page contents

Identification and Possible Treatment of PANDAS

Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infections

Kathryn Seraphin

 The Pennsylvania State University – College of Medicine

June 2003


Introduction: There has been a subset of young children who have been noted to abruptly develop Obsessive Compulsive Disorder (OCD) and/or tic disorders, such as Tourette’s Disorder, in association to a recently documented Group A Beta-hemolytic Streptococcal (GABHS) infection. It was found that these children have a condition termed Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infections (PANDAS), which has unique criteria and characteristics differentiating it from classic childhood OCD or tic disorders.


PANDAS as a separate identity:

         The working criteria for the diagnosis of PANDAS was modified through a study which identified the first 50 cases of PANDAS:

“Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infections: Clinical Description of the First 50 Cases” by Susan E. Swedo et al Am J Psychiatry 155:2, Feb 1998.


         The five criteria established are as follows:

  1. Presence of OCD and/or tic disorder – the patient must meet lifetime diagnostic criteria (DSM V) for OCD or tic disorder
  2. Pediatric onset – symptoms first evident between ages 3 and beginning of puberty
  3. Episodic course of symptom severity – clinical course consists of abrupt onset psychiatric symptoms or dramatic symptom exacerbation
  4. Association with GABHS infection – lifetime pattern of symptom exacerbation must be temporally related to GABHS infection (diagnosed via throat culture or rise in antibody titers)
  5. Association with neurological abnormalities – abnormal neurological exam (i.e. – choreiform movements or tics) during exacerbation


In this report, children were recruited that met the above criteria for PANDAS.

These 50 children were then systematically evaluated clinically to further evaluate any unique characteristics of PANDAS.   Evaluation consisted of medical and psychiatric assessment, including structured psychiatric interview, standard neurological exam, and baseline labs.

Results of clinical investigation revealed the following:

  • The average age of onset of PANDAS is much earlier than classic childhood OCD or tic disorder. Mean age PANDAS subset of OCD 7.5 years (SD = 2.7), and mean age PANDAS subset of tic disorder 6.3 years (SD = 2.7); these mean ages are nearly 3 years younger than previous groups of childhood onset OCD or tic disorders.
  • The PANDAS children were evenly divided between those with a primary tic disorder (52%) and primary OCD (48%)
  • Demographic data showed males outnumbered females with PANDAS, 2.6:1
  • PANDAS clinical course: acute and dramatic relapsing and remitting course (parents could pin-point date neuropyschiatric symptoms erupted) with significant comorbidity accompanying the exacerbations: emotional lability, separation anxiety, nighttime fears, cognitive defects, oppositional behaviors
  • Frequent association with motoric hyperactivity, impulsivity, and distractibility. Several patients met criteria for ADHD, except the onset of symptoms frequently occurred after age 6 years.
  • Each child had at least one symptom exacerbation that was preceded (within 6 weeks) by a documented GABHS infection. Though, of the total 144 neuropsychiatric -exacerbations of the 50 children it was noted that 23 % (33 cases) were not associated with any sign of GABHS infection within the preceding month.


Author’s conclusion: The working diagnostic criteria appear to accurately characterize a homogeneous patient group in which psychiatric symptom exacerbations are triggered by GABHS infections. The identification of such a subgroup will allow for testing of models of pathogenesis, as well as development of novel treatment and prevention strategies.


PANDAS as a separate identity based on pathophysiology:

         The proposed theory of the pathophysiology of PANDAS is an immune-mediated model of molecular mimicry, the same mechanism suggested for rheumatic fever. After exposure to antigens on the surface of Group A beta-hemolytic streptococcus during a streptococcal infection, the body develops antibodies to combat the bacteria. In susceptible individuals, the antigenic area of the bacteria is similar enough to endogenous host tissue, causing a cross-reaction of the antibody to the host; therefore, formulating an autoimmune attack on it’s own tissue. This causes a local immune reaction to the targeted tissue; i.e. – brain, joints, heart valves.   In rheumatic fever, this causes heart valve damage, arthritis, and abnormal neurological movements termed Sydenham’s chorea. In PANDAS the antibodies are thought to cross-react with neuronal tissue of the CNS, in particular the basal ganglia of the brain, which is responsible for movement and behavior – resulting in tics and/or OCD.


Studies supporting this theory:


  1. “Identification of Children with Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infections by a Marker Associated with Rheumatic Fever”, by Susan E. Swedo et al Am J Psychiatry 1997; 154:110-112.

         In this article, a highly specific and sensitive trait marker used to identify patients susceptible to rheumatic fever, D8/17 (found on subset of DR+ cells in the peripheral circulation), was found to also identify children susceptible to PANDAS accurately.

There were 70 children enrolled, separated into three groups: 9 children with Sydenham’s chorea served as the positive comparison group, 24 healthy children served as the negative comparison group, and 27 meeting PANDAS criteria served as the experimental group.

Results: The frequency of D8/17- positive individuals was significantly higher in both patient groups, Sydenham’s chorea at 89 % and PANDAS at 85 %, than healthy volunteers at 17 %. Also, the mean number of D 8/17 + cells in the serum samples from the patient groups were significantly higher than the healthy volunteer group.

These results suggest that the PANDAS subgroup of children with OCD and Tourette’s disorder are susceptible to a post-streptococcal autoimmune phenomenon, similar to that of Sydenham’s chorea (the neurological manifestation of rheumatic fever). A trait marker, D8/17, that has been associated with rheumatic fever can now be used to identify children with PANDAS.


  1. “MRI Assessment of Children with Obsessive-Compulsive Disorders and Tics Associated with Streptococcal Infection”, by Jay N. Giedd et al Am J Psychiatric Association: 157(2); 281-283

         Since the basal ganglia is perceived to be the target of the autoimmune cross-reactivity in PANDAS, a study was designed to compare brain MRIs of children with PANDAS associated OCD and tics to healthy comparison children (matched for age and sex) to determine if there was a significant difference in the size of basal ganglia, reflecting inflammation.

It was found that the average sizes of the caudate, putamen, and globus pallidus, but not the thalamus or total cerebrum, were significantly greater in the group of children with streptococcal-associated OCD and/or tics than in healthy children. The differences were similar to those found in previous studies of subjects with Sydenham’s chorea compared to normal subjects. There was no direct correlation of increase in basal ganglia size to severity of OCD and/or tics.

The findings of significant enlargement of the basal ganglia in children with PANDAS, similar to those found in children with Sydenham’s chorea, suggest that an inflammatory processes is occurring, most likely secondary to a post-streptococcal cross-reactive autoimmune reaction. This inflammation of the basal ganglia is likely to be the physical factor causing of the neurologic symptoms of OCD and tic disorders.  


Treatment of PANDAS based on pathogenesis of post-streptococcal autoimmunity

The autoimmune pathophysiology proposed for PANDAS and Sydenham’s chorea, would support the use of immuno-modulation as possible therapy for children with PANDAS, such as plasma exchange, IV immunoglobulin, or corticosteriods. Corticosteriods are an unfavorable treatment modality because they worsen OCD and tic behavior. Therefore, plasma exchange (which filters out the antibodies) and IVIG (which binds receptors with pooled antibodies) were studied compared to placebo (sham IVIG) in reducing severity of neuropsychiatric symptoms in the following study:


“Therapeutic Plasma Exchange and Intravenous Immunoglobulin for Obsessive Compulsive Disorders and Tic Disorders in Childhood”, by Susan J. Perlmutter et al, The Lancet 1999, 345 (9185):1153-58.

         In this partial double-blind randomized trial, 30 children with severe infection-triggered exacerbations of OCD or tic disorders were randomly assigned treatment with plasma exchange (5 single volume exchanges over 2 weeks), IVIG (1 g/kg daily on 2 consecutive days), or placebo (IV saline solution over 2 days). Psychiatric symptom severity was rated at baseline, at 1 month, and at 1 year after treatment by standard assessment scales for OCD, tic disorder, anxiety, depression, and global function.

{Randomized by randomization chart}

{Investigators and study participants were blinded to whether the child received IVIG or placebo, but were obviously aware of who received plasma exchange}

After symptom ratings at 1 month were completed the IVIG/placebo masking was broken, and open treatment to IVIG and plasmapheresis was available for those not improving on saline (all of the subjects), therefore, there was no more placebo 1 year follow-up ratings.


  • 29 children completed the trial:10 received plasma exchange, 9 received IVIG, and ten placebo
    • o at baseline, the three study groups were similar in age, primary diagnosis, duration of exacerbation, use of psychotropic medications, presence of antistreptococcal titers.
  • 1 month follow up: striking improvements seen in plasma exchange and IVIG groups concerning obsessive-compulsive symptoms, anxiety, depression, emotional liability and global functioning.
    • o Global change scores for children in the plasma exchange and IVIG groups improved by 48% and 41 % respectively, in contrast to placebo which showed no change in overall symptom severity
    • o Improvement in obsessive-compulsive symptoms, rated by Yale-Brown OCD scale, seen in plasma exchange (33% improvement) and IVIG (35% improvement) groups
    • o Anxiety improvement, as rated by the NIMH anxiety scale, significant in plasma exchange and IVIG groups, 31 % and 47% improvement respectively
    • o Plasma exchange group showed significant improvement in tic severity over placebo, but IVIG did not
  • 1 year follow up: psychiatric symptoms remained improved from baseline on all measures.
    • o The most clinically meaningful improvements occurred in obsessive-compulsive symptoms (58 % and 70 % improvement from baseline; IVIG and plasmapheresis respectively), tic severity (53 % improvement plasmapheresis), and global measures of symptom severity (26 % and 45 % improvement) and psychological functioning (26 % and 47 % improvement).
    • o The symptoms ratings and clinical impression of treatment of OCD favored plasmapheresis over IVIG
    • o Parents report “ my child’s back to their old self again”, and children reported “things are a lot easier now”. Kids went from “symptom impairments in several social areas” to now “good functioning in all areas”.


Conclusion: Both plasma exchange and IVIG were effective in significantly reducing the symptom severity of OCD and tic disorder for children with PANDAS. Possible mechanisms would be through blocking (via IVIG) or removing (via plasma exchange) the antistreptococcal antibodies that were cross-reacting to the neuronal tissue. Supporting the idea that PANDAS is a separate subgroup of OCD/tic disorder and may require a unique treatment plan.

Also separating PANDAS from classic childhood onset OCD is the remission of symptoms after a single course of IVIG or plasma exchange. Typically, patients with OCD will have response to SSRIs and/or behavioral modification; however, the response is only partial and relapse is common after medication is discontinued.

Further studies: Trials with more selective and specific immunomodulatory agents for children with PANDAS.


Further Observations of treatment

         It has been documented in several case studies and a prospective study at a community pediatric practice that children with PANDAS respond readily to antibiotic treatment. The prospective study:

“Prospective identification and treatment of children with pediatric autoimmune neuropsychiatric disorder associated with group A streptococcal infection (PANDAS)”, by Marie L. Murphy et al, Archives of Pediatric and Adolescent Medicine 156(4):356-61.

         Though this study only had a sample population of 12, all cases had documented active GABHS infections (by throat swab or culture, and/or rise GABHS Ab titers) associated with the initial episode of neuropsychiatric symptoms, and these neuropsychiatric symptoms (OCD, anxiety, ADHA, tic disorder) disappeared during antibiotic treatment for streptococcal-infected throat. The recurrences of neuropsychiatric behavioral symptoms were directly related to the recurrence of streptococcal throat infections, which again resolved with appropriate antibiotic treatment (amoxicillin or cephalosporin). There were no instances of new recurrence of OCD in the absence of new GABHS infection (throat cultures were also taken after initial treatment of antibiotics, when symptom free, and were negative).

This study supports the linkage of GABHS infection to acute neuropsychiatric disorders, as described in the PANDAS sub-category; however, it slightly undermines the theory that PANDAS is caused by an autoimmune cross-reactivity to brain tissue, because rapid treament (within 2-14 days) with antibiotics does not target antibodies designed against GABHS/brain tissue.

This leads one to believe there are several components to the pathophysiology of PANDAS, including genetic susceptibility of the host, an autoimmune reaction, and possibly a reaction to the toxins of the bacteria themselves or a certain strain of GABHS.

Each of the delineated points can be the focus of diagnostic testing and therapeutic intervention. Currently, the NIMH is investigating the role of antibiotic prophylaxis against GABHS and its affects on the recurrence of neuropsychiatric symptoms in children with PANDAS.



Garvey MA, Giedd J, Swedo SE: PANDAS: The Search for Environmental Triggers of Pediatric Neuropsychiatric Disorders, Lessons from Rheumatic Fever.  J Child Neurology 1998, 13(9); 413-23 

Garvey MA, Perlmutter S, Allen AJ, et al: “A Pilot of Penicillin Prophylaxis for Neuropsychiatric Exacerbations Triggered by Streptococcal Infections” Society of Biological Psychiatry 1999, 45: 1564-71

Giedd, JN et al: “MRI Assessment of Children with Obsessive-Compulsive Disorders and Tics Associated with Streptococcal Infection”, Am J Psychiatric Association: 157(2); 281-283

 Murphy, ML et al. “Prospective identification and treatment of children with pediatric autoimmune neuropsychiatric disorder associated with group A streptococcal infection (PANDAS)”, Archives of Pediatric and Adolescent Medicine 156(4):356-61.

Perlmutter SJ, Leitman SF, Swedo SE, et al: “A Case of Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infection”, Am J Psychiatry 1998 : 155 (11) 1592-98

Perlmutter SJ, Leitman SF, Swedo SE, et al: “Therapeutic Plasma Exchange and Intravenous Immunoglobulin for Obsessive- Compulsive Disorder and Tic Disorder in Childhood”. The Lancet 1999, 354(9185): 1153-58

Snyder LA, Swedo SE: Pediatric Obsessive-Compulsive Disorder JAMA 2000, 284(24): 3104-06

Swedo SE, Leonard H, Garvey MA, et. al: Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infections: Clinical Descriptions of First 50 Cases Am J Psychiatry 1998, 155(2): 264-71

Swedo, SE et al: “Identification of Children with Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infections by a Marker Associated with Rheumatic Fever Am J Psychiatry 1997; 154:110-112

Share This