Heart Rate Variability in Individuals With Duchenne Muscular Dystrophy

A cross-sectional study was carried out, in which 40 boys, aged 11 to 18 years, were evaluated. The recruitment of groups was carried out at the neuromuscular disease outpatient clinic of the Federal University of São Paulo (UNIFESP). The recruited individuals were divided into 4 groups, namely: DMD that used deflazacort (DMD-D) with n=11; DMD that used Prednisone/Prednisolone (DMD-P), with n=9; DMD Control with no corticoid use (DMD-C), and n=10 and Controls with typical development (CTD) with n=10. The protocol was applied during the evaluation that was carried out at outpatient follow-up visits. Anthropometry was analyzed in 4 groups and functionality was assessed in 3 groups with DMD. To assess the functionality of each patient, the Vignos scales were used to characterize the sample and the Motor Function Measure (MFM) for association with HRV indices.

Initially, the resting ECG was analyzed to verify the existence of sinus rhythm and to exclude individuals with arrhythmias and blocks. The measurements at rest were performed immediately before and after the HRV assessment, including systolic (SBP) and diastolic (DBP), heart rate (HR), Respiratory Rate (RF - blood pressure) to ensure that at the beginning and at the end of the collection FR remains between 9 - 24rpm, in the range of 0.15 - 0.40Hz) and partial oxygen saturation (SO2). Heart rate was recorded by the cardiofrequency meter (RS800CX, Polar). And the partial oxygen saturation by means of a digital oximeter (DX2010, Dixtal) connected to the participant's index finger or hallux, through a sensor of age-appropriate size, in room air. Hemoglobin saturation by oxygen was recorded after the first minute of stabilization, as the value that remains most constant during the second minute.

All heart rate records were performed using a cardiofrequencymeter (V800, Polar). After placing the brace and monitor, the individuals were placed in the supine position and remained at rest spontaneously breathing for 25 minutes. For HRV analysis, indexes obtained by linear methods, in the domain of time and frequency, and non-linear methods were used.

In the time domain, for linear analyses, each normal RR interval (sinus beats) was verified during a certain time interval and, using statistical and non-linear methods, the translating indexes of fluctuations in the duration of the cardiac cycles were calculated. With this, the indices expressed in ms were obtained: SDNN (Standard deviation of the normal RR intervals recorded in a time interval); rMSSD (square root of the mean of the square of the differences between adjacent normal RR intervals in a time interval).

In the frequency domain, the spectral power density is more used, mainly when treating individuals in resting conditions. This analysis divides HRV into fundamental oscillatory components, which were used the main: High Frequency normalized unite (HFnu) component and Low Frequency normalized unite (LFnu) component. O algoritmo utilizado para a análise espectral foi a transformada rápida de Fourier - FFT (janela de 256 s com 50% de overlap).

Among the nonlinear methods used for HRV analysis, we can mention the Detrended Fluctuations Analysis (DFA), Visual Recurrence Analysis (VRA) and symbolic analysis (SA), the three of which were verified in this study.

DFA is used to quantify the fractal property of RR interval time series, being used to detect possible abnormalities present in a subject, based on α coefficients. For this, short memory parameters α1, which corresponds to a period of 4 to 11 beats, and long memory α2, which corresponds to the period from 64 to 1024 beats, were used.

The VRA is used to study the time dependence of a series, that is, in the study of stationarity35. The recurrence graph makes it possible to visualize the behavior of trajectories in the phase space and, in addition, to show the times in which a state of a dynamic system is repeated, besides these factors can be confirmed regarding the quantitative analysis of this, wich presents such indices: Mean, SD, PerRec, PerDet, PerLam, TrapTim, Ratio.

The evaluation of the symbolic analysis is based on the quantification of the information carried in a time series, in the transformation of the previously selected iRRs into integers from zero to six, from which sequences of 3 symbols (symbolic patterns) are constructed. For this, all possible patterns will be grouped according to the number and type of variations between the symbols, shown subsequently. The patterns were (1) patterns, without variation (0V), three identical symbols; (2) patterns with a variation (1V), that is, two subsequent symbols that are the same and a different one; (3) patterns with two similar variations (2LV) that is, the three symbols form a ramp; (4) patterns with two different variations (2ULV), that is, the three symbols form a peak or a valley.