Imaging of movement disorders
May. 19, 2022
At vero eos et accusamus et iusto odio dignissimos ducimus qui blanditiis praesentium voluptatum deleniti atque corrupti quos dolores et quas.
Sydenham chorea is the prototype of chorea resulting from immune mechanisms. Although its incidence has steadily declined in the last decades, it remains the most common cause of acute chorea in childhood worldwide and is still an endemic condition in developing areas of the world. There is still interest related to the possibility that a similar pathogenic mechanism may be responsible for a subset of patients with tics and other movement disorders, as well as behavioral abnormalities. The author reviews clinical features, pathogenesis, and management of this condition.
• Sydenham chorea is the most common cause of acute chorea in children worldwide although there is a decline of its incidence worldwide.
• Vascular chorea is the most important differential diagnosis of Sydenham chorea.
• Evidence suggests that Sydenham chorea results from Streptococcus-induced antibodies that cross-react with central nervous system antigens.
• Neuropsychiatric symptoms, such as obsessions, compulsions, hyperactivity, and attention disorder, as well as depression, are often present in patients with Sydenham chorea.
• Genetic conditions such as mutations of ACDY5, NKX2-1, PDE10A, and PDE2A may mimic Sydenham chorea.
"Chorea" (derived from the Latin choreus meaning "dance") describes abnormal involuntary movements that are brief, usually distal, and without purpose. First described in the Middle Ages, the most common illness was perhaps a psychogenic movement disorder, but some cases were probably the postinfectious chorea known now as Sydenham chorea. For many years, chorea was the term applied to any hyperkinetic syndrome. Some chorea (including Sydenham chorea) has been called St. Vitus dance (58). The reader is referred to the Hayden text for an excellent historical review of the various choreic disorders (75) and a review of the “dancing mania” (86). Although Thomas Sydenham first described postinfectious choreic movements of children in 1686, the casual relationship of this form of chorea with streptococcal infection was only firmly established 45 years ago (126). There is also evidence indicating that the famous composer Gustav Mahler had Sydenham chorea (37). A review of all notes of inpatients seen by Sir William Gowers at the National Hospital of London from 1878 to 1911 showed that almost all chorea cases were caused by Sydenham chorea (138). The incidence of Sydenham chorea dropped drastically in North America and Western Europe after World War II. Interest in this condition, however, has been fueled lately by its persistence as an important health problem in developing countries, the resurgence of Sydenham chorea in the United States and Australia in the 1990s, the recent outbreaks in Central Europe (84), and the hypothesis that streptococcus-induced antibodies targeted at basal ganglia neurons might account for tics and behavioral abnormalities among children (117; 29; 38; 42). Sydenham chorea is also known as rheumatic chorea or minor chorea.
Sydenham chorea is the most common movement disorder associated with bacterial infection. The usual age at onset of Sydenham chorea is 8 to 9 years, but there are reports on patients who developed chorea during the third decade of life. In most series, there is a female preponderance (43; 122). Typically, patients develop this disease 4 to 8 weeks after an episode of group A beta-hemolytic streptococcal pharyngitis. It does not occur after streptococcal infection of the skin. The chorea, characterized by a random and continuous flow of contractions, spreads rapidly and becomes generalized, but 20% of patients remain with hemichorea (105; 43). This movement disorder is characterized by a random and continuous flow of contractions. Patients display motor impersistence, particularly noticeable during tongue protrusion and ocular fixation. The muscle tone is usually decreased; in severe and rare cases (8% of all patients seen at the Movement Disorders of the Federal University of Minas Gerais, Brazil), this is so pronounced that the patient may become bedridden (chorea paralytica). Carefully assessing adult subjects with a Sydenham chorea in remission, we found that 64% of the 25 individuals had residual bradykinesia. This finding may suggest that there is nigro-striatal lesion in Sydenham chorea, which explains the proclivity of these patients to develop drug-induced parkinsonism when treated with neuroleptics (14).
Patients often display other neurologic and nonneurologic symptoms and signs. Although there are reports of common occurrence of tics in Sydenham chorea, this author finds it virtually impossible to distinguish simple tics from fragments of chorea. Even vocal tics, found in 70% or more of patients with Sydenham chorea in 1 study, are not of simple diagnosis in patients with hyperkinesias (98). Those physicians experienced with movement disorder patients are well aware that involuntary vocalizations may result from dystonia or chorea of the pharynx and larynx. This has been reported, for instance, in subjects with oromandibular dystonia or Huntington disease (78). Under these circumstances the vocalization lacks the subjective feeling (premonitory urge or sensory tic) so characteristic of idiopathic tic disorders such as Tourette syndrome. In a cohort of 108 Sydenham chorea patients carefully followed up at our unit, we have identified vocalizations in only 8% of subjects. We have avoided the term “tic” because there was no premonitory sign or complex sound and, conversely, the vocalizations were associated with severe cranial chorea. Taken together, these findings suggest that involuntary sounds present in a few patients with Sydenham chorea result from choreic contractions of the upper respiratory tract muscles rather than true tics (130). There is evidence that many patients with active chorea have hypometric saccades, and a few of them also show oculogyric crisis. Dysarthria is common, and during the 19th century Gowers had already recognized that Sydenham chorea patients present with a “disinclination to speak.” In fact, a case-control study of patients described a pattern of decreased verbal fluency that reflected reduced phonetic, but not semantic, output (52). This result is consistent with dysfunction of the dorsolateral prefrontal-basal ganglia circuit. Studying adults with Sydenham chorea, we have extended this finding, showing that many functions dependent on the prefrontal area are impaired in these patients. The conclusion of this study is that Sydenham chorea should be included among the causes of dysexecutive syndrome (36; 16). The prosody is also affected in Sydenham chorea. In some studies, the prosody of patients with Sydenham chorea was compared to patients with rheumatic fever without chorea as well as healthy controls (40; 41; 107). Sydenham chorea patients had voice with decreased range of fundamental frequency, higher intensity, and decreased speed. Interestingly, these findings are similar to those observed in Parkinson disease (07).
In a survey of 100 patients with rheumatic fever, half of whom had chorea, we found that migraine is more frequent in Sydenham chorea (21.8%) than in normal controls (8.1%, p=0.02) (136). This is similar to what has been described in Tourette syndrome (88). In the older literature, there are also references to papilledema, central retinal artery occlusion, and seizures in a few patients with Sydenham chorea.
Attention has also been drawn to behavioral abnormalities associated with Sydenham chorea. Swedo and colleagues found obsessive-compulsive behavior in 5 of 13 Sydenham chorea patients, 3 of whom met criteria for obsessive-compulsive disorder, whereas no patient of the rheumatic fever group presented with obsessive-compulsive behavior (125). In another study of 30 patients with Sydenham chorea, Asbahr and colleagues demonstrated that 70% of the subjects presented with obsessions and compulsions, whereas 16.7% of them met criteria for obsessive-compulsive disorder (03). None of 20 patients with rheumatic fever without chorea had obsessions or compulsions (03). These results, however, were roughly replicated by a study that found that patients with rheumatic fever without chorea had more obsessions and compulsions than healthy controls (97). The text of this study, however, is not clear about the actual percentage of Sydenham chorea patients who had obsessive-compulsive disorder. It indicates that about 12% of the 22 Sydenham chorea patients met criteria for this condition. Mercadante and colleagues also tackled the issue of hyperactivity and attention deficit disorder in Sydenham chorea and found that 45% of their 22 patients met criteria for this condition. Maia and colleagues investigated behavioral abnormalities in 50 healthy subjects, 50 patients with rheumatic fever without chorea, and 56 patients with Sydenham chorea (93). The authors found that obsessive-compulsive behavior, obsessive-compulsive disorder, and attention deficit and hyperactivity disorder were more frequent in the Sydenham chorea group (19%, 23.2%, 30.4%) than in the healthy controls (11%, 4%, 8%) or in the patients with rheumatic fever without chorea (14%, 6%, 8%). In this study, the authors demonstrated that obsessive-compulsive behavior displays little degree of interference in the performance of the activities of daily living. One note of caution must be made regarding the interpretation of data of hyperactivity in Sydenham chorea. Comparing patients with acute and persistent Sydenham chorea, attention deficit hyperactivity disorder was significantly more common in the latter (50% vs. 16%). There was also a trend toward more obsessive-compulsive behavior and obsessive-compulsive disorder among patients with more prolonged forms of Sydenham chorea, but the difference failed to reach statistical significance. As there is no biological marker used in the current diagnostic criteria, it is not always easy to differentiate restlessness associated with chorea from true hyperactivity of hyperactivity attention deficit disorder. Another study compared the phenomenology of obsessions and compulsions of patients with Sydenham chorea with subjects diagnosed with tic disorders (02). The authors demonstrated that the symptoms observed among the Sydenham chorea patients were different from those reported by patients with tic disorders but were similar to those previously noted among samples of pediatric patients with primary obsessive-compulsive disorder. An investigation comparing healthy controls with patients with rheumatic fever showed that obsessive-compulsive behavior is more commonly seen in patients with Sydenham chorea with relatives who also have obsessions and compulsions (76). This study makes clear that there is interplay between genetic factors and environment in the development of behavioral problems in Sydenham chorea. We reported that although rarely, Sydenham chorea may induce psychosis during the acute phase of the illness (135). We also found that 1 of our patients with Sydenham chorea developed trichotillomania (87). Interestingly, another study of our group suggests that Sydenham chorea is not a cause of nonspecific behavioral problems; when comparing patients and controls, there was no difference on a rating scale of anxiety symptoms (127). In a careful investigation of psychiatric comorbidities in 50 patients with Sydenham chorea, it was found that the most frequent psychiatric disorders observed in Sydenham chorea patients were: major depression (14%), generalized anxiety disorder (16%), social phobia (24%), and obsessive-compulsive disorder (24%). The frequency of psychiatric disorders did not differ between Sydenham chorea patients in remission when compared to patients with persistent chorea, except for depressive disorders, which were more frequent in the latter (101). We have also demonstrated that a subset of patients with a previous history of Sydenham chorea is left with problems on tests assessing attention and executive functions (16). These results were confirmed by another group of investigators (45). A systematic review confirms that all studies of behavioral changes in patients with Sydenham chorea have found an association with obsessive-compulsive disorder (18). Another systematic review confirms that obsessive-compulsive disorder is more commonly seen in individuals with Sydenham chorea than controls. Interestingly, the authors also concluded that the evidence supporting the association with tic disorders is rather weak (113).
There are data demonstrating that the peripheral nervous system is not targeted in Sydenham chorea (32; 140). Finally, it must be kept in mind that Sydenham chorea is a major manifestation of rheumatic fever. Sixty percent to 80% of patients display cardiac involvement, particularly mitral valve dysfunction, in Sydenham chorea, whereas the association with arthritis is less common, seen in 30% of patients; however, in approximately 20% of patients, chorea is the sole finding (43; 60). A prospective follow up of patients with Sydenham chorea with and without cardiac involvement in the first episode of chorea suggests that the heart remains spared in those without lesion at the onset of the rheumatic fever (109). There is also evidence suggesting that Sydenham chorea may be associated with other autoimmune disorders such as Takayasu arteritis (139; 10).
The current diagnostic criteria of Sydenham chorea are a modification of the Jones criteria; chorea with acute or subacute onset and lack of clinical and laboratory evidence of an alternative cause are mandatory findings. The diagnosis is further supported by the presence of additional major or minor manifestations of rheumatic fever (Guidelines for diagnosis of rheumatic fever, Jones criteria 1992; 43; 44). In an investigation from New Zealand, the authors employed a slightly modified Jones criteria, allowing as major criteria subclinical carditis (ie, solely based on echocardiography), carditis, and monoarthritis if the patient is on anti-inflammatory drugs. This led to a 16% increase of the number of diagnosis of acute rheumatic fever (144). There is a validated scale to rate Sydenham chorea. The Universidade Federal de Minas Gerais Sydenham Chorea Rating Scale was designed to provide a detailed quantitative description of the performance of activities of daily living, behavioral abnormalities, and motor function of patients with Sydenham chorea. It comprises 27 items, and each is scored from 0 (no symptom or sign) to 4 (severe disability or finding) (133).
The finding that behavioral problems are common in patients with rheumatic fever and chorea contributed to the notion that Sydenham chorea is a model for childhood autoimmune neuropsychiatric disorders (124). PANDAS, pediatric autoimmune neuropsychiatric disorders associated with streptococcus, is a controversial hypothesis. It proposes that infection with group A beta-hemolytic streptococci may induce tics, obsessive-compulsive behavior, and other neuropsychiatric disturbances. The following working diagnostic criteria for this condition have been proposed: (1) presence of obsessive-compulsive disorder or a tic disorder; (2) prepubertal symptom onset; (3) episodic course of symptom severity; (4) association with group A beta-hemolytic streptococci infections; and (5) association with neurologic abnormalities. According to a description of 50 patients who met these criteria, the onset of tics and obsessive-compulsive disorder were at a mean age of 6.3 and 7.4 years, respectively. The same study also noted “significant psychiatric comorbidity”: emotional lability, separation anxiety, nighttime fears and bedtime rituals, cognitive deficits, and oppositional behaviors (125). There is a growing list of neurologic symptoms and signs related to streptococcus infection: dementia, dystonia, encephalitis lethargica-like syndrome, motor stereotypies, myoclonus, opsoclonus, parkinsonism, paroxysmal dyskinesia, restless leg syndrome, and tremor (32). At the present time, however, there is no conclusive evidence that antibasal ganglia antibodies induced by streptococcus play a significant role in the pathogenesis of tic disorders. In fact, a population-based epidemiological survey performed in London failed to demonstrate a significant relationship between streptococcal infection and motor or behavioral syndromes (66; 119). In fact, a new study again failed to find immunologic abnormalities that distinguish controls from patients with Tourette syndrome and who meet criteria for PANDAS (102). However, a published systematic review and meta-analysis showed that a subset of obsessive-compulsive disorder patients have a high titer of anti-basal ganglia antibodies (112). A systematic review reiterates that there is no compelling evidence supporting the association of auto-immune disorders with obsessive-compulsive and tic disorders (113).
Nevertheless, the group who originally described the existence of PANDAS continues to follow up with individuals who have met diagnostic criteria for this condition (90). Interestingly, most of them had a benign course with mild, if any, motor or behavioral findings on a long run. The authors suggest that these individuals might in fact have Sydenham chorea.
The older literature describes Sydenham chorea as a rather benign, self-limited condition that comes into remission after a few months (105). Careful prospective follow-up of patients in the last few years demonstrates, however, that in up to half of patients’ chorea remains active 2 years after its onset. Moreover, despite regular use of secondary prophylaxis, recurrences of the movement disorders are observed in up to 50% of subjects (44; 85). Interestingly, in many of the recurrences there is lack of association either with streptococcus infection or even antibasal ganglia antibodies (73; 85; 142). One study from Turkey reviewed the outcome of 90 patients with Sydenham chorea (70). The authors found that recurrence was found in 16% of individuals and was associated with persistence of the first episode for more than 1 year well as, unlike previous studies, irregular use of antibiotics prophylaxis. The most worrisome problem in patients with Sydenham chorea is the occurrence of valvulopathy and other cardiac problems. The importance of this complication is illustrated by the finding that in areas where rheumatic fever is endemic, 70% of cardiac surgeries are performed to treat its complications (29). A study of patients with Sydenham chorea in Slovenia emphasizes its association with carditis, which was found in all patients with chorea (84). This was also found in a retrospective review of 375 Turkish patients seen at an academic center over 25 years (60). An investigation has demonstrated that adults with a previous history of rheumatic fever with or without chorea have scores on scales designed to assess obsessions and compulsions that are similar to controls (04).
An 8-year-old girl suddenly developed pain and swelling of right wrist, knee, and ankles associated with movement disorder in all limbs. One week later, the patient was unable to speak, and as a result of worsening of the involuntary movements, she became bedridden. Twenty days after onset, she was admitted to the Movement Disorders Clinic of the Federal University of Minas Gerais. The most important findings on examination were anarthria, severe and generalized chorea, and decreased muscle tone, as well as inability to walk and even sit without assistance. Her past medical history was remarkable for the presence of repeated episodes of sore throat, and 2 years before she had developed arthritis, carditis, and chorea. She was diagnosed with rheumatic fever and Sydenham chorea. The latter was controlled with haloperidol, but the patient did not comply with penicillin prophylaxis.
During the admission to treat the episode of chorea paralytica, cardiac evaluation (echocardiography) showed mild mitral insufficiency. Lab work-up included antistreptolysin 336 UI/mL (normal range less than 250), C-reactive protein 7.45 mg/mL (normal range less than 8), and sedimentation rate 35 mm. One week of 30 mg/kg per day of valproic acid failed to improve the patient. Because of the severity of the chorea, this anticonvulsant was discontinued, and she was started on pimozide (4 mg twice a day). Fifteen days after her admission, the patient was discharged. On that occasion, there was still anarthria, mild chorea, and severe decreased muscle tone, but she was able to sit and walk if assisted. On a follow-up visit 3 weeks later, she was still on 8 mg per day of pimozide; the main clinical findings were dysarthria, decreased verbal output (she stated 6 animal names in 1 minute. Normal range for her age and educational level is 11 to 13), moderately decreased muscle tone, but no chorea was noticed, and she sat and walked unassisted. A few weeks later, a gradual decrease in pimozide dosage was started, and pimozide was discontinued 4 months after discharge.
Forty-two months after the episode of chorea paralytica, the patient remained without neurologic abnormalities and received penicillin prophylaxis.
Group A beta-hemolytic streptococci are the causative agents of Sydenham chorea and related disorders.
Taranta and Stollerman established the casual relationship between infection with group A beta-hemolytic streptococci and the occurrence of Sydenham chorea (126). Based on the assumption of molecular mimicry between streptococcal and central nervous system antigens, it has been proposed that the bacterial infection in genetically predisposed subjects leads to formation of cross-reactive antibodies that disrupt the basal ganglia function. Several studies have demonstrated the presence of such circulating antibodies in 50% to 90% of patients with Sydenham chorea (77; 29). A specific epitope of streptococcal M proteins that cross-reacts with basal ganglia has been identified (24). In 1 study of patients it was demonstrated that all patients with active Sydenham chorea have antibasal ganglia antibodies demonstrated by ELISA and Western Blot. In subjects with persistent Sydenham chorea (duration of disease greater than 2 years despite best medical treatment) the positivity was about 60% (47). An investigation from another group confirmed that antibodies from patients with Sydenham chorea bind to neuronal surface. Interestingly, antibodies from subjects with Tourette syndrome or who met criteria for PANDAS did not have this binding pattern (22). It must be emphasized that the biological value of the antibasal ganglia antibodies remains to be determined. One study suggests that they may interfere with neuronal function, however. Kirvan and colleagues demonstrated that IgM of 1 patient with Sydenham chorea induced expression of calcium/calmodulin-dependent protein kinase II (CaMKII) in a culture of neuroblastoma cells (83). Although an interesting finding, this study has 3 limitations: (1) it is an in vitro investigation, employing an artificial paradigm that does not necessarily reflect the situation observed in human patients; (2) the antibody was obtained from a single patient; and, (3) the authors studied IgM, whereas all investigations of antibasal ganglia antibodies in Sydenham chorea have detected IgG. Nevertheless, in a more recent study this same group of authors have found circulating serum antibodies in patients with Sydenham chorea, targeting tubulin, lysoganglioside GM1, and dopamine receptors D1 and D2 (46). These antibodies are reported as capable of increasing the activity of CaMKII. Our finding of a linear correlation between the increase of intracellular calcium levels in PC12 cells and antibasal ganglia antibody titer in the serum from Sydenham chorea patients suggests that these antibodies have a pathogenic value (132). However, an in vivo study failed to demonstrate that antibodies from Sydenham chorea patients infused in the basal ganglia of rodents failed to induce behavioral changes although they were found to bind to a ∼50-kDa molecule in the striatum extract (19). One possible explanation for the finding of this study is that a low titer of the antibodies prevented the occurrence of detectable behavioral manifestations. More data lend support, however, to the molecular mimicry hypothesis. First, we have demonstrated that infusion of sera of Sydenham chorea patients in rodents with 6-OH-dopamine-induced unilateral lesion of the nigro-striatal system caused circling behavior similar to apomorphine (57). This finding suggests that the circulating antibodies act on dopamine receptors. Another study gives weight to this hypothesis. Rats exposed to Streptococcus antigens not only developed behavioral abnormalities reminiscent of Sydenham chorea, but also had IgG that reacted with tubulin, D1 and D2 receptors, causing elevated calcium/calmodulin-dependent protein kinase II signaling (23). This same group just demonstrated that the injection of these antibodies in the striatum of naïve rats led to behavioral and immunologic abnormalities similar to those found in animals exposed to Streptococcus antigens. Specifically, the antibodies bind to D1 and D2 as well as 5HT-2A and 5HT-2C receptors (91). There is indeed growing evidence that the pathogenesis of Sydenham chorea involves changes of dopamine transmission (57): monoclonal antibodies derived from Sydenham chorea patients target dopaminergic neurons in transgenic mice and signal D2 receptors (50); and there is a correlation between rates on the Universidade Federal de Minas Gerais Sydenham Chorea Rating Scale with the ratio between anti-D1 and anti-D2 receptors antibodies (20). Data cast doubt on the so-called “dopamine hypothesis” of the pathogenesis of Sydenham chorea because other groups have failed to replicate some of the findings described above (35). One article provides a comprehensive review of the molecular mimicry hypothesis of the pathogenesis of Sydenham chorea (51). The author emphasizes the notion that antibodies targeting lysoganglioside of group A Streptococcus cross react with dopamine D1 and D2 receptors.
It remains unclear why up to 50% of patients with Sydenham chorea develop a persistent course of the illness (35). In this subset of individuals, the titers of antibasal ganglia antibodies are low. Taking into account this finding, as well as our observation that serum BDNF levels are high in this group of patients, one may hypothesize that the acute immune process causes structural brain lesions resulting in permanent dysfunction of the basal ganglia (128).
Although some investigations suggest that susceptibility to rheumatic chorea is linked to human leukocyte antigen expression (06), there are studies failing to identify any relationship between Sydenham chorea and human leukocyte antigen class I and II alleles (56). An investigation has shown, however, an association between HLA-DRB1*07 and recurrent streptococcal pharyngitis and rheumatic heart disease (74). The genetic marker for rheumatic fever and related conditions would be the B-cell alloantigen D8/17 (62). Despite repeated reports of the group that developed the essay claiming its high specificity and sensitivity (59; 72), findings of other authors suggest that the D8/17 marker lacks specificity and sensitivity. For instance, Kaur and colleagues demonstrated that the discriminating power of monoclonal antibody against D8/17 was relatively low among patients with rheumatic fever of North Indian ethnic origin (81). Studying Caucasians in the United States, Murphy and colleagues showed that 65.6% of their patients with obsessive-compulsive disorder or chronic tic disorder and 8.3% of controls tested positive for D8/17 (104). In the Netherlands Jansen and colleagues found that just a minority of their patients with post-group A beta-hemolytic streptococcal arthritis has an elevation of D8/17-positive lymphocytes (79). Another suggested genetic risk factor for development of acute rheumatic fever but not Sydenham chorea is polymorphisms within the promoter region of the tumor necrosis factor-alpha gene (116).
Due to the difficulties with the molecular mimicry hypothesis to fully account for the pathogenesis of Sydenham chorea, there have been studies that address the role of immune cellular mechanisms in this condition. Investigating sera and CSF samples of patients of the Movement Disorders of the Federal University of Minas Gerais, Church and colleagues found elevation of cytokines that take part in the Th2 (antibody-mediated) response, interleukins 4 (IL-4) and 10 (IL-10), in the serum of acute Sydenham chorea in comparison to persistent Sydenham chorea (48). They also described interleukin 4 in 31% of the CSF of acute Sydenham chorea, whereas just interleukin 4 was raised in the CSF of persistent Sydenham chorea. The authors concluded that Sydenham chorea is characterized by a Th2 response. However, as they have found an elevation of interleukin 12 in acute Sydenham chorea and as we described an increased concentration of chemokines CXCL9 and CXCL10 in the serum of patients with acute Sydenham chorea (131), it can be concluded that Th1 (cell-mediated) mechanisms may also be involved in the pathogenesis of this disorder. Another investigation confirmed that cellular immune mechanisms may be relevant to the pathogenesis of Sydenham chorea because there is a dysfunction of monocytes (137).
Some authors have suggested that streptococcal infection induces vasculitis of medium-sized vessels, leading to neuronal dysfunction. Such vascular lesions could be produced by antiphospholipid antibodies. Interestingly, although essentially all patients with Sydenham chorea are negative for antiphospholipid antibodies, a study demonstrates many immunologic similarities between primary antiphospholipid antibody syndrome and Sydenham chorea exist (21). There is also a suggestion that cellular immune mechanisms participate in the pathogenesis of streptococcus-related movement disorders. However, most of these findings have not been replicated to date. Currently, the weight of evidence suggests that the pathogenesis of Sydenham chorea is related to circulating cross-reactive antibodies. It has been demonstrated that streptococcus-induced antibodies can be associated with a form of acute disseminated encephalomyelitis characterized by a high frequency of dystonia and other movement disorders as well as basal ganglia lesions on neuroimaging (53). Antineural and antinuclear antibodies have also been found in patients with Tourette syndrome but their relationship with prior streptococcus infection remains equivocal (103).
From the neurophysiological point of view, it was widely accepted that choreas in general, including Sydenham chorea, were characterized by increased excitability of the motor cortex. However, 1 study of 16 patients with Sydenham chorea has demonstrated that there is a decreased output of the motor cortex similarly to what is found in Huntington disease. The reasons for this paradoxical finding remain to be determined (71; 82).
The incidence of rheumatic fever and Sydenham chorea in the United States and Western Europe has declined since World War II as result of improved health care, increased antibiotic usage, and lower virulence of streptococcal strains (115). This fall is demonstrated by the finding that the annual age-adjusted incidence rate of initial attacks of rheumatic fever per 100,000 children declined from 3.0 in 1970 to 0.5 in 1980 in Fairfax County Virginia, United States (120). Furthermore, Nausieda and colleagues showed that Sydenham chorea accounted for 0.9% of children admissions to hospitals in Chicago before 1940, whereas this number dropped to 0.2% during the period between 1950 and 1980. Despite the fall in the incidence, Sydenham chorea remains the most common cause of acute chorea in children. Outbreaks of rheumatic fever with occurrence of chorea have been identified in the United States and Australia (05; 117). Variable manifestations may occur among people of differing ethnic backgrounds, with the Australian aboriginals being a group at high risk of developing rheumatic fever, with and without chorea (26). The importance of Sydenham chorea even in developed areas is shown by a study performed in the pediatrics unit of a university hospital in Pennsylvania that showed Sydenham chorea accounted for of 96% of all patients with chorea seen during the 1980 to 2004 period (145). Studies from Australia confirm that it is a relatively common cause of acute chorea in children (54; 123). A national survey of all cases of rheumatic fever in Australia identified 151 children with this condition in a 3-year period. Sydenham chorea was diagnosed in 19% of individuals. Not surprisingly, the majority (86.7%) were in Indigenous Australians. However, the authors identified 10 cases in non-Indigenous Australians, of whom 3 had atypical presentations. The authors concluded that rheumatic fever may be more common than previously thought among low-risk children (106). There is a report of an increase of cases of acute rheumatic fever in Slovenia with an annual incidence of 1.25 cases per 100,000 children during the period of January 2008 to December 2014 (84). On the other hand, rheumatic fever has remained a significant public health problem in developing areas, particularly within the low-income population. At the Movement Disorders Clinic of the Federal University of Minas Gerais, Brazil, for instance, Sydenham chorea accounted for 64% of all patients with chorea. This has changed with a substantial reduction of new cases of Sydenham chorea. In the top end of the Northern Territory in Australia, an area predominantly inhabited by Aboriginal people, the point prevalence of rheumatic fever was 9.6 per 1000 people aged 5 to 14 years in 1995 (27). Sydenham chorea occurs in about 26% of patients with rheumatic fever in our center (43) whereas in Turkey the proportion falls to 13.5% (60). In the outbreak of rheumatic fever in Slovenia, chorea was found in 32% of patients (84). However, despite the lack of community-based studies, we and other investigators from areas in the past plagued by rheumatic fever (Africa and the Asian subcontinent) have observed a steep decline of the incidence of Sydenham chorea in our units. A study from Turkey confirms the notion that in the last decade there is also a substantial decline of new cases (60).
Prompt treatment of streptococcal pharyngitis with appropriate antibiotics has lowered the incidence of Sydenham chorea. Once the diagnosis of rheumatic chorea is established, the patient must receive secondary prophylaxis with penicillin, or for patients with an allergy to penicillin, sulpha drugs. This has been shown to effectively decrease the risk of neurologic or cardiac problems with additional streptococcal infections (94). The recommendation of the World Health Organization is to maintain the secondary prophylaxis up to 21 years of age. In instances where the diagnosis of Sydenham chorea is made after this age, the policy is less clear. Because of the potential seriousness of cardiac lesion, our own recommendation is to maintain prophylaxis indefinitely. Patients with a history of Sydenham chorea should be informed of the possible re-emergence of chorea during pregnancy or with use of oral contraceptives.
Sydenham chorea is the most common cause of acute chorea in children worldwide but several other conditions can cause present with chorea in this age range (09).
Several conditions may present with clinical manifestations similar to Sydenham chorea (Table 1) (31). The most important differential diagnosis is systemic lupus erythematosus, where up to 2% of patients may develop chorea. From a clinical point of view, the majority of subjects with this condition will have other nonneurologic manifestations such as arthritis, pericarditis, and other serositis as well as skin abnormalities. Moreover, the neurologic picture of systemic lupus erythematosus tends to be more complex and may include psychosis, seizures, other movement disorders, and even mental status and consciousness level changes (08). Only in rare instances will chorea, with a tendency for spontaneous remissions and recurrences, be an isolated manifestation of systemic lupus erythematosus. The difficulty in distinguishing these 2 conditions is increased by the finding that up to 20% of patients with Sydenham chorea display recurrence of the movement disorder. Eventually, patients with systemic lupus erythematosus will develop other features, meeting diagnostic criteria for this condition (12). Primary antiphospholipid antibody syndrome is differentiated from Sydenham chorea by the absence of other clinical and laboratory features of rheumatic fever as well as the usual association with repeated abortions, venous thrombosis, other vascular events, and the presence of typical laboratory abnormalities. Anti-NMDA receptor encephalitis is a condition that has become increasingly diagnosed. In children and adolescents, it is usually a paraneoplastic disorder related to ovarian teratoma. Patients present with a combination of psychiatric problems and a mixed movement disorder that may include chorea. However, the severe and proteiform clinical presentation readily distinguishes this condition from Sydenham chorea (11). Encephalitides, either as a result of direct viral invasion or by means of an immune-mediated postinfectious process, can cause chorea (114). This usually happens, however, in younger children; the clinical picture is more diversified to include seizures, pyramidal signs, and impairment of the psychomotor development, and there are laboratory abnormalities suggestive of the underlying condition. Drug-induced choreas are readily distinguished by careful history demonstrating temporal relationship between onset of the movement disorder and exposure to the agent. Although vascular disease is uncommon in the first 2 decades of life, it can occur and cause chorea (95). This may happen in 3% of patients with moyamoya disease (01). Benign hereditary chorea (BHC) is another condition that causes chorea in children and may mimic Sydenham chorea. It is clear now that benign hereditary chorea is a genetic syndrome associated with multiple mutations. The most common cause is mutation of NKX2-1. It is an autosomal dominant disorder; the family history may remain unnoticed due to the variable expression or the possibility a de novo mutation. These patients display chorea and other movement disorders (ie, tics, myoclonus), as well as behavioral disorders (34). It usually starts in early childhood and progresses until the second decade, after which time it remains static or even spontaneously improves. Patients may have a slight hypotonia and motor delay because of chorea; they may also have mild gait ataxia, dystonia, tics, handwriting impairment, and even drop attacks. However, the disorder is self-limiting after adolescence in most cases, although it may persist as a mild chorea beyond 60 years of age. Benign hereditary chorea has been linked to mutations in the NKX2-1 (previously called TITF1) gene on chromosome 14q13, coding for a transcription essential for the organogenesis of the lung, thyroid, and basal ganglia (110). Benign hereditary chorea should be considered in children and adults with chorea, mental retardation, congenital hypothyroidism, and chronic lung disease; hence, the term brain-thyroid-lung syndrome has been proposed for this disease. About half of the patients spontaneously improve in adulthood, but levodopa can also be beneficial. There are reports suggesting that thyroid hormone replacement therapy and methylphenidate may be effective (63; 121). Another form of “benign” chorea with relatively intact cognitive function is an autosomal dominant syndrome associated with ACDY5 mutation, but the phenotype of this disorder is expanding, and it may include dystonia, myoclonus, axial hypotonia, and episodic and fluctuating painful spasms associated with falls, sleep disturbance, and other manifestations (96). More rare mutations of the genes PDE10A and PDE2A may also result in a clinical picture similar to benign hereditary chorea. However, these are autosomal recessive conditions, often associated with seizure disorder and bilateral striatal necrosis (99; 118). Other genes to be considered in this scenario are FOXG1, GNAO1, GPR88, SLC2A1, SQSTM1, ATP8A2, or SYT-1. One review describes the differential diagnosis of autoimmune choreas in detail (35) and a more recent article tackles the differential diagnosis of chorea in children (09).
• Systemic lupus erythematosus
• Subacute bacterial endocarditis
• Tardive dyskinesia
• Anoxic encephalopathy
Children and young adults with chorea should undergo complete neurologic examination and diagnostic testing to assess the various causes of chorea because there is no specific biological marker of Sydenham chorea. The aim of the diagnostic workup in patients suspected to have rheumatic chorea is 3-fold: (1) to identify evidence of recent streptococcal infection or acute phase reaction; (2) to search for cardiac injury associated with rheumatic fever; and (3) to rule out alternative causes. Tests of acute phase reactants, such as erythrocyte sedimentation rate, C-reactive protein, leukocytosis; other blood tests like rheumatoid factor, mucoproteins, protein electrophoresis; and supporting evidence of preceding streptococcal infection (increased antistreptolysin-O, antiDNAse-B, or other antistreptococcal antibodies; positive throat culture for group A Streptococcus; or recent scarlet fever) are much less helpful in Sydenham chorea than in other forms of rheumatic fever due to the usual long latency between the infection and onset of the movement disorder. Elevated antistreptolysin O titer may be found in populations with a high prevalence of streptococcal infection. Furthermore, the antistreptolysin O titer declines if the interval between infection and rheumatic fever is greater than 2 months. Anti-DNase-B titers, however, may remain elevated up to 1 year after strep pharyngitis. Heart evaluation (ie, Doppler echocardiography) is mandatory because the association of Sydenham chorea with carditis is found in up to 80% of patients. Cardiac lesions are the main source of serious morbidity in Sydenham chorea. Serologic studies for systemic lupus erythematosus and primary antiphospholipid antibody syndrome must be ordered to rule out these conditions. EEG may show generalized slowing acutely or after clinical recovery. Spinal fluid analysis is usually normal, but it may show a slight increased lymphocyte count. It is important to highlight that although autoantibodies are important for the pathogenesis of Sydenham chorea, they have no role in establishing the diagnosis in clinical practice. This is related to their lack of specificity (28).
In general, neuroimaging will help rule out structural causes such as Moyamoya disease. CT scan of the brain invariably fails to display abnormalities. Similarly, head MRI is often normal, although there are case reports of reversible hyperintensity in the basal ganglia area. Very rarely, patients may develop necrotic lesions of the basal ganglia (25). In a study, Giedd and colleagues showed increased signal in just 2 of 24 patients, although morphometric techniques revealed mean values for the size of the striatum and pallidum larger than controls (65). Unfortunately, these findings are of little help on an individual basis because there was an extensive overlap between controls and patients. PET and SPECT imaging may prove to be useful tools in the evaluation, revealing transient increases in striatal metabolism (68; 143; 89; 108). In fact, Barsottini and colleagues showed that 6 of 10 patients with Sydenham chorea have hyperperfusion of the basal ganglia (15). A study confirmed the existence of increased striatal metabolism in Sydenham chorea (67). This contrasts with other choreic disorders (such as Huntington disease) that are associated with hypometabolism. Of note, however, 1 investigation showed hyperperfusion in 2 patients with Sydenham chorea whereas the remaining 5 had hypometabolism (49). It is possible that the inconsistencies in these studies reflect heterogeneity of the population of patients. It was found that subjects with Sydenham chorea in motor remission are left with hyperperfusion of the left putamen on SPECT (17). A volumetric study showed that putamen is enlarged in acute phase of Sydenham chorea whereas other areas of the basal ganglia remain unchanged (80).
Increasing interest is now directed to autoimmune markers that may be useful for diagnosis. The test of antineuronal antibodies, however, is not commercially available, being just performed for research purposes. Preliminary evidence, moreover, suggests that these antibodies are not specific for Sydenham chorea. Similarly, the low sensitivity and specificity of the alloantigen D8/17 renders it unsuitable for the diagnosis of this condition.
There are few controlled studies of symptomatic treatment of Sydenham chorea (33; 100). The first choice of the author is valproic acid with an initial dosage of 250 mg per day that is increased during a 2-week period to 250 mg 3 times a day. If the response is not satisfactory, dosage can be increased gradually to 1500 mg per day. As this drug has a rather slow onset of action, it is prudent to wait 2 weeks before concluding that a regimen is ineffective. Carbamazepine and levetiracetam have also been reported as effective to control chorea in these patients in uncontrolled open-label studies (55). If the patient fails to respond to these antiepileptic medications, the next option is to prescribe neuroleptics. Neuroleptics can also be prescribed as a first line treatment in patients who present with chorea paralytica. Risperidone, a relatively potent dopamine D2 receptor blocker, is usually effective in controlling the chorea. The usual initial regimen is 1 mg twice a day. If, 2 weeks later, the chorea is still troublesome, the dosage can be increased to 2 mg twice a day. Haloperidol and pimozide are also occasionally used in the management of chorea in Sydenham chorea. However, they aren't as well tolerated as risperidone. Dopamine D2 receptor blockers must be used with great caution in patients with Sydenham chorea. After the observation of development of parkinsonism, dystonia, or both in patients treated with neuroleptics, we performed a case-control study comparing the response to these drugs in patients with Sydenham chorea and Tourette syndrome. We demonstrated that 5% of 100 patients with chorea developed extrapyramidal complications, whereas these findings were not observed among patients with tics matched for age and neuroleptics dosage (129). Because haloperidol, pimozide, and other neuroleptics act by blocking dopamine receptors and, therefore, may cause tardive dyskinesia, many clinicians use tetrabenazine, a dopamine-depleting drug, for the symptomatic treatment of Sydenham chorea as this drug has a very low risk of tardive dyskinesia.
There are no published guidelines concerning the discontinuation of antichoreic agents. My personal policy is to attempt a gradual decrease of the dosage (25% reduction every 2 weeks) after the patient remains free of chorea for at least 1 month. Finally, the most important measure in the treatment of patients with Sydenham chorea is secondary prophylaxis with penicillin or, if there is allergy, with sulfa drugs, up to 21 years of age. In case the onset occurs after this age, the recommendation is to maintain prophylaxis indefinitely (30). One study demonstrated that carbamazepine (15 mg/kg per day) is as effective as valproic acid (20 to 25 mg/kg per day) to induce remission of chorea (64).
Some controversy exists as to the role of immunosuppression in the management of Sydenham chorea. Despite mentions of the effectiveness of prednisone in suppressing chorea, this drug is only used when there is associated severe carditis. A placebo-controlled study showed that oral prednisone only accelerates the control of chorea; rates of remission and recurrence were not changed by the active treatment (111). Few reports describe the usefulness of plasma exchange or intravenous immunoglobulin in Sydenham chorea. An open-label study of 10 children treated with conventional treatment associated with immunoglobulin showed a better outcome than 10 patients who just received conventional management (141). Because of the efficacy of other therapeutic agents described in the previous paragraph, potential complications, and the high cost of the latter treatment modalities, these options are not usually recommended. In my practice, intravenous methylprednisolone is reserved for patients with persistent disabling chorea refractory to antichoreic agents. We reported that 25 mg/kg per day in children and 1 g/day in adults of methylprednisolone for 5 days followed by 1 mg/kg per day of prednisone is an effective and well-tolerated treatment for patients with Sydenham chorea refractory to conventional treatment with antichoreic drugs and penicillin (39; 13; 134). One study claims that oral prednisone (2 mg/kg for 14 days followed by gradual tapering) may indeed increase the speed of recovery as well as the rate of remission (61). However, the data should be interpreted with caution because of several issues: open-label nature, unmatched groups, lack of use of rating scale to asses chorea, and small number of patients.
Patients with previous history of Sydenham chorea may develop chorea gravidarum as a result of recurrence of the movement disorder during pregnancy. The onset of chorea is usually in the first trimester with spontaneous improvement or even remission in the third semester. If the movement disorder is not severe, no treatment is advisable. However, if necessary, low doses of neuroleptics are safe to mother and child. There is no risk of teratogenesis, but the 19th century literature describes fetal loss related to severe chorea. It is important to emphasize that there are also other causes of chorea gravidarum such as systemic lupus erythematosus (29). A study from our unit has shed light on the behavior of patients with Sydenham chorea during pregnancy. We found that 75% of women with Sydenham chorea in remission who became pregnant developed chorea gravidarum. A worrying finding of this study was an abortion ratio of 13% among patients with chorea during pregnancy (92).
Francisco Cardoso MD PhD
Dr. Cardoso of the Federal University of Minas Gerais has no relevant financial relationships to disclose.See Profile
Robert Fekete MD
Dr. Fekete of New York Medical College received consultation fees from Acadia, Acorda, Adamas, Amneal/Impax, Kyowa Kirin, Lundbeck, Neurocrine, and Teva.See Profile
Nearly 3,000 illustrations, including video clips of neurologic disorders.
Every article is reviewed by our esteemed Editorial Board for accuracy and currency.
Full spectrum of neurology in 1,200 comprehensive articles.
May. 19, 2022
May. 11, 2022
May. 01, 2022
Apr. 20, 2022
Apr. 20, 2022
Apr. 20, 2022
Apr. 03, 2022
Mar. 18, 2022