28 Symptomatic (secondary) tic syndromes

Disorders causing secondary tics can be divided into congenital and acquired types (Table 28.1). The former include dominantly and recessively inherited neurological disorders in which tic occasionally occurs. Many older descriptions refer to tic-like movements in conditions that are now recognized as primarily choreic, dystonic, myoclonic, and the like. If vocalizations occurred in such conditions there was the temptation to describe them as examples of secondary tics. Even in more recent literature confusion exists (Weingarten 1968). Thus, although tic has occasionally been reported in Huntington's disease, generalized idiopathic torsion dystonia, and so forth, they are not further elaborated here. In choreoacanthocytosis some cases have both motor and vocal tics in addition to other involuntary movements (see Chapter 21 in section on ‘Chorea’). In addition, there is evidence to suggest that obsessive-compulsive behaviour, which is commonly seen in Gilles de la Tourette's syndrome, can occur secondary to pathology involving the basal ganglia or frontal lobes (Laplane et al. 1984, Ward 1988, Williams et al. 1988, Laplane et al. 1989, Rapoport 1990). Mejia and Jankovic (1995) investigated this aspect and found that similar to Tourette's syndrome, about half of the patients with secondary tics had obsessive-compulsive features. However, in contrast to Tourette's syndrome only a minority had attention deficit (present in one third of patients), behavioral (present in one fifth of patients), or sleep (present in only 5% of patients) problems. As discussed in Chapter 27, distinguishing between such behaviour and complex tics is not easy and the dividing line is somewhat arbitrary. None-the-less, obsessive-compulsive behaviour unaccompanied by other evidence of tics is not further considered in this chapter.

Childhood onset Gilles de la Tourette's syndrome with focal tics, cursing, and obscene gestures has been reported in association with tuberous sclerosis. Although both conditions may be apparent in the same patient, tuberous sclerosis may occur alone in other family members (Matthews 1981).

Citrullinaemia is a rare recessively inherited disorder of intermediate metabolism caused by arginosuccinate synthetase deficiency with resultant elevation of serum ammonia due to disturbance of the urea cycle. Intermittent ataxia may occur, but in addition recurrent tic and involuntary vocalization have been noted during bouts of hyperammonaemia (Drury et al. 1986).

In 1982 Sacks mentioned a patient with ‘Mongolism who developed Tourettism at the age of 30 years, coincident with the appearance of thyrotoxicosis’. Although this association initially seemed fortuitous, other patients with Down's syndrome have been noted to have Gilles de la Tourette's syndrome (Barabas et al. 1986, Karlinsky et al. 1986, Kerbeshian and Burd 2000). In some cases tics commenced during early childhood but in others they were not noted until adolescence or adulthood. It seems likely that in some cases they may be considered merely part of the abnormal behaviour associated with mental retardation and thus overlooked (Barabas et al. 1986). Both motor and vocal tics with coprolalia occur.

Trisomy 21 and mosaicism have been reported (Barabas et al. 1986). Another 21 chromosomal abnormality which has been associated with Gilles de la Tourette's syndrome is t(3;8)(p21;q24) (Gurling and Brett 1989).

A Gilles de la Tourette-like syndrome has been reported in association with several sex chromosome abnormalities including XYY karyotype (Merskey 1974), triple X 9p and mosaicism (Singh et al. 1982), the fragile-X syndrome (Kerbeshian et al. 1984, Schneider et al. 2008), and Duchenne's muscular dystrophy (Louis and Bertorini 1982). The disorder has also been found with 9 p monosomy (Taylor et al. 1991) and translocations involving 18q22, as mentioned in Chapter 27. These cases are probably best considered here, rather than as causes of idiopathic Gilles de la Tourette's syndrome, although the phenotype may be identical (also see Chapter 27 for genetics in Tourette's syndrome). Other genetic and chromosomal disorders rarely associated with tics reported in the literature include corpus callosum dysgenesia, mental retardation due to craniosynostosis, Arnold-Chiari malformation, Klinefelter's disease, Rett syndrome and neurofibromatosis.(Mejia and Jankovic 1995).

Gilles de la Tourette's syndrome has been noted in a small number of autistic patients (Kerbeshian and Burd 1986). In a study of 447 students in special school for autism, 6.5% had Tourette's syndrome, a figure that was more than expected by chance alone (Baron-Cohen et al. 1999). While prior neuroleptic drug administration might account for some cases (Stahl 1980, Barabas and Matthews 1983), it was not a feature in others (Realmuto and Main 1982). A significant association has been reported between autism and the c-Harvey-Ras gene. Comings et al. (1996) confirmed an association with this gene in autistic children but did not find it in those with Tourette's syndrome. Autism may also occur in the fragile-X syndrome, which normally results in mental retardation in males (Brown et al. 1982, August and Lochart 1984, Schneider et al. 2008). It is uncertain whether propensity to tics is related to autism, the fragile-X syndrome, or both. The idiot Savant syndrome, with calendrical calculating abilities, is characteristically associated with autism. The former has also been noted in Gilles de la Tourette's syndrome without autism and comment made on the common occurrence of obsessive-compulsive behaviour (Moriarty et al. 1993) and echolalia and echopraxia in both conditions (Ford 1989).

Cerebral haemorrhage and infarction can produce involuntary movements, but these are usually choreic or dystonic (see sections on ‘Chorea’, Section 5 and ‘Dystonia’, Section 8). Reports in the older literature occasionally refer to tic-like activity, but descriptions tend to be imprecise and of doubtful validity. Only rarely, documented stroke has been followed by a Gilles de la Tourette-like disorder or simple (phonic) tics (Bleeker 1978, Gomis et al. 2008) or by contralateral tics with other neurological deficits (Sacks 1982). Jankovic (1993) described a single case with imaging. This 8-year-old boy had a migrainous cerebral infarction of the right middle cerebral artery involving the head of the caudate and 2 weeks later had left hemidystonia followed by multifocal facial motor tics. Some cases in the literature are doubtful. For example, a 16-year-old boy with a left frontal haemorrhage from an arteriovenous malformation developed complex motor tics with head turning to the right, vocalization, and left arm clonic jerks (Yochelson et al. 2000). There was, however, a possibility that these were seizures. Against this argues that there was, however, no response to carbamazepine but to clonidine. Although transient tics occurring in older people may result from underlying vascular accident, evidence of cause is usually lacking (Sutula and Hobbs 1983).

Tic appears after perinatal brain damage. Kjaer et al. (1986) reported the onset of motor and vocal tics from 6 years of age after premature birth with left-sided seizures and hemiparesis. A large right tempero-parieto-occipital porencephalic cyst involving the basal ganglia and thalamus was present.

Increased prevalence of left-handedness and minor neurological signs in Gilles de la Tourette patients has prompted some authors to speculate that perinatal brain damage may be related to the disorder (Sweet et al. 1973). Definite evidence, however, is lacking.

Tics have occasionally been reported following head injury in children and adults. Erisson and Persson (1969) described onset of motor tics shortly after depressed temporal skull fracture in a 3 year old. Vocal tics and self-mutilatory behaviour subsequently followed. Family history of facial tics, however, makes the significance of the injury uncertain. Fahn (1982) reported the occurrence of tics within 3.5 months of closed head injury with ‘post-concussion syndrome’ at 18 years of age. There were dystonic movements and tics with vocalizations, but no word production. Other cases have been reported (Gaul 1994, Krauss and Jankovic 1994, Alegre et al. 1996). Adult onset tics are often associated with a history of either head or peripheral trauma (Factor and Molho 1997, Chouinard and Ford 2000). Chouinard and Ford (2000) found that three of their 13 patients with new onset tics after age 21 had onset following trauma, two of whom had head injury due to a motor vehicle accident. In most cases of tics with craniocerebral trauma due to motor vehicle accidents MRI brain imaging has been normal (Krauss and Jankovic 1994, Chouinard and Ford 2000). The latency from trauma to onset has been reported to vary from days to several months. Tics caused by peripheral trauma, for instance after neck strain (Choiunard and Ford 2000, Erer and Jankovic 2008), appear to be single, non-varying isolated tics, usually in the injured part itself or in the immediate vicinity. The pathogenesis of post-traumatic tics remains unclear and, as mentioned, usually there does not appear to be obvious structural damage. One view is that trauma may be decompensating a latent tic disorder in those who may be genetically or otherwise predisposed to develop Tourette's syndrome and earlier may even have had some features (apart from tics) of the same on careful enquiry, as exemplified by the case of Kumar and Lang (1997). They warned of a serendipitous association, given the infrequent reports, possible medicolegal implications, lack of sound biological association, and frequency of trauma (Kumar and Lang 1997).

An unusual disorder of late onset, induced by an unexpected startling stimulus, has been reported to occur after physical trauma (Tijssen et al. 1999). The startle reflex itself was, however, normal. The authors suggested that this was distinct from psychogenic myoclonus.

Although tics have been described in a variety of infectious and parainfectious disorders, interpretation of the type of involuntary movement is open to question. Sudden choreic or myoclonic jerks may be misinterpreted, making it difficult to evaluate reports of tics in disorders such as Sydenham's chorea and Jakob–Creutzfeldt disease. Encephalitis lethargica, however, seems to have frequently been accompanied by tics in addition to chorea (see under Chapter 24). Von Economo (1917–18, 1929) described occasional tics during the acute phase but noted that they were more common in the chronic stage. Such tics include simple motor movements as well as sniffing, coughing, yawning, clicking, and the like (Van Bogaert and Nyssen 1925, Wilder and Sulbermann 1927, Weingarten 1968). Once developed, post-encephalitic tics do not progress and may be localized to the face more than in Gilles de la Tourette's disease. True vocalization may occur but coprolalia is infrequent (Benedek 1925, Van Bogaert 1934, Weingarten 1968). Klazomania (compulsive shouting) tends to be associated with oculogyric crises (Schwab et al. 1951, Weingarten 1968). There may also be bouts of obsessive-compulsive activity, including forced thinking and counting. Whether attacks of forced ocular deviation should just be regarded as another form of compulsive motor activity (Devinsky 1983) remains uncertain. It has been postulated that these symptoms may arise from severe damage to substantia nigra pars reticulata and ventral tegmental area (Devinsky 1983) (see later under ‘Drug-induced tic’ later).

A Gilles de la Tourette's-like syndrome has been reported following a presumptive (Northam and Singer 1991) and biopsy proven (Turley 1988) herpes simplex encephalitis involving the temporal lobe. Multiple tics have also been noted following cerebral malaria (Davis and Knezevic 1994). A 40-year-old man with advanced HIV infection was reported to have developed motor and vocal tics (McDaniel and Summerville 1994) [see Cardoso (2002) for review of HIV-related movement disorders], cognitive impairment, and a blunted affect. Investigations showed elevation of CSF protein, non-specific slowing on electroencephalography, and a normal MRI brain scan. The pathophysiological basis of the tic disorder was a presumed encephalitis due to HIV.

As mentioned in Chapter 27, the onset of tics in some children is associated with the presence of antineuronal antibodies and serological evidence of beta-haemolytic streptococcal infection, in much the same way as occurs in Sydenham's chorea (see Chapter 22). Some such cases are said to have developed features indistinguishable from Gilles de la Tourette's syndrome, including attention deficit hyperactivity disorder and obsessive-compulsive behaviour. This has led to the the notion of a spectrum of immune-mediated neurobehavioural syndromes termed PANDAS (standing for paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection). Possibly, the first report about this was from Japan. An 11-year-old developed a Tourette-like syndrome soon after a febrile illness with elevated antistreptolysin antibody titres and had dramatic improvement of the tics with steroid treatment but not with haloperidol (Kondo and Kabasawa 1978). Following this were other similar reports (Kerbeshian et al. 1990, Matarazzo 1992). Kiesling et al. (1993) described a temporal relationship between streptococcal infection and onset of tics with the presence of antineuronal antibodies to the human caudate nucleus in 45% of such cases versus 20% of controls. The situation thus has some parallels with Sydenham's chorea, which is discussed in Chapter 24. Also a high rate of obsessive-compulsive disorder was found in Sydenham's chorea (Swedo et al. 1989). Furthermore, onset or worsening of obsessive-compulsive disorder and tics with evidence of recent streptococcal infection was reported by the same group in two boys with immediate benefit from plasmapheresis and intravenous immunoglobulin (Allen et al. 1995). Others have described the same sequence of events (Allen et al. 1995, Budman et al. 1997). Similarly, Tucker et al. (1996) reported a case with tics, obsessive-compulsive disorder, and circulating antistriatal antibodies who responded to penicillin and plasma exchange therapy but not conventional drugs. Basal ganglia enlargement, particularly of the caudate nucleus, has been noted in Sydenham's chorea, post-streptococcal tics, and obsessive-compulsive disorders (Tucker et al. 1995, Giedd et al. 2000). It is thought to represent an antibody immune mediated inflammatory reaction (Giedd et al. 2000). The patient reported by Tucker et al. (1995), mentioned above, had reduction of the caudate volumes following treatment in parallel with clinical improvement. Other studies also suggest that both streptococcal and other infection may be a stimulus for a tic and obsessive-compulsive disorder (Singer et al. 2000).

Using the clinical criteria mentioned above, Swedo and colleagues (1998) gave the clinical description of 50 such paediatric cases. The children's symptom onset was acute and dramatic, typically triggered by group A beta-haemolytic streptococcal infections at a very early age (mean = 6.3 years for tics; mean = 7.4 years for obsessive-compulsive disorder). The PANDAS clinical course was characterized by a relapsing–remitting symptom pattern with significant psychiatric comorbidity accompanying the exacerbations. Emotional lability, separation anxiety, night-time fears, bed-time rituals, cognitive deficits, oppositional behaviours, and motor hyperactivity were particularly common. Symptom onset was triggered by group A beta-haemolytic streptococcal infection for 22 (44%) of the children and by pharyngitis (no throat culture obtained) for 14 others (28%). Among the 50 children, there were 144 separate episodes of symptom exacerbation with either documented group A beta-hemolytic streptococcal infection or with symptoms of pharyngitis or upper respiratory infection. The authors felt that these diagnostic criteria appeared to accurately characterize a homogeneous patient group in which symptom exacerbations were triggered or worsened by infections (Swedo et al. 1998).

It has also been suggested that these patients may have an inherited susceptibility to produce an abnormal immune response to the pathogen (Murphy et al. 1997, Swedo et al. 1997). In this regard, using a monoclonal antibody, 85% patients with PANDAS had the D8/17, a B cell surface marker. This was also found in 89% of patients with Sydenham's chorea but not in controls (17%), suggesting an abnormal susceptibility or reaction to streptococcal infection in both conditions (Murphy et al. 1997, Swedo et al. 1997, Kurlan, 1998). Lougee et al. (2000) made the interesting observation that the rate of tic and obsessive-compulsive disorders in the first-degree relatives of paediatric probands with PANDAS was higher that that in the general population. These studies need to be replicated, but may suggest a genetic predisposition interacting with environmental factors.

Not all agree that tic and obsessive-compulsive disorders are related to streptococcal antibodies. Peterson et al. (2000) found that attention deficit hyperactivity disorder, but not chronic tic or obsessive-compulsive disorders, was associated with a significantly increased antistreptococcal antibody titre. When MRI basal ganglia volumes were included in the analysis the relationship between antibody titres and basal ganglia volumes were significantly different in the obsessive-compulsive and attention deficit cases compared with either those with tics or controls. Only in these two groups were higher antibody titres associated with larger putamen and globus pallidus volumes (Peterson et al. 2000). Further studies are therefore required. It would seem that not all cases of Tourette's syndrome fall in the category of PANDAS and it is likely that these post-infectious immune mediated mechanisms are relevant to only a small subset of such patients (Kurlan 1998). Kurlan (1998) reviewed some of the objections to the streptococcal infection theory in the pathogenesis of Tourette's syndrome. Unlike Sydenham's chorea there is no clear correlation between the presence or degree of auto-antibodies and the severity of symptoms in Tourette's syndrome. Also antineuronal antibodies are found in a substantial number of controls and are not specific for Tourette's syndrome and/or obsessive-compulsive disorder. Unlike Sydenham's chorea, other manifestations of rheumatic fever, like valvular heart dysfunction and polyarthritis, are never seen in Tourette's syndrome. Also, the key criteria of PANDAS, as per definition, do not appear to differentiate the condition from typical Tourette's syndrome. Prepubertal age of onset can be true for both. Tics can appear to be sudden in onset in typical Tourette's. A waxing and waning course is seen in Tourette's and exacerbations and remissions, as defined for PANDAS, may be difficult to differentiate, but severe explosive exacerbations are not seen in typical Tourette's syndrome. While only a small fraction of Tourette's cases, mostly children, may be in the PANDAS category, it is important to recognize these because they may benefit from immune modifying treatment. Perlmutter et al. (1999) showed that plasma exchange or intravenous immunoglobulin were effective in lessening symptom severity in children with infection triggered tic and obsessive-compulsive disorders. The authors mentioned that further studies were required to determine which children would benefit from such therapies (Perlmutter et al. 1999). They are currently not recommended for all cases with Tourette's syndrome because of both cost and potential risk of complications (Kurlan 1998).

Gilles de la Tourette's syndrome with simple and complex motor tics, vocalizations, and coprolalia has been reported in association with pinealoma. Additional small areas of calcification in walls of the third ventricle, lateral thalamus, and mesencephalic grey matter ventral to the aqueduct of Sylvius may have been involved in production of tics (Lakke and Wilmink 1985).

Tic occasionally develops in elderly patients and can involve complex actions plus vocalizations. It may thus present as full-blown Gilles de la Tourette's syndrome and respond to dopamine receptor blocking drugs (Sutula and Hobbs 1983). Aetiology is uncertain although gradual onset argues against cerebral infarction or haemorrhage. However, Chouinard and Ford (2000) in a review of adult onset tics, including some elderly cases, pointed out that new onset tics in this population were more likely to be symptomatic or secondary. In their series, infection, trauma, cocaine use, and neuroleptic exposure were the most common causes.

Carbon monoxide poisoning can produce motor and vocal tics in addition to a more diffuse encephalopathy. The tics may resemble Gilles de la Tourette's syndrome with repetitive movements and involuntary speech, including coprolalia, echolalia, palilalia, klazomania, and inappropriate sexual activity. In a case reported by Pulst et al. (1983) CT scan showed ventricular enlargement and low density areas in the region of the basal ganglia.

Motor and vocal tics in a 5-year-old boy have been reported following use of Chinese medicinal herbal spray called ‘watermelon frost’ which was found to have a very high unorganic mercury content (Li et al. 2000). There was elevation of blood mercury content. The tics resolved after 4 weeks with the return of blood mercury level to normal. It was postulated that mercury may combine with the sulphydryl group of S-adenosylmethionine, which acts as a cofactor for catecholamine-O-methyltransferase (COMT), leading to accumulation of catecholamines with resultant tics (Li et al. 2000).

As mentioned above, dopamine receptor blocking drugs decrease tics in Gilles de la Tourette's syndrome. In addition, administration of l-dopa may increase tic frequency (Sweet et al. 1976). In keeping with this concept that dopamine receptor under-activity improves and over-activity exacerbates tics, Shale et al. (1986) reported loss of tics in Gilles de la Tourette's syndrome subsequent to the development of Parkinson's disease and recurrence of tics following l-dopa treatment. However, as mentioned in Chapter 27, the same was not observed by Kumar and Lang (1997), who described patients with tics and coincident Parkinson's disease without worsening of tics after levodopa treatment.

Although tics and obsessive-compulsive behaviour may occur in association with oculogyric crises in post-encephalitic Parkinson's disease, response to l-dopa is variable. Improvement and deterioration of both aspects have been reported. Sometimes, however, tics first appear after initiation of therapy (Calne et al. 1969, Yahr et al. 1969, Sacks and Kohl 1970, Duvoisin et al. 1972, Sacks 1976). l-dopa has also been reported to induce motor and vocal tics after brain injury (Klempel 1974).

Occasional patients with Gilles de la Tourette's syndrome on dopamine receptor blocking drugs have developed tardive dyskinesia (Mizrahi et al. 1980, Shapiro and Shapiro 1982) or akathisia (Weiden and Bruun 1987). This may exacerbate the condition and necessitate withdrawal of this therapy. The proportion of cases developing this complication, however, seems to be very small. More importantly, a tardive form of Gilles de la Tourette's syndrome has been described in a small number of patients taking neuroleptic therapy for other reasons, often schizophrenia (Klawans et al. 1982, Mueller and Aminoff 1982, Bharucha and Sethi, 1995). This side effect is also rare and in a review Kumar and Lang (1997) mention only 15 such cases in the literature. Motor tics and involuntary vocalizations have in some cases been accompanied by choreic or dystonic movements, involving particularly the face and upper limbs. In most cases neuroleptic medication has been taken for a number of years prior to these involuntary movements. Like the choreic variety of tardive dyskinesia, tics may be worsened or appear for the first time on neuroleptic withdrawal and are improved by increasing or reinstituting this therapy. Although tics have sometimes cleared within several months of neuroleptic withdrawal, they have persisted during follow-up, lasting several years in other patients (Lal and Al Ansari 1986). Tardive syndromes are uncommon with atypical neuroleptics particularly clozapine. There is, however, a report of tics developing with clozapine treatment (Lindenmeyer et al. 1995). Tics were said to have occurred 2–5 months after the institution of and during treatment, but it was unclear whether it was withdrawal of typical neuroleptics or the clozapine itself that caused the appearance of tics (Lindenmeyer et al. 1995).

Rarely, anticholinergics and antihistaminics can worsen or sometimes induce tics (Tanner et al. 1982, Shafi, 1986). The onset of, or exacerbation of tics due to cocaine has been reported (Mesulam 1986, Factor et al. 1988, Pascual-Leone and Dhuna 1990, Attig et al. 1994, Daniels et al. 1996, Chouinard and Ford 2000). Whether this drug acts directly or via an effect on the dopaminergic system is uncertain. The role of central nervous system stimulating drugs in exacerbating or triggering tics is mentioned in Chapter 27 under ‘Gilles de la Tourette's syndrome’.

Onset or exacerbation of pre-existing mild tics has been reported with carbamazepine given as anti-epileptic therapy. Electroencephalography during tics has not shown associated paroxysmal discharges, and CT and MRI brain scans have been normal (Robertson et al. 1993). These changes have occurred without clinical or laboratory evidence of anticonvulsant intoxication. Tics have occurred within days or weeks of commencing carbamazepine. In some (Robertson et al. 1993) but not all (Neglia et al. 1984) cases the tics have settled following drug withdrawal. In others, the movements have resolved despite continuing and even increasing the dose of medication (Robertson et al. 1993). Similarly, phenytoin can rarely induce tics (Drake and Cannon 1985). Motor and vocal tics have also been noted with phenobarbital (Burd et al. 1986, Sandyk 1986).

There has been a suggestion in the earlier literature that tricyclic antidepressants, like imipramine, could precipitate or worsen tics (Fras 1978), but there has been some doubt as the same effect was not found in controlled trials of clomipramine and other tricyclics (Caine et al. 1979). The newer category of serotonin reuptake inhibitor antidepressants can rarely cause worsening of pre-existing tics or de novo tic disorder (Cunningham et al. 1990, Eisenhauer and Jermain 1993, Gatto et al. 1994, Hauser and Zesiewicz 1995, Bharucha and Sethi 1996).

A tic disorder with some encephalopathic features including echolalia, echopraxia, orofacial, and limb automatisms and amnesia has been reported in a 71-year-old man with ofloxacin, a fluoroquinolone (Thomas and Reagan 1996). It resolved with discontinuation of the ofloxacin.

Transient recurrence of previously quiescent tics after alcohol withdrawal in patients with alcoholism has been reported (Sandyk and Gillman 1985).