Abstract
We compared the frequency of migraine among Sydenham's chorea (SC) patients, rheumatic fever (RF) patients without neurological symptoms and matched controls. Migraine was more frequent in SC patients (12/55, 21.8%) than in controls (9/110, 8.1%) and as common as in the RF group (10/55, 18.2%). Our data are in agreement with previous studies reporting higher frequency of migraine in other basal ganglia disorders, such as essential tremor and Tourette's syndrome.
Introduction
Rheumatic fever (RF) is a systemic autoimmune inflammatory disorder following streptococcal β-haemolytic infection. When RF affects the central nervous system, particularly the basal ganglia, the patients may develop a neuropsychiatric disorder called Sydenham's chorea (SC) (1). SC, the most common cause of chorea in children worldwide, is more frequent in girls, being characterized not only by movement disorders (chorea, tics, decreased muscle tone) but also behavioural disturbances such as obsessions, compulsions, hyperactivity and attention deficit (1).
In our patients with SC we have observed a high occurrence of migraine headache. In addition to the classical symptoms of SC, migraine may be an important cause of disability, resulting in missed school days and impairment of the activities of daily life in these patients. Moreover, clinical data support the view of a higher frequency of migraine in patients with basal ganglia disorders, such as essential tremor and Tourette's syndrome (2, 3). We decided, therefore, to compare the frequency of migraine among SC patients, RF patients without neurological symptoms and age- and gender-matched controls.
Methods
One hundred and ten subjects diagnosed as having RF according to the modified Jones criteria (4) and 110 healthy children ranging in age from 6 to 18 years from a Brazilian school were enrolled in this study performed from January, 2003 to December 2003. RF subjects were diagnosed at the Paediatric Cardiology Clinic of the Federal University of Minas Gerais, Belo Horizonte, Brazil, and exhibited signs of carditis demonstrated by echocardiographic studies. Fifty-five of the 110 RF patients had SC diagnosed according to criteria described elsewhere (5) and are currently followed in the Movement Disorders Clinic at the Federal University of Minas Gerais, Belo Horizonte, Brazil. The remaining RF patients had no neurological symptoms. Students in the age range of RF patients attending a local public school were invited to participate in the study as healthy controls. Children answered a questionnaire applied by a neurologist to investigate the presence of migraine. This structured questionnaire contained questions related to all criteria of migraine according to the International Headache Society (IHS) (6). Further evaluation of SC patients included a structured interview performed by one of us (D.P.M.). The interview was based on items of the Yale-Brown Obsessive Compulsive Scale (7) and the Leyton Obsessional Inventory-Child Version (8) to assess the presence of obsessive-compulsive disorder (OCD) according to DSM-IV criteria (9). This was done to test the hypothesis that OCD is associated with greater frequency of migraine. Comparisons among groups were made using Fisher's exact test with the level of significance set at P < 0.05.
Results
The mean age of SC, RF and control groups were, respectively, 13.8 ± 3.8, 14.1 ± 3.5 and 13.7 ± 3.0 years (P = 0.8). The groups were also gender matched (P = 0.5). The frequency of migraine was 21.8% (12/55) in SC patients, 18.2% (10/55) in RF patients and 8.1% (9/110) in controls. These results are displayed in Table 1. The difference in frequency of migraine headache between SC and controls was statistically significant (P = 0.02). There was a trend towards statistical significance when comparing the frequency of migraine with aura in SC patients (7/55, 12.7%) and RF subjects (1/55, 1.8%) (P = 0.06).
Comparison among Sydenham's chorea (SC) patients, rheumatic fever (RF) patients and controls
Of the 55 SC patients, nine (16.3%) also met DSM-IV criteria for OCD. OCD was not more common in SC patients with migraine (2/12, 16.7%) than in those without migraine (7/43, 16.3%, P > 0.99).
Discussion
Studies of paediatric migraine have shown different prevalence estimates, ranging from 3.2 to 10.6% depending on the diagnostic criteria, the age distribution of the sample and other methodological features (10). Only a few investigations have been conducted using IHS criteria and just one in Brazil (11). Our study provides original data on the occurrence of migraine in schoolchildren from Brazil. The obtained frequency of 8.1% is within the expected range worldwide (10). We also demonstrated that the frequency of migraine was more common in SC patients when compared with control subjects.
The presumable mechanism of SC is basal ganglia dysfunction mediated by autoantibodies that cross-react with Streptococcus pyogenes antigens (1). In fact, all patients with acute SC from our centre so far tested are positive for antibasal ganglia antibodies (ABGA) (12). Our results are thus in accordance with previous studies demonstrating increased prevalence of migraine in basal ganglia disorders, such as essential tremor and Tourette's syndrome (2, 3). Indeed, experimental and clinical studies have shown that basal ganglia are involved in the processing of noxious information, representing a gating mechanism for the regulation of nociceptive inputs to higher centres, and may participate in the pathophysiology of migraine (13).
Despite not exhibiting chorea or other neurological signs, RF patients could also have their basal ganglia circuits affected by autoimmune mechanisms. Such an hypothesis is consistent with our finding of a similar frequency of migraine in the RF and SC patients. Moreover, others have shown that OCD is more frequently detected in RF patients than in controls (14). The finding of ABGA in our patients with RF further strengthens such an hypothesis (12). It is obvious, however, that basal ganglia dysfunction is more severe in patients with SC in comparison with those with RF without chorea. Not only the absence of an overt movement disorder in RF but also the discrepancy in the positivity of ABGA (100% vs. 13%) support this assumption (12). In the light of these findings, one may speculate that the trend towards aura (a more specific feature of migraine) being more common in SC than in RF is consistent with the hypothesis that the basal ganglia play a part in the pathogenesis of migraine.
We also found that 16.3% of our SC patients had OCD, which is comparable to previous studies (14, 15). As OCD is commonly diagnosed in SC and similar pathophysiological mechanisms have been proposed for OCD and migraine (3), we compared the incidence of OCD between SC patients with and without migraine. There was no difference in the frequency of this behavioural disorder between SC patients with and without migraine. Recently, Kwak et al. (3) also failed to demonstrate that Tourette's syndrome patients with migraine had greater obsessive-compulsive features.
Our study presents the first survey of migraine in SC, demonstrating its higher occurrence in these patients. Our study also adds further clinical evidence in support of the involvement of the basal ganglia in migraine pathophysiology.
Footnotes
Acknowledgements
This work was supported by grants from Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Fundação de Amparo à Pesquisa de Minas Gerais (FAPEMIG), Brazil. We also thank the students of Escola Estadual Ana de Carvalho (Belo Horizonte, Brazil) as well as its principal, Mr Edir de Jesus, for participating in the study.