Symptom flares after COVID-19 infection versus vaccination among youth with PANS/PANDAS

Abstract

Background:

Flares of autoimmune conditions can happen after coronavirus disease 2019 (COVID-19) infection and after COVID-19 vaccines. Patients and clinicians confront difficult decisions about vaccine safety when considering efforts to balance the risks of disease exacerbation after vaccines versus the protection that vaccines offer to potential serious life-threatening complications of COVID-19 infection.

Objective:

To examine symptom flares after COVID-19 infection and vaccines in the case of a form of autoimmune encephalitis that primarily affects children and young adults (pediatric acute-onset neuropsychiatric syndrome [PANS] / pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections [PANDAS]).

Method:

A cross-sectional study that used a survey distributed through PANS/PANDAS groups, organizations, and clinics.

Results:

Surveys were completed by 496 parents and/or caregivers of children with PANS/PANDAS. Among the children reported to have had a COVID-19 infection (n = 178), 43% reported severe flares of PANS/PANDAS symptoms, 23% reported mild flares, and 30% reported no symptom flares. Among those who had received COVID-19 vaccines (n = 181), 65% observed no changes in PANS/PANDAS symptoms after the vaccine, 19% reported mild flares, and 15% reported severe flares. Paired sample t-tests showed that, after COVID-19 infections, children with PANS/PANDAS were significantly more likely to experience a severe symptom flare than a mild flare or no change in symptoms. In contrast, after receiving COVID-19 vaccines, children were significantly more likely to experience no change in PANS/PANDAS symptoms than to endure a mild or severe symptom flare. In addition, children who recovered from PANS/PANDAS were significantly more likely to have no symptom change versus children who were managing the condition, both after COVID-19 infection (92% versus 25%) and vaccine receipt (100% versus 66%).

Conclusion:

These results support COVID-19 vaccination for most individuals with PANS/PANDAS given that the frequencies of symptom flares and setbacks after COVID-19 infection were significantly higher than after the COVID-19 vaccine. Nonetheless, further studies are needed to fully assess the risk-benefit balance and to provide a more individualized approach to disease prevention in people with immune vulnerabilities.

Notwithstanding the undisputed safety and importance of vaccines in both clinical medicine and public health, new vaccines, such as coronavirus disease 2019 (COVID-19) vaccines, are made available with uncertainty about their effectiveness and safety,¹ with patients and providers trying to make decisions based on the rapid changes in recommendations by public health and scientific bodies.² Human genetic diversity results in variations in antibody responses among the general population³ and could explain the great variability in people’s responses to COVID-19 infections and vaccines.² For example, analyses of data bases indicate that young men and boys might have a higher risk of developing myocarditis after message RNA (mRNA) COVID-19 vaccines,^4,5 with a larger spacing between the vaccine’s first and second dose, potentially reducing that risk.⁶ With recognizing that hospitalizations, surgeries, and deaths associated with vaccines can happen, governments have created vaccine injury reporting mechanisms and compensation programs.

The safety of new vaccines might be particularly relevant for patients already prone to dysregulated immune responses,^7,8 given the molecular mimicry of the vaccine and the potential of vaccine mRNA to activate several pro-inflammatory cascades.^8,9 Although several studies conclude that COVID-19 vaccines are safe or have an “acceptable” safety profile for people with autoimmune conditions,^10–12 a global survey and a review study indicate that disease flares can be triggered by COVID-19 vaccines, especially in cases of patients with high disease activity and in cases of patients with systemic lupus erythematosus, psoriatic arthritis, or polymyalgia rheumatica.^13,14 Indeed, several studies of patients with rheumatic disease found that fear of adverse effects, relapse, or worsening of preexisting rheumatic disease contribute to COVID-19 vaccine hesitancy, although, importantly, a recent review concluded that the flare rate after severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection was notably greater than the flare rate after vaccination.^14–16 Hence, our research aims to contribute to the literature contrasting the potential worsening of symptoms due to the reactogenicity to COVID-19 infection in contrast to mRNA vaccines in individuals with dysregulated immune responses,⁸ with a focus in youth.

We created a cross-sectional study that focused on pediatric acute-onset neuropsychiatric syndrome (PANS) / pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections (PANDAS) (see description below), a particular form of autoimmune encephalitis that primarily affects children. This study aimed to address the following questions: (1) how often are COVID-19 infections followed by symptom flares in children and young adults with PANS/PANDAS and are frequencies of flares (or no flares) significant, (2) how often are COVID-19 vaccines followed by symptom flares in children and young adults with PANS/PANDAS and are frequencies of flares (or no flares) significant, and (3) are frequencies of flares different for children who are fully recovered from PANS/PANDAS versus children who have not recovered?

Short Description of PANS/PANDAS

PANDAS infection was first described by Swedo et al.¹⁷ at the National Institutes of Health in 1998 after a subgroup of children were identified as exhibiting a severe and sudden onset of obsessive-compulsive disorder and other neuropsychiatric symptoms after group A β-hemolytic streptococci infections.^17–19 Swedo²⁰ and Swedo et al.²¹ postulated that, similar to Sydenham chorea, a misdirected autoimmune response to the basal ganglia could explain the condition. In 2010, PANS was proposed as a broader diagnostic category to capture children with similar severe and abrupt symptoms that result from a variety of viral and bacterial infections as well as environmental triggers.^18,22 There already are reports of children who developed PANS after COVID-19 infections.²³

Symptoms of PANS and PANDAS include anxiety; tics; obsessive-compulsive disorder; emotional lability or depression; attention difficulties; rages and/or aggression and irritability; age-inappropriate behavior and/or regression; enuresis or urinary frequency; deteriorations in school performance related to attention-deficit/hyperactivity disorder, memory deficits, and cognitive changes; and sleep disturbances and new bedtime fears and/or rituals.^17,22,24 Although there is consistency in the acuity of the onset of the co-occurring symptoms, there is extensive variability in the severity, types of symptoms, and disease trajectory, including a relapsing-remitting, chronic static, or chronic-progressive course.^22,25 Results of studies found that antineuronal autoantibodies are elevated in patients with PANS/PANDAS when experiencing higher symptomatology,²⁶ which, along with the many neural circuits that connect the basal ganglia’s motor coordination functions with higher-order executive functioning (including cognitive, attention, and emotional brain regions),²⁷ support the pathophysiologic understanding of the condition as a form of autoimmune encephalitis. Besides anti-infectious agents and mental health treatments, immune modulatory agents are considered key in the treatment of the condition, particularly in moderate-to-severe cases.²⁵

METHODS

Study Design

For this cross-sectional study, we designed a survey that was completed online by using Qualtrics (Qualtrics XM, University of Connecticut, Storrs, CT) and was distributed through PANS/PANDAS groups, organizations, and clinics. There were two main waves of survey completion: (1) August 9, 2021, through June 7, 2022 (n = 308), and (2) February 29, 2023, through April 28, 2023 (n = 188). Wave 1 coincided with vaccine availability in most countries for ages ≥ 16 years. By the time wave 2 began, vaccines were available in most countries for all children ages ≥ 6 months. The recruitment flyer invited families to participate in a survey about experiences of families affected by PANS/PANDAS and did not indicate that there would be questions about COVID-19 infection or vaccination. The 64 questions of the survey were created based on the literature and a previous qualitative phase in which we interviewed 20 families with children diagnosed with PANS/PANDAS. The final survey included the following sections: (1) sociodemographic information, including age, sex, and geographic region; (2) clinical characteristics and diagnoses (PANS, PANDAS, psychiatric, and common comorbid conditions) at onset and over time; (3) developmental trajectories (e.g., emotional, social, cognitive, identity and/or self-concept); (4) treatments and experiences accessing care; (5) questions about COVID-19; and (6) expectations for the child’s future. The COVID-19 questions were toward the end of the survey and were completed by 88% of the 496 respondents. Of the 58 respondents who did not complete the COVID-19 questions, 56 had stopped completing the survey before the COVID-19 section. In the current analysis, we used demographic and COVID-19 data (see below). The study was approved by the UConn-Storrs Institutional Review Board within the Human Research Protection Program. Written informed consent was obtained from the participants at the beginning of the survey.

Participants

All the parents or caregivers of children with PANS/PANDAS who gave consent to the online survey were included in the sample. There was no follow-up. Given that the survey was distributed via social media, we could not report on a denominator or response rate. For the full sample (N = 496), sociodemographic data reported for the child and/or individual with PANS/PANDAS included current age (mean 12.5 years), age of PANS/PANDAS onset (mean 6.9 years), sex (56% boys, 43% girls, 1% other), and country of residence (70% United States, 11% Canada, 7% United Kingdom, 4% Australia, 3% Sweden 1% Italy, and <1% from each of the following countries: Argentina, Croatia, Denmark, France, Japan, New Zealand, Norway, Philippines, Poland, Serbia, Spain, Turkey, and United Arab Emirates.

Sociodemographic and clinical characteristics of the subsamples who reported having been infected with COVID-19 (n = 178) and who received the COVID-19 vaccine (n = 181) were similar to the full sample, although the children’s current age for the vaccinated group was 2 years older on average (Table 1). Within the vaccinated group, 19% received one dose, 62% received two doses, and 19% received three or more doses. Of the children who received one dose only, the vaccines included 28 Pfizer (Pfizer-BioNTech, New York, and Mainz, Germany), 4 Jansen (Leiden, Netherlands and Beerse, Belgium, subsidiaries of Johnson&Johnson, New Brunswick), and 3 Moderna (Cambridge). Of the children who received two doses, the vaccines included 103 Pfizer, 7 Moderna, 1 Jansen, and 1 Sinopharm, Beijing (Institute of Biological Products, Beijing, China), with an additional three children whose two doses included a combination of Pfizer, Moderna, Jansen, and Oxford/AstraZeneca (Oxford University, Oxford, England and Cambridge Biomedical Campus, Cambridge, England). Of those children who received three or more doses, 25 children received Pfizer, 3 received Moderna, and 7 received three or more doses that were a combination of Pfizer, Moderna, and Oxford/AstraZeneca. For the full sample (N = 496), 5% of the children were reported to be fully recovered, 54% were improving, 41% were not improving (acute phase, new flare, worsening, or chronic-static).

Table 1.

Demographic data for the full sample, the subsample of children who were infected with COVID-19, and the subsample of children who received the COVID-19 vaccine, including current age, age of PANS/PANDAS onset, and sex

COVID-19 = Coronavirus disease 2019; PANS = pediatric acute-onset neuropsychiatric syndrome; PANDAS = pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections; SD = standard deviation.

Variables

COVID-19 infection questions included whether the child and/or individual with PANS/PANDAS had ever been positive for COVID-19 infection and whether there had been postinfection PANS/PANDAS symptom changes.

COVID-19 vaccination questions included whether the child with PANS/PANDAS had received a COVID-19 vaccine, which vaccine, the number of doses, and whether there had been postvaccine PANS/PANDAS symptom changes. The options for reporting vaccines received used the commercial name of each vaccine given that people are generally unfamiliar with the generic names or vaccine ingredients. The options for reporting postinfection and postvaccine symptom changes included (1) it was fine, caused no change in PANS symptoms; (2) created a mild flare; (3) created a flare that set the child back (severe flare); or (4) improved PANS symptoms.

Statistical Analysis

Descriptive statistics were computed to determine the demographic characteristics of the study sample, and frequencies were calculated to examine the percentage of respondents who reported changes or no change in PANS/PANDAS symptoms after COVID-19 infection and COVID-19 vaccination. Then, the means were compared via paired sample t-tests to determine if there were significant differences in the number of caregivers who reported different changes in symptoms (no change, mild symptom flare, and severe symptom flare) after COVID-19 infection and COVID-19 vaccination. In addition, for both COVID-19 infections and vaccine receipt, a χ² test of independence was performed to test the relationship between changes in PANS/PANDAS symptoms and whether the child had recovered from PANS/PANDAS.

RESULTS

COVID-19 Infection

Based on caregivers’ reports on their children with PANS/PANDAS, 178 (36% of the sample) had been infected with SARS-CoV-2 infection. The majority were reported to have experienced a post-COVID flare in PANS/PANDAS symptoms (66%), including 23% having a mild flare and 43% having a severe flare. Thirty percent of the caregivers reported that their children had no change in PANS/PANDAS symptoms, and 5% reported an improvement in symptoms after COVID-19 infection (Table 2). Paired sample t-tests revealed significant differences between the percentages of children who experienced a severe flare versus a mild flare, t(177) = –3.326, p < 0.001, two-tailed, and the percentages of children who experienced a severe flare versus no change in PANS/PANDAS symptoms, t (177) = –2.043, p = 0.043, two-tailed. Differences between the percentages of children who experienced no change in PANS/PANDAS symptoms versus a mild flare were not significant (Table 3). In addition, the Pearson χ² test of independence showed a significant relationship between change in PANS/PANDAS symptoms after COVID-19 infection and whether a child had recovered from PANS/PANDAS, χ² (3) = 6.30, p < 0.001, Cramer V = 0.384. Children who were reported to be fully recovered from PANS/PANDAS were more likely to have no symptom change after a COVID-19 infection (92%) than children who were managing the condition (25%). Children who were currently managing PANS/PANS were more likely to have a mild (25%) or severe flare (46%) than children who were fully recovered from PANS/PANDAS (0%, 8%) (Table 4).

Table 2.

Reports of PANS/PANDAS symptoms after COVID-19 infections and vaccinations

PANS = Pediatric acute-onset neuropsychiatric syndrome; PANDAS = pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections; COVID-19 = coronavirus disease 2019.

Table 3.

Paired sample t-test results for parent-reported changes in their child’s PANS/PANDAS symptoms after COVID-19 infection (n = 178), comparing no change, mild flare, and severe flare

PANS = Pediatric acute-onset neuropsychiatric syndrome; PANDAS = pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections; COVID-19 = coronavirus disease 2019.

Table 4.

Contingency table and χ² test for the association between PANS/PANDAS symptoms after COVID-19 infections and PANS/PANDAS recovery^*

PANS = Pediatric acute-onset neuropsychiatric syndrome; PANDAS = pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections; COVID-19 = coronavirus disease 2019.

^*χ² (3) = 6.30, p < 0.001, Cramer V = 0.384.

COVID-19 Vaccination

COVID-19 vaccines were received by 36% of the sample (n = 181), with the majority reported to have experienced no change in PANS/PANDAS symptoms (65%). Post-vaccine PANS/PANDAS symptom changes were reported as 65% no changes, 19% mild flare, 15% severe flare, and 2% improvement in symptoms (Table 2). Paired sample t-tests demonstrated that there were significant differences between the percentages of children who experienced no change in symptoms versus a mild flare; t(180) = 7.788, p < 0.001, two-tailed; and the percentages of children who experienced no change in PANS/PANDAS symptoms versus a severe flare t(180) = 9.009, p < 0.001, two-tailed. Differences between the percentages of children who experienced a severe versus a mild flare were not significant (Table 5). In terms of PANS/PANDAS recovery, the Pearson χ² test of independence showed a significant relationship between the change in PANS/PANDAS symptoms after vaccine receipt and whether a child had recovered from PANS/PANDAS, χ² (3) = 8.30, p = 0.04, Cramer V = 0.214. Children who were reported to be fully recovered from PANS/PANDAS were more likely to have no symptom change after COVID-19 vaccine receipt (100%) than children who were managing the condition (62%). Children who were currently managing PANS/PANS were more likely to have a mild (20%) or severe flare (16%) than children who were fully recovered from PANS/PANDAS (0%, 0%) (Table 6).

Table 5.

Paired sample t-test results for parent-reported changes in child’s PANS/PANDAS symptoms after COVID-19 vaccination (n = 181), comparing no change, mild flare, and severe flare

PANS = Pediatric acute-onset neuropsychiatric syndrome; PANDAS = pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections; COVID-19 = coronavirus disease 2019.

Table 6.

Contingency table and χ² test for the association between PANS/PANDAS symptoms after COVID-19 vaccinations and PANS/PANDAS recovery^*

PANS = Pediatric acute-onset neuropsychiatric syndrome; PANDAS = pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections; COVID-19 = coronavirus disease 2019.

^*χ² (3) = 8.30, p = 0.04, Cramer V = 0.214.

DISCUSSION

The results of this study demonstrated significant differences in postinfection and postvaccine PANS/PANDAS symptom changes, with children most likely to have a severe flare of PANS/PANDAS symptoms after a COVID-19 infection and most likely to experience no change in symptoms after receiving the COVID-19 vaccine. However, we did find higher rates of postvaccine flares relative to other autoimmune conditions (33% in this study in comparison with rates of 2.47% to 20% for adults and 12.1% for adolescents with other autoimmune conditions),^14,15,28 which suggests that the immune vulnerability to disease exacerbation after vaccines is higher for PANS/PANDAS than for other autoimmune conditions. Nonetheless, other studies with adult populations with a range of autoimmune conditions reporting low percentage of flares, primarily mild, do also report variability in the severity of flares, and some serious postvaccine complications^{11,12, 29} These different reactions (e.g., in the present study, 19% mild flares and 15% severe flares) might be explained by the great heterogeneity of antibody responses to both COVID-19 infection and vaccines.² Alternatively, some studies that focused on autoimmune conditions reported that higher disease activity was likely to be associated with higher flare rates after COVID-19 vaccines, and controlled disease was a protective factor for flares,^14,15 and, in our study, children who recovered from PANS/PANDAS were significantly less likely to experience flares (after both COVID-19 vaccines and infections) in comparison with children who were still managing the condition.

Similar to the review by Xie et al.¹⁴ that focused on adults with rheumatic disease, which found the flare risk to be much greater after SARS-CoV-2 infection (>40% in some studies), our findings indicate that the risks of symptom flares and setbacks with SARS-CoV-2 infection are much higher than after the vaccine. The frequency of flares after a COVID-19 infection were particularly high (43% severe and 23% mild) in the present study. Even though research that fully explains the pathophysiologic mechanisms of PANS/PANDAS is still lacking, these high flare rates might be explained by heterogeneous mechanisms in which some genetic mutations may disrupt peripheral and central immune functions, neurotransmission, and/or the blood-cerebrospinal fluid/brain barriers after stressors such as infections.²⁴ Another significant factor that differs from other childhood autoimmune conditions is the heightened permeability of the blood-cerebrospinal fluid/brain barrier in PANS/PANDAS and its association with gut dysbiosis.^24,30

It should be noted that this study had several limitations. First, we were not able to calculate a response rate given that the survey was primarily distributed through postings on PANS/PANDAS organizations’ web sites and social media pages (i.e., lack of a denominator). The study was also retrospective in nature, which requires parents and caregivers to recall any changes in their child’s symptoms after COVID-19 infection or COVID-19 vaccine receipt. However, parents’ reports of severe flares after a range of immune triggers are well known in the literature.^31–33 In addition, there could be some self-selection bias in the families who opted to complete a lengthy questionnaire and/or who are actively engaging with PANS/PANDAS organizations’ web sites and social media pages. In both cases, it is likely that this would lead to overrepresentation of children severely affected; however, the survey captured a range of severity, with 5% of parents and/or caregivers reporting that their children were fully recovered, 54% reporting that their children were improving, and 41% reporting that their children were not improving and/or severely affected. Given the survey methodology used in the study, we also relied on caretakers’ reports and cannot confirm if their children had a positive COVID diagnostic test result. Although a strength of this study was that it included data from different countries, a limitation was that the sample size for most countries was not large enough to allow for cross-national comparisons. Nonetheless, the results across countries seem consistent.

Despite these limitations, we believe that our study is the first one to offer comparative data between exacerbations after SARS-CoV-2 infection versus vaccine in a population of people with an autoimmune condition. Our study supports the recommendation made by review studies¹⁴; global surveys¹³; and the COVID-19 vaccination in autoimmune disease survey protocol,³⁴ an ongoing international collaborative study that involves 29 countries and > 110 investigators, which strongly encourages vaccination in these populations, given that the benefits of vaccination outweigh the risks of adverse effects and disease flares. Future studies should investigate whether PANS/PANDAS vulnerability to symptom flares depends on the progression and severity or management of the condition, as in other autoimmune conditions¹⁴ and whether there are underlying immunophenotypes that predispose certain patients to disease flares or setbacks,¹³ particularly in light of a recent discovery of some rare genetic mutations in PANS/PANDAS.²⁴ In an age of individualized medicine, we join other researchers in arguing that more research needs to be developed to indicate which vaccine components and modifications (combination of specific vaccines antigens, antigen-generating products, and adjuvants) might offer significant protection against severe disease while reducing the risk of flares and setbacks in populations with immune vulnerabilities and specific immunophenotypes.^8,14,35

CONCLUSION

This study demonstrated that, in contrast to other autoimmune diseases, children and adolescents with PANS/PANDAS have a higher vulnerability to exaggerated inflammatory responses that result in symptom flares after COVID-19 infection and vaccines, but flares were more frequent and more severe after infections than after vaccines. Based on our research, a risk-benefit analysis supports COVID-19 vaccination, given the risks of flares and significant setbacks that the SARS-CoV-2 infection can cause in individuals with this particular autoimmune encephalitis. Nonetheless, further studies are needed to have a full assessment of the risk-benefit balance and to provide a more individualized approach to disease prevention in people with immune vulnerabilities.