PANDAS: Nonexistent or Simply Rare?

June 1, 2005

At first glance, PANDAS (Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infection) has little in common with the cuddly bear that roams the bamboo forests of southwest China. But, in fact, they share 2 important features: both are rare and both are threatened with extinction.

At first glance, PANDAS (Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infection) has little in common with the cuddly bear that roams the bamboo forests of southwest China. But, in fact, they share 2 important features: both are rare and both are threatened with extinction. A handful of ongoing PANDAS studies challenge its survival, and the outcomes of those studies will have an impact on how neurologists and other physicians evaluate and treat children with obsessive-compulsive symptoms and tics.

The idea that streptococcal infections can precipitate neuropsychiatric disorders in children, including obsessive-compulsive symptoms and tics, has surfaced intermittently for more than 100 years. It was not until 1998 that this notion was fully articulated. In that year, Susan E. Swedo, MD, and her colleagues1 published their findings on 50 such cases, provided criteria for making the diagnosis, and gave the syndrome the name and acronym that has stuck.

Swedo was a fellow at the Child Psychiatry Branch of the National Institute of Mental Health (NIMH) when she started investigating the relationship between streptococcal infections and obsessive-compulsive disorder (OCD) in children. Still at NIMH but now directing the Division of Pediatric Translational Research and Treatment Development and doing part-time research, Swedo explained that she began the studies that led to the identification of PANDAS because her mentor at the NIMH, Judith Rapaport, MD, was investigating the brain pathology underlying OCD. Swedo found a compelling association between Sydenham chorea, the only neurologic disorder known to be associated with streptococcal infection, and OCD symptoms.

As far back as the late 19th century, William Osler (1849-1919) noted "a certain perseverativeness of behavior" in children with Sydenham chorea.1 Others made similar observations. Swedo and colleagues found a high prevalence of OCD symptoms. Swedo also carried out longitudinal studies of children with OCD. She found that a subgroup of such children had a pattern of abrupt onset of OCD, an episodic course, and exacerbations that often preceded group A b-hemolytic streptococcal (GABHS) infections.


Swedo and colleagues suggested that these children represent a unique subgroup defined by: (1) OCD and/or a tic disorder; (2) onset between age 3 and the beginning of puberty; (3) episodic course characterized by abrupt onset of symptoms or dramatic symptom exacerbations; (4) temporal association with GABHS infection; and (5) neurologic abnormalities (eg, adventitious movements) during symptom exacerbations. They postulated that as in Sydenham chorea, GABHS infections, through a process of molecular mimicry, trigger in susceptible children an autoimmune response targeted to neurons. The presence in some children with PANDAS of the same antineuronal antibodies found in Sydenham chorea provided support for this hypothesis. Swedo points out that children with PANDAS, unlike those with Sydenham chorea, don't get frank chorea or the cardiac and other manifestations of rheumatic fever. She suggests that the causative agent may be the same in both conditions but that the "dose" is lower in PANDAS.

The PANDAS concept captivated pediatricians and child psychiatrists. It provided an explanation for a condition that heretofore seemed to arise out of nowhere; the proposed etiology and pathophysiology made sense, not only on its own merits but also because it followed the established and widely accepted model of Sydenham chorea. Most compelling of all, it promised a simple, effective, and safe treatment.

Tics can be suppressed, but the drugs commonly used to treat them (antipsychotics and others) are beset with side effects.2 Also, the mainstays of OCD treatment, selective serotonin reupake inhibitor antidepressants and cognitive behavioral therapy, bring meaningful improvement to far fewer than half of those treated. The idea that these conditions would clear up with a course of antibiotics had tremendous appeal not only to the clinicians struggling to help these kids but, not surprisingly, to their parents as well.

Despite the paucity of data in support of it and the uncertainty about its prevalence, the PANDAS concept took hold. Current pediatric and psychiatric textbooks refer to it as an established, albeit possibly rare, phenomenon. Although a rash of studies call the concept into question, and although many clinicians have never seen PANDAS, few are ready to abandon the idea of it.

Joseph Friedman, MD, chief of neurology at Memorial Hospital of Rhode Island, doesn't treat kids and has never seen PANDAS, but as is typical of many in the field, he has a generous approach to the concept. It hasn't been proved, Friedman said, and it's probably overdiagnosed, but PANDAS makes sense: "I'm willing to believe it exists."

The concept of PANDAS has not been confined to medical dialogue. In January 2004, USA Today carried an article on OCD in children. PANDAS, according to this article, accounts for 1 in 10 cases of OCD in children, and if treated promptly with antibiotics, "symptoms can in many cases disappear in days."3


If only this claim were so. The clinical value of PANDAS rests on the promise of antibiotic treatment, and here the results of controlled trials have been, at best, inconclusive. Penicillin prophylaxis in one controlled study didn't prevent exacerbations of tics and OCD, but it didn't prevent streptococcal pharyngitis either.4 Swedo and her colleagues have recently completed a penicillin prophylaxis study, as yet unpublished, which, she said, shows that the antibiotic does relieve PANDAS symptoms. For now, the most compelling case for the value of antibiotic treatment comes from an uncontrolled study of 12 kids who met PANDAS criteria and improved with antibiotics,5 and the testimonials of clinicians and investigators who have seen tics and OCD symptoms disappear in individual children treated with antibiotics. But clinical observations of this sort, convincing as they might appear, mislead at least as often as they point to useful information. Clinicians who have given antibiotics to children who meet PANDAS criteria have not been uniformly impressed.

Henry Sachs, MD, a child psychiatrist at Bradley Hospital in East Providence, RI, believes that PANDAS exists and when he sees a child with acute onset of tics or OCD, he gets a throat culture and checks antistreptolysin O (ASO) and anti-DNase B titers. He has seen elevated ASO and anti-DNase titers in association with exacerbations of tics but points out that since he usually doesn't have a baseline, he doesn't know whether these elevations represent a recent increase in titers and that it's the change in titers that reliably indicates recent infection.

Nevertheless, when these lab tests raise the possibility of recent or current streptococcal infection, he prescribes antibiotics. He admitted, though, "I haven't seen much benefit from them." Despite his skepticism, Sachs, like many clinicians, is reluctant to abandon the promise of antibiotic treatment. He said that the next time he comes across a child with an acute onset of tics or OCD and signs of streptococcal infection, he will again treat with antibiotics but "without a great expectation that it will help."

Neurologists have, in general, been more critical of the PANDAS concept than pediatricians and child psychiatrists. Part of the reason may be that neurologists see more children with tics than with OCD. Typical OCD has a gradual onset and a stable course and is quite resistant to treatment. An acute onset or exacerbation of OCD, as described in kids with PANDAS, is different from the usual OCD course. But the course of PANDAS tics is not all that different from that of ordinary tics. As Roger Kurlan, MD, a neurologist at the University of Rochester School of Medicine, points out, tics have a waxing and waning course and get worse with any sort of stress, including that of illness. That tics worsen in the context of streptococcal infection doesn't necessarily mean that the pathogen has a specific etiologic role.


Although Swedo and colleagues focused in their own research on the relationship between streptococcal infection and OCD, they included tics in PANDAS both because tics often occur with OCD and also because tics have long been noted to worsen after infections. Tics were first described as "a postinfectious sequela in the early half of the last century," Swedo said, and then in the 1980s, Louise Kiessling, MD, "made the observation that tics were increased during strep season among children in her ADHD [attention-deficit hyperactivity disorder] clinic" at Memorial Hospital of Rhode Island. Swedo and others have suggested, in fact, that a broad range of neuropsychiatric symptoms in children, including attentional problems and anxiety symptoms, might also be streptococci-related and ultimately included in PANDAS.

That the boundaries of PANDAS are not well defined troubles Kurlan. He believes that the criteria for the syndrome are not sufficiently delineated or justified. In addition to the uncertainty about which neuropsychiatric conditions should be included in PANDAS, he points to the lack of clarity regarding the temporal relationship between GABHS infection and symptom onset and the difficulty in ascertaining acute onset or exacerbation, particularly with respect to tics. Kurlan worries that the PANDAS concept has been prematurely accepted, particularly by pediatricians, who are under pressure from parents to do something for their kids and who are likely to order throat cultures and titers and prescribe antibiotics.

In part because so little data and, as yet, no definitive studies exist about PANDAS, almost everything about the concept is a matter of controversy. Critics and advocates of PANDAS draw different conclusions from the same data. The 1999 Lancet paper describing the results of immunomodulatory therapy is a case in point.6 Twenty-nine children with PANDAS were treated with either plasmapheresis or intravenous immunoglobulin (IVIG). The IVIG, but not the plasmapheresis, was placebo-controlled.

The children showed dramatic improvement in tic and OCD symptoms. According to Swedo, who was one of the study investigators, improvement was far greater than that achieved with standard treatments.

To advocates of the PANDAS concept, this study provided important proof of the autoimmune pathophysiology. But Kurlan and other skeptics call attention to flaws in the study's design, including the highly selected patient population, concurrent use of psychotropic drugs, and the limited controls. They think that this study may prove little more than that the symptoms in these kids fluctuate over time and are highly placebo-responsive.

A recent study by Eliana Perrin, MD, and her colleagues7 seemed to deal a near-fatal blow to PANDAS as a diagnosis. Perrin is an assistant professor of pediatrics at the University of North Carolina, Chapel Hill, School of Medicine. They observed 814 children in a large pediatric practice. Half developed GABHS infections, a quarter developed viral illnesses, and a quarter remained well. No children developed full-blown PANDAS, and the children with streptococcal infections were no more likely to develop OCD symptoms or tics than those who remained well or had presumed viral illnesses. Swedo, one of the report's authors, is undeterred by these results. She explained that PANDAS is rare--she estimates that it accounts for fewer than 5% of children with OCD and tics--and she thinks that a larger sample may be required to find it.

That PANDAS is rare does seem to be one point on which both researchers and clinicians agree. In one pediatric practice, among 4000 children with streptococcal infection seen over a 3-year period, only 12 had PANDAS.6 Alison Days, MD, a pediatrician and medical director of Texas Tech's Child Wellness Center in El Paso, learned about PANDAS during her pediatrics residency. "My colleagues have seen 2 kids who may have had PANDAS, she said, but I haven't seen any, and I see loads of kids with strep. I guess I believe in it but it's so rare I don't think of it very often."

Several large-scale prospective studies of PANDAS funded by the NIH are now under way (Kurlan is running one of them). They should help resolve the controversy both about whether PANDAS is real--whether a subgroup of kids with OCD and tics have their symptoms precipitated by GABHS--and whether prophylactic penicillin treatment of these children can prevent relapses.

The clinical stakes are high. If PANDAS proves as resilient as its bamboo-chewing namesake, clinicians will have a firm rationale for checking throat cultures and titers in children with abrupt OCD and tic onset. More important, a subgroup of children who suffer with these symptoms will have the promise of rapid symptom relief with antibiotics. Conversely, if these studies do not support PANDAS, clinicians will have little justification for diagnostic tests for streptococcal infection.

Inconclusive as the existing data are, a PANDAS "signal" does come through. Ongoing prospective studies may show PANDAS to be so rare that it is clinically insignificant. However rare it might be, says Swedo, PANDAS provides a window into the "neurocircuitry of OCD and tics," and, she continued, "we never thought of it as anything else." *


1. Swedo SE, Leonard HL, Garvey M, et al. Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections: clinical description of the first 50 cases. Am J Psychiatry. 1998;155:264-271.

2. Bruun RD. Gilles de la Tourette Syndrome. In: Rakel RE, Bope ET, eds. Conn's Current Therapy. Philadelphia: Elsevier; 2005.

3. Manning A. Obsessive-compulsive disorder; early intervention helps kids who need treatment before rituals are ingrained. USA Today. January 15, 2004;D10.

4. Garvey MA, Perlmutter SJ, Allen AJ, et al. A pilot study of penicillin prophylaxis for neuropsychiatric exacerbations triggered by streptococcal infections. Biol Psychiatry. 1999;45:1564-1571.

5. Murphy ML, Pichichero ME. Prospective identification and treatment of children with pediatric autoimmune neuropsychiatric disorder associated with group A streptococcal infection (PANDAS). Arch Pediatr Adolesc Med. 2002;156:356-361.

6. Perlmutter SJ, Leitman SF, Garvey MA, et al. Therapeutic plasma exchange and intravenous immunoglobulin for obsessive-compulsive disorder and tic disorders in childhood. Lancet. 1999;354:1153-1158.

7. Perrin EM, Murphy ML, Casey JR, et al. Does group A b-hemolytic streptococcal infection increase risk for behavioral and neuropsychiatric symptoms in children. Arch Pediatr Adolesc Med. 2004;158:848-856.