a review of 1500 patients with TS, 48% were diagnosed as having ADHD, a figure that is consistent with the results of other studies (Comings and Comings, 1988b). The symptoms of ADHD may be the initial manifestations of TS and may precede the onset of motor and phonic tics by about 3 years. During this time, therapy with stimulant drugs may trigger the onset of tics and may precipitate the emergence of other TS symptoms (Price et al., 1986).

There are three types of ADHD: predominantly inattentive, predominantly hyperactive-impulsive, and combined (Dulcan AACAP Works Group on Quality Issues, 1997) (Table 16.5). ADHD is one of the most common neurobehavioral disorders, affecting 3–10% of children and 4% of adults (Wilens et al., 2004; Rappley, 2005). Adults with ADHD often have childhood histories of educational and discipline problems; and during adulthood, they usually have lower socioeconomic status, lower rates of professional employment, and higher rates of marital problems, driving violations, and other life failures. In a genome scan of 106 families, including 128 affected sib-pairs with estimated heritability of 60–80%, evidence of a gene locus was found on chromosomes 4, 9, 10, 11, 12, 16, and 17 (Smalley et al., 2001). Although attention deficit is certainly one of the most common and disabling symptoms of TS, in many patients the inability to pay attention is due to not only a coexistent ADHD but also uncontrollable intrusions of thoughts. Some patients are unable to pay attention because of a compulsive fixation of gaze. For example, while they are sitting in a classroom or a theater or during a conversation, their gaze becomes fixated on a particular object, and despite concentrated effort, they are unable to break the fixation. As a result, they miss the teacher’s lesson or a particular action in a play. Another reason for impaired attention in some TS patients is mental concentration exerted in an effort to suppress tics. Yet another cause for inattention is the sedative effect of anti-TS medications. It is therefore important to determine which mechanism or mechanisms are most likely to be responsible for the patient’s attention deficit. This is particularly important in selecting the best therapeutic approach. Despite growing publicity about ADHD, there is little evidence of widespread overdiagnosis or overtreatment of ADHD (Goldman et al., 1998). One study showed that children and adolescents with ADHD as compared to those without ADHD are more likely to have major injuries and asthma, and their 9-year medical costs are double (Leibson et al., 2001). The mechanism of ADHD with or without TS is not well understood, and there are no specific pathologic abnormalities identified. However, by using high-resolution MRI, abnormal morphology was observed in the frontal cortices, reduction in the anterior temporal cortices, and increased gray matter in the posterior and inferior parietal cortices in children and adolescents with ADHD (Sowell et al., 2003).

Table 16.5 Attention-deficit hyperactivity disorder/hyperkinetic disorder (HKD) (ICD-10, DSM-IV)

Neurophysiology

Although the pathogenic mechanisms of TS are still unknown, the weight of evidence supports an organic rather than psychogenic origin, probably involving the basal ganglia circuitry (Leckman et al., 1997b; Palumbo et al., 1997; Jankovic, 2001a; Leckman, 2002; Berardelli et al., 2003; Albin and Mink, 2006) (Fig. 16.3). Despite the observation that some tics may be voluntary, at least in part, physiologic studies suggest that tics are not mediated through normal motor pathways utilized for willed movements (Obeso et al., 1982). Using back-averaging techniques, Obeso and colleagues (1982) observed normal Bereitschaftspotential in six subjects who voluntarily simulated tic-like movements, but no such premovement potential was noted in association with an actual tic. The common absence of premotor potentials in simple motor tics suggests that tics are truly involuntary or that they occur in response to some external cue (Papa et al., 1991). Karp and colleagues (1996), however, documented premotor negativity in two of five patients with simple motor tics. Although the investigators could not correlate the presence of Bereitschaftspotential with the premonitory sensation, the physiology of the premovement phenomenon requires further studies. Transmagnetic stimulation used to study cortical excitability in 11 TS patients found evidence of motor-cortical disinhibition at rest (Heise et al., 2010).

Figure 16.3 Basal ganglia circuitry relevant to pathophysiology of tics. GPi, globus pallidus interna; STN, subthalamic nucleus; SNr, substantia nigra pars reticulata.

Conventional neurophysiologic investigations have found that TS patients have defective inhibitory mechanisms, as is suggested by the increased duration of the late response of the blink reflex and reduced inhibition at paired pulse testing (Smith and Lees, 1989; Berardelli et al., 2003). TS patients also have exaggerated audiogenic startle response (Gironell et al., 2000). About 20% patients with TS have exaggerated startle responses, which may fail to habituate with repetition (Stell et al., 1995).

As was noted before, functional MRI showed decreased neuronal activity during periods of suppression in the ventral globus pallidus, putamen, and thalamus and increased activity in the right caudate nucleus, right frontal cortex, and other cortical areas that are normally involved in the inhibition of unwanted impulses (prefrontal, parietal, temporal, and cingulate cortices) (Peterson et al., 1998a). In another study of three patients with TS, functional MRI showed marked reduction or absence of activity in secondary motor areas while the patient attempted to maintain a stable grip-load force control (Serrien et al., 2002). The authors interpreted the findings as an ongoing activation of the secondary motor reflecting patients’ involuntary urges to move. In a study of children with ADHD, functional MRI showed increased frontal activation and reduced striatal activation on various tasks and an enhancement of striatal function after treatment with methylphenidate (Vaidya et al., 1998). Functional MRI studies show decreased neuronal activity during periods of suppression in the ventral globus pallidus, putamen, and thalamus and increased activity in the caudate, frontal cortex, and other cortical areas that are normally involved in the inhibition of unwanted impulses (prefrontal, parietal, temporal, and cingulated cortical areas) (Peterson, 2001). By using event-related functional MRI (fMRI) in 10 patients with TS, a brain network of paralimbic areas such as anterior cingulate and insular cortex, supplementary motor area (SMA), and parietal operculum (PO) was identified that predominantly activated before tic onset, whereas at the beginning of tic action, significant fMRI activities were found in sensorimotor areas including superior parietal lobule bilaterally and cerebellum (Bohlhalter et al., 2006). Using resting-state functional connectivity MRI (rs-fcMRI) in 33 adolescents with TS, Church et al. (2009) found anomalous connections primarily in the frontoparietal network, suggesting widespread immature functional connectivity, particularly in regions related to adaptive online control.

Transcranial magnetic stimulation studies have demonstrated a shortened cortical silent period and defective intracortical inhibition (determined in a conditioning test paired-stimulus paradigm) in patients with TS (Orth et al., 2005) and OCD (Greenberg et al., 1998), thus providing a possible explanation for intrusive phenomena. Subsequent studies utilizing the same technique have demonstrated that patients with tic-related OCD have more abnormal motor cortex excitability than OCD patients without tics (Greenberg et al., 2000). Transcranial magnetic stimulation studies have also demonstrated that TS children have a shorter cortical silent period but that their intracortical inhibition was not different from that of controls, although intracortical inhibition is reduced in children with ADHD (Moll et al., 2001). There is evidence of additive inhibitory deficits, as demonstrated by reduced intracortical inhibition and a shortened cortical silent period in children with TS and comorbid ADHD. In another study, ADHD more than tics were associated with short-interval intracortical inhibition (Gilbert et al., 2004). Both short-interval intracortical inhibition and short-interval afferent inhibition, modulated by nicotinic receptors, were reduced in eight patients with TS (ages 24–38 years) as compared to ten matched healthy controls (Orth et al., 2005). Comprehensive discussion of the neuroscience of ADHD is beyond the scope of this chapter, but the reader is referred to some recent reviews of this topic (Castellanos and Tannock, 2002).

Sleep studies have provided additional evidence that some tics are truly involuntary. Polysomnographic studies in 34 TS patients recorded motor tics in various stages of sleep in 23 patients and phonic tics in 4 patients (Glaze et al., 1983; Jankovic and Rohaidy, 1987). Additional sleep studies have suggested that some patients with TS have alterations of arousal, decreased percentage (up to 30%) of stage 3/4 (slow wave) sleep, decreased percentage of REM sleep, paroxysmal events in stage 4 sleep with sudden intense arousal, disorientation and agitation, restless legs syndrome, and periodic leg movement in sleep (Voderholzer et al., 1997; Chokroverty and Jankovic, 1999; Picchietti et al., 1999). Restless legs syndrome has been reported to be present in about 10% of patients with TS and in 23% of parents (Lesperance et al., 2004). In this regard, it is of interest that in a study in which 14 single-nucleotide polymorphisms were typed spanning the three genomic loci in 298 TS trios, 322 TS cases (including 298 probands from the cohort of TS trios), and 290 control subjects, it was found that the same variant of the gene BTBD9 that increases the risk of restless legs syndrome also increases the risk of TS without obsessive-compulsive disorder or attention-deficit disorder (“pure” TS) (Rivière et al., 2009). Other sleep-related disorders that are associated with TS include sleep apnea, enuresis, sleepwalking and sleep talking, nightmares, myoclonus, bruxism, and other disturbances (Rothenberger et al., 2001; Hanna and Jankovic, 2003).

Neuroimaging

Although standard anatomic neuroimaging studies in TS are unremarkable, by using special volumetric, metabolic, blood flow, ligand, and functional imaging techniques, several interesting findings have been reported that have strong implications for the pathophysiology of TS (Peterson, 2001; Albin and Mink, 2006; Frey and Albin, 2006) (Fig. 16.3). Careful volumetric MRI studies have suggested that the normal asymmetry of the basal ganglia is lost in TS (Peterson et al., 1993; Singer et al., 1993; Peterson, 2001). In one study, the normal left > right asymmetry (Singer et al., 1993) in the volume in the right anterior brain, right caudate, and right pallidum was reversed in TS subjects (Castellanos et al., 1996). Caudate volumes have been reported to correlate significantly and inversely with the severity of tic and OCD in early adulthood (Bloch et al., 2005). The volumetric studies, however, have not always produced consistent results (Moriarty et al., 1997). In another volumetric MRI study, Frederickson and colleagues (2002) found evidence of smaller gray matter volumes in the left frontal lobes of patients with TS, further supporting the findings of loss of normal left > right asymmetry. Quantitative MRI studies have found a subtle, but possibly important, reduction in the volume of caudate nuclei in patients with TS. In 10 pairs of monozygotic twins, the right caudate was smaller in the more severely affected individuals, providing evidence for the role of environmental events in the pathogenesis of TS (Hyde et al., 1995). In contrast, the corpus callosum has been found to be larger in children with TS than in normal controls (Baumgardner et al., 1996). Subsequent study showed that this finding was gender-related and was present only in boys with TS (Mostofsky et al., 1999). An MRI-DTI study of monozygotic twins showed that the mean fractional anisotropy values were significantly lower, particularly in the posterior portion of the corpus callosum, in the twin affected with TS (Cavanna et al., 2010). Using voxel-based morphometry and high-resolution MRI in 31 TS patients, increased gray matter was found in the left mesencephalon compared to 31 controls (Garraux et al., 2006). This finding, however, could not be replicated in a smaller study of 14 boys with TS using the same technique (Ludolph et al., 2006). This latter study found increased gray matter volumes in bilateral ventral putamen and regional decreases in gray matter volumes in the left hippocampus gyrus. The difference was attributed to younger mean age in the latter study compared to the previous study (12.5 vs. 32 years). Diffusion-tensor MRI (DT-MRI) used to investigate the structural integrity of basal ganglia and thalamus in 23 children with TS found increased mean water diffusivity bilaterally in the putamen and decreased anisotropy in the right thalamus, indicating impairment of white matter integrity in the fronto-striatal-thalamic circuit at a microstructural level (Makki et al., 2008). Using tractography of the fronto-striato-thalamic circuit, the same group showed that TS patients had significantly lower probability of connection between caudate nucleus and anterior-dorsolateral-frontal cortex on the left and that OCB was negatively associated with connectivity score of the left caudate and anterior dorsolateral-frontal cortex, and was positively associated with connectivity score for the subcallosal gyrus and for the lentiform nucleus (Makki et al., 2009). Additional imaging studies have identified frontal and parietal cortical thinning, most prominent in ventral portions of the sensory and motor homunculi in patients with TS (Sowell et al., 2008). Dopaminergic hyperfunction (Albin et al., 2003), possibly as a result of overproduction of synapses in mid-childhood (Giedd et al., 1999), might be associated with increased striatal volume in childhood which may be lost in adulthood.

Positron emission tomography (PET) scanning has shown variable rates of glucose utilization in basal ganglia as compared to controls. In one study, [¹⁸F]fluorodeoxyglucose PET has shown evidence of increased metabolic activity in the lateral premotor and supplementary motor association cortices and in the midbrain (pattern 1), and decreased metabolic activity in the caudate and thalamic areas (limbic basal ganglia–thalamocortical projection system) (pattern 2) (Eidelberg et al., 1997). Pattern 1 is reportedly associated with tics, and pattern 2 correlates with the overall severity of TS. In a follow-up study, involving 12 TS adult patients (untreated for >2 years) and 12 controls, the investigators found a TS-related metabolic pattern which was characterized by increased premotor cortex and cerebellum activity, and reduced resting activity of the striatum and orbitofrontal cortex (Pourfar et al., 2011). They also provided evidence for an OCD-related pattern which was characterized by reduced activity of the anterior cingulate and dorsolateral prefrontal cortical regions associated with relative increases in primary motor cortex and precuneus. This pattern significantly correlated with the severity of the OCD symptoms.

In contrast to dystonia, which is characterized by lentiform nucleus-thalamic metabolic dissociation, attributed to overactivity of the direct striatopallidal inhibitory pathway, the pattern of TS is characterized by concomitant metabolic reduction in striatal and thalamic function. The authors suggested that this pattern can be explained by a reduction in the indirect pathway, resulting in reduction in subthalamic nucleus (STN) activity. This is in part consistent with another study that found evidence of increased activation in the direct pathway, but the activity in the prefrontal cortex and STN has been found to be increased presumably as a result of compensatory activation (Baym et al., 2008). Using event-related [¹⁵O]H₂O PET combined with time-synchronized audiotaping and videotaping of six patients with TS, Stern and colleagues (2000) found increased activity in the sensorimotor, language, executive, paralimbic, and frontal-subcortical areas that was temporarily related to the motor and phonic tics and the irresistible urge that precedes these behaviors. Using a similar technique Lerner et al. (2007) found robust activation of cerebellum, insula, thalamus, and putamen during tic release. Rauch and colleagues (1997) showed bilateral medial temporal (hippocampal/parahippocampal) activation on PET in patients with OCD as compared to normal controls and absence of activation of inferior striatum, seen in normal controls. Various neuroimaging studies have also demonstrated moderate reduction in the size of the corpus callosum, basal ganglia (particularly caudate and globus pallidus), and frontal lobes (Filipek et al., 1997) and striatal hypoperfusion (Vaidya et al., 1998) in patients with ADHD.

An autopsy study of three TS brains found consistent increases in dopamine transporter (DAT) and D2 receptor as well as D1 and α-2A density, suggesting that dopaminergic hyperfunction in the frontal lobe may play a role in the pathophysiology of TS (Yoon et al., 2007). Another pathological study showed a marked increase in total number of neurons in the globus pallidus interna (GPi) and decreased number in the globus pallidus externa (GPe) and in the caudate nucleus of brains of patients with TS (Kalanithi et al., 2005). Furthermore, an increased number and proportion of the GPi neurons were positive for the calcium-binding protein parvalbumin in tissue from TS subjects, whereas lower densities of parvalbumin-positive interneurons were observed in both the caudate and putamen of TS subjects. These abnormalities have been interpreted as indicating a developmental defect in the migration of some GABAergic neurons.

Neurochemistry

An alteration in the central neurotransmitters has been suggested chiefly because of relatively consistent responses to modulation of the dopaminergic system (Harris and Singer, 2006). Dopamine antagonists and depletors generally have an ameliorating effect on tics, whereas drugs that enhance central dopaminergic activity exacerbate tics (Jankovic and Rohaidy, 1987; Jankovic and Orman, 1988). Low cerebrospinal fluid homovanillic acid, coupled with a favorable response to dopamine receptor-blocking drugs, has been interpreted as evidence in support of the notion that tics and TS are due to supersensitive dopamine receptors (Singer, 2000). Postmortem binding studies of dopamine receptors, however, have failed to provide support for this hypothesis (Singer, 2000).

Neurochemical studies of TS have been hampered by the lack of available postmortem brain tissue. Biochemical abnormalities in the few postmortem brains that have been studied include low serotonin, low glutamate in the medial globus pallidus, and low cyclic AMP in the cortex (Singer, 2000). Haber and Wolfer (1992) reported that in a blind rating of five TS brains, three had low dynorphin immunoreactivity in the ventral portion of the medial globus pallidus. A defect in tryptophan oxygenase in TS has been proposed by Comings (1990), who analyzed 1400 blood samples from patients with TS or ADHD and their relatives and controls; he found decreased platelet serotonin and low blood tryptophan levels in TS patients and their parents. In support of the “tryptophan hypothesis” is a study that used alpha-[¹¹C]methyl-L-tryptophan (AMT) positron emission tomography (PET) to assess global and focal brain abnormalities of tryptophan metabolism in 26 children with TS and nine controls. The findings indicate that tryptophan uptake is significantly decreased in the dorsolateral prefrontal cortex and increased in the thalamus of TS patients (Behen et al., 2007).

One intriguing hypothesis, supported partly by the increased ³H-mazindol binding to the presynaptic dopamine uptake carrier sites, suggests that TS represents a developmental disorder resulting in dopaminergic hyperinnervation of the ventral striatum and the associated limbic system (nucleus accumbens) (Singer et al., 1991). Using [¹¹C]raclopride PET and amphetamine stimulation, Singer and colleagues (2002) found evidence for increased dopamine release in the putamen of patients with TS. They postulate that in TS, there is increased activity of the dopamine transporter leading to increased dopamine concentration in the dopamine terminals and stimulus-dependent increase in dopaminergic transmission. Support for this hypothesis has been provided by the imaging studies of Albin and colleagues (2003). The authors used PET with the vesicular monoamine transporter type 2 ligand [¹¹C]dihydrotetrabenazine that binds to type 2 vesicular monoamine transporter (VMAT2) to quantify striatal monoaminergic innervation in patients with TS (n = 19) and control subjects (n = 27). With voxel-by-voxel analysis, the investigators found increased [¹¹C]dihydrotetrabenazine binding in the ventral striatum (right > left) in patients with TS as compared to age-matched controls. However, a subsequent PET study involving 33 adults with TS and utilizing not only [¹¹C]dihydrotetrabenazine but also [¹¹C]methylphenidate, a ligand for DAT, binding, found no differences between subjects with TS and controls (Albin et al., 2009). In a study of 8 patients with TS and 8 controls [¹¹C]FLB 457 PET in conjunction with an amphetamine challenge used to evaluate extrastriatal D2/D3 receptor binding and DA release, TS patients showed decreased [¹¹C]FLB 457 binding potentials bilaterally in cortical and subcortical regions outside the striatum, including the cingulate gyrus, middle and superior temporal gyrus, occipital cortex, insula, and thalamus (Steeves et al., 2010). Furthermore, amphetamine challenge induced widespread increased DA release in TS patients, which extended more anteriorly to involve anterior cingulate and medial frontal gyri. The authors suggested that “reductions in D2/D3 receptor binding in both frontal cortex and thalamus are consistent with recently published preliminary data demonstrating similar abnormalities of D2/D3 binding in TS patients using a different PET ligand.” A postmortem examination of two brains of patients with typical childhood-onset TS and one with adult-onset tics showed that the prefrontal cortex rather than the striatum showed most abnormalities, including increased D2 receptor protein, as well as increases in dopamine transporter, VAMP2, and α-2A (Minzer et al., 2004).

Although this idea is highly speculative, it is possible that the genetic defect in TS somehow interferes with the normal regulation of the neuronal progenitor cells during development, thus resulting in the increased innervation of the ventral striatum (Ikonomidou et al., 1999; Itoh et al., 2001; Hanashima et al., 2004). This implies that the genetic defect somehow interferes with the programmed cell suicide that is needed to control cell proliferation in normal development and growth. The ventral striatum is the portion of the basal ganglia that is anatomically and functionally linked to the limbic system. The link between the basal ganglia and the limbic system might explain the frequent association of tics and complex behavioral problems, and a dysfunction in the basal ganglia and the limbic system seems to provide the best explanation for the most fundamental behavioral disturbance in TS, namely, loss of impulse control and a state of apparent “disinhibition.” The notion that deficits in inhibitory functions are at the core of the clinical phenotype associated with TS is supported by studies by Baron-Cohen and colleagues (1994) showing that patients with TS are not able to appropriately edit their intentions and by Swerdlow and colleagues (1996) showing that TS patients demonstrate deficits on the visuospatial priming tasks.

Functional neuroimaging studies have been used to aid in the understanding of neurotransmitter and receptor alterations in TS. Using [¹²³I]β-carboxymethoxy-3β-(4-iodophenyl) tropane (CIT) single photon emission computed tomography scans, Malison and colleagues (1995) demonstrated a mean 37% increase in binding of this dopamine transporter ligand in the striatum in five adult patients with TS as compared to age-matched controls. In contrast, Heinz and colleagues (1998) found no difference in [¹²³I] β-CIT binding in the midbrain, thalamus, or basal ganglia between 10 TS patients and normal control subjects. There was, however, a significant negative correlation between the severity of phonic tics and β-CIT binding in the midbrain and thalamus. In another study involving 12 TS adult patients, β-CIT scans showed evidence of increased dopamine transporter binding (Müller-Vahl et al., 2000). Combining single photon emission computed tomography and MRI, Wolf and colleagues (1996) found 17% greater binding of IBZM, a D2 receptor ligand, in the caudate (but not putamen) nucleus in five of the more affected monozygotic twins who were discordant for TS. It is important to note, however, that two of the five subjects were taking neuroleptics for up to 6 weeks prior to the single photon emission computed tomography studies. These findings, if confirmed by other studies of neuroleptic-naive patients, support the notion that the presynaptic dopamine function is enhanced in TS. This might, in turn, lead to a reduced inhibitory pallidal output to the mediodorsal thalamus. The observation that in patients with Parkinson disease the severity of childhood-onset tics was not influenced by the development of parkinsonism or by its treatment with levodopa, however, argues against the role of dopamine in the pathogenesis of TS symptoms (Kumar and Lang, 1997a). This is supported by the results of PET ligand studies showing normal D2 receptor density (Turjanski et al., 1994). Furthermore, Meyer and colleagues (1999) used PET imaging of (+)-α-[¹¹C]dihydrotetrabenazine to determine the density of vesicular monoamine transporter type 2, a cytoplasm-to-vesicle transporter that is linearly related to monoaminergic nerve terminal density unaffected by medications, in 8 TS patients and 22 controls. This study showed no significant difference in terminal density between patients and controls, thus failing to provide support for the concept of increased striatal innervation. However, these studies do not exclude the possibility of abnormal regulation of dopamine release and uptake. Subsequent study involving 19 adult patients with TS showed that the [¹¹C]dihydrotetrabenazine-binding potential is significantly increased, thus supporting the notion that striatal monoaminergic innervation is increased in the ventral striatum (right > left) of TS patients (Albin et al., 2003). The results of this study contrasted with previous findings of 30–40% increase in dopamine transporter (Müller-Vahl et al., 2000; Singer et al., 2001). Furthermore, in a small sample of TS patients, PET studies, have demonstrated a 25% increase in accumulation of fluorodopa in the left caudate (P = 0.03) and a 53% increase in the right midbrain (P = 0.08) (Ernst et al., 1999). These findings indicate possible dopaminergic dysfunction in the cells of origin and in the dopaminergic terminals, suggesting increased activity of dopa decarboxylase.

Functional imaging has provided insights into the mechanisms not only of TS but also of ADHD. In 53 adults with ADHD (without tics) when compared to 44 healthy controls, a reduction in dopamine markers was demonstrated, using ¹¹C-cocaine as a marker for dopamine transporter and ¹¹C-raclopride, a ligand for D2/D3 receptors (Volkow et al., 2009).

Despite some limitations and inconsistencies, the imaging, ligand, and biochemical studies provide support for the hypothesis that the corticostriatal-thalamic-cortical circuit plays an important role in the pathogenesis of TS and related disorders (Witelson, 1993; Peterson, 2001). The dorsolateral prefrontal circuit, which links Brodmann’s area 9 and 10 with the dorsolateral head of the caudate, appears to be involved with executive functions (manipulation of previously learned knowledge, abstract reasoning, organization, verbal fluency, and problem solving; it is closely related to intelligence, education, and social exposure) and motor planning. An abnormality in this circuit has been implicated in ADHD. The lateral orbitofrontal circuit originates in the inferior lateral prefrontal cortex (area 10) and projects to the ventral medial caudate. An abnormality to this circuit is associated with personality changes, mania, disinhibition, and irritability. Last, the anterior cingulate circuit arises in the cingulate gyrus (area 24) and projects to the ventral striatum, which also receives input from the amygdala, hippocampus, medial orbitofrontal cortex, and entorhinal and perirhinal cortex. A variety of behavioral problems, including OCD, may be linked to an abnormality in this circuit.

Reduced metabolism or blood flow to the basal ganglia, particularly in the ventral striatum, most often in the left hemisphere, has been demonstrated in the majority of the studies involving TS subjects. These limbic areas are thought to be involved in impulse control, reward contingencies, and executive functions, and these behavioral functions appear to be abnormal in most patients with TS. The radioligand studies have been less consistent, but they provide some support for increased D2 receptor density in the caudate nucleus. Imaging studies of presynaptic markers such as dopa decarboxylase, dopamine, and dopamine transporter have produced results that are even less consistent. Future imaging and ligand studies should include children, since this population has been largely excluded because of ethical considerations. The studies should also rigorously characterize comorbid disorders and should take into consideration potential confounding variables, such as the secondary effects of chronic illness and medications.

Immunology

The potential role of immunologic mechanisms and specifically antineuronal antibodies is currently being explored in a variety of neurologic disorders, including TS (Allen, 1997; Hallett and Kiessling, 1997; Hallett et al., 2000; Morshed et al., 2001; Church et al., 2003; Harris and Singer, 2006; Singer et al., 2008; Martino et al., 2009). Several studies have suggested that exacerbations of TS symptoms correlated with an antecedent group A β-hemolytic streptococcus (GABHS) infection (demonstrated by elevated antistreptococcal titers) and the presence of serum antineuronal antibodies (Kiessling et al., 1993). Epitopes of streptococcal M proteins have been found to cross-react with the human brain, particularly the basal ganglia, and may be pathogenetically important in various neurologic disorders, such as Sydenham disease, TS-like syndrome, dystonia, and parkinsonism (Bronze and Dale, 1993; Dale et al., 2001). In 10 patients with poststreptococcal acute disseminated encephalomyelitis (PSADEM) following exposure to GABHS, Dale and colleagues (2001) showed antibasal ganglia antibodies in all with three (60, 67, and 80 kDa) dominant bands. Furthermore, MRI showed hyperintense basal ganglia in 80% of the patients. The B-lymphocyte antigen D8/17 is considered to be a marker for rheumatic fever but is also frequently overexpressed in patients with tics, OCD, and autism (Murphy et al., 1997; Swedo et al., 1997; Hoekstra et al., 2001). In one study, children and adults with TS had significantly higher serum levels of antineuronal antibodies against the putamen, but not the caudate or globus pallidus, as compared to controls (Singer et al., 1998). The potential relevance of this finding has been questioned, however, since there is no relationship between the presence of the antineuronal antibodies and age at onset, severity of tics, or the presence of comorbid disorders. Trifiletti and Packard (1999) have confirmed the presence of a specific brain protein at an apparent molecular weight of 83 kDa that is recognized by antibodies in the serum of 80–90% of patients with TS or OCD (Trifiletti and Packard 1999). They concluded that there may be a subset of patients with TS and OCD, perhaps up to 10% of all cases, in whom a streptococcal infection triggers the onset of symptoms (Trifiletti and Packard, 1999). In a large case-control study of 150 patients with tics compared to 150 controls, Cardona and Orefici (2001) found a correlation between the occurrence of tics and prior exposure to streptococcal antigens and that the severity of tic disorder correlates with the magnitude of the serologic response to streptococcal antigens measured by antistreptolysin O (ASO) titers (38% of the children with tics compared with 2% of the control subjects had ASO titers ≥500 IU, (P < 0.001)). In another study involving 25 adult patients with TS and 25 healthy controls, increased antibody titers against streptococcal M12 and M19 proteins were found in the TS group as compared to the healthy controls (Müller et al., 2001). In yet another study involving 81 patients with TS, 27 with Sydenham disease, 52 with autoimmune disorders, and 67 normal controls, Morshed and colleagues (2001) found elevated titers of IgG antineuronal antibodies in diseased individuals compared to controls, but there was no relationship between the antibody titers and the age, severity of TS, or comorbid disorders. Church and colleagues (2003) studied 100 patients with TS, 50 children with neurologic disease, 40 with recent uncomplicated streptococcal infection, and 50 healthy adults. They found elevated ASO titers in 64% of TS children compared with 15% of pediatric controls (P < 0.0001) and in 68% of adults with TS compared with 12% of adult neurologic controls and 8% of adult healthy controls (P < 0.05). Evidence of basal ganglia antibodies were found in 20–27% of TS patients, compared to 2–4% of controls. Similar to the proposed antigen in Sydenham disease, the most common antigen in TS was 60 kDa protein. Furthermore, 91% of patients with positive basal ganglia antibodies were found to have elevated ASO titers, whereas only 57% of those without such antibodies had high ASO titers. The authors concluded that these findings “suggest a pathogenic similarity between Sydenham disease and some patients with TS.” In another study, 65% of 65 patients with atypical movement disorders, including dystonia and tics, had antibasal ganglia antibodies (Edwards et al., 2004b). Edwards and colleagues (2004a) also reported four adult cases of tic disorder and stereotypies associated with the presence of antibasal ganglia antibodies. There was no difference when clinical features of TS patients (53 children and 75 adults) with and without antinuclear antibodies, determined by Western immunoblot showing bands for at least 1 of 3 reported striatal antigens (40, 45, and 60 kDa), were compared (Martino et al., 2009). When brain MRI T1 and diffusion tensor (DTI) imaging was compared in 9 adults with TS who had antibasal ganglia antibodies with 13 without detectable antibodies, the voxel-based morphometry analysis failed to detect any significant difference in gray matter density between the two groups, again suggesting lack of relevance of the antibasal ganglia antibodies to the pathogenesis of TS. Although there are many inconsistencies in the reported results related to autoantibodies in TS, and increased activation of immune responses in TS has been suggested by some investigators, further studies are needed before it can be concluded that TS is an autoimmune disorder (Martino et al., 2009).

Development of dyskinesias (paw and floor licking, head and paw shaking) and phonic utterances has been reported in rodents after the microinfusion of dilute IgG from TS subjects into their striatum (Hallett et al., 2000). They extended their initial observations by demonstrating that intrastriatal microinfusion of TS sera or gamma immunoglobulins (IgG) in rats produced stereotypies and episodic utterances, analogous to involuntary movements seen in TS, and confirmed the presence of IgG selectively bound to striatal neurons (Hallett et al., 2000). Peterson and colleagues (2000) found that ADHD was associated significantly with titers of two distinct antistreptococcal antibodies, ASO and antideoxyribonuclease B, but no significant association was seen between antibody titers and a diagnosis of either chronic tic disorder or OCD. When basal ganglia volumes were included in these analyses, the relationships between antibody titers and basal ganglia volumes were significantly different in OCD and ADHD subjects compared with other diagnostic groups. Higher antibody titers in these subjects were associated with larger volumes of the putamen and globus pallidus nuclei.

Variably referred to as pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections (PANDAS) or pediatric infection-triggered autoimmune neuropsychiatric disorders (PITANDS), this area is one of the most controversial topics in pediatric neurologic and psychiatric literature (Kurlan, 1998b; Trifiletti and Packard, 1999; Hoekstra et al., 2002; Murphy and Pichichero, 2002; Church et al., 2003; Snider and Swedo, 2003; Kurlan, 2004; Gabbay et al., 2008; Gilbert and Kurlan, 2009; Schrag et al., 2009). The following are the diagnostic criteria for pediatric infection-triggered autoimmune neuropsychiatric disorders, which are similar to those for PANDAS except that they define an episode of illness more specifically:

The PANDAS concept has some other shortcomings: (1) The streptococcal infection may cause stress, which can exacerbate tics; (2) some patients with PANDAS may simply have a variant of Sydenham disease; and (3) there is no established temporal link between antecedent streptococcal infection and the exacerbation of tics. One piece of evidence supporting the notion that PANDAS simply represents one spectrum of Sydenham disease is the frequent occurrence of TS comorbidities in patients with Sydenham disease. One study compared 56 patients with Sydenham disease with 50 rheumatic fever patients and 50 healthy subjects; it found that obsessive-compulsive behavior (OCB) was present in 19%, OCD in 23.2%, and ADHD in 30.4% of Sydenham disease patients, all significantly more frequent than in the rheumatic fever or healthy controls (Maia et al., 2005). Nevertheless, the concept of postinfectious OCD is gradually seeping into the literature, even though definite proof is still lacking (Leonard et al., 1999). One observation, yet to be confirmed, that has been used in support of the autoimmune hypothesis of TS is that infusion of sera from patients with TS who have high titers of antibodies against nuclear or neural protein into ventrolateral striata of rats was associated with a higher rate of oral stereotypies as recorded by “blinded” raters (Taylor et al., 2002). This observation, however, could not be reproduced by other investigators who infused sera from patients with TS who had antiputaminal antibodies (n = 9) and from patients with PANDAS (n = 8) into the striatum of rats (Loiselle et al., 2004). No stereotypic behavior or abnormal movements were observed. Furthermore, there was no difference in antibasal ganglia antibodies between patients with PANDAS and patients with TS as measured by enzyme-linked immunosorbent assay (Singer et al., 2004). In a study of 40 subjects prospectively evaluated with intensive laboratory testing for GABHS for an average of 2 years with additional testing at the time of any clinical exacerbations, Kurlan et al. (2008) found that 5 of 64 exacerbations were temporally associated (within 4 weeks) with a GABHS infection, which was significantly higher than the number that would be expected by chance alone (1.6). In another study no correlation between a variety of immune markers and clinical exacerbations in children with PANDAS was found, arguing against autoimmunity as a pathophysiologic mechanism in this disorder (Singer et al., 2008). In a study that involved consecutive ratings of tics and behavioral symptoms in 45 cases and 41 matched control subjects over a 2-year period, only a small minority of children with TS and early-onset OCD were found to be sensitive to antecedent GABHS infections (Lin et al., 2010). In a case-control study of a large primary care database of 678 862 patients with an average follow-up of 5.08 years, no support was found for a strong relationship between streptococcal infections, neuropsychiatric syndromes such as OCD or TS, or PANDAS (Schrag et al., 2009). Although, in contrast to the Schrag et al. (2009), similar health services database studies in the United States found modest statistical associations between streptococcal infections and tic or OCD diagnoses, “current evidence indicates that for the ordinary TS/OCD phenotype of waxing and waning symptoms, GABHS infection does not seem to be an important etiologic factor and therefore not an appropriate target for assessment or therapy” (Gilbert and Kurlan, 2009).

Untreated GABHS infection is often complicated by rheumatic fever within 10–14 weeks and by Sydenham disease within several months. Several studies have provided evidence for an overlap between TS and Sydenham disease, tics and OCD being manifested in both disorders (Church et al., 2003). It is therefore not clear whether TS and OCD are independent sequelae of GABHS or whether the observed symptoms of TS and OCD are manifestations of Sydenham disease. In a 3-year prospective study of 12 children, 5 to 11 years old, with documented acute GABHS tonsillopharyngitis, Murphy and Pichichero (2002) noted a sudden onset of OCD symptoms in all patients and tics in 3 patients concurrently with the acute infection or within 4 weeks. These symptoms resolved in all patients within 2 weeks of initiation of antibiotic therapy. Although this intriguing hypothesis requires further studies, plasmapheresis, intravenous IgG, and immunosuppressant therapies are currently being investigated in the treatment of TS (Allen, 1997). In a study of 30 children in whom OCD or tics were presumably triggered or exacerbated by GABHS, there were striking improvements in various measures of OCD after intravenous immunoglobulin (IVIg) and in tics after plasma exchange (Perlmutter et al., 1999). Twenty-nine children with PANDAS were randomized in a partially double-blind fashion (no sham plasmapheresis) to an IVIg, IVIg placebo (saline), and plasmapheresis (PEX) group. One month after treatment, the severity of obsessive-compulsive symptoms (OCS) was improved by 58% and 45% in the PEX and IVIg groups, respectively, compared to only 3% in the IVIg control group. In contrast, tic scores were only improved after PEX treatment (i.e., reductions of 49% (PEX), 19% (IVIg), and 12% (IVIg placebo)). Improvements in both tics and OCS were sustained for 1 year. However, there was no control PEX group, and the control comparisons were limited to the 1-month visit. Furthermore, there was no relationship between rate of antibody removal and therapeutic response. Until the results of this study are confirmed, these treatment modalities are not justified in patients with TS. Furthermore, because of uncertainties about the possible cause-and-effect relationship between GABHS and tics and OCD, an antibiotic treatment for acute exacerbations of these symptoms is currently considered unwarranted (Kurlan, 1998b). Some studies have suggested that encephalitis lethargica is also part of the spectrum of poststreptococcal autoimmune diseases (Dale et al., 2004; Vincent, 2004).

Other hypotheses

Although direct evidence is still lacking, TS is currently viewed as a disorder of synaptic transmission involving disinhibition of the corticostriatal-thalamic-cortical circuitry. Several studies have provided evidence in support of the notion that the basal ganglia, particularly the caudate nucleus, and the inferior prefrontal cortex play an important role in the pathogenesis of not only TS but also comorbid disorders, particularly OCD (Baxter et al., 1992; McGuire, 1995; Swoboda and Jenike, 1995). For example, Laplane and colleagues (1989) described eight patients with bilateral basal ganglia lesions (anoxic or toxic encephalopathy) who showed stereotyped activities and OCB; extrapyramidal signs were mild or absent. Several imaging studies (reviewed earlier in the chapter) have implicated the ventral striatum–limbic system complex as playing an important role in the primitive reproductive behavior. A disturbance in sex hormones and certain excitatory neurotransmitters that normally influence the development of these structures may be ultimately expressed as TS (Kurlan, 1992). This hypothesis may explain the remarkable gender difference in TS, with males outnumbering females by 3 to 1; the exacerbation of symptoms at puberty and during the estrogenic phase of the menstrual cycle (Schwabe and Konkol, 1992); the characteristic occurrence of sexually related complex motor and phonic tics; and a variety of behavioral manifestations with sexual content. According to this hypothesis, the gene defect in TS results in an abnormal production of gonadal steroid hormones and increased trophic influence exerted by the excitatory amino acids, causing disordered development and increased innervation of the striatum and the limbic system. This is consistent with the finding of increased presynaptic dopamine uptake sites in brains of patients with TS (Singer et al., 1991).

Although there are no animal models of TS, studies of stereotypies in animals may provide insight into the pathogenesis of habits, rituals, tic-like, and impulsive behaviors in humans (Graybiel, 2008; Cools, 2008). These studies have shown that a broad spectrum of repetitive behaviors and rituals can become habitual and stereotyped through learning, partly as a result of experience-dependent neuroplasticity. Several families of horses with equine self-mutilation syndrome have been described with features that resemble human TS (Dodman et al., 1994). These horses exhibit a variety of stereotypic behaviors, such as glancing, biting at the flank or pectoral areas, bucking, kicking, rubbing, spinning, rolling, and vocalizing (squealing). Similar to TS, this condition is much more common in young males, is typically exacerbated by stress and during restful nonvigilance or boredom, and may be triggered by head trauma and relieved by castration. Genetic factors, however, seem to be most important.