Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcus infections (PANDAS), describe a subset of childhood obsessive-compulsive disorders (OCD) and tic disorders triggered by group-A ß-hemolytic Streptococcus pyogenes (GABHS) infection(1). The proposed five diagnostic criteria of PANDAS are: 1) The presence of OCD or a tic disorder or both; 2) Pediatric onset; 3) Episodic course of symptom severity with abrupt onset or dramatic symptom exacerbations; 4) Temporal association with GABHS infection; 5) Association with neurological abnormalities during symptom exacerbations(2).
The proposed pathophysiological mechanism of PANDAS involves a susceptible host’s immune system, producing antibodies against GABHS which cross-react with cellular components of the basal ganglia, particularly in the caudate nucleus and putamen(2). The obsessions, compulsions, tics, and other neuropsychiatric symptoms are postulated to arise from this interaction.
PANDAS continues to be a controversial diagnosis(3). A major point of debate surrounding the PANDAS diagnosis is the difficulty in establishing a temporal association between GABHS infection and neuropsychiatric symptoms. GABHS infections occur commonly in children and the association with clinical exacerbations could therefore be due to chance(4). Moreover, exacerbations following GABHS may be attributed to a non-specific stress response rather than a specific disease mechanism. Furthermore, there is a lack of cohort studies to confirm a definitive temporal relationship between antecedent GABHS infection and PANDAS symptom exacerbations.
Evidence supporting a temporal relationship points to two case-control studies, which found that patients with OCD, Tourette’s syndrome (TS) or tic disorders were more likely than controls to have had prior streptococcal infections in the past 3 months(5) or past year(6). There is abundant evidence that argues against a relationship between GABHS and PANDAS symptoms. Two prospective blinded cohort studies examined subjects following GABHS infection and found no differences in exacerbations between PANDAS and non-PANDAS patients following a GABHS infection(4,7). Another case-control study compared the rate of possible streptococcal infections in patients with OCD, TS and tics with matched controls and did not find an association between strep infections and neuropsychiatric syndromes(8). A larger prospective cohort study that followed 814 children seen for sore throat or well-child care, half with a symptomatic GAS infection and treated with antibiotics, and the other half divided between presumed viral throat infection or well-child found no differences between the three groups in the development of new PANDAS symptoms. Of note, ill children (GAS and presumed viral) had more PANDAS symptoms at baseline(9).
There has been data found to support that there are circulating antineuronal antibodies related to PANDAS. One study found that sera from patients with PANDAS contained antibodies that induced calcium-calmodulin-dependent protein (CaM) kinase II activity(10). Another study found that children with PANDAS may have a specific D8/17 B cell marker which is generally associated with rheumatic fever(11). Case reports have also shown improvement in symptoms through the treatment of antibiotics(12), and immune-modulating therapies and plasma exchange(13) in the treatment of PANDAS. Despite this, opposing data has also been found that compared serum antibodies between patients with PANDAS, Tourette’s syndrome, and age-matched controls and found no differences, challenging the hypothesis that PANDAS develops secondary to anti-neuronal antibodies(14).
PANDAS diagnosis is further complicated by the potential inaccuracy of cultures and serum collections in detecting recent strep infection(4). One study revealed that in some cases, GABHS throat cultures remained positive and anti-streptolysin O and anti-DNase B titers remained elevated for the entire 2-year course of the study. To enhance the accuracy of infection definition, the authors suggested collecting sequential samples and correlating titer increases with GAS infection(15). Leckman et al. suggested that throat cultures are unnecessary in patients with PANDAS in the absence of pharyngitis(7). Leckman et. al did find that patients with PANDAS were more likely to have a positive history of rheumatic fever and concluded that PANDAS patients may represent a subgroup of patients with TS or OCD that are vulnerable to GABHS.
Alternative terms for PANDAS have been created to encompass a wider presentation of symptoms. In 2012, the term pediatric acute-onset neuropsychiatric syndrome (PANS) was formulated by Swedo et. al, which was defined as acute OCD and/or acute-onset eating disorder associated with at least two cognitive, behavioral, or affective symptoms such as anxiety, depression, or irritability(16). Childhood acute neuropsychiatric symptoms (CANS), was also proposed by Singer to incorporate patients with PANDAS symptoms without a GABHS infection(17) but also included other infectious agents as causal agents such as Borrelia Burgdorferi, Herpes Simplex Virus, Varicella-zoster(18), HIV, Mycoplasma pneumonia(19), and the common cold(20). This diagnosis does not limit symptom onset to a specific age range, because prepubertal onset is not specific; tic disorders and OCD develop in many children within this age range. This diagnosis also does not require a recurrence of symptoms but does require a dramatic onset. Similar to CANS, this diagnosis proposes that PANDAS should include a broader set of neuropsychiatric symptoms such as general anxiety, phobias, regression, and emotional lability.
Treatments
Primary treatments for OCD include cognitive behavioral therapy and selective serotonin reuptake inhibitors (SSRIs)(21). Suggested supplemental treatments when suspecting PANDAS include tonsillectomy and antibiotics. Immune-based therapies such as intravenous immunoglobulin (IVIG) and total plasma exchange (TPE) are used as well. Overall, systemic reviews for the treatment of PANDAS have found that studies performed had a high risk of bias with inconclusive results(22).
Studies that have examined the effect of antibiotics on PANDAS have been inconclusive. One crossover study with 37 children showed that the prophylactic use of penicillin V did not significantly affect OCD and tic symptoms or prevent recurrent infections when compared to placebo(23). On the other hand, a later study examined both penicillin and azithromycin prophylaxis and found that both decreased the number of neuropsychiatric exacerbations as well as recurrent streptococcal infections(24). More recently, a randomized controlled trial showed that patients who received azithromycin had greater reductions in OCD severity when compared to the placebo group(25). More recently, Leckman et. al found that the use of antibiotics following PANDAS diagnosis did not influence subsequent changes in tic or obsessive-compulsive symptom severity and therefore recommended against antibiotic prophylaxis(7).
Research examining immune-based therapies is limited and includes mostly retrospective case studies. One retrospective case series published in 2015 examined 12 children treated with IVIG for their PANDAS symptoms, who all subsequently had improvements in their PANDAS symptomatology(26). On the contrary, a 2017 randomized controlled trial examining IVIG in children with PANDAS showed that IVIG was not superior to placebo for PANDAS symptom reduction(27).
Surgical intervention by tonsillectomy has been explored as a treatment for PANDAS as well. One observational study compared patients with PANDAS who had tonsillectomy performed to non-surgical controls, and showed that surgery did not affect OCD or tic-disorder symptomatology, progression, streptococcal and neuronal antibodies, or the severity of neuropsychiatric symptoms(28). The existing evidence for treatments for PANDAS is inconsistent and limited, highlighting the need for further research.