Is cardiac autonomic modulation influenced by beta blockers in adolescents with Duchenne Muscular Dystrophy?

Participants

A cross-sectional study was carried out, in which 90 participants aged between 11 and 18 years were evaluated, divided into three groups: (1) 30 people with a medical diagnosis of DMD undergoing treatment with beta-blocker (DMDB); (2) 30 people diagnosed with DMD who were not currently using beta-blockers and had never used them previously (GDMD) and (3) 30 people with typical development (TD), matched for age and sex to DMD groups. The recruitment of the DMD groups was carried out at the Brazilian Association of Muscular Dystrophy (ABDIM) and at the Neuromuscular Diseases Outpatient Clinic of the Federal University of São Paulo (UNIFESP). The protocol was applied during the assessment carried out in the follow-up outpatient visits.

Inclusion and exclusion criteria

The inclusion criteria for DMD groups required participants to have: (1) a confirmed diagnosis of Duchenne muscular dystrophy established through molecular methods and/or skeletal muscle protein expression analysis; (2) age between 11 and 19 years; (3) absence of cardiac arrhythmias and atrioventricular block; (4) no congenital anomalies including congenital heart disease or pulmonary malformations; and (5) no current use of antidiabetic medications. The inclusion criteria of the participants in the healthy control group we consider adolescents aged between 11 and 19 years, matched by age and sex to the DMD participants. They were required to have no history of neuromuscular, cardiovascular, pulmonary, or metabolic disorders, including diabetes mellitus assessed by an anamnese done by the experimenter. All participants had to be free from cardiac arrhythmias and congenital anomalies, and not taking any medications known to influence autonomic or cardiovascular function. Written informed consent was obtained from all participants’ parents or legal guardians. As exclusion criteria, we consider errors in the analysis of HRV above 5 %.

Study design and data acquisition

For data collection, a form was used and filled out from the patient’s medical record to obtain information about medications and age. For the assessment of patients’ motor function, the Vignos scale was applied [17]; and to classify and assess the functionality of each patient, the Egen Klassification (EK) [18], 19] and Motor Function Measure (MFM) [20] scales were used.

The control of ambient temperature (temperature between 21 °C and 23 °C), humidity (humidity between 40 and 60 %) and the preparation of the equipment used were carried out before the arrival of the participants at the place destined for the tests. A minimum number of people was allowed to circulate around the room during the collections in order to reduce the anxiety of these participants.

The participants were oriented not to practice high physical effort, drink coffee or eat before the measurements. The capture strap was placed on the volunteers’ chest and, in their wrist, the heart rate receiver (RS800CX, Polar), previously validated equipment for beat-to-beat heart rate capture and the use of their data for HRV analysis [21]. After placement of the strap and monitor, the participants were placed in dorsal decubitus and remained at rest breathing spontaneously for 25 min.

For data collection of HRV data in the dorsal decubitus position, the 15-min period with the greatest signal stability was selected. To verify signal stability was evaluated by calculating the percentage difference in the mean and standard deviation between consecutive segments of the time series, allowing to identify the most homogeneous portion for analysis [22].

The HRV acquisition protocol used was selected based on prior validation studies that demonstrated improved reliability and stability of nonlinear indices with longer recordings. While short-term recordings (5 min) are common for time and frequency domain metrics, nonlinear analyses – such as Detrended Fluctuation Analysis (DFA) and Symbolic Dynamics (SD) – require longer (25 min), artifact-free RR series for robust interpretation [23]. Additionally, the supine posture was chosen for its ability to minimize sympathetic activation and promote parasympathetic tone, thereby optimizing the detection of autonomic changes in a clinical population with limited mobility [24].

R-R interval data were collected using heart rate monitors (Polar), which apply a proprietary filtering process – specifically, the standard moderate filter with a 6-beat protection zone [22]. However, as this filtering may not fully remove ectopic beats, artifacts, or noise, visual inspection and manual correction were performed using Microsoft Excel to ensure data quality. In this manual process, artifacts and ectopic beats were eliminated without replacement [22]. Only series with more than 95 % sinus beats and containing 1,000 consecutive RR intervals were included in the analysis [19]. Kubios HRV software was also used for additional filtering – applying a medium threshold, as recommended for younger populations – and to calculate HRV indices [19].

HRV analysis was conducted using linear methods (time and frequency domains) and non-linear methods (chaos domain).

Heart rate variability

The linear indices of time and frequency domain were calculated using the Kubios software (v. 2.2, Kuopio, Finland). Time-domain analysis was performed by RMSSD which is Square root of the mean of squared differences between successive beat intervals (RR), PNN50 that correspond to the percentage of adjacent RR intervals that differ by more than 50 ms, SDNN that corresponds to the standard deviation of all normal RR intervals, and also Mean RR and Mean HR [25], [26], [27], [28], [29].

In the frequency domain, the analysis included the spectral components of low frequency (LF: 0.04–0.15 Hz) and high frequency (HF: 0.15–0.40 Hz), measured in both milliseconds squared (ms²) and normalized units (nu). The spectral assessment was conducted using the Fast Fourier Transform algorithm [25].

Non-linear methods were employed for HRV analysis, including Detrended Fluctuation Analysis (DFA) and Symbolic Dynamics (SD).

DFA quantifies the presence or absence of fractal correlation properties in RR intervals and has been validated using time series data. It calculates the root mean square fluctuation of the integrated and detrended time series, allowing for the detection of intrinsic self-similarity within non-stationary signals [30]. The DFA plot is not strictly linear; rather, it consists of two distinct regions separated by a transition point. These represent two scaling exponents: a short-term fractal exponent (α1), typically reflecting fluctuations over 4–11 beats (or sometimes 4–13), and a long-term exponent (α2) for fluctuations beyond 11 beats [31].

The SD Analysis estimates the complexity of sympathetic and parasympathetic modulation on cardiac autonomic control and is well-suited for analyzing short RR interval sequences. The RR intervals are grouped into symbolic patterns based on their sequential variation, classified into four categories [32]: 0V%: No variation (e.g., 4–4–4), 1V%: One variation (e.g., 4–4–2), 2LV%: Two like variations in the same direction (e.g., 1–2–4), 2UV%: Two unlike variations forming a peak or valley (e.g., 2–4–2). In addition, Shannon Entropy (SE) quantifies the complexity of these symbolic patterns [33], 34].

Data analysis

The dependent variables (RR Mean, HR Mean, SDNN, RMSSD, pNN50, LF, HF, LF/HF, alfa 1, alfa 2, alfa 1/alfa 2, Sham Entropy, 0V%, 1V%, 2LV% and 2ULV%) were submitted to a 3 (groups: DMDB, GDMD, TD group) factor Multiple Analysis of Variance (MANOVA). The Least Significant Difference (LSD) was used as a post-hoc test. The partial eta-squared ( η p 2 ) was reported to measure the effect size and interpreted as small (effect size>0.01), medium (effect size>0.06), or large (effect size>0.14) [35].

For the comparison between the groups of the characterization variables (Table 1), such as age, body mass, body height and BMI, one-way ANOVA was used. Student’s t-test was used to compare the scales. Values of p<0.050 were considered significant. The software used was SPSS (Statistical Product and Service Solutions), 20.0 (Chicago, Illinois, USA).

Participants characteristics

Initially 78 patients with DMD were recruited according to the inclusion criteria and 18 were excluded from the analysis for presenting an error above 5 % on Kubios software. The characterization of the sample is shown in Tables 1 and 2, in which the participants are matched by age between groups. The weight was not statistically significant, however the height and BMI showed significant differences (TD group is taller than the participants of the two DMD groups, and TD group presented BMI significantly lower than those in the DMDB). There were no significant differences in motor scales.

Heart rate variability indices

Comparison between DMDB and GDMD

The comparison between DMD patients with and without beta-blockers (DMDB vs. GDMD) revealed significant differences in Mean RR and mean HR values, with the DMDB group showing higher Mean RR and lower mean HR. These findings are consistent with the known pharmacological effects of beta-adrenergic blockade, which typically results in heart rate slowing and longer RR intervals. This observed difference in autonomic parameters, although not leading to full normalization, aligns with previous studies demonstrating the effects of heart rate reduction [36].

Individuals with DMD experience marked dysfunction in cardiac autonomic regulation, characterized by reduced parasympathetic activity and heightened sympathetic dominance, which tends to worsen with advancing age and disease progression [32], 37]. In a review of patients with heart failure, Borovac et al. [33] highlighted that sympathetic nervous system overactivity, commonly seen in certain pathological conditions, is associated with impaired baroreceptor and chemoreceptor reflexes, increased neuronal and circulating catecholamine levels, diminished parasympathetic tone, and elevated sympathetic output to the heart, kidneys, and skeletal muscles. When these sympathoexcitatory effects persist over time, they contribute to the development and exacerbation of heart failure by promoting cardiomyocyte apoptosis, maladaptive remodeling of the myocardium and vasculature, the occurrence of arrhythmias, and ultimately, a poorer prognosis [38].

Based on these mechanisms, we propose that beta-blocker therapy in patients with DMD may be associated with more favourable cardiovascular outcomes by potentially mitigating autonomic dysfunction and its associated risks.

Although no statistically significant differences were observed in other HRV indices between the DMD groups, the DMDB group consistently demonstrated values that aligned more closely with those of typically developing adolescents compared to the GDMD group, which showed significant differences from the TD reference group (Table 3). While this cross-sectional observation cannot establish causation, it suggests that beta-blocker therapy may be associated with attenuated autonomic dysfunction in DMD, warranting further investigation through longitudinal studies.

Comparison with typically developing controls

When examining both DMD groups relative to TD participants as a reference for normal autonomic function, distinct patterns emerged. Participants with DMD, regardless of beta-blocker use, showed lower HRV and reduced complexity in nonlinear dynamics compared to healthy controls, consistent with previous literature on autonomic impairment in neuromuscular diseases.

Importantly, using the TD group as a reference point for normal physiological values revealed that the DMDB group exhibited HRV patterns that more closely approximated normal ranges compared to the GDMD group. In the linear analysis, the DMDB group demonstrated parasympathetic-related indices – such as RMSSD, pNN50, HF, and LF/HF ratio – as well as global HRV measures like SDNN and LF, with values that did not differ significantly from those observed in the TD reference group. This observation is clinically relevant given the established association between higher HRV and better cardiovascular prognosis.

Regarding non-linear analysis, DFA results indicated a loss of fractal properties (α2) in the long-term range for the GDMD group when compared to TD controls, whereas the DMDB group showed α2 values that did not differ significantly from the TD reference group. Since a reduction in α2 is associated with increased cardiac risk [32], the observation that DMDB values remained within the normal range suggests a potential protective association with beta-blocker therapy, though longitudinal studies are needed to confirm this relationship.

Shannon entropy values were higher (although not significantly) in the DMDB group, indicating greater complexity and variability of heart rate patterns. Given that higher entropy is associated with healthier autonomic regulation [34], this pattern suggests a possible beneficial association with beta-blocker therapy.

Symbolic dynamics analysis revealed lower sympathetic modulation (0V%) in the DMDB group, with values comparable to those observed in the TD reference group. However, the parasympathetic-related indices (2LV% and 2ULV%) differed significantly among all three groups, with the highest values recorded in the TD group and the lowest in the GDMD group.

Previous studies have shown that the symbolic dynamics indices 0V% and 2LV% may reflect shifts in the dynamic balance between sympathetic and vagal modulation in response to stress in healthy individuals [39], 40]. Therefore, this method may serve as a promising non-invasive tool for evaluating cardiac beta-adrenergic regulation, given its sensitivity to heart rate fluctuations mediated by the sympathetic nervous system [39], [40], [41], [42].

Nevertheless, the interpretation of these results requires caution due to the cross-sectional nature of our study. While the DMDB group consistently showed HRV indices that approximated normal values more closely than the GDMD group, direct comparisons between DMD groups did not reach statistical significance. The pattern observed – where DMDB values frequently fell between GDMD and TD values – provides valuable clinical context but cannot establish causal relationships. We acknowledge that future studies with larger sample sizes and longitudinal designs are essential to better elucidate the true impact of beta-blockers on autonomic modulation in adolescents with DMD.

Clinical relevance

The clinical significance of these findings extends beyond statistical comparisons to practical implications for patient care. Our results suggest that beta-blocker therapy in adolescents with DMD may be associated with autonomic parameters that more closely approximate normal physiological ranges, potentially indicating improved cardiac autonomic regulation.

Specific Clinical Implications:

1. Treatment Monitoring: The observed patterns suggest that HRV analysis, particularly nonlinear indices like DFA α2 and symbolic dynamics, could serve as non-invasive biomarkers for monitoring autonomic function in DMD patients receiving beta-blocker therapy. This could help clinicians assess treatment response beyond traditional echocardiographic parameters.

2. Risk Stratification: Given that reduced HRV and altered fractal dynamics are associated with increased cardiovascular risk, the observation that beta-blocker-treated patients showed values closer to normal ranges may indicate lower cardiovascular risk profiles. This information could inform clinical decision-making regarding the timing and intensity of cardiovascular interventions.

3. Therapeutic Decision-Making: While our cross-sectional design cannot establish causation, the consistent pattern of improved autonomic parameters in the beta-blocker group provides preliminary evidence supporting current clinical practice guidelines that recommend beta-blocker therapy for DMD patients with cardiac involvement.

4. Exercise and Rehabilitation Programs: The preservation of autonomic complexity in beta-blocker-treated patients may have implications for exercise prescription and physical therapy programs, as better autonomic regulation could indicate improved exercise tolerance and cardiovascular stability during physical activities.

This aligns with recent research by Oliveira et al. [43] and Dias et al. [32], showing that different pharmacological therapies, including corticosteroids, exert varying levels of cardiac protection. The application of nonlinear HRV analysis tools, traditionally used in exercise physiology, highlights the translational potential of these methods for monitoring disease progression and therapeutic responses in clinical populations with reduced mobility.

Future research directions

Longitudinal studies evaluating HRV changes before and after the initiation of beta-blocker therapy are crucial to establish causal relationships and optimize therapeutic strategies for cardiac autonomic dysfunction in DMD. Such studies would strengthen the evidence base for clinical decision-making and potentially establish HRV monitoring as a standard component of cardiovascular care in DMD patients.

Limitations

This study has several limitations. First, participants were taking additional medications such as ACE inhibitors and corticosteroids, which may have influenced HRV. Second, detailed information on beta-blocker dosage and duration was unavailable, limiting dose-response analysis. Third, while pubertal status can affect HRV, we minimized variability by selecting an age-matched sample (11–18 years) but did not directly assess Tanner stages. Fourth, although our artifact correction threshold (≤5 %) aligns with standard HRV guidelines, stricter thresholds (<3 %) could further improve signal quality. Fifth, although none of the participants were engaged in structured exercise programs, DMD participants routinely attended physiotherapy sessions (1–3 times/week), which could subtly affect autonomic outcomes despite no significant differences in motor function between groups. Finally, the cross-sectional design limits causal inference, and future longitudinal studies assessing HRV before and after beta-blocker initiation are needed to better understand therapeutic effects.