Introduction
The diagnosis of dysautonomia as a clinical entity is questioned in modern medicine, and scientific research on this subject has been neglected.1 However, dysautonomia may present with a number of frequent seemingly unrelated symptoms, including syncope, dizziness, gastrointestinal dysmotility, headaches, irritable bowel syndrome, chronic fatigue syndrome, fibromyalgia, Raynaud’s phenomena, and palpitations. There is a close relationship between dysautonomia and emotional diseases and eating disorders likely due to neuronal pathways that connect the limbic system to the autonomic nervous system. Julian Thayer first introduced a model of neurovisceral integration in emotion regulation and dysregulation2 with a clear historical reference to Claude Bernard.3 This model is focused on vagal regulation derived from measurement of heart rate variability (HRV) as a marker of health risks4 and prognosis related to cardiovascular disease.5 However, other investigators argue that heart rate alone is sufficient to explain the impact of the autonomic nervous system on life expectancy and the cardiovascular system,6 explaining the use of therapeutic approaches like beta blocker therapy in heart failure.
We investigated the impact of macro- and micro nutrition on heart rate for a more complete understanding of dysautonomia in children with anorexia nervosa,7 obesity,8 and attention deficit disorder9 to understand heart-brain interactions in children with psychosomatic diseases. Our data clearly indicate a bidirectional interaction between heart rate and emotional regulation as a window of opportunity for new therapeutic approaches. We observed a further increase in cases of dysautonomia in children and adults during the corona virus disease 19 (COVID-19) pandemic that may be related to the impact of SARS-CoV-2 infections on the autonomic nervous system.10 As such, dysautonomia has been linked to autoimmunity11 in the context of SARS-CoV-2 infections in adults12,13 and children.10 In addition, new diseases such as long COVID are overwhelming our health system.
In the current study, we conducted a retrospective analysis of clinical routine data and found that pharmacotherapy with low dose propranolol or ivabradine and supplementation with omega-3-fatty acids (O3-FA) impacted heart rate in children with dysautonomia. We also present measurable success of our therapeutic approaches in the treatment of dysautonomia within the last 5 years, which could be beneficial for the treatment of COVID-19 patients.
Methods
This work follows the Standards for Reporting Diagnostic Accuracy Studies (STARD) guidelines (Supplementary File 1). The study protocol was reviewed and approved by the Institutional Review Board of Landesärztekammer Baden Württemberg F-2012-0056. The authors are accountable for all aspects of the work and ensure that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Declaration of Helsinki (as revised in 2013). Written informed consent was obtained from all patients.
Patients
We retrospectively analyzed the data of 181 consecutive children who had an active standing test within the last five years at the pediatric department of Caritas Hospital in Bad Mergentheim and at the author’s private practice in Forchtenberg. Of the 181 children, 131 had an additional 24-h Holter echocardiogram (ECG). The children suffered from dysautonomia due to postural orthostatic tachycardia (POTS) and an increased average heart rate of more than 35 beats per minute (bpm) while standing (n = 48), an inappropriate sinus tachycardia (IST) with a mean 24-h heart rate ≥ 95 bpm (n = 74), or vasovagal syncope (VVS) (n = 12). The data were stored in a database generated by a software system (HRV Scanner™, BioSign GmbH; Germany). Children with heart disease (congenital heart defects, arrhythmias, or heart failure) were excluded from the study.
G-protein coupled receptor (GPCR) autoantibody measurements
Whole blood samples from nine subjects were allowed to clot at room temperature and then centrifuged at 2,000 g for 15 m in a refrigerated centrifuge. Serum was purified and stored at −35°C. The anti-alpha and beta adrenergic receptors (α1, α2, β1, β2), anti-muscarinic receptors (M1-M5), anti-endothelin receptor type A, and anti-angiotensin II type 1 receptor autoantibodies were measured in serum samples using a sandwich enzyme linked immunosorbent assay (ELISA) kit (CellTrend GmbH; Luckenwalde, Germany). The microtiter 96-well polystyrene plates were coated with GPCR antibodies. To maintain the conformational epitopes of the receptor, 1 mM calcium chloride was added to all buffers. Duplicate samples of a 1:100 serum dilution were incubated at 4°C for 2 h. After washing, plates were incubated for 60 m with a 1:20,000 dilution of horseradish-peroxidase–labeled goat anti-human IgG used for detection. To obtain a standard curve, plates were incubated with test serum from an anti-GPCR autoantibody positive index patient. The ELISAs were validated according to the Federal Drug Administration’s Guidance for industry: Bioanalytical method validation.
HRV
For short HRV analysis of two 5 m intervals while lying and standing during the active standing test, we used the HRV Scanner™ (BioSign GmbH; Germany). The method was validated in a large group of children, with a limit value for a normal increase in heart rate (≤ 35 bpm) as recently published.14 For 24-h HRV analysis, we used a 12-bit digital ECG recorder at 1,024 scans/s (Reynolds Pathfinder II, Spacelabs; Germany). Measurement and interpretation of HRV parameters in the current sample were standardized according to the Task Force Guidelines.15 The following time domain parameters are included in the analysis: 1) average heart rate [bpm], 2) standard deviation of NN (SDNN [ms]) to reflect global HRV, 3) percent of NN intervals that differ more than 50/20 ms from the prior interval (pNN50, pNN20 [%]), and root mean square of differences between successive NN intervals (rMSSD [ms]). rMSSD, pNN50, and pNN20 reflect the parasympathetic influence. The stress index reacts sensitively to shifts in the vegetative balance between the sympathetic and parasympathetic nerves and is calculated as follows:
Stress index=Amo2×Mo×MxDMn
Mo, Modal value, most common value of the RR intervals; Amo, number of RR intervals corresponding to the mode as a percentage of the total number of all readings; MxDMn, variability width, difference between the maximum and minimum RR intervals.For frequency domain analysis using the Fourier transformation, the HRV signals were divided into three frequency bands: 1) very low frequency power (VLF = 0,00–0,04 Hz, [ms2]), 2) low-frequency power (LF = 0,04–0,15 Hz, [ms2]), and 3) high-frequency power (HF = 0,15–0,4Hz, [ms2]) that represent respiratory sinus arrhythmia, mediated by alternating levels of parasympathetic tone. Total power (TP) (ms2) measures the total variance in HRV.
Pharmacotherapy and nutritional supplementation
Many patients received pharmacotherapy (e.g., psychostimulants, antidepressants) with published effects on HRV. For the current analysis, we investigated the effect of a new pharmacotherapy with low dose propranolol, ivabradine, and midodrine in children with POTS in the active standing test: propranolol 10–10–0 up to 20–20–0 mg (n = 15), ivabradine 5–5–0 mg (n = 11), and midodrine 2.5–2.5–0 mg (n = 6). This therapy was based upon a Consensus Statement of the Heart Rhythm Society published in 2015.16 Patients treated with fludrocortisone (n = 2), mestinon (n = 1), and saline infusions (n = 2) were excluded from statistical analysis.
We further investigated the impact of O3-FA supplementation (n = 7) ± nutritional refeeding in children with anorexia nervosa (n = 6) on heart rate increase in the active standing test.17 As recently published, we introduced O3-FA supplementation in children with sinus tachycardia after showing a significant reduction of the mean heart rate in 24-h ECG18 in accordance with a recent meta-analysis.19 Patients usually purchased products based upon 1–2 g fish oil per day from a retail store. The following dose recommendations were provided: children up to 8 years of age should receive at least 400 mg eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) as a suspension per day. Children who were able to swallow capsules should receive at least 800 mg EPA and DHA per day.
In the first visit, children with POTS were provided lifestyle advice, including increased fluid and salt intake, low dose exercise, O3-FA supplementation, and avoiding known triggers like prolonged standing. If lifestyle intervention was not successful, we introduced pharmacotherapy with either low dose propranolol, ivabradine, or midodrine.
Statistics
All analyses were performed using IBM SPSS Statistics software, (IBM Corp. IBM SPSS Statistics for Windows, Version 26.0; Armonk, NY, USA). Data are expressed as mean ± standard deviation. The HRV parameters of the 5-m segments in the lying and supine positions during the active standing test in the three groups with dysautonomia were compared to the healthy control group in an unpaired t-test. The treatment data before and after pharmacotherapy and O3-FA supplementation were analyzed with a paired t-test of each patient. The following p-values indicated statistical significance: *p < 0.05; **p < 0.01; ***p < 0.001.
Results
Compared to healthy controls anthropometric data showed that children with POTS were significantly taller (Table 1), and children with IST tended to have higher body mass indices (BMIs) consistent with recently published findings.18 Children with VVS were younger compared to the other groups, and we did not analyze the anthropometric data in this group. Blood pressures were in the normal range in all groups.
Table 1Anthropometric data of the patient groups
| Healthy control | Postural orthostatic tachycardia | Vagovasal syncope | Inapproriate sinus tachycardia |
---|
Patients | 47 | 48 | 12# | 74 |
Age [Years] | 14.2 ± 3.8 | 14.9 ± 2.1 | 9.9 ± .4.4 | 13.2 ± 2.2 |
Height [cm] | 160.1 ± 14.2 | 167.5 ± 10.6** | 142.9 ± 30.4 | 155.0 ± 12.3 |
Weight [kg] | 52.6 ± 14.3 | 54.0 ± 12.4 | 44.8 ± 26.1 | 58.5 ± 24.3 |
BMI [kg/sqm] | 20.1 ± 3.0 | 19.1 ± 3.5 | 20.0 ± 6.3 | 24.0 ± 8.6 |
SBP [mmHg] | 114.5 ± 9.2 | 117.2 ± 9.3 | 117.3 ± 19 | 122.0 ± 16.0 |
DBP [mmHg] | 61.7 ± 11.2 | 64.9 ± 12.2 | 58.9 ± 10.5 | 65.1 ± 11.1 |
Based on the definition, 24-h heart rates were significantly elevated in children with IST (99.9 ± 5.0 bpm) compared to healthy controls (80.8 ± 13.5 bpm) but not in the other two groups with dysautonomia. Based on the second definition, heart rate increases in the active standing test were significantly enhanced in children with POTS (43.5 ± 8.6 bpm) compared to healthy controls (16.2 ± 7.1) but not in the other two groups with dysautonomia. In children with POTS heart rate and HRV in the lying position were normal, but HRV suddenly decreased and heart rate increased while standing. These data indicated that a sudden postural collapse of vagus activity was induced by the significant decrease of rMSSD (from 67.1 ± 42.6 ms to 15.7 ± 8.1 ms***), pNN50 (from 38.7 ± 24.7 to 2.8 ± 7.1%***), and HF (from 2,229 ± 4,238 to 111 ± 115 ms2***), which represents the vagal tone. When LF showed no significant changes, the LF to HF ratio, which represents the sympatho-vagal balance, significantly increased to 7.4 ± 8.4 compared to 2.5 ± 1.9 in healthy children. In contrast, we found a global decrease in HRV and increase in heart rate in the lying and standing position in children with IST that may explain global dysautonomia in the 24-h Holter ECG (Table 2). The small group of 12 children with VVS did not show a clear pattern of dysautonomia and most values ranged between the two other groups.
Table 2Heart rate and HRV during active standing test and 24-h Holter ECG
| Healthy control | Postural orthostatic tachycardia | Vasovagal syncope | Inappropriate sinus tachycardia |
---|
Patients | 47 | 48 | 12 | 74 |
24-hours Holter ECG |
24h mean HR | 80.8 ± 13.5 | 78.5 ± 9.9 | 84.2 ± 11.1 | 99.9 ± 5.0*** |
Night HR | 64.8 ± 11.4 | 60.0 ± 6.9 | 70.3 ± 8.2 | 87.2 ± 10.3*** |
Min HR | 51.7 ± 8.8 | 50.9 ± 7.1 | 58.0 ± 7.1 | 69.2 ± 8.7*** |
Day HR | 84.5 ± 13.7 | 86.5 ± 11.0 | 98.7 ± 19.3 | 107.8 ± 7.0*** |
Max HR | 158.4 ± 29.3 | 164.4 ± 21.0 | 151.2 ± 29.8 | 167.6 ± 20.6** |
SDNN | 164.7 ± 45.4 | 184.5 ± 42.2 | 138.3 ± 34.4 | 105.4 ± 31.0*** |
rMSSD | 48.8 ± 15.8 | 47.2 ± 19.7 | 38.4 ± 8.5 | 22.5 ± 7.4*** |
Active standing test |
HR Increase | 16.2 ± 7.1 | 43.5 ± 8.6*** | 17.7 ± 15.1 | 15.3 ± 6.9 |
Lying HR | 73.6 ± 12.5 | 72.0 ± 11.1 | 78.8 ± 10.4 | 96.0 ± 5.3*** |
Standing HR | 89.8 ± 13.2 | 115.3 ± 15.4*** | 96.5 ± 16.3 | 110.8 ± 7.4*** |
rMSSD Lying | 85.1 ± 56.2 | 67.1 ± 42.6 | 39.6 ± 19.7** | 40.4 ± 22.7*** |
rMSSD Standing | 40.4 ± 22.7 | 15.7 ± 8.1*** | 36.6 ± 43.8 | 33.6 ± 33.5 |
pNN50 Lying | 43.3 ± 22.8 | 38.7 ± 24.7 | 20.0 ± 17.2*** | 7.2 ± 9.6*** |
pNN50 Standing | 15.7 ± 14.3 | 2.8 ± 7.1*** | 10.9 ± 15.2 | 5.3 ± 6.1 |
Stress Index Lying | 97.8 ± 85.1 | 113.3 ± 106.8 | 177.3 ± 137.6 | 288.2 ± 162.3*** |
Stress Index Standing | 168.5 ± 116.5 | 568.0 ± 438.2*** | 214.4 ± 171.7 | 358.5 ± 177.1*** |
HF Power Lying | 2,920 ± 4,403 | 2,229 ± 4,238 | 794 ± 1,017 | 470 ± 735* |
HF Power Standing | 949 ± 1,222 | 111 ± 115*** | 850 ± 2,080 | 394 ± 469* |
LF Power Lying | 1,518 ± 2,795 | 1,381 ± 1,705 | 757 ± 566 | 471 ± 329 |
LF Power Standing | 1,331 ± 1,115 | 558 ± 628*** | 1,637 ± 2,148 | 970 ± 1,091 |
VLF Power Lying | 1,553 ± 2,182 | 1,123 ± 1,820 | 689 ± 443 | 559 ± 565* |
VLF Power Standing | 1,299 ± 1,506 | 451 ± 432*** | 4,355 ± 6,570** | 606 ± 554* |
Total Power Lying | 5,819 ± 6,203 | 4,718 ± 6,273 | 2,240 ± 1,749* | 1,501 ± 1,475** |
Total Power Standing | 3,579 ± 3,012 | 1,116 ± 967*** | 6,842 ± 9,660* | 1,970 ± 1,866* |
LF/HF Lying | 0.97 ± 1.10 | 1.65 ± 3.68 | 1.82 ± 1.48* | 1.93 ± 1.46** |
LF/HF Standing | 2.54 ± 1.95 | 7.40 ± 8.42*** | 4.18 ± 2.53* | 3.84 ± 2.70* |
Figure 1 demonstrates the effect of pharmacotherapy, O3-FA supplementation, and refeeding in children with POTS. We found significant effects on postural heart rate increases after pharmacotherapy with beta blockers and ivabradine but also after O3-FA supplementation and nutritional refeeding in six patients with anorexia nervosa. In contrast to ivabradine, we found a significant increase in the decreased vagus parameter RMSSD (from 11.9 ± 2.7 ms to 35.1 ± 9.7 ms*) in the standing position that was accompanied by a significant decrease in elevated stress index (from 947 ± 171 to 443 ± 103**) in patients who received a low dose beta blocker (propranolol 10–10–0 or 20–20–0 mg) (Table 3). Low dose propranolol reduced heart rates while lying and standing compared to ivabradine and O3-FA supplementation that reduced heart rate only in the standing position (Table 3). Nutritional refeeding together with O3-FA supplementation in patients with anorexia nervosa increased heart rate while lying and decreased postural heart rate while standing, explaining the significant effect on the postural heart rate increase (from 50.0 ± 13.9 to 33.0 ± 9.9**). We did not find significant effects in the small group of patients who were treated with Midodrine.
Table 3The effect of treatment of postural orthostatic tachycardia on heart rate and heart rate variability in the active standing test
| Betablocker (N=15) | Ivabradine (N=11) | Midodrine (N=6) | Omega-3-FA(N=7) | Refeeding +O3-FA (N=6) |
---|
HR Increase | 41.7 ± 17.5 | 29.5 ± 17.8** | 37.5 ± 16.4 | 23.6 ± 8.12* | 31.9 ± 8.4 | 27.4 ± 10.5 | 44.0 ± 11.9 | 25.6 ± 8.4* | 50.0 ± 13.9 | 33.0 ± 9.9** |
HR Lying | 88.4 ± 20.5 | 75.9 ± 12.3* | 80.6 ± 17.9 | 77.9 ± 14.5 | 65.1 ± 6.7 | 67.5 ± 13.7 | 82.2 ± 17.6 | 84.3 ± 13.9 | 71.3 ± 13.4 | 81.6 ± 16.6 |
HR Standing | 130.1 ± 17.1 | 104.1 ± 20.8** | 118.2 ± 15.0 | 101.4 ± 16.9** | 97.1 ± 12.0 | 94.9 ± 16.0 | 126.2 ± 11.4 | 109.9 ± 16.9* | 121.4 ± 26.7 | 114.6 ± 17.1 |
RMSSD Lying | 38.3 ± 6.8 | 70.4 ± 16.5 | 52.0 ± 17.0 | 44.0 ± 6.0 | 91.9 ± 68 | 90.1 ± 45.3 | 87.6 ± 35.0 | 54.6 ± 12.4 | 51.6 ± 14.2 | 42.0 ± 41.7 |
RMSSD Standing | 11.9 ± 2.7 | 35.1 ± 9.7* | 17.5 ± 3.8 | 24.5 ± 9.5 | 24.1 ± 14.2 | 44.9 ± 38.0 | 11.8 ± 3.2 | 27.2 ± 12.6 | 17.4 ± 6.2 | 18.0 ± 8.0 |
Stressindex Lying | 354 ± 145 | 115 ± 106 | 232 ± 73 | 185 ± 38 | 63.7 ± 36.2 | 90.4 ± 84.4 | 192 ± 182 | 146 ± 111 | 206 ± 77 | 216 ± 59 |
Stressindex Stand | 947 ± 171 | 443 ± 103** | 609 ± 123 | 548 ± 197 | 239 ± 178 | 238 ± 222 | 855 ± 160 | 561 ± 217 | 959 ± 334 | 944 ± 283 |
Analysis of GPCR autoantibodies in nine patients with POTS showed many elevated values (Table 4). There was no clear pattern of autoimmunity except in one female patient with anorexia nervosa, and most patients showed elevated autoantibodies against vasoconstrictive receptors (anti-angiotensin 1 receptor, anti-endothelin receptor, anti-α1 adrenergic receptor) that may explain postural hypotension as well as the compensatory heart rate increase to maintain blood pressure. Some anti-mucarinergic autoantibodies may explain postural tachycardia and the collapse of the vagal tone in the standing position (Table 4).
Table 4Autoantibodies in children with postural orthostatic tachycardia
Patient | Anti AT1R | Anti ETAR | Anti α1 adrenerg | Anti α2 adrenerg | Anti β1 adrenerg | Anti β2 adrenerg | Anti MC R1 | Anti MC R2 | Anti MC R3 | Anti MC R4 | Anti MC R5 |
---|
Reference [U/mL] | <10 | <10 | <7 | <15 | <15 | <8 | <9 | <9 | <6 | <10,7 | <14,2 |
POTS (Post Lyme Disease) | 9.3 | 9.2 | 29 | 21 | 9.8 | 8.8 | 4.1 | 3.8 | 12.0 | 8.7 | 11.5 |
POTS (Emery Dreifuss myopathy) | 21.2 | 20.6 | 14.2 | 22.6 | 21.5 | 19.3 | 9.0 | 7.5 | 12.0 | 9.0 | 9.0 |
POTS (Chronic Fatigue Syndrome) | 6.6 | 6.5 | 18.3 | 10.6 | 8.2 | 6.5 | 2.7 | 1.6 | 7.8 | 3.5 | 7.3 |
POTS (Dizziness) | 10,8 | 8,2 | 13,5 | 12,4 | 5,1 | 3,3 | 1,3 | 2,5 | 3,5 | 4,1 | 4,6 |
POTS (Anorexia Nervosa) | 10,3 | 8,2 | 3,9 | 9,8 | 5,6 | 13,0 | 4,2 | 1,6 | 6,4 | 15,3 | 12,4 |
Long COVID after SARS-CoV-2 infection |
POTS | 13.0 | 11.8 | 7.8 | 8.8 | 8.8 | 6.4 | 9.9 | 12.8 | 5.4 | 4.7 | 11.4 |
POTS | 14,7 | 14,4 | 13,9 | 8,6 | 11,9 | 21,7 | 3,6 | 3,6 | 12,3 | 10,4 | 3,6 |
POTS | 24,8 | 43,7 | 28,5 | 5 | 42,8 | 41,7 | 4,2 | 7,7 | 38,4 | 24,1 | 7,4 |
SARS-CoV-2 Vaccination (Cominaty™) |
POTS post Vaccination | 20,7 | 18,1 | 10,7 | 8,4 | 25,7 | 27,5 | 6,1 | 6,6 | 8,7 | 10,6 | 11,0 |
Case 1: Dysautonomia after COVID-19 infection (long COVID)
After a mild COVID-19 infection in April 2020, a 16-year-old girl presented clinical symptoms of long COVID with palpitations, headache, and low physical performance for more than a year. In-hospital diagnostics did not lead to a therapeutic approach. We diagnosed the patient with IST and the 24-h Holter ECG showed a significant decrease in mean heart rate from 104 bpm to 85 bpm after low dose propranolol (Fig. 2). The patient did not tolerate ivabradine to treat headache and showed little clinical improvement with propranolol to induce overshooting vagus activity indicated by RMSSD and pNN20 (Fig. 3). The patient was dissatisfied with performance at 16 months after COVID-19 infection, which may be related to Hashimoto thyroiditis (additional elevated antithyroid peroxidase antibodies and thyrotropin receptor antibodies). Angiotensin 1 receptor, endothelin receptor, α1 adrenergic receptor, muscarinic cholinergic receptor-2, and muscarinic cholinergic receptor-3 autoantibodies were elevated.
Case 2: Dysautonomia after SARS-CoV-2 mRNA vaccination
A 16-year-old girl suffered from palpitation, dizziness, and deterioration in performance after a second SARS-CoV-2 mRNA vaccination. She could not attend school and in-hospital diagnostics did not lead to a therapeutic approach. We diagnosed the patient with POTS and started ivabradine therapy one month after vaccination. At baseline, the mean 24-h heart rate was elevated to 96 bpm with normal blood pressure during the day and night. After ivabradine treatment, heart rate decreased to 77 bpm and blood pressure remained unchanged (Fig. 4). As shown in Figure 3, HRV during the active standing test progressively improved within the next 10 weeks; due to the clinical improvement, the patient was able to return to school. However, stopping ivabradine treatment after 3 months was not successful. Angiotensin 1 receptor, endothelin receptor, α1 adrenergic receptor, β1 adrenergic receptor, β2 adrenergic receptor, and muscarinic cholinergic receptor-2 autoantibodies were elevated.
Case 3: Dysautonomia in post-traumatic stress disorder in a Syrian refugee
Data of a 16-year-old girl with post-traumatic stress disorder after escaping the Syrian civil war to Germany are shown in Figure 3. Therapy with low dose propranolol20 was not well tolerated, but the patient completely recovered after ivabradine treatment, which has not been used in this setting. HRV data (Fig. 3) showed an overshooting vagus activity (pNN50/RMSSD) after low dose propranolol that may explain the discomfort of this therapy and why the change to ivabradine had a slightly better effect on the heart rate increase in the active standing test.
Case 4: Dysautonomia in boy with anorexia nervosa
Figure 5 shows the clinical course of a 15-year-old boy with severe anorexia nervosa who became bedridden under nasogastric tube feeding and showed clinical symptoms of severe depression despite psychopharmacological therapy. After diagnosis and treatment of severe POTS with metoprolol and midodrine, the patient recovered immediately and showed an above-average weight gain following discharge.
Discussion
The current study demonstrates that dysautonomia could be used in diagnosis to improve treatment of psychosomatic diseases and COVID-19-related disorders in children. Our data show that an objective diagnosis is possible in children with dysautonomia using the 24-h Holter ECG and the active standing test. We propose a mean heart rate ≥ 95 bpm in the Holter ECG18 to diagnose an IST and an increase in the average of the heart rate after 5 m lying to 5 m standing of ≥ 35 bpm14 to diagnose POTS. Lowering heart rate in the active standing test was achieved using pharmacotherapy with low dose propranolol (from 41.7 ± 17.5 to 29.5 ± 17.8 bpm **) and ivabradine (from 37.5 ± 16.4 to 23.6 ± 8.12 bpm*), as shown in Table 3. However, as recently shown in children with IST,18 our data clearly showed a significant reduction in heart rate while standing after O3-FA supplementation (from 44.0 ± 11.9 to 25.6 ± 8.4 bpm*) and additional nutritional refeeding in patients with anorexia nervosa (from 50.0 ± 13.9 to 33.0 ± 9.9 bpm**). There was no significant effect on heart rate increase in the active standing test after midodrine treatment (Fig. 1). In accordance with Lin J et al,21 we observed POTS more often in tall adolescents (Table 1) and patients with anorexia nervosa17 but less in those with IST who were obese, had small stature, and attention deficit disorder.18 If heart rate in patients with anorexia nervosa is low, we propose O3-FA supplementation during nutritional refeeding to prevent an overshooting heart rate increase that is not well tolerated by the patient.8
Analysis of HRV as shown in Tables 2 and 3 may help to understand the underlying pathophysiology but is not necessary for diagnosis. In contrast to IST with a global decrease in HRV, children with POTS have normal HRV in the 24-h Holter ECG (Table 2) and in the lying position (Table 3). In accordance with a previously published meta-analysis,22 we found a postural collapse of HRV as causal for an overshooting heart rate increase while standing in children with POTS. However, this collapse involves both the time domain and the frequency domain analysis with a clear emphasis on the vagus parameters RMSSD, pNN50, HF, and LF/HF. In summary, a vagus weakness seems to be common in children with dysautonomia and IST and only in upright position in children with POTS.
In our analysis, children with postural VSS had normal 24-h mean heart rates and a normal heart rate increase in the active standing test. However, Medow et al23 reported a heart rate increase in patients with VSS of 39.8 ± 2.1 bpm that was significantly greater (p < 0.001) than in the healthy controls. In this study, an increase in heart rate ≥ 40 bpm by 5 m and 10 m or before fainting with a head-up tilt, occurred in 26% and 44% of patients with VSS, respectively, but not in controls. This difference could be attributed to the following. We regularly observed an increase in heart rate before a sudden drop of heart rate caused by the vagus reflex before we had to stop the test in children with VSS. If we were able to completely analyze the standing test, we calculated a normal mean heart rate on average. However, the vagus is at the center of the pathophysiology in children with postural VSS and may be recognized by low rMSSD and pNN50 values in the lying position, as shown in Table 2. Furthermore, there are some children whose clinical presentation in the active standing test changed between POTS, IST, and VSS. However, a sick vagus remains the focus of therapy.
Based on the literature, we found a significant effect on postural heart rate increases after pharmacotherapy with beta blockers and ivabradine but also after O3-FA supplementation and nutritional refeeding in six patients with anorexia nervosa (Fig. 1, Table 3). However, we did not find significant effects in a small group of patients who were treated with midodrine, which is in contrast to Chen et al., who found a significant decrease in the heart rate increase in the standing test from 36 ± 1.3 bpm to 25 ± 1.9 bpm** in 19 children using a standardized protocol.24
There is important additional information to be gained from understanding the effects of therapy on HRV using our data in Table 3 and our additional published data from 24-h Holter ECG, as shown in Table 5.24,25 First, beta blockers reduce heart rate while lying and standing, together with a significant increase of the vagus parameter RMSSD and a significant reduction in the stress index, which could be due to the fact that heart rate decreases during day and to a lesser degree at night. Secondly, there were nearly no significant effects on HRV in the 24-h Holter ECG (Table 5). In contrast, O3-FA supplementation exclusively reduced heart rate while standing with no significant effects on HRV in the active standing test. However, there were significant effects of O3-FA supplementation on heart rate during the day and night, as well the global HRV in the 24-h Holter ECG (Table 5). This effect was completely reversed by nutritional refeeding in 10 children with anorexia nervosa, who had significant heart rate increases during the day and night, as well as a significant decrease in elevated HRV in the 24-h Holter ECG (Table 5). However, increasing heart rate in the lying position and decreasing heart rate while standing combined with nutritional refeeding and O3-FA supplementation induced a significantly lower postural heart rate increase in children with anorexia nervosa in the active standing test (from 50.0 ± 13.9 to 33.0 ± 9.9 bpm**). Ivabradine (n = 11), midodrine (n = 6), O3-FA (n = 7), and refeeding + O3-FA (n = 6) had no significant effect on the stress index in the active standing test.
Table 5The effect of the pharmacotherapy and Supplements on heart rate and heart rate variability in 24-hours Holter-ECG
| Children with somato-form Disorder (N=10)
| Adults with Cardio-myopathy (N=48)24
| Children with somatoform Disorders (N=19)
| Children with Anorexia Nervosa (N=10)
| Adults with vitamin D deficiency (N=52)25
|
---|
Baseline | Betablocker | Baseline | Ivabradine | Baseline | Omega-3-FA | Baseline | Refeeding+O-3-FA | Baseline | Vitamine D |
---|
24h HR [bpm] | 81 ± 10 | 74 ± 12*** | 83.6 ± 6.0 | 64.6 ± 5.8*** | 96 ± 11 | 85 ± 10*** | 59 ± 8 | 70 ± 8*** | 77.4 ± 6.6 | 78.08 ± 6.1 |
HR Day [bpm] | 90 ± 10 | 80 ± 13*** | 89.2 ± 8.9 | 68.6 ± 7.5*** | 102 ± 13 | 95 ± 12* | 69 ± 10 | 80 ± 10** | | |
HR Night [bpm] | 69 ± 9 | 65 ± 11* | 79.7 ± 7.7 | 61.4 ± 5.5*** | 85 ± 16 | 71 ± 9*** | 44 ± 6 | 54 ± 9** | | |
SDNN [ms] | 154 ± 36 | 161 ± 57 | 56.2 ± 15.7 | 87.9 ± 19.4*** | 127 ± 49 | 148 ± 32** | 289 ± 66 | 225 ± 86** | 68.6 ± 13.5 | 119.9 ± 28.3*** |
RMSSD [ms] | 43 ± 22 | 50 ± 25* | 13.5 ± 4.6 | 17.8 ± 5.4*** | 27 ± 16 | 38 ± 11*** | 73 ± 13 | 55 ± 14** | 23(12/3) | 58(46/92)*** |
TP 24h | 5,736 ± 3,570 | 7,726 ± 5,879 | | | 3,201 ± 2,246 | 4,608 ± 1,883*** | 9,132 ± 3,036 | 7,412 ± 4,357 | | |
VLF 24h | 3,618 ± 2,488 | 5,206 ± 4,464 | | | 1,926 ± 1,462 | 2,634 ± 1,410** | 5,533 ± 2,795 | 4,636 ± 370 | | |
LF 24h | 1,367 ± 712 | 1,675 ± 1,089 | | | 836 ± 614 | 1,224 ± 577** | 2,114 ± 492 | 1,566 ± 838 | 62.9 ± 15.6 (normalized) | 59.4 ± 13.9 (normalized) |
HF 24h | 689 ± 450 | 779 ± 467 | | | 396 ± 288 | 688 ± 312**** | 1,442 ± 450 | 1,148 ± 483* | 19.6 ± 10.2 (normalized) | 39.4 ± 11.7*** (normalized) |
HF/LF 24h | 0.48 ± 0.13 | 0.48 ± 0.13 | | | 0.5 ± 0.26 | 0.61 ± 0.26* | 0.73 ± 0.34 | 0.88 ± 0.55 | | |
If the effect of ivabradine on HRV in the 24-h Holter ECG was not well documented, we had to use the data from adults with nonischaemic dilated cardiomyopathies (Table 5).25 In those patients, ivabradine significantly reduced heart rate during the day and night and significantly improved the reduced HRV.
With respect to our sick vagus theory in children with dysautonomia, we concluded that O3-FA supplementation improves parameters of HRV, indicating vagus activity as determined by 24-h RMSSD (from 27 ± 16 to 38 ± 11 ms***) and 24-h HF (from 396 ± 288 to 688 ± 312 ms2***). Beta blockers improved RMSSD while standing (from 11.9 ± 2.7 to 35.1 ± 9.7 ms*) and in the 24-h Holter ECG (from 43 ± 22 to 50 ± 25 ms*) but had no significant effect on frequency domain analysis. Thus, ivabradine may improve vagus activity as shown in adults with cardiomyopathies25 (RMSSD from 13.5 ± 4.6 to 17.8 ± 5.4 ms***), but these data are missing in children with dysautonomia. Nutritional refeeding in children with anorexia nervosa reduced highly elevated HRV with an overshooting increase of low heart rates that was not well tolerated by patients but should be treated. However, prospective trials are needed to clarify if O3-FA supplementation and/or ivabradine/beta blockers can improve the success of nutritional refeeding in children with anorexia nervosa (as presented in Case 4).
Management of dysautonomia in the COVID-19 pandemic
Modern medicine is characterized by major pathophysiological concepts, such as the neurovisceral integration concept from Claude Bernard,3 which have been abandoned as current medicine becomes more detailed. However, since medicine is becoming increasingly expensive and more people are being excluded from its benefits, this policy is being revisited during the pandemic to provide patients with effective, cheap therapies. This explains the re-emergence of vitamin D26 and O3-FA in the ongoing COVID-19 pandemic, specifically in countries with a low budget for health care. As shown in Table 5, these therapies have significant effects on objectively measured parameters that represent vagus activity that likely plays key roles in the pathophysiology of COVID-19-related disease (e.g., elevated heart rates, reduced rMSSD, and SDNN).10
There is growing evidence that post-acute complications after SARS-CoV-2 infections are related to dysautonomia.27 In children with dysautonomia after SARS-CoV-2 infections/vaccination, we found elevated functional autoantibodies against G-protein coupled receptors (Table 4, Figs. 2 and 4). Our data are in accordance with Wallukat et al., who found these autoantibodies in adults with persistent long COVID-19 symptoms.13 In summary dysautonomia may be one of multiple autoimmune diseases triggered by the spike protein of the SARS-CoV-2 virus or more precisely with immune responses to SARS-CoV-2 proteins.12 These autoantibodies have been found in other patients with POTS11 and myalgiac encephalomyelitis/chronic fatigue syndrome28 and may be treated with immunoabsorption.29 However, since immunoabsorption is an expensive therapy that is not an option for many patients during the pandemic, we propose well-known pharmacotherapies such as low dose propranolol, ivabradine, and O3-FA supplementation that are effective in adolescents with IST due to long COVID (Fig. 2) or POTS after SARS-CoV-2 mRNA vaccination (Fig. 4).
Improvement in physical performance is the most important therapeutic goal in patients with POTS in the context of chronic fatigue syndrome caused by long COVID and/or dysautonomia. There is evidence from prospective randomized trials that low dose propranolol30 as well as ivabradine31 improve performance in patients with POTS.
Limitations
This retrospective study included data from clinical routine exams. All included children suffered from dysautonomia caused by different diseases such as nutritional disorders, post-traumatic stress disorder, or more recently COVID-19-related disorders. All children received the established standard therapies, which often did not lead to clinical improvement. These therapies were not considered in this analysis. The evidence and symptoms associated with COVID-19 are still very new and we need to determine if the long-term problems of this disease are actually an expression of dysautonomia.
Future directions
Based upon clinical routine data, the current study demonstrates that dysautonomia could be a diagnostic tool to improve therapy for many psychosomatic disease and COVID-19 related disorders in childhood and adults. Based upon the bidirectional interaction between heart rate and emotions, we can more objectively diagnose and treat poorly defined psychosomatic diseases like long COVID. These methods should increasingly be used by physicians in clinical practice. However, where clinicians are still hesitant to use this diagnostic tool, patients can use the widely available wearables to measure parameters themselves. We demonstrated significant effects of cheap and widely available cardiovascular drugs including propranolol, ivabradine, and O3-FA supplementation. However, these novel therapeutic approaches to psychosomatic diseases have to be further tested in prospective randomized trials, as it should be noted that some patients did not respond to this therapy. Future research is needed to clarify the impact of autoimmunity on dysautonomia to improve ongoing symptoms that do not improve with cardiovascular medication.
Conclusions
Our data show that an objective diagnosis is possible in children with dysautonomia using the 24-h heart rate and Holter ECG and the heart rate increase in the active standing test. Based upon HRV analysis, a vagus weakness is common among children with dysautonomia compared to those with IST and those with POTS only in upright position. In some children, dysautonomia may be related to G-protein coupled receptor autoantibodies against receptors of the autonomous nervous system.
Lowering heart rate increases in the active standing test is possible with the use of low dose propranolol (from 41.7 ± 17.5 to 29.5 ± 17.8 bpm**) and ivabradine (from 37.5 ± 16.4 to 23.6 ± 8.12 bpm*). Furthermore, O3-FA supplementation significantly reduced the heart rate increase while standing (from 44.0 ± 11.9 to 25.6 ± 8.4 bpm*). Beta blockers reduced heart rate while lying and standing together with a significant increase in the vagus parameter RMSSD. O3-FA supplementation exclusively reduced heart rate while standing with highly significant effects on heart rates during the day and night, as well as global HRV in the 24-h Holter ECG. We presented four case studies showing clinical improvements with treatment of dysautonomia in adolescents with the following diagnoses: 1) long COVID syndrome, 2) long COVID-like symptoms after SARS-CoV-2 mRNA vaccination, 3) POTS in a Syrian refugee, and 4) anorexia nervosa.
Based on a bidirectional pathway between heart rate and emotions via the autonomic nervous system, it is possible to modulate emotions with therapies that improve vagus activity.
Abbreviations
- BMI:
body mass index
- COVID-19:
corona virus disease 2019
- DBP:
diastolic blood pressure
- DHA:
docosahexaenoic acid
- ECG:
electrocardiogram
- ELISA:
enzyme-linked immunosorbent assay
- EPA:
eicosapentaenoic acid
- FFA:
fast fourier analysis
- GPCR:
G-protein coupled autoantibody
- HF:
high frequency power
- HR:
heart rate
- HRV:
heart rate variability
- IST:
inappropriate sinus tachycardia
- LF:
low frequency power
- LF/HF:
low to high frequency power ratio
- mRNA:
messenger ribonucleic acid
- O3-FA:
omega-3-fatty acids
- POTS:
postural orthostatic tachycardia
- pNN50/pNN20:
number of pairs of adjacent NN intervals differing by more than 50/20ms divided by the total number of all NN intervals
- rMSSD:
square root of the arithmetic mean of the squared deviation of successive normal RR intervals
- SARS-CoV-2:
severe acute respiratory syndrome coronavirus 2
- SBP:
systolic blood pressure
- SDNN:
standard deviation of all normal RR intervals in a time frame
- TP:
total power
- VLF:
very low frequency power
Declarations
Acknowledgement
The author thanks Dr. Harald Heidecke (CellTrend GmbH Luckenwalde, Germany) for measuring G-coupled receptor antibodies in the 9 patients (Table 4) when these measurements were not covered by the patient’s health insurance.
Ethical statement
The study was reviewed and approved by the Institutional Review Board of Landesärztekammer Baden Württemberg F-2012-0056. The authors are accountable for all aspects of the work and ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Declaration of Helsinki (as revised in 2013). Written informed consent was obtained from all patients.
Data sharing statement
The data used to support the findings of this study are available from the corresponding author.
Funding
None.
Conflict of interest
The author has no conflicts of interest related to this publication.
Authors’ contributions
RB was the sole author.