On the Possible Relationship Between Anti-Streptolysin-O Titer and Neuropsychiatric Disorders Other than PANS

To The Editor:

Renewed interest in the relationship between autoimmune diseases and psychiatric disorders has been raised by a recent consensus paper describing pediatric acute-onset neuropsychiatric syndrome (PANS) (Chang et al. 2014), which also includes the previously known pediatric autoimmune neuropsychiatric disorder associated with streptococcal infections (PANDAS) (Swedo et al. 1998). Although the diagnostic process is still debated and differential diagnosis may be challenging, several features of both disorders have been described. PANDAS is characterized by an infection-mediated (by group A-β hemolytic Streptococcus) abrupt, dramatic onset of neuropsychiatric symptoms mainly including, but not limited to, tics and obsessive-compulsive behaviors, whereas PANS may be caused by several triggers, not only postinfectious ones, and may present, together with obsessive-compulsive behaviors, with the predominant feature of acute-onset food refusal, rather than tics (Murphy et al. 2014). Anti-streptolysin O (ASO) titration is supportive of the diagnosis of PANDAS or PANS, and should be performed when they are suspected (Kiessling et al. 1993; Murphy et al. 2015; Pozzi et al. 2014). In the case of isolated measurements, the diagnostic threshold has been set at 400 U/mL, whereas the normality threshold is generally 200 U/mL, although it may be adjusted by age (Chang et al. 2014). It is also known that elevated ASO titers may be linked with a plurality of conditions that are not always pathological; however, scant data are available on the possible association of group A Streptococcus and ASO titers with the occurrence of neuropsychiatric manifestations other than Sydenham's chorea and PANDAS or PANS (Pozzi et al. 2014).

Autoimmunity-based etiological hypotheses are increasingly being probed for neuropsychiatric diseases, such as autism spectrum disorders (ASD) (Keil et al. 2010). We have explored the distribution of ASO titers in a wide pediatric neuropsychiatric population, aiming to check for differences among groups of patients with different psychiatric diagnoses.

We conducted a one-time observation on a randomly assembled cohort of 203 neuropsychiatric outpatients, 11.0±3.4 years of age, collecting data from their last control visit; all neuropsychiatric outpatients with an ASO titration from the previous 4 months were included. We recorded psychiatric diagnoses (rated according to Diagnostic and Statistical Manual of Mental Disorders [DSM-IV] with structured clinical interviews) (American Psychiatric Association 1994) and ASO titers. Only patients with the following diagnoses were considered further: Attention-deficit/hyperactivity disorder (ADHD) as prevalent diagnosis (n=40, age 10.9±2.3 years, 20% females); ASD (n=28, 10.5±3.7 years, 3.6% females); psychosis spectrum disorders (n=11, 12.9±3.1 years, 36.4% females); externalizing disorders (comprising oppositional defiant disorders, conduct disorders, impulse control disorders) (n=31, 11.2±2.5 years, 12.9% females); and internalizing disorders (anxiety disorders and depressive disorders) (n=76, 11.3±3.5 years, 39.5% females). The presence of either tics or obsessive-compulsive behaviors constituted an exclusion criterion, as these are known symptoms of PANDAS/PANS (n=17). Although a formally sound analysis was beyond the purpose of this preliminary work, we controlled for possible differences in age and gender among groups, finding only one significant discrepancy (female patients in the ASD group, p=0.04).

Observed ASO levels were reported in terms of positivity/negativity with respect to the normality threshold of 200 U/mL and to the suggested diagnostic threshold of 400 U/mL. P values refer to χ² tests (ASO positivity vs. diagnosis).

With the 200 U/mL normality threshold, 77.5% patients with ADHD were found positive (p=0.07), as compared with 70.0% with externalizing disorders (p=0.15), 64.3% with ASD (p=0.86), 57.9% with internalizing disorders (p=0.21) and 33.3% with psychosis spectrum disorders (p=0.05). Regarding the 400 U/mL diagnostic threshold, we found that 50.0% patients diagnosed with ASD (p=0.10), 41% with externalizing disorders (p=0.10), 35.0% with ADHD (p=0.91), 27.6% with internalising disorders (p=0.15), and 8.3% with psychosis spectrum disorders (p=0.10) were positive.

Behavioral traits may be responsible for the differences we observed in ASO titers. The presence of externalizing disorders, as well as ADHD, may increase the frequency and intensity of the interactions with the environment and other people, hence contributing to an increased risk of streptococcal infection (which triggers the elevation of ASO titers). On the contrary, patients with internalizing disorders may have less social and environmental interactions, leading to lower ASO titers. Therefore, the attribution of possible causality in the context of ADHD and externalizing or internalizing disorders could be challenging.

In our sample, patients diagnosed with ASD displayed ASO titers above the range of normality in most of the cases and, surprisingly, half of these patients also exceeded the diagnostic threshold. The development of rheumatic fevers in parents has been previously linked to the presence of ASD in offspring (Keil et al. 2010), which may support this observation, although the role of immunity in the pathogenesis of ASD is still debated (Mead and Ashwood 2015). A novel and interesting observation concerned psychosis, for which autoimmune origins have also been put forward (Bergink et al. 2014). In this sample it was associated with a low percentage of ASO-positive patients, considering both the normality and the diagnostic thresholds. The limited sample size may have prevented the achievement of statistical significance, whereas the absence of a control group did not allow the definition of a predictive power. Furthermore, given the exploratory nature of this observation, we did not take into consideration confounding factors. Overall, we observed possible interesting differences that may serve to better direct the choice of performing ASO titration in neuropsychiatry. We hope that this letter, not aimed at defining causal relationships, stimulates systematic research on the possible association of alterations in ASO titers with novel psychiatric diagnoses, not comprised in the current spectrum of PANDAS and PANS.