Sydenham’s Chorea

Lisa Bilston

Sydenham’s chorea (SC) is a neurodevelopmental disorder that manifests itself in children aged 7-12, who have previously experienced rheumatic fever (RF) [1] . The primary symptom of SC is motor chorea, as well as a series of other behavioural, neuropsychiatric and cognitive disturbances (symptoms differ between individuals). The etiology of this disorder is still controversial. The predominant theory proposes that it results from inflammation of the basal ganglia (BG), which leads to disruptions in cell signalling, and the balance of the BG's neurotransmitter systems (specifically GABAergic, cholinergic, and dopaminergic) [2] . SC is an important disorder to study as rates have been increasing in the past decade, and it remains the most common form of acquired chorea[3] (for other choreic disorders see Huntington's disease). It is also pivotal to study, as it has been found to re-occur throughout adulthood with an increasing degree of pathology. In addition, it is valuable to understand SC as it has been hypothesised that the genetic or familial vulnerabilities underlying the onset of SC, also predispose development of other neuropsychiatric conditions (eg. OCD and ADHD). As a result, better understanding of SC can improve the prognosis of impacted individuals, as well as further the understanding of these other conditions.


Sydenham's chorea is one of the major diagnostic symptoms of rheumatic fever (RF), where 25-60% of RF patients develop SC four to eight weeks after infection[1]. SC primarily affects children between the ages of 7-12[1], and is twice as common in females than males[4] . Prevalence rates of RF and therefore SC differ between developed and developing nations. In the US and Western Europe, rates of RF have been in decline since the 1960s, a trend that has been attributed to improvements in sanitary conditions, and the widespread use of penicillin[3]. Despite this, localized outbreaks of RF continue to be reported (eg. 1980s Ohio, and 1990s Pennsylvania), and a recent unexplained increase in RF cases has sparked renewed interest in the subject[3]. RF is more prevalent in developing countries, where it remains a severe problem[4]. For both developed and developing nations, SC is the most common form of acquired chorea.


The prognosis of Sydenham’s chorea is largely positive, with half of all acute cases being resolved within 2-6 months of onset[5] . The remaining fifty percent develop a more persistent form of chorea that can last for up to two years. Symptoms can recur within a year or two of remission. Several factors impact an individual's prognosis including family history, a personal history of behavioural problems, and degree of compliance with medications (eg. penicillin)[1].


Sydenham’s chorea is associated with motor, cognitive, and neuropsychiatric symptoms, as well as changes in neuroanatomy. In addition, it has been found to co-occur with other medical complications. Specifically, 60-80% of patients develop rheumatic heart disease, and 30% develop arthritis[6] .


The primary symptom of SC is chorea, which is defined as 'involuntary, purposeless and non-rhythmic movements' of the body[7] . Other motor symptoms include hypotonia[7](decreased muscle tone), disturbances in gait, facial grimacing[5], and hypometric saccades[2]. In rare cases tics have been found to co-occur with SC, although they have a distinct phenomenology to those found in other tic disorders; they are simple sounds believed to result from involuntary contractions of the pharynx and the larynx, and are not specific to certain situations or stimuli (like in Tourette’s syndrome)[8] .


Dysexecutive syndrome- a syndrome associated with deficits in attention and executive control- has been reported in patients with a history of SC[9] . Speech problems have also been identified. Specifically, phonemic verbal fluency, which is a physical correlate of the ease and intensity of word production, is disrupted[10] . In addition, prosody impairment has been identified, where melody, articulation time, and vocal energy of the speaker deviate from the norm, resulting in slow monotone speech[11] .

Neuropsychiatric and Emotional

SC is highly co-morbid with other neuropsychiatric disorders including anxiety, depression[12] , ADHD, OCD and OCSD (obsession compulsion spectrum disorders- tic disorder, body dysmorphic disorder and grooming behaviours)[13] . These concomitant neuropsychiatric disorders are more prevalent in persistent than acute SC[12]. Although rare, psychotic episodes have also been found to occur after the onset of acute SC, where the patient experiences both hallucinations and delusions[3]. Finally, emotional lability is a common feature in SC patients.


Fig. 1: Left- control, Right- increased size of the BG in SC patient (3).
SC is primarily associated with an acute increase in size of the basal ganglia (BG) (Fig. 1)[3], although studies have also identified changes in the cerebral cortex and thalamus[1]. In addition to increased size, the basal ganglia exhibits increases in signalling intensity, and perfusion/blood flow[14] . Furthermore, magnetic resonance spectroscopy has shown biochemical abnormalities in the BG of acute SC patients, which is an indication of changes in BG metabolism; two abnormal peaks were identified, which were associated with abnormal levels of lipids and amino acids, thought to be the by-products of neuron damage[15] . In addition to neuronal damage, changes in these brain regions include: decrease in neuronal cell number, changes in cell signalling, gliosis (astrocyte proliferation in damaged areas of the CNS), swelling of the endothelial cells, infiltration of perivascular round cells and petechial haemorrhages[1]. The
permanence of these changes is controversial (see Long-Term Effects), as is the relationship between these changes and SC onset, although some theories have emerged.


Immune-Mediated Hypothesis

Fig. 2: Infection of the pharynx with streptococci in RF

Sydenham’s chorea is one of the major symptoms of RF, which develops following infection of the pharynx by one of the group A beta-haemolytic streptococci (Fig. 2)[5]. The predominant etiological theory of SC is that it is an autoimmune disorder resulting from the cross reaction of IgG antibodies raised against the M protein of the streptococci, with endogenous antigens found on the basal ganglia (specifically the caudate and subthalamic nuclei)[16] , a process known as molecular mimicry[7]. The exact molecular substrate of these anti-basal ganglia antibodies (ABGA) is still controversial, although it has been proposed that they react with lysoganglioside receptors, resulting in improper activation of intracellular signalling[17] . This activation is believed to interfere with normal basal ganglia functioning, leading to the hyperkinetic phenotype associated with SC.

Three lines of evidence exist in support of this inflammatory theory of SC:[2]
  1. Symptoms of SC improve after exposing patients to immunomodulatory therapies. Examples of these therapies include steroids and intravenous antibody injections, which act as immunosuppressants[5].
  2. ABGAs have been identified in 100% of acute SC cases, and 69% of persistent SC cases[5] . It has also been shown that exposure of neuronal cells to these antibodies increases intracellular calcium levels after cellular depolarisation by KCl[2], resulting in the initiation of calcium dependent protein kinase II activity[18] (i.e. evidence exists in support of the hypothesis that ABGA can induce changes in intracellular signalling).
  3. Th2 cytokines have been found to be up-regulated in the serum and CSF of SC patients, further implicating the immune system in SC pathogenesis[2].

The Dopamine Hypothesis
Fig 3. Pathways of the BG (21).

The basal ganglia is an important structure that modifies movement via the direct (‘go’) and indirect (‘stop’) pathways (Fig. 3)[19] . It receives excitatory input from the cortex, which is then relayed through the GABAergic neurons of the basal ganglia. Varying levels of inhibitory output are then sent to the thalamus, which relays messages back to the cortex, allowing for the modification of movement. The activity of the basal ganglia is modulated by dopaminergic neurons of the substantia nigra, where dopamine works to increase movement. Acetylcholine is another neurotransmitter used by the basal ganglia; compared to other brain regions, cholinergic neurons are found at very high concentrations in the BG.

Another etiological theory of SC is that symptoms arise when damage to the basal ganglia results in an imbalance of its neurotransmitter systems[1]. The dopamine hypothesis of SC states that damage to the basal ganglia (potentially by an immune-mediated mechanism) results in increased levels of dopamine (DA) and decreased levels of GABA and acetylcholine activity[20] . This imbalance in the neurotransmitter systems is theorised to result in a hyperactive 'go' pathway and a hypoactive 'stop' pathway resulting in chorea. Pharmacological evidence supports this hypothesis, where anti-dopaminergic medications (eg. haloperidol and pimozide), and GABA stimulating drugs (valproic acid) help in alleviating symptoms of SC[1]. In addition, studies have revealed that 20% of patients with a history of SC are hypersensitive to agents that stimulate the DA system[5]. The finding of co-morbid psychosis also supports this theory, as onset of psychosis has also been previously associated with higher-than-normal DA levels[3].

Animal Models

Until recently, evidence in support of a
Fig. 4: Immunostaining of IgG in the rat BG (20).
n immune-mediated etiology of SC and the DA hypothesis has been highly speculative, and based on retrospective case studies reporting correlations between certain factors and SC. A recent study conducted in 2012 worked to establish a causal link between the presence of ABGA in patients with SC and the onset of the disorder20. This study was necessary to prove that the antibodies were a causative agent of SC, as opposed to a by-product of an inflamed basal ganglia. To do this, antibodies were extracted from the serum of SC patients and stereotaxically injected into the basal ganglia of mice (Fig. 4). Changes in motor behaviour and neurotransmitter systems were then monitored, but no changes were detected. This lack of positive results indicated that small concentrations of these ABGA were not sufficient to elicit the symptoms/features of SC. Two possible explanations exist for these results. Firstly, other factors could be involved in the onset of SC, and secondly onset might only occur after continuous exposure to high levels of the antibodies. This is a pivotal study as it was the first to try and establish a causal link between ABGA and SC using animal models, and it tests the validity of the etiological theories previously proposed.

Co-morbid disorders and Genetic Implications

Infection with streptococci is a common problem in humans, but only a small proportion of cases lead to the onset of rheumatic fever, of which SC is a major diagnostic symptom[21] . Research has been aimed at trying to identify factors that confer susceptibility to developing RF/SC, but with little success. Genetic factors,for example, only account for the development of 3% of RF cases, and no specific genes have been implicated in increased risk of SC development.
Most of the research into the genetics of SC is based on the observation that there is a high comorbiditybetween SC and other neuropsychiatric disorders such as OCD, ADHD and depression[12]. The relationship between SC and these disorders is difficult to reconcile, as two possible explanations exist for the high co-morbidity rates[17]:
  1. Long-term damage incurred by the basal ganglia during the acute phases of SC might increase susceptibility to other neuropsychiatric disorders.
  2. These neuropsychiatric disorders and RF/SC have a shared genetic/familial etiology. Note: this is not to say that long-term damage does not occur, only it is not the factor responsible for the co-morbidity rates.

Prevalence of Co-morbid Neuropsychiatric Disorders

Ridel and colleagues (2010) conducted a study that examined the prevalence of ADHD, depression and anxiety disorders prior to, during, and after the onset of chorea in a cohort of 28 patients with a history of SC[12]. Both ADHD and anxiety disorders saw only a modest increase in prevalence rates after the onset of chorea: rates increased from 30% to 37% for ADHD, and 71% to 79% for anxiety. In contrast, the prevalence of depression increased dramatically with the onset of chorea (13% to 31%), and remained high after onset (27%) (Fig. 5). At all time periods, the prevalence rates of neuropsychiatric disorders in SC patients were higher than in control populations, indicating high comorbidity.

This study revealed that onset of chorea was not a strong predictor of the de novo development of ADHD or anxiety disorders. As a result, the high co-morbidity of these disorders with SC indicated that SC patients have an inherent susceptibility towards their development, a finding in support of point 2 [12] .The relationship between SC and these disorders is unclear, but these findings suggest a potential genetic or familial link (a link that was further explored using family studies). A second study that aimed to further reconcile the relationship between SC and ADHD, further implicated a history of ADHD as a risk factor predicting SC onset when infected with streptococci[16] . In terms of depression, development of chorea was a strong predictor of depression onset, a correlation that could be attributed to either long term damage incurred during SC (point 1), or the social impacts of the disorder (eg. onset of a motor disorder could impact ones personal life, resulting in development of depression)[12].
Fig. 5: Prevalence of co-morbid neuropsychiatric disorders, prior to-, during, and after onset of chorea (12).

Family studies: Anxiety Disorders

OCD/OCSDs, a sub-category of the anxiety disorders, have the highest rate of comorbidity with SC of all the neuropsychiatric disorders[1]. This high comorbidity is found in both the acute phase of SC, and after remission[22] . Family studies, which examined the relationship between OCSDs and SC, yielded results in support of the second hypothesis; they showed that first degree relatives of patients with RF/SC have higher rates of OCSDs compared to controls[17]. These results provided good evidence for a genetic link between OCSDs and SC.
Two explanations exist as to why OCSDs and RF have become genetically linked[17]:
  1. Evidence has shown that OCSDs as well as SC can result from a dysfunctional immune reaction against the basal ganglia[1]. This shared dysfunctional reaction could result from genes that regulate the immune response being abnormally expressed.
  2. An association between OCD and an immune response could be adaptive, and therefore selected for over the course of evolution[17]. Compulsive behaviours typically consist of grooming and cleaning rituals, which become extreme in OCD. Increasing these behaviours (i.e. development of OCD) after an infection could be beneficial in alleviating the root cause of the disease (eg. unsanitary conditions).
Generalised anxiety disorder has also been found to be more prevalent in first degree relatives (FDRs) of SC patients, than in FDRs of healthy controls, suggesting a common genetic/familial basis for anxiety disorders in general and SC[13].
These family studies in association with Ridel's study led to the proposal that SC is a multifaceted disorder that results from the interaction between environmental factors and multiple genes, each of which confer a slight vulnerability towards development of SC after streptococci infection[17].

Persistent SC and Long-Term Effects

The long-term effects of SC and therefore the causes of persistent and recurrent SC are still controversial. Two potential explanations exist as to how these chronic SC cases can arise[7]:
  1. They could result from a continuous/recurrent autoimmune response against the basal ganglia (i.e. no long-term damage).
  2. or They could result from permanent structural/functional changes to the brain as a result of the initial infection (i.e. long-term damage).
Evidence for and against these two points is inconclusive and varies between subgroups of SC patients.

One line of evidence showing long-term BG damage, is based on a subset of patients with a history of SC and OCD[23] . In these patients, NAA (an important CNS metabolite)/creatin ratios were much lower than in control groups (patients with no history of SC or OCD) (Fig. 6), indicating a decrease in basal ganglia metabolism relative to controls. This decrease has not been found in patients with a history of SC who do not have a concomitant history of OCD. This suggests that permanent basal ganglia damage occurs in a subset of SC patients who also have a history of other neuropsychiatric disorders.

Fig. 6: MRS in SC patient in remission (left) and in SC patient with OCSD (right). NAA/CR is lower in the co-morbid condition (23).
In addition, it has been shown that having pervasive SC as a child (prolonged case with multiple relapses) could result in irreversible basal ganglia damage[5] . Elderly individuals (age 70-80) with a history of persistent SC, who were suffering from a late recurrence of SC, were found to have no ABGA in their serum. The absence of these antibodies makes it unlikely that the onset of the disorder was triggered by an autoimmune reaction against the basal ganglia, but instead resulted from chronic damage to the BG as a result of SC, leading to the BG having an overactive DA system, which causes chorea. From a clinical perspective, this study shows how important it is to treat SC at a young age and to prevent multiple recurrences, in order to decrease the possibility of long term damage, and therefore recurrence in old-age.

Brain derived neurotrophic factor (BDNF) has also been implicated in persistent SC, where it is found in lower levels in patients with persistent as opposed to acute SC[7] . As BDNF promotes the activity and survival of striatal neurons, it has been hypothesized that neurodegeneration of the striatal neurons is involved in persistent SC, and results in the SC phenotype.

Although indirect evidence seems to indicate that SC causes long-term damage to the BG, which can then lead to a more chronic form of SC, there remains a lack of direct evidence in favour of this theory. Although research conducted in the 1920s indicated that widespread damage to the CNS existed[24] , more recent studies show that structural/functional changes incurred during the acute phase of SC disappear with remission[7]. In addition, in the absence of a co-morbid neuropsychiatric disorder, no difference has been found in basal ganglia metabolism in patients with or without a history of SC[23]. The long-term effects of SC are therefore still speculative, and the mechanism underlying persistent/recurrent SC unknown.


  1. ^ Gordon, N. (2009). Sydenham’s chorea, and its complications affecting the nervous system. Brain & Development 31: 11-14.
  2. ^ Teixeira, L.A., Guimaraes, M.M., Romano-Silva, M.A., & Cardoso, F. (2005). Serum from Sydenham’s chorea patients modifies intracellular calcium levels in PC12 cells by a complement-independent mechanism. Movement Disorders 20(7): 843-845.
  3. ^ Cardoso, F., Seppi, K., Mair, K.J., Wenning, G.K., & Poewe, W. (2006). Seminar on choreas. Lancet Neurol. 5: 589-602.
  4. ^ Teixeira, L.A., Maia, D.P., & Cardoso, F. (2007). Psychosis following acute Sydenham’s chorea. Eur. Child Adolesc. Psychiatry 16: 67-69.
  5. ^ Harrison, N.A., Church, A., Nisbet, A., Rudge, P., & Giovannoni, G. (2004). Late recurrences of Sydenham’s chorea are not associated with anti-basal ganglia antibodies. J Neurol. Neurosurg. Psychiatry 75: 1478-1479.
  6. ^ Walker, K.G., & Wilmshurst, J.M. (2010). An update on the treatment of Sydenham’s chorea: the evidence for established and evolving interventions. Ther. Adv. Neurol. Disord. 3(5): 301-309.
  7. ^ Teixeira, L.A., Bretas, T.L., Kummer, A., Melo, L.C., Baraldi, A., et al. (2010). Serum levels of brain-derived neurotrophic factor in Sydenham’s chorea. Neurol. Sci. 31: 399-401.
  8. ^ Teixeira, A.L., Cardoso, F., Maia, D.P., Sacramento, D.R., Mota, C.C.C., et al. (2009). Frequency and significance of vocalizations in Sydenham’s chorea. Parkinsonism & Related Disorders 15: 62-63.
  9. ^ Cardoso, F., Beato, R., Siqueira, C.F., & Lima, C.F. (2005). Neuropsychological performance and brain SPECT imaging in adult patients with Sydenham’s chorea. Neurology 64: A76.
  10. ^ Cunningham, M.C.Q.S., Maia, D.P., Teixeira, A.L., & Cardoso, F. (2006). Sydenham’s chorea is associated with decreased verbal fluency. Parkinsonism and Related Disorders 12: 165-167.
  11. ^ Oliveira, P.M., Cardoso, F., Maia, D., Cunningham, M.C.Q., Teixeira, A.L., et al. (2010). Acoustic analysis of prosody in Sydenham’s chorea. Arq. Neuropsiquiatri. 68(5): 744-748.
  12. ^ Ridel, K.R., Lipps, T.D., & Gilbert, D.L. (2009). The prevalence of neuropsychiatric disorders in Sydenham’s chorea. Pediatric Neurology 42(4):243-248.
  13. ^ Seixas, A.A.A., Hounie, A.G., Fossaluza, V., Curi, M., Alvarenga, P.G., et al. (2008). Anxiety disorders and rheumatic fever: is there an association? CNS Spectr. 13(12): 1039-1046.
  14. ^ Barsottini, O.G., Ferraz, H.B., Seviliano, M.M. & Barbieri, A. (2002). Brain SPECT imaging in Sydenham’s chorea. Braz. J. Med. Biol. Res. 35:431-436.
  15. ^ Castillo, M., Kwock, L., Arbelaez, A. (2000). Sydenham’s chorea: MRI and proton spectroscopy. Neuroradiology 41: 943-945.
  16. ^ Margari, L., Ventura, P., Portoghese, C., Presicci, A., Buttiglione, M., et al. (2005). Brain magnetic resonance spectroscopy in Sydenham’s chorea and ADHD. Pediatric Neurology 34(6): 467-473.
  17. ^ Hounie, A.G., Pauls, D.L., Rosario-Campos, M.C., Mercadante, M.T., Diniz, J.B., et al. (2007). Obsessive-compulsive spectrum disorders and rheumatic fever: A family study. Biol. Psychiatry 61: 266-272.
  18. ^ Kirvan, C.A., Swedo, S.E., Heuser, J.S., & Cunningham, M.W. (2003). Mimicry and autoantibody-mediated neuronal cell signalling in Sydenham chorea. Nat. Medicine 9: 914-920.
  19. ^ Garrett, E., & Crutcher, M.D. (1990). Functional architecture of basal ganglia circuits: neural substrates of parallel processing. TINS 13(7): 266-271.
  20. ^ Ben-Pazi, H., Sadan, O., & Offen, D. (2012). Striatal microinjection of Sydenham’s chorea antibodies: Using rat model to examine the dopamine hypothesis. J. Mol. Neurosci. 46: 162-166.
  21. ^ Bessen, D.E. (2001). Genetics of childhood disorders: XXXII autoimmune disorders, part 5: Streptococcal infection and autoimmunity, an epidemiological perspective. Development and Neurobiology 40(11): 1346-1348.
  22. ^ Mercadante, M.T., Diniz, J.B., Hounie, A.G., Ferrao, Y., Alvarenga, P., et al. (2005). Obsessive-compulsive spectrum disorders in rheumatic fever patients. J. Neuropsychiatry Clin. Neurosci. 17: 544-547.
  23. ^ Alkan, A., Kutlu, R., Kocak, G., Sigiric, A., Emul, M., et al. (2004). Brain MR spectroscopy in children with a history of rheumatic fever with a special emphasis on neuropsychiatric complications. Eur. J. Radiol. 49: 224-228.
  24. ^ Greenfield, J.G., & Wolfsohn, J.M. (1922). The pathology of Sydenham’s chorea. Lancet. 2: 603-606.