Rheumatic fever, an autoimmune disorder triggered by a streptococcal infection, and its central nervous system (CNS) expression Sydenham’s chorea are associated with psychiatric symptoms.1,2 Autoantibodies and mimicry mechanisms have been considered as possible explanations for this association.2 Currently, several studies support the association between Sydenham’s chorea and obsessive-compulsive disorder (OCD) tic disorders, and/or attention deficit hyperactive disorder (ADHD).1,2
It has been speculated that OCD, tic disorder, body dysmorphic disorder, trichotillomania, and eating disorders belong to the same spectrum,3,4 with possible common underlying genetic mechanisms.5 Spectrum disorders may share clinical, psychopathological, and pathophysiological characteristics.4
Due to the lack of agreement about the relationship between nonactive rheumatic fever and OCD or obsessive-compulsive spectrum disorders (OCSD), we decided to report five cases with documented rheumatic fever and at least one of the putative OCSD (e.g., BDD, trichotillomania,5 and eating disorders6 in addition to tic disorder and ADHD.
The five cases described were assessed by certified psychiatrists through semistructured interviews: Structured Clinical Interview for DSM—IV or the Kiddie-SADS-E.7
Mr. A, age 33, developed rheumatic fever with migratory arthritis and carditis at age 10 after scarlet fever. Body dysmorphic disorder started shortly thereafter characterized by excessive concerns about an imagined gynecomastia and a small penis. He developed obsessive-compulsive symptoms at age 8, characterized by concerns related to contamination, religion and symmetry. His obsessive-compulsive symptoms worsened after the onset of rheumatic fever. He was first treated with clomipramine at age 20 and his body dysmorphic disorder and OCD remitted. His OCD became severe again at age 30 and remitted with fluoxetine. He also met criteria for recurrent major depression and chronic motor tic disorder for 5 years after he was 18.
Ms. B, age 50, had rheumatic fever at age 5 with migrant polyarthritis and elevated acute-phase markers and streptococcal antibody titers. Body dysmorphic disorder started at age 14 with exaggerated concern of having a big nose. Her symptoms had partial remission in the beginning of adulthood with a 3-year period exacerbation at her early thirties. OCD first appeared at age 10 including ordering and symmetry, and persisted mild until her second gestation, at her late twenties, when symptoms worsened. She also reported skin-picking behavior and met criteria for social phobia and ADHD in childhood. She also presented all symptoms of anorexia nervosa except for amenorrhea.
Ms. C, age 25, had rheumatic fever at age 5 with migrant polyarthritis and subcutaneous nodules, fever, elevation in the acute-phase markers and streptococcal antibody titers, after a streptococcal throat infection. Body dysmorphic disorder first appeared at age 14 as an overvalued preoccupation about having large and asymmetrical breasts and compulsively sought reassurance. Her body dysmorphic disorder had a waxing and waning course, never reaching full remission. OCD started at age 7 and included hoarding, verifying, and repeating rituals and was never treated. She also met criteria for current major depressive disorder and social phobia and past ADHD in childhood.
Ms. D, age 15, had Sydenham’s chorea at age 11 diagnosed by an experienced child neurologist. Concurrent with Sydenham’s chorea she reported hair-pulling behavior, spending hours arranging her hair at the mirror, and the need of taking many showers a day to relieve dirty feelings in her body. She was thus diagnosed with trichotillomania and OCD. These symptoms remitted along with Sydenham’s chorea. Since that period, she has met criteria for chronic vocal tics (sniffing). More recently, when she was 18 years old, she reported intermittent binge eating not meeting full criteria for bulimia nervosa.
Mr. E, age 37, had rheumatic fever, presenting migrant polyarthritis, fever, elevation in the acute-phase markers and streptococcal antibody titers, after streptococcal throat infection at age 13. Two years later he started hair-pulling behaviors. He began pulling out his eyebrows, and afterward eyelashes and many other regions. Before pulling, he felt his skin itching around a specific hair. He attributed this to allergy and wore glasses and a cap to hide his hair loss. He feels irritated if not able to pull out his hair and reports no remission since the disorder started. Therefore, he meets criteria for trichotillomania.
This article reports five cases of rheumatic fever (only one with Sydenham’s chorea) with OCSD such as body dysmorphic disorder, trichotillomania, and eating disorders in addition to tic disorder and ADHD. Tic disorder and ADHD were already described in Sydenham’s chorea.1 Based on these cases and other recent reports of trichotillomania9 and body dysmorphic disorder10 in rheumatic fever without Sydenham’s chorea we propose to expand the range of psychiatric disorders that may be etiologically related to rheumatic fever.
The literature on this topic has focused solely on the relationship between Sydenham’s chorea and OCD/tic disorder, and streptococcal infection and OCD/tic disorder (PANDAS).1 Until today, only a single systematic study reported a relationship between OCD/tic disorder and rheumatic fever without Sydenham’s chorea.11 For sure, these distinct associations have pathophysiological and genetic implications. In addition, the relationship of other OCSDs and rheumatic fever has not been explored in this field yet.
The existence of an OCSD has been supported by genetic family studies of OCD that reported a higher frequency of tic disorder, body dysmorphic disorder, and grooming behaviors (trichotillomania, skin picking, and onicophagia)5 in first-degree relatives of OCD probands. Conversely, OCD has also been reported more frequently in first-degree relatives of probands with eating disorders.6
In addition to genetic studies, neuroanatomical models suggest that OCSDs involve dysfunction within cortico-striatal pathways. As recently delineated, there exist several parallel segregated cortico-striatal pathways, each subserving different types of functions.3 Thus, in the case of OCSD and rheumatic fever association, the phenotypic presentation along the OCSD could reflect the topography of dysfunction caused by immunological abnormalities within these circuits.12
The immunological mechanisms by which rheumatic fever could trigger these psychiatric manifestations are still unknown. However, there are intriguing hypotheses about the feasibility of complex behaviors being modulated by the immune system.12 Several models can be applied to explore how the immune system might influence the CNS. For example, the antibody-antigen complex, besides interacting with myosin (generating carditis), might have a role as a neuronal activator agent, binding to specific membrane molecules inducing intracellular signaling events. Recently, a study showed that monoclonal antibodies from Sydenham’s chorea patients bind to glucosamine (a streptococcal epitope) as well as to mammalian lysoganglioside, inducing a calcium/calmodulin-dependent protein (CaM) kinase.13 Based on these data, a scenario can be thought where the presence of activated immune cells, after crossing the blood-brain barrier and finding antigenic stimulus, would be generating a complex environment with autoantibodies inducing intracellular signaling, as well as the releasing of different cytokines, which have their own neuronal receptors. All together, it is possible to speculate that the activation of different striatum cell populations (such as striosomes and the matrix) might generate disequilibrium in the balance of compartments activity causing repetitive behaviors.14
Interestingly, in this case series body dysmorhpic disorder occurred after the OCD onset. This could be due to the late onset of body dysmorphic disorder compared to the usually early onset of the OCD related to rheumatic fever,12 but it is also possible to speculate that more exposure to the immunological mechanisms would be necessary to affect the body dysmorphic disorder neurobiological substrates or to trigger the expression of vulnerability genes.
Notwithstanding, one can argue that the association of disorders reported in this article could be coincidental, due to the high prevalence of rheumatic fever in Brazil (close to 3.6/1000.15 This hypothesis is mitigated, however, by the fact that the estimated risk of a random association among these disorders would probably be as low as 7.2/100,000, taking OCD as an example, whose prevalence is close to 2%. Furthermore, our sample was not actively ascertained nor obtained from larger rheumatic fever systematic studies, reinforcing the higher risk hypothesis.
In addition, an intriguing question emerges from the description of some of these cases. Which abnormal immune mechanism could explain the long interval between the acute phase of rheumatic fever and the manifestations of the OCSD? One possibility is that a late abnormal immune process occurs, mediated by the T-independent B cell response, when the presence of the T-independent antigenic stimulation could trigger a temporary antibody production. In this model, as soon as the stimulus becomes silent the immunological response goes down. Few B memory cells are kept, even for years, which can be reactivated again by new trigger agents.
The immediate clinical relevance of the presence of OCSD in RF patients is to alert clinicians of this association. Of course, if this association continues to be reported, RF criteria could eventually include OCSD as part of the clinical picture.
Recently, a group of researchers proposed five criteria to diagnose cases of pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections (PANDAS). The current diagnostic criteria for PANDAS include only OCD and tic disorder and have as exclusionary criteria the presence of rheumatic fever or Sydenham’s chorea.16 It has been discussed whether PANDAS are or are not part of the RF definition.1 The lack of other RF features such as carditis and arthritis in PANDAS cases has been one of the strongest arguments in supporting this distinction.16 Future studies, using epidemiological approaches, such as case-control studies, as well as clinical and laboratory investigations, including gender differentiation, in samples of OCSD patients should address this issue to confirm or reject the possible association among OCSD, rheumatic fever, and PANDAS.
In conclusion, although we still need a better understanding of the association between streptococcal infections and psychiatric symptoms, these cases support the hypothesis that rheumatic fever, which is a late complication of a streptococcal infection, may be associated with a broad range of psychiatric disorders such as OCD, tic disorder, body dysmorphic disorder, trichotillomania, eating disorders, and ADHD.
The authors thank James F. Leckman and Virginia W. Eicher for their careful review.
This study was supported in part by grants from MackPesquisa to Dr. Mercadante, and Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP): #99/08560-6 to Dr. Mercadante, #98/15013-9 to Dr Hounie, #99/13005-1 to Dr. Diniz, and #99/08560-6 to Dr. Miguel. Also supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPQ), Brazil (grant #521369/96-7) to Dr. Miguel.