Cyclic vomiting syndrome (CVS) is a disorders of gut-brain interaction (DGBI) that manifests by recurrent episodes of relentless nausea and vomiting (1). The illness is often disabling due to difficulties managing the relentless nausea, vomiting, and abdominal pain during attacks. Quality of life is in fact worse than in children with other DGBI (2). The impact on mental health in children with CVS is considerable and needs better management. Attacks are most commonly precipitated by emotional arousal (3, 4). Comorbidities of anxiety, depression, and sleep disturbances are highly prevalent and underlie the poor quality of life (5, 6). One study documented that comorbid anxiety rather than vomiting severity predicted reduced quality of life (6). Over time, recurrent episodes and failing drug trials result in progressive anxiety and depression. There is thus a need to better address the mental health comorbidities in pediatric CVS with more targeted therapies.

Despite young patient age, limited scientific evidence, and lack of regulatory approval, there is a high rate of antidepressant use in children with CVS (7). These children are at risk of polypharmacy and serious side effects including suicidal ideation (8). There is an urgent and unmet need for safer, non-pharmacological therapies. Percutaneous electrical nerve field stimulation (PENFS) has emerged as a promising, noninvasive auricular neurostimulation therapy with demonstrated efficacy and regulatory approval in children with pain-associated DGBI (9, 10). A recent study of adolescents with DGBI found positive effects on several non-gastrointestinal comorbidities including sleep and anxiety (11). A low vagal tone was documented to predict PENFS response, suggesting that poor dynamic parasympathetic control may underlie certain DGBI (12). Several small studies have documented suboptimal autonomic regulation in both children and adult CVS patients (3, 13–15). A recent study demonstrated that children with CVS have abnormal cardiac vagal regulation during their inter-episodic well phase as compared to healthy controls (16).

There are no prior studies assessing the effects of PENFS on abdominal pain and extraintestinal comorbidities in CVS. By targeting autonomic pathways via stimulation of the outer ear, auricular neurostimulation may regulate triggers and comorbidities driven by altered sympathovagal balance. This study aimed to investigate the efficacy of PENFS on pain, anxiety, sleep, and quality of life in children with CVS. We hypothesized that in children with CVS, PENFS has positive effects on pain and comorbidities and, consequently, also improves quality of life.

Thirty-two subjects were enrolled. One subject dropped out after first visit due to transportation issues and one was excluded from analyses due to incorrect diagnosis (constipation). Thirty subjects were included in the final analyses. Eighteen (60%) were female and median (IQR) age was 10.5 (8.5–15.5) years. The majority (n = 29, 97%) were Caucasian and one subject was Hispanic. Twenty-two subjects (73%) suffered from CVS attacks 1–2 times per month, while the rest had episodes as often as three times (n = 1) or four times (n = 7) per month. The median (IQR) duration of episodes was 48.0 (24.0–96.0) h. All participants had failed previous pharmacological interventions to control the CVS episodes with a median (IQR) of 2.5 (1.0–4.0) prophylactic medications failed prior to entering the trial. The median (IQR) time to extended follow-up was 4.4 (3.1–6.8) months.

The frequency of episodes/month decreased from monthly median (IQR) 2.0 (1.0–3.0) at baseline to 0.5 (0.2–1.13) at the extended follow-up (p < 0.0001).

This is the first study evaluating the effects of a non-implantable, auricular neurostimulation device on comorbidities and triggers associated with cyclic vomiting syndrome in children. This open-label, prospective trial demonstrates benefits of PENFS on pain, sleep, and several quality of life indices, including physical function, anxiety, fatigue, and pain interference. More importantly, there were long-term improvements in pain and anxiety measures at extended follow-up 4–6 months after study completion. The combination of short-term improvement in comorbidities and quality of life measures with a durable reduction in anxiety suggests that the therapy may have sustained alterations on the course of this disease.

Existing literature demonstrates a strong influence of mental health conditions in children with CVS. Anxiety is not only highly prevalent in children with CVS but also the main factor that predicts a reduced quality of life (5, 6). Anxiety and emotional arousal during positive and negative life events are the most common triggers for CVS attacks in children (3). A baseline higher autonomic reactivity and sensitivity to any disruptions of internal regulation is a postulated explanation for the rapid deterioration after an inciting trigger (22). Autonomic dysfunction has been implicated in the pathophysiology based on several smaller studies of both children and adults with CVS (3, 13–15). One small study documented reduced parasympathetic tone in response to a social stress test (3). This study further noted that majority (82%) of children with CVS met the criteria for anxiety and that acute (state) anxiety (but not trait anxiety) correlated negatively with parasympathetic tone. These findings support a link between acute stress responses and altered vagal reactivity in children with CVS. Recently published data demonstrate an underlying autonomic imbalance with inefficient cardioinhibitory vagal regulation in children with CVS during their interictal well phase (16). These findings were generated in comparison to a larger cohort of age-, gender-, and size-matched healthy controls and corrected for medication confounders. Although speculative, anxiety may thus induce an unimpeded sympathetic activation that is chronically dampened by PENFS therapy, which likely acts via stimulation of the auricular branch of the vagus nerve (12). Thus, the long-term effects on anxiety measures in this study corroborate prior data and a plausible therapeutic mechanism that targets the underlying autonomic dysregulation. PENFS may have dual mechanistic roles by (1) influencing autonomic nervous system (ANS) and thus targeting the underlying pathology of CVS (16) and secondarily reducing anxiety/worries about subsequent episodes, and/or (2) direct anti-anxiety effects via vagal neurostimulation as reported in both human and animal studies (23, 24) and thus minimizing the triggering events. It is interesting to note that CVS has been linked to altered brain network connectivity. In adults with CVS, a functional MRI study suggested increased connectivity between the salience processing network and the mid/posterior insula, a key region for processing nausea and viscerosensory stimuli (25). There was also decreased sensorimotor connectivity in both CVS and migraine patients, suggesting a common pathophysiology. This provides another potential mechanism of excessive neuronal excitability with a lower threshold for activation by triggering events. Vagal neuromodulation via PENFS may induce neural plasticity and alter brain connectivity (26), thereby dampening this hyperexcitability. In a study of adults with fibromyalgia undergoing PENFS therapy, improved pain interference and sleep correlated with altered resting state functional network connectivity (27).

There is a great demand for effective, non-pharmacological therapies for children with CVS. Recent data show that children with DGBI incur significant healthcare utilization due largely to the lack of targeted and effective therapies (28). CVS patients, in particular, incur a substantial healthcare burden due to repeated emesis attacks (29). Costs related to hospitalizations for children with acute episodes of CVS have more than doubled over the past two decades and totaled $84 million/year (30). The need for repeated high acuity medical interventions due to rapid dehydration is likely to escalate the mental health comorbidities, explaining why anxiety is the main contributor to poor quality of life and disability (6). While there are scant supportive data and no regulatory approval, children with CVS are frequently treated with antidepressants that incur the risks of serious side effects and polypharmacy (8, 31, 32). A Cochrane review found no evidence to support the use of antidepressants for children with DGBI (7). Further, these drugs have strong anticholinergic effects that exert direct, inhibitory effects on cardiac vagal signaling (33). Finding effective, nondrug alternatives is thus paramount to improving care for this disabling condition.

The strengths of our study include the prospective study design and the long-term and multidimensional assessment of disease comorbidities and quality of life indices. Yet, our study has several important limitations. The lack of a sham-controlled study design is an important limitation and placebo effects may underlie some of the improvement as subjects received weekly attention. Placebo effects may be particularly high with non-pharmacological, noninvasive interventions such as the PENFS that are often highly sought by parents of CVS sufferers after failing numerous drug therapies. Bias in the report of symptom improvement may thus be a factor. Yet, the device placements were performed by certified research nurses who are not part of the research team and were instructed to refrain from interactions with subjects. The noted long-term improvements are also less likely due to placebo effects. Assessing an episodic illness with standardized patient-reported outcome surveys designed for chronic conditions is problematic. The pain assessment (API), in particular, is a survey of pain symptoms over past week, and long-term improvement in this measure may be less relevant to an episodic illness. However, progressive pain improvement throughout and after the treatment intervention based on several measures coupled with long-term reduction in episode frequency suggests that PENFS therapy influences the disease course. Additionally, the inclusion of comprehensive and long-term assessments of several comorbidities and quality of life measures in relation to the CVS episodes lessens this confounder. State anxiety assessment could also be confounded by the assessment at single points in time. The state anxiety or PROMIS anxiety subscales did not show a clear relationship. This could be due to a small sample size and insufficiently powered study. A longer duration of follow-up is likely needed to thoroughly assess the impact on episode frequency. However, anxiety indices were improved on both PROMIS and STAI-C (state) scales, suggesting long-term improvement in this common comorbidity. Further investigations of the impact of mental health comorbidities, including antecedent trauma, would be important for a more comprehensive understanding of this patient cohort. Finally, that lack of standardization of protocols, the type of stimulation (transcutaneous vs. percutaneous), and stimulation parameters limits the generalizability of research in the field of noninvasive neuromodulation (34). Reliable biomarkers of parasympathetic activity are needed to optimize stimulation protocols and individualize therapy.

In summary, this pilot study shows promising effects of noninvasive auricular neurostimulation via PENFS in children with CVS. While short-term effects were noted for several pain and quality of life measures during and after therapy, there were important long-term improvements on several anxiety measures. As a sympathetically mediated disorder strongly governed by heightened arousal and stress states, this therapy shows promise as an emerging non-pharmacological and targeted intervention for CVS.