|Home | About | Journals | Submit | Contact Us | Français|
Lesch–Nyhan disease (LND) is caused by deficiency of the purine salvage enzyme hypoxanthine-guanine phosphoribosyltransferase (HPRT). Affected individuals exhibit over-production of uric acid, along with a characteristic neurobehavioural syndrome that includes mental retardation, recurrent self-injurious behaviour and motor disability. Prior studies involving relatively small numbers of patients have provided different conclusions on the nature of the motor disorder. The current study includes the results of a multi-centre international prospective study of the motor disorder in the largest cohort of patients studied to date. A total of 44 patients ranging from 2 to 38 years presented a characteristic motor syndrome that involved severe action dystonia superimposed on baseline hypotonia. Although some patients also displayed other extrapyramidal or pyramidal signs, these were always less prominent than dystonia. These results are compared with a comprehensive review of 122 prior reports that included a total of 254 patients. Explanations for the differing observations available in the literature are provided, along with a summary of how the motor disorder of LND relates to current understanding of its pathophysiology involving the basal ganglia.
Lesch–Nyhan disease (LND) is an X-linked recessive disorder caused by deficiency of the purine salvage enzyme hypoxanthine-guanine phosphoribosyltransferase (HPRT). Affected individuals have hyperuricaemia and a characteristic neurobehavioural phenotype that includes mental retardation, self-injurious behaviour and motor disability (Lesch and Nyhan, 1964; Jinnah and Friedmann, 2000). A number of detailed studies have addressed the genetic (Jinnah et al., 2000; Jinnah et al., 2004), metabolic (Seegmiller, 1989; Jinnah and Friedmann, 2000; Puig et al., 2001), cognitive (Anderson et al., 1992; Matthews et al., 1995; Schretlen et al., 2001) and behavioural (Nyhan, 1976; Anderson and Ernst, 1994; Robey et al., 2003; Schretlen et al., 2005) features of this disease. In comparison, fewer studies have specifically addressed the motor disorder.
In their original report of two affected brothers in 1964, Lesch and Nyhan described choreoathetosis and increased muscle tone as the salient motor abnormalities (Lesch and Nyhan, 1964). Subsequently, several retrospective studies described larger numbers of patients. The largest of these included 19 patients, all of whom were described as having choreoathetosis, with a minority also having dystonia (Christie et al., 1982). All of these patients were described as also showingsigns of corticospinal dysfunction, such as scissoring of the legs or extensor plantar reflexes. The second largest series included 14 patients, most of whom were described as having choreoathetosis and spasticity (Mizuno, 1986). In keeping with these reports, most modern textbooks of paediatric neurology emphasize choreoathetosis and pyramidal signs in their descriptions of LND. However, one of the most detailed studies that included eight patients described dystonia rather than choreoathetosis as the major motor abnormality (Watts et al., 1982). All of these patients also were described as having hypotonia rather than hypertonia, and pyramidal signs were uncommon. Another five patients were described as showing a combination of choreoathetosis, dystonia, ballismus and ataxia (Jankovic et al., 1988). Pyramidal signs were not noted in these patients. Finally, many reports of smaller numbers of cases have described various combinations of choreoathetosis, dystonia, ballismus, spasticity, hypotonia or ataxia. The reasons for these incongruent descriptions of the motor disorder of LND remain unclear.
To further delineate the motor disorder of LND, 44 patients from 40 different families were evaluated systematically by the same group of investigators using defined criteria for the diagnosis of motor disorders. The evaluations revealed a relatively uniform motor phenotype that consists of severe generalized dystonia superimposed on hypotonia, sometimes with less prominent choreoathetosis or spasticity. These results are compared with a comprehensive review of the published literature, and the potential reasons for the differing descriptions are considered. The motor phenotype is also considered in the light of current knowledge of pathophysiology in LND.
This study included 44 LND patients. They were recruited after referral for clinical evaluation and management, after referral to centres devoted to the long-term care of patients with developmental disabilities or metabolic diseases, or after participation in other studies. A detailed assessment of the motor syndrome has not been reported for the current patients, excepting no. 29 (De Gregorio et al., 2005). Some of the results have been summarized in abstract form (Jinnah et al., 1998; Visser et al., 2004).
The focus of this study was on the motor features of classic LND due to complete HPRT deficiency. A definitive molecular (hprt gene mutation predicting null enzyme function) or biochemical (<1% residual HPRT enzyme activity in blood or fibroblast samples) diagnosis was established for 40 of 44 patients. Of the remaining four patients, three declined laboratory confirmation and one died before testing could be arranged. Among these four patients were three who met clinical criteria for LND with expression of the full syndrome including mental retardation, self-injurious behaviours, motor disability and over-production of uric acid. The remaining patient (no. 3) exhibited the full syndrome except for self-injurious behaviours. This patient was included because he had a cousin who was diagnosed with LND on the basis of the full clinical syndrome that included self-injury. Because the focus of this study was on the classic form of the disease due to complete HPRT deficiency, attenuated or partial forms of the disease known to be due to partial HPRT deficiency were excluded. Only one patient with classic LND who was evaluated for these studies was excluded, because of an incomplete record.
The first author directly evaluated 33 of the 44 patients. The remaining nine patients were evaluated by clinicians who had experience in movement disorders or had worked with the first author during evaluation of other LND cases. Detailed videotapes available for three of these patients were reviewed by the first author to confirm diagnostic impressions. The evaluation began with a complete history. Before their evaluation for this study, many patients had been evaluated only once by a general neurologist and subsequently followed in institutions or clinics devoted to paediatrics or metabolic diseases. Therefore, historical records on the evolution of neurological symptoms and signs were not always available.
The examination consisted of a standardized written protocol with specific attention to the motor features. Video recordings of the examination were made for 33 patients. All movement disorders and assessments of muscle tone were defined according to standard published criteria (Barbeau et al., 1981; Fahn, 1988; Jankovic and Fahn, 1998; Sanger et al., 2003). Ballism was defined as involuntary rapid flinging movements, typically of the limbs. Choreoathetosis was defined as involuntary non-rhythmic and unpatterned movements with variable timing and distribution. Dystonia was defined as involuntary, sustained, patterned and sometimes repetitive contractions of opposing muscles often leading to twisting movements or abnormal posturing. Spasticity was defined as an abnormality of muscle tone with a velocity-dependent increase in resistance to externally imposed movement, often worse in extension than flex-ion. The determination of spasticity was made independent from other pyramidal signs. Dystonia was discriminated from spasticity by the lack of velocity-dependence, exaggeration by attempted action and frequently marked variation in tone with emotional state (Sanger et al., 2003).
This study is primarily descriptive, but to obtain an estimate of the severity of individual motor defects for comparative purposes, we applied a semi-quantitative rating scale that was modelled after one used in a prior study of LND (Jankovic et al., 1988). The scale was defined as follows: 0 = motor abnormality absent; 1 = motor abnormality mild and insufficiently severe to be significantly disabling; 2 = motor abnormality moderate with significant disability but most functions still possible with added effort; 3 = motor abnormality severe enough to preclude meaningful function.
The Medline database through December 2005 was reviewed for reports that included the keywords Lesch–Nyhan or hypoxanthine-guanine phosphoribosyltransferase. Additional patients were identified by surveying the reference lists of these and other articles. A list of the 122 reports reviewed is available from the authors by request. The clinical data were recorded in a database, and information for patients reported more than once was combined into a single entry wherever possible. Because the focus of this study is on the classic form of the disease, reports of attenuated variants due to partial enzyme deficiency were again excluded. Because the disorder is X-linked and recessive, virtually all cases were male, and the rarely reported female cases were excluded.
The 44 patients ranged from 2 to 38 years (Table 1). They were followed for an average duration of 3.3 years (range 0–18 years). All but one of the patients were male. The female patient had a clinical phenotype indistinguishable from that of the males.
All but one patient were receiving allopurinol for control of uric acid. Twelve were receiving medications to alkalinize the urine to promote uric acid solubility. Eleven were taking medications for gastro-oesophageal reflux.
Seven were taking medications to reduce muscle tone (baclofen, dantrolene, paracetamol). Nineteen were taking benzodiazepines to alleviate both increased muscle tone and mood lability (diazepam, clonazepam, alprazolam). Seven were taking dopamine receptor antagonists (risperidone, thioridazine, olanzepine) for control of self-injury and two were taking metaclopramide to promote gastrointestinal motility. Eleven were taking medications for mood stabilization or depression (paroxetine, gabapentin, carbamazepine, clomipramine, sertraline). Though this study was not designed to assess treatment responses, none of these medications was clearly associated with a dramatic improvement of the motor disorder. Motor abnormalities among those taking dopamine receptor antagonists were similar to those who had never taken them.
Information concerning presenting signs was available from the histories in 39 patients. All but two first came to medical attention for motor delay (Table 1). Motor delay became apparent most often at 3–6 months of age as hypotonia, with failure to sit upright unsupported. Overt involuntary movements most often were noted between 6 and 12 months of age, although abnormal movements sometimes were appreciated by parents or caretakers earlier, or delayed for up to 4 years.
After some evolution during the first few years, the severity of the motor disorder appeared relatively static rather than progressive in most cases. The majority of patients were reviewed more than once over a period of several years, and progression of the motor disorder was not significant beyond 5–6 years of age. The oldest patients did not appear to be more seriously affected than the youngest ones. However, several patients evidenced signs of increasing disability from long-standing abnormal movements. For example, three patients had histories noting objective signs of worsening motor function, such as loss of grasp or ability to stand with assistance, often in a stepwise fashion. Six developed severe scoliosis, and one required a surgical procedure to stabilize the spine.
Only two patients presented first with difficulties attributable to excessive production of uric acid. One presented with lactic acidosis, presumably due to nephrolithiasis with obstruction of urine flow and renal failure. The other presented with orange crystals in the diapers, most probably due to crystalluria combined with microhaematuria. No patient presented with self-injurious behaviours, which consistently emerged later than the other problems, usually between 1 and 8 years of age.
Abnormal movements were relatively minor at rest in all patients, and became most obvious with stress, excitement, anticipation or voluntary movement. The most prominent feature of the motor syndrome in all patients was dystonia (Table 2, Fig. 1). Essentially, all parts of the body were affected. Multidirectional cervical dystonia was universal. Truncal twisting and arching was present in all, particularly with efforts to stand. Dystonia of the upper limbs prevented their use for most tasks such as feeding or grasping in all patients. All regularly used wheelchairs because lower limb dysfunction prevented them from walking or standing unassisted. Oromandibular and lingual dystonia were evident during speaking or eating in most. Several also exhibited blepharospasm, most prominently during ocular testing. Several developed fixed abnormal postures of the hands or feet, and fixed contractures of the hamstring muscles with incomplete extension at the knee were common. Muscle hypertrophy resulting from long-standing dystonia was evident in several patients in the neck and arms.
Severe opisthotonus or truncal arching was observed directly in 11 (25%), and movements compatible with opisthotonus were described frequently in the medical records as severe or prolonged ‘arching’ or ‘backward bending’ of the trunk. Sudden and rapid backward thrusting of the head without concomitant truncal involvement was observed directly in four and described in the records of several others. Such spasms were absent at rest, but emerged when patients attempted to stand or were transferred from the seated to supine position. Severe, sustained truncal arching with dystonic tremor sometimes resembled a generalized epileptic seizure, except that consciousness was preserved and there was no post-ictal depression of consciousness.
Although dystonia was universal and always the most severe extrapyramidal disorder, choreoathetosis was present in 22 (50%) and ballism was observed in 13 (30%). Ballism was most common in the arms, though it occasionally appeared in the legs. These movements were seen almost exclusively with excitement or agitation. Ballism was often triggered by the presence of an object near enough to strike, raising the question of whether this represented impulsive or voluntary acts of aggression. However, the movements otherwise appeared typical of ballism, and their speed and fluidity contrasted with the marked impairment in all other voluntary movements.
When fully relaxed, 40 (91%) patients were considered to have hypotonia (Table 2; Fig. 1). Spasticity was noted in 10 (23%) patients. In six of these the spasticity was limited to the legs, and in three others it was markedly asymmetrical. Rigidity was noted in six (14%), but cogwheeling was not apparent.
A total of 23 patients (52%) had pathologically brisk muscle stretch reflexes (Table 2). Pathological reflexes were limited to the legs in 17 of these patients, and they were asymmetrical in 5. Clonus was observed in 14 (32%) cases, and in 13 of these cases it was limited to the legs.
Several patients demonstrated an abnormal resting posture of the great toe. In some, the toe was plantar flexed and deviated laterally at the metacarpophalangeal joint in a fixed position above or below the adjacent toe. In others, the toe remained in extension for lengthy periods without plantar stimulation, but eventually returned to a normal position. Plantar stimulation often led to varied responses, with an extensor on some occasions and a flexor on others. These phenomena were considered more characteristic of a striatal toe response than a classic Babinski sign, but distinction between these signs was not attempted.
Speech was delayed in all patients for whom developmental histories were available. Most began using words by 2–4 years of age. A few were not speaking at much older ages, such as one who was still not using words by age 14 years.
When speech occurred, it was always dysarthric. In all patients, dysarthria was severe enough that speech was difficult to understand without frequent repetition. One was anarthric because of extreme cervical and oromandibular dystonia but communicated effectively with a keyboard. Most communicated using only single words or very short phrases. Speech had the typical characteristics of striatal dysarthria, with slow and irregular enunciation associated with excessive activation of pharyngeal, lingual, and perioral muscles. During casual conversation, a trailing volume was common and sufficiently soft to be considered pathologically hypophonic in 10 (23%) patients, and 2 (4%) of these also exhibited stuttering without obvious jaw dystonia. Spasmodic dysphonia with intermittent breathiness, intermittent strained-strangled quality or phonic breaks was not apparent. None had a high-pitched, nasal or strained-strangled quality suggestive of spastic dysarthria. None displayed poor modulation of tempo or volume suggestive of cerebellar dysarthria.
Chewing and swallowing was laboured in most patients. Many also had gastro-oesophageal reflux with frequent emesis (Table 1). Together, these problems were sufficiently severe to require gastrostomy tube in nine (20%) patients.
Respiratory abnormalities were evident in several patients. Two patients had prominent inspiratory stridor without cyanosis or apnoea during examination and another had a history of recurrent ‘noisy breathing’ (Table 2). In all three patients, the respiratory problems were noted during wakefulness and not during sleep. Two other patients had frequent episodes of apnoea and cyanosis without stridor, also during wakefulness but not during sleep. Extensive pulmonary evaluations were normal for both, and EEG studies provided no evidence for epilepsy. Two other patients each had at least one episode of cardiopulmonary arrest leading to multiple hospitalizations, and extensive evaluations failed to disclose a cause. One of these patients expired without obvious cause, presumably due to an unwitnessed respiratory event. Another case with no history of cardiopulmonary events expired during sleep with no apparent cause.
Three patients had histories compatible with generalized tonic–clonic epileptic seizures, though none was witnessed by the evaluators, and the possibility of prolonged opisthotonic posturing with dystonic tremor could not be excluded. One case had episodic and briefly sustained upwards or lateral conjugate eye deviation resembling ocular tics. Two cases with a history of neuroleptic use exhibited repetitive orobuccolingual stereotypies typical of tardive dyskinesia. Other tics or stereotypies were absent. None of the patients had evidence for cerebellar dysfunction such as appendicular or axial ataxia or ocular dysmetria. Tremor and myoclonus were absent.
A total of 122 reports describing 254 LND patients reported from 1959 to 2005 were examined. Overall, the average age at presentation was 6.1 months, with a range of 1 day to 7 years. The average age at the time of case description was 9.5 years, with a range of 1 day to 33 years.
At least one presenting problem was recorded for 158 patients, and 130 of these included neurological problems (Table 3). The most common presenting problems were signs of motor delay, abnormalities of muscle tone or ‘cerebral palsy’. These descriptions are not mutually exclusive, as all indicate abnormal motor development. Less common neurological presentations included opisthotonus, apparent epileptic seizures, ‘spastic torticollis’, fisting of the hands, scissoring of the legs, nystagmus and microcephaly.
Presenting problems among the remaining patients were due mostly to excessive production of uric acid (Table 3). Most commonly, these included renal failure or uric acid crystals in the urine. Gout was uncommon, being noted as the presenting feature for only three patients. Only two patients presented with self-injury. A few patients were noted to present with such non-specific problems as ‘colic’, ‘irritability’ or difficulty feeding. A few pre-symptomatic patients also were identified because of a known affected relative.
No prospective studies of the evolution and natural history of LND have been reported. However, in one retrospective study of eight patients followed for up to 8 years, most presented with hypotonia and/or motor delay between 3 and 9 months of age, and then developed dystonia between 6 and 24 months of age (Watts et al., 1982). Another study of six patients followed over several years provided a similar picture of early motor delay followed by development of athetosis and spasticity (Michener, 1967). Limited information concerning the natural history of neurological features was reported for a total of 69 patients. The most frequently noted initial neurological problems included hypotonia and delayed motor development. Involuntary movements or spasticity usually were not present at birth, but emerged at an average of 16 months (range 1 month to 6 years). Involuntary movements or spasticity then appeared to remain static, as few were described as showing progressive motor dysfunction beyond 2 years.
A summary of all neurological features noted at the time each case was reported is provided in Table 4. The neurological features were described with varying degrees of detail. No information concerning the neurological features was provided for 34 patients. A detailed description of the neurological features was available for only 42 patients. Among the 220 patients for whom any neurological information was provided, various extrapyramidal signs were described. These included chorea, choreoathetosis, athetosis, dystonia, opisthotonus and ballismus. Most patients were reported to have more than one extrapyramidal sign. Other features commonly reported as pyramidal signs included spasticity, hyperreflexia, an extensor plantar reflex, scissoring of the legs and clonus.
Descriptions of muscle tone varied. Among the 96 patients for whom muscle tone was described, hypertonia was noted in 69%, hypotonia in 46% and both hypotonia and hypertonia in 14 of these patients. Increased muscle tone was most often described as spasticity or dystonia, although such non-specific terms as ‘hypertonic’ or ‘stiffness’ were used for several cases. Only three patients were described as having rigidity.
Uncommonly reported neurological features included apparent epileptic seizures, ataxia, unspecified dyskinetic movements and myoclonic-like limb jerks. Varying degrees of self-injurious behaviours were documented for nearly all patients but other motor stereotypies and motor or vocal tics were not apparent. None was reported to have tremor.
Brain CT scans have been reported for 22 patients, including 12 that were part of a retrospective study (Table 5). Atrophy was reported for four, with the remainder being interpreted as normal. Results of brain MRI have been reported for 25 patients. Only three were noted to have visible atrophy. However, a 17% reduction of total cerebral volume and 34% reduction in basal ganglia volumes were found in a quantitative study when seven patients were compared with normal controls (Harris et al., 1998).
Although most reports use the term ‘atrophy’ to describe cerebral volume loss, concurrent enlargement of the cranial sinuses in one imaging study suggested that the volume loss reflects poor brain development rather than a degenerative process (Harris et al., 1998). Therefore, the term ‘dystrophy’ might be more appropriate.
Autopsies including brain have been reported for a total of 23 LND patients (Table 6). These studies have not revealed any consistent neuropathological changes. Notably, complete neuropathological studies were entirely unremarkable for six patients, including electron microscopic studies for at least two. Though multiple different abnormalities have been reported, the lack of consistency suggests that many of the findings may be incidental or related to other events occurring at the time of death.
Although a literature review suggests considerable variation in the motor disorder of LND, our evaluation of the largest series of patients currently available demonstrates a characteristic motor syndrome with only minor phenotypic variability. The motor disorder characteristically evolves in a manner similar to that of dyskinetic cerebral palsy, with hypotonia and/or delayed acquisition of motor skills within the first 3–6 months of age. Involuntary movements usually develop between 6 and 24 months of age. Once fully developed, the clinical course remains relatively static with severe action dystonia superimposed on a baseline of hypotonia. Many also display choreoathetosis or ballismus, but these features are always less severe than dystonia. Although many have pathological reflexes or spasticity, few have clonus. When these pyramidal signs are evident, they typically are mild, limited to the legs or asymmetrical.
The relatively consistent motor syndrome observed in our prospectively evaluated patients contrasts with the varied descriptions of motor abnormalities reported previously and reviewed in this study. One possible explanation for this discrepancy is that most cases were reported in non-neurological journals, often with only brief neurological evaluations made by examiners with little or no formal training in the diagnosis of motor disorders. It is widely recognized that the terminology of movement disorders has been applied using different criteria by different medical subspecialties for many years, and many of the reports on LND patients were published before the establishment of standard definitions and criteria for many motor disorders (Barbeau et al., 1981). Some uncertainties concerning nomenclature, particularly those relating to muscle tone in children, have persisted even after these definitions were established. These uncertainties have prompted recent efforts to clarify them further (Delgado and Albright, 2003; Sanger et al., 2003). To minimize difficulties associated with nomenclature, each case in the current series was examined by at least one neurologist with special training in movement disorders, using currently accepted criteria for diagnosis of abnormal motor syndromes.
Another reason for the inconsistent descriptions of the motor syndrome in LND is that some manifestations of dystonia closely resemble other neurological signs often associated with dysfunction of the corticospinal motor systems. Several major signs have been cited as evidence for corticospinal dysfunction in LND, including spasticity, the extensor plantar reflex, scissoring of the legs, hyperreflexia and clonus. The increased muscle tone of dystonia is readily mistaken for spasticity without special attention to their discriminating features. The extensor plantar reflex, or Babinski sign, is not a reliable indicator of corticospinal dysfunction among patients with extrapyramidal disease, because it cannot be reliably discriminated from the striatal toe, which is a form of dystonia (Nausieda et al., 1980; Jan, 2004; Ashour et al., 2005). Scissoring of the legs is typical in corticospinal dysfunction but also occurs with severe dystonia (Jankovic and Fahn, 1998; Furukawa et al., 2001). The many different manifestations of dystonia are not as well recognized as the more common manifestations of spasticity among evaluators without training in movement disorders, and they therefore are often mistaken as evidence for corticospinal dysfunction (Watts et al., 1982; Jan, 2004).
The third factor contributing to the varied descriptions of the motor syndrome in the published literature on LND is related to mood lability and anxiety. Patients with LND are more anxious than other children with developmental disabilities. In some, emotional lability is pronounced, including some with overt panic attacks. Anxiety and mood lability are particularly prominent when the patient encounters a stranger, such as a new care-giver or examiner. The heightened anxiety makes choreoathetosis and ballismus more apparent. Anxiety and mood lability lessen as the patients become accustomed to the evaluators, so the examination features may change during a long visit or over several visits. Consequently, even a single examiner can form different impressions on the basis of serial evaluations of the same patient. This problem was minimized in the current series because evaluations were conducted over a lengthy visit or the patients encountered the same evaluator during multiple visits over several years.
A last reason for the inconsistent descriptions of motor abnormalities in LND is that some characteristics of the disease may vary with development and ageing. There are no studies that have specifically addressed the natural history with longitudinal studies, but the available evidence suggests a stereotypical pattern of progression with hypotonia and developmental delay early in the first year of life. Other involuntary movements emerge later, between 6 and 24 months of age. Once established, significant progression of the severity of the motor syndrome with further ageing is uncommon, though progression of disability from long-standing dystonia became more apparent, for example as contractures or scoliosis.
The evolution of the motor syndrome parallels the evolution of the cognitive disability in LND. Patients with LND exhibit mild or moderate mental retardation with slow learning in school, but there is no evidence for progressive dementia with ageing (Anderson et al., 1992; Matthews et al., 1995; Solan et al., 1997; Schretlen et al., 2001). The evolution of motor and cognitive impairments in LND resembles that seen in cerebral palsy. The early changes of the motor features may reflect the response of the developing brain to a static insult rather than an ongoing pathological process (Almli and Finger, 1984; Finger and Almli, 1984). Although LND is often classified in texts and reviews of movement disorders among the ‘heredodegenerative’ dystonias, it is more appropriately classified among the ‘dystonia-plus’ syndromes, in view of the absence of obvious progression and the lack of evidence for a degenerative process.
Several clinical phenomena observed in LND may be considered in the context of other movement disorders where dystonia is well recognized. Severe opisthotonus with coarse whole-body shaking was frequent in the current series and in the literature on LND. This phenomenon is sometimes mistaken for epileptic seizures, but it is also seen in many movement disorders, where it is best characterized as a truncal dystonia with dystonic tremor (Fricka et al., 2001; Jan, 2004). The oromandibular and lingual dystonia in LND is similar to that seen in neuro-acanthocytosis (Hardie, 1989) or tardive syndromes due to chronic exposure to dopamine receptor antagonists (Tan and Jankovic, 2000). It is interesting that self-injurious tongue biting, a nearly universal phenomenon in LND, is seen occasionally in neuro-acanthocytosis or tardive syndromes too (Jankovic, 1988). The inspiratory stridor that occurs during wakefulness but not sleep in several LND patients is similar to that reported as laryngeal breathing dystonia (Grillone et al., 1994) and in some cases of multiple system atrophy (Merlo et al., 2002). This observation may provide a clue for understanding LND patients who exhibit recurrent apnoea or even sudden death. Finally, patients with LND exhibit ocular motor apraxia characterized by delayed initiation of voluntary saccades with facilitation of reflexive saccades similar to that observed in several other extrapyramidal disorders (Jinnah et al., 2001). These additional manifestations further emphasize the extra-pyramidal nature of the motor disorder of LND.
The motor syndrome of LND described in the present series is consistent with current knowledge regarding the pathophysiology of the disorder. Overall, the most frequent abnormality among the imaging (Table 5) and autopsy studies (Table 6) has been a reduction in brain volume. This reduction in volume, however, seems often small enough to escape notice in routine studies. Routine imaging studies are therefore not particularly helpful in the evaluation of these patients.
Although consistent anatomical defects in LND brains have not been apparent, biochemical and PET studies have demonstrated abnormalities of basal ganglia dopamine systems. Biochemical studies of five LND brains have documented a profound loss of dopamine and its metabolites in the basal ganglia (Lloyd et al., 1981; Saito et al., 1999). Dopamine was not significantly reduced in the midbrain, fuelling speculation that midbrain dopamine neurons were preserved but that their axonal projections to the basal ganglia failed to develop or degenerated. In support of this proposal, midbrain dopamine neurons identified by Nissl stains or by immunostaining for tyrosine hydroxylase did not appear to be reduced in numbers in two cases (Saito et al., 1999). Furthermore, two PET studies have confirmed severe abnormalities of dopaminergic fibres in the basal ganglia. In one study, the accumulation of [18F]fluorodopa into monoaminergic axons in the striatum was reduced by 60–70% in 11 LND patients (Ernst et al., 1996). In the other study, the binding of [11C]WIN 35 428 to dopamine uptake sites on dopamine axons was reduced by a similar amount in seven other patients (Wong et al., 1996). These studies suggest a severe loss of dopaminergic axons in LND, though histological evidence is currently lacking.
These biochemical and PET data are consistent with the prominent dystonia observed in LND (Visser et al., 2000). A profound loss of striatal dopamine most often causes parkinsonism in adults, but more frequently causes dystonia in children. For example, early dopamine loss associated with inherited deficiency of GTP cyclohydrolase or tyrosine hydroxylase is more often associated with dystonia rather than parkinsonism (Perlmutter and Mink, 2004). Experimental studies have also shown that the age at which striatal dopamine depletion occurs has a dramatic influence on motor function in rodents. In adult rats, destruction of 95% of nigrostriatal dopamine neurons results in a motor syndrome resembling parkinsonism (Hirsch et al., 2003). In contrast, the same lesion in neonatal rats results in spontaneous hyperactivity and aggressiveness without signs of parkinsonism (Moy et al., 1997).
Isolated dysfunction of striatal dopamine systems does not account for the pyramidal signs that appear in some cases. Even recognizing the possibility of misclassification of neurological signs, another possibility is that HPRT deficiency causes dysfunction in corticospinal motor systems, although evidence currently is lacking. Another possibility is that corticospinal tract signs reflect an indirect, acquired process. The observation that the corticospinal tract signs frequently are limited to the legs suggests the possibility of cervical myelopathy, perhaps secondary to degenerative changes of the cervical spine caused by cervical dystonia with retrocollis. In fact, cervical instability leading to myelopathy has been documented for two LND patients (Watts et al., 1982). Violent retrocollis caused atlantoaxial dislocation in another case (Hoefnagel et al., 1965). A non-traumatic high cervical fracture in one case, and an os odontoideum in another, could both have resulted from chronic and forceful involuntary movements of the neck (Shewell and Thompson, 1996).
The possibility that cervical myelopathy results from involuntary neck movements is not unique to LND, but has been described in several other conditions, including cervical dystonia, generalized dystonia, paroxysmal dystonia, dyskinetic cerebral palsy and Tourette syndrome (Anderson et al., 1962; Levine et al., 1970; Hirose and Kadoya, 1984; Ebara et al., 1989; El-Mallakh et al., 1989; Adler et al., 1996; Harada et al., 1996; Krauss and Jankovic, 1996; Rosenfeld and Friedman, 1999). Since baseline motor disability is often quite severe in LND, the potential development of a secondary myelopathy can be difficult to identify. Further studies are needed to clarify the source of the pyramidal signs in LND. Until this information is available, it is prudent to perform imaging studies of the cervical or thoracic spine in LND patients who develop true corticospinal signs.
Although several trials have addressed the treatment of self-injury, there have been no controlled trials devoted to treatment of the motor disorder in LND. Anecdotal reports of dopamine replacement with levodopa in 11 patients described mixed results (Mizuno and Yugari, 1974; Mizuno and Yugary, 1975; Watts et al., 1982; Manzke et al., 1986; Jankovic et al., 1988; Hunter et al., 1996). Mild improvement was noted for six, mild worsening was noted for three and no effect was reported for two. There is a single report of two patients involving the use of a direct-acting dopamine agonist, bromocriptine, again with mixed results (Jankovic et al., 1988). The interpretation of these results is complicated by two issues. First, the maximal doses employed and the duration of therapy were recorded only rarely. Second, several investigators have suggested that some of the dyskinetic movements and even self-injury in LND might reflect a form of ‘endogenous’ dopamine-related dyskinesia (Proctor and McGinness, 1970; Taira et al., 2003). It is therefore possible that dopaminergic drugs reduce the severity of dystonia while simultaneously increasing other involuntary movements or distressing behaviours, rendering a negative overall impression of the treatment response. Further trials with more detailed observations of the influence of specific doses of these agents on different aspects of the condition are warranted.
Others have tried the dopamine-depleting agent tetrabenazine (Watts et al., 1974; Jankovic et al., 1988). Efforts to control self-injurious behaviours often lead to the use of dopamine receptor antagonists including pimozide, haloperidol, fluphenazine and Risperdal (Watts et al., 1982; Goldstein et al., 1985; Jankovic et al., 1988; Allen and Rice, 1996). Their benefits on self-injury are unreliable, and they do not appear to have a significant influence on dystonia. We prefer to avoid these drugs as they carry the risk of causing tardive syndromes that make assessment and management of the pre-existing motor disorder unnecessarily complicated.
Oral baclofen and benzodiazepines are often prescribed for patients with LND. While they may be useful for controlling spasticity and clonus in the minority of patients who have these problems, their value in the management of the extrapyramidal features is unclear. Two of our patients received intrathecal baclofen pumps. Both reported good results, though objective assessments were not possible because the procedures were performed before our evaluations. There was a notable paucity of reports describing results of treatment with trihexyphenidyl or other anticholinergics. In our experience, it is not useful in non-sedating doses. However, botulinum toxins can be used to treat some of the most discomforting manifestations, such as cervical or limb dystonia.
Finally, there are two case reports of thalamotomy producing little or no improvement (Michener, 1967; Bunn et al., 1975) and another case report describing 30% improvement in dystonia following chronic deep brain stimulation of the globus pallidus in LND (Taira et al., 2003). In the last case, an unexpected benefit was complete elimination of self-injurious behaviours for at least one year. Further studies of these procedures are required before they can be recommended to all patients.
The motor disorder of LND characteristically presents with hypotonia and/or delayed acquisition of motor skills within the first 3–6 months of age. Additional involuntary movements, most prominently dystonia, usually develop between 6 and 24 months of age. Other extrapyramidal and pyramidal signs may occur, but these are usually minor compared with dystonia. The more varied descriptions of the motor syndrome in the literature probably reflect differences in the use of movement disorder terminology, poor recognition of the many different manifestations of dystonia and influences related to age and anxiety at the time of evaluation. The processes whereby a defect in HPRT-mediated purine recycling leads to the neurobehavioural manifestations remain largely unknown, but a growing body of evidence has pointed to dysfunction of basal ganglia dopamine systems as a potential underlying contributor to dystonia. Current treatments are not satisfactory. Increasing our understanding of the pathophysiology of LND may allow for the rational design of new approaches for treatment.
We gratefully acknowledge the patients and their families for participating in these studies. We appreciate the expertise of Patrick O’Neill in verifying the mutations for some patients. We thank Harvey Singer for helpful comments on the manuscript. Support was provided by the Lesch–Nyhan Syndrome Children’s Research Foundation, the Fondo de Investigaciones Sanitarias REDEMETH G03/054 and National Institutes of Health grants NS01985 and HD33095.
For Permissions, please email: email@example.com