1. Introduction
Post-infectious autoimmune and neuro-inflammatory events have been recognized to cause acute childhood neuropsychiatric disorders and have been designed with different terms, but all presented with similar clinical-pathogenetic manifestations. These disorders were previously indicated as paediatric infection-triggered autoimmune neuropsychiatric disorder (PITND); pediatric autoimmune neuropsychiatric disorder associated with streptococcal infection (PANDAS); paediatric acute-onset neuropsychiatric syndrome (PANS); and childhood acute neuropsychiatric syndrome (CANS) [
1,
2,
3,
4,
5].
The clinical recommendations of the PANS Consensus Conference clarified the clinical evaluation and diagnostic criteria of patients with young pediatric acute-onset neuropsychiatric syndrome (PANS) [
3].
PANS describes a clinical situation defined by the sudden and dramatic onset of obsessive-compulsive disorders or severely restricted food intake, associated with acute onset neuropsychiatric symptoms such as anxiety, emotional lability and/or depression, irritability, aggression, and/or strongly oppositional behavior behavioral regression (development), deterioration in school performance or memory impairment. On the other hand, the PANDAS subgroup is defined by an acute prepubertal onset of tics or OCD symptoms associated with GAS infection and specific neuropsychiatric symptoms. It is distinguished from PANS by a sudden onset, and episodic course and tics have an “off/on” and increasing/decreasing course. While PANDAS has a specific infectious pathogen responsible, PANS foresee different microbes possibly implicated in the genesis of the disorders postinfectious neurological such as H1N1 influenza, Epstein Barr virus, and
Borrelia burgdorferi (Lyme) disease [
4,
6,
7,
8].
The term PANS/PANDAS has been connected with a clinical condition in children and adolescents presenting with a sudden onset of various neuropsychiatric disorders, including obsessive-compulsive disorder (OCD), severely restricted food intake, anxiety, and inattention deficit hyperactivity disorder (ADHD). Therefore, diagnostic criteria have been proposed in order to allow a clear identification of individuals affected by PANS/PANDAS and consist of the onset of childhood/adolescent-related obsessive-compulsive disorder or severe restrictive eating, associated with at least two of the following neuropsychiatric disorders such as anxiety, emotional lability, and depression, irritability, aggression or strongly oppositional behavior, behavioral and developmental regression [
5]. Other disturbances may include a deterioration in school performance, sensory or motor difficulties, somatic signs or symptoms, including sleep disturbance, enuresis, or increased urinary frequency. These disturbances should not be better explained or be related to a known neurologic or medical disorder [
4,
5,
6,
7,
8,
9,
10].
The simultaneous presence of additional neuropsychiatric symptoms has been associated with similarly severe and acute onset, such as anxiety, emotional lability or depression, irritability, aggression, and severely oppositional behaviors, and sensory or motor difficulties [
11,
12]. In addition, somatic signs or symptoms may also be reported, including sleep disturbances, enuresis, or urinary frequency [
12].
At the diagnosis of PANS/PANDAS, several types of treatment have been proposed and used according to the prevalent clinical signs and the severity of the disturbances. Antibiotics (penicillin V, azithromycin), anti-inflammatory drugs (cyclooxygenase (COX) inhibitors, corticosteroids), immunomodulating treatments (intravenous immunoglobulin –IVIG. plasma exchange) are the most applied treatment singularly or in association [
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24,
25,
26,
27].
2. Treatment of Obsessive-Compulsive Disorder in PANS/PANDAS in Children
The PANDAS concept was first stated by Sweedo et al. in 1998 [1]. Among the various diagnostic hypotheses reported in the literature, streptococcal and other infections have been shown to trigger the development of symptoms such as OCD, tics and behavioral disturbances [2,3,5]. Therefore, numerous studies have been performed on neurological outcomes in PANDAS patients undergoing medical or surgical treatments, with the principle of adenotonsillar infection as the primary target [13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,32,33]. Pavone et al. examined whether adenotonsillectomy could impact both disease remission and affect the clinical course, streptococcal antibody titers, neuronal antibodies, or the clinical severity of the obsessive-compulsive disorder (OCD) [13]. Surgery did not affect symptom progression, streptococcal and neuronal antibodies or clinical severity of neuropsychiatric symptoms with comparable results for remission (17 surgical vs. 14 non-surgical; p = 0.29) and disease recurrence (39 surgical vs. 50 controls; p = 0.09) [14]. Conversely, promising results have been reported later by Demesh et al. [22]. The authors noted that nine surgically treated patients achieved clinically significant relief, including those who had no response to antibiotic therapy alone (p = 0.03). However, the sample of enrolled patients constitutes a limitation for the reported study, reducing the significance of the evidence demonstrated.
Prasad et al. attempted to compare the outcomes achieved through surgical treatment alone versus those obtained from the combined surgical approach with intravenous immunoglobulins [23]. However, the authors enrolled a small group of patients, dividing them into three different treatment arms: tonsillectomy and adenoidectomy (AT) (n = 28), AT plus intravenous immunoglobulin (IVIG) (n = 22), or non-surgical treatment (n = 10). Although caregivers did not report a decrease in symptom frequency depending on the type of treatment except choreiform movement (p = 0.0296), TA was shown to be the treatment with the greatest symptom impact for patients (p = 0.05).
Another issue of concern remains the effectiveness of surgical treatment on disease prevention by influencing the onset of neuropsychiatric symptoms. Nevertheless, when examining the literature, the only quantifiable outcome remains the change in OCD before and after treatment as reported in our systematic review, not allowing a quantitative analysis of the results and, therefore, not drawing valid conclusions.
In this regard, Murphy et al. reported in 2013 poor prevention of neuropsychiatric symptoms (e.g., OCD or tics) based on patient surgical status (p = 0.71), inferring a mismatch between tonsillectomy and course of OCD/tics or concentration of streptococcal antibodies [32].
Using the same etiopathogenetic principle, antibiotic therapy has been proposed as a treatment for PANDAS patients, boasting a dual purpose, acting both on the prevention of the disease and relapses [15,16,20,21]. Murphy et al. in 2002 reported promising medical therapy results, with a rapid resolution of the symptoms of OCD, anxiety and tics that occur after 14 days of appropriate antibiotic treatment in 6/12 patients with PANDAS [21]. The authors put forward an interesting point of view regarding the management of related symptoms. GABHS serological tests have been used as an objective evaluation of response to treatment and the eradication of the germ by antibiotics has shown efficacy in the resolution of OCD symptoms as well as any relapse after acute streptococcal infection.
In reverse, different studies instead focus on the limits of the efficacy of antibiotics in prophylaxis, affirming instead the therapeutic benefits during acute episodes. Snider et al. reported a significantly reduced number of exacerbations of neuropsychiatric symptoms in both penicillin and azithromycin-treated patients (p < 0.01 for both) [15]. Subsequently, the possible role of antibiotic therapy was hypothesized by a study comparing the severity of OCD on the Clinical Global Impression Scale (CGI-S). However, it should be noted that the authors in this case did not correctly correlate the number of exacerbation and the eradication of the germ to the reduction of specific symptoms.
In 2017 Murphy et al. comparing two different medical treatment for PANDAS patients, demonstrated significant reductions in the azithromycin group (
n = 17) matched to the placebo group (
n = 14) (
p = 0.003) [
16].
Other authors have instead evaluated the effects of reducing oropharyngeal and nasosinus inflammation after treatment with non-steroidal anti-inflammatory drugs in an attempt to improve neuropsychiatric symptoms, reporting unpromising results [
18,
19]. The lack of efficacy has probably been interpreted in the role of the eradication of the germ in the resolution of the pathology, regardless of the reduction of the episodes of inflammation.
In response to a Group A Streptococcal infection, cross-reactive antibodies have been hypothesized in the etiology and pathogenesis of PANDAS in susceptible individuals by reacting to cell wall components and neuronal proteins of the basal ganglia [
34,
35,
36,
37,
38]. For this reason, some studies have tested the possible role of intravenous immunoglobulins (IVIG) on the clinical course of the disease [
19,
24].
Although Williams et al. demonstrated that IVIG was safe and well-tolerated, differences between groups in the double-blind comparison did not demonstrate the superiority of IVIG over placebo [
19,
24]. In this regard, the reported outcomes certainly suffered from the disadvantage of small sample sizes and the lack of specific biomarkers predicting a positive response to immunotherapy.
Exploiting the same concept of higher concentrations of cross-reactive antibodies in acute serum samples, therapeutic plasmapheresis (TPE) has been proposed as an alternative treatment. However, the support for the plasmapheresis use remains limited to a few reports and controlled-placebo trials with a lessened population, demonstrating symptom improvements mainly in the short-term follow-up.
Latimer et al. proposed therapeutic plasmapheresis (TPE) to treat 35 severely ill children and adolescents with PANDAS disorders [
38]. The authors reported an average improvement of 65% at six months post-TPE and 78% at long-term follow-up. However, the sample enrolled by the authors was not homogeneous, with possible confounding variables not adequately identified.
The therapeutic effect of corticosteroids, also proposed in the treatment of PANDAS, seems to be affected by the timing of drug administration. Indeed, while early administration appears to be associated with shorter flare-up periods (
p < 0.001) [
39], longer administrations are associated with better control of neuropsychiatric symptoms (
p = 0.014) [
18]. The probable therapeutic hypothesis is the stabilization of the chronic inflammatory state induced by the streptococcal infection and the maintenance of a low antibody titre, although in the literature there are insufficient data in this regard.
At the overall pooled analysis performed for the subgroup of medical patients, the improvement in OCD was recorded in 150/263 (57.03%) patients. However, at the subsequent meta-analysis of the data by subgroup fixed effects modeling although an OR of 1.86 was found [95% CI 0.81, 4.28], the overall effect Z score 1.47 (p < 0.14) heterogeneity I2 = 0% (p = 0.41) were not significant. It should be noted that several medical approaches have been adopted without any standardized protocol, which represents a major limitation for the pooled analysis of the studies included.
3. Conclusions
Although PANS/PANDAS are disorders with well-defined diagnostic criteria, the proposed therapeutic protocols report conflicting data to date. Furthermore, the effectiveness of medical and surgical approaches on movements and behavioral disorders often does not reach significant differences and is frequently affected by the timing of administration. Finally, it is necessary to consider that the evidence in the literature does not follow a unified therapeutic protocol, leading to poor enrolled samples and unsatisfactory results.
Encyclopedia
