ABC | Volume 110, Nº2, February 2018

Original Article Amaral et al Autonomic and vascular control in prehypertensive patients Arq Bras Cardiol. 2018; 110(2):166-174 high frequency (HF). Due to the short registration period, the VLF component of iRR variability does not present well-established physiological explanation. 21 While the VLF of SBP variability seems to be related to myogenic vascular function. 22 The LF component of iRR variability reflects, predominantly, cardiac sympathetic modulation and the HF component, synchronized with respiration, cardiac parasympathetic modulation. 21 In the variability of SBP, the LF component quantifies the vasomotor sympathetic modulation, whereas the HF reflects the mechanical effect of respiration in the heart and vessels and does not represent an autonomic index. 23 The spectral power of each component of the variability of iRR and SBP was calculated in absolute terms and in normalized units. 21 The ratio between the LF and HF components of the iRR was calculated to quantify the cardiac sympathovagal balance. Arterial baroreflex control The gain and the time delay of response of the baroreflex control of the heart rate were measured by the analysis of the transfer function analysis using the bivariate autoregressive identification procedure. 24 This procedure allowed the quantification of coherence, phase shift and gain among the time series of the iRR (output signal) and the SBP (input signal) as described by Freitas et al. 24 In this study, the gain was calculated whenever the coherence between the signals was greater than 0.5 and the phase shift negative in the LF band, which indicates that the changes in the SBP preceded the changes in the iRR. In addition, it should be noted that the coherence, phase shift, gain and time delay of baroreflex control of heart rate were quantified at the central frequency corresponding to the maximum coherence within the LF band. Experimental protocol The evaluations were performed at the University Hospital of the Federal University of Juiz de Fora (HU-CAS), always in the morning. The volunteers were instructed not to ingest alcohol and / or caffeine and not to undertake vigorous physical activities within 24 hours prior to the evaluations as well as not eating fatty foods on the day of data collection. The volunteers responded to the anamnesis that included the clinical data of the patients and their parents and were submitted to anthropometric evaluation. After the volunteers remained in the supine position for 10 minutes, simultaneous recording of heart rate, respiratory rate and BP was started for 15 minutes at rest. Then, the muscular blood flow of the forearm was measured during three minutes of rest and three minutes of reactive hyperemia. Statistical analysis Data were presented as mean ± standard deviation of the mean or as median and interquartile range. To verify the normality of the distribution of all variables analyzed, the Shapiro-Wilk test was used. In addition, the assumption of homogeneity of variance was also verified by the Lèvene test. The distribution of the sexes between the groups was presented in absolute and percentage values. Fisher's exact test was used to verify the possible difference between the proportions of the sexes and of volunteers with both hypertensive parents in the groups. The possible differences related to the demographic, clinical and autonomic characteristics of the groups were verified through the unpaired Student t test for the data that presented normal distribution and Mann-Whitney U for the variables that violated this assumption. Two-way analysis of variance for repeated measures was used to test for possible differences between groups in vascular conductance during resting and reactive hyperemia. The main and interaction effects were analyzed with Bonferroni confidence interval adjustment. All statistical analyzes were performed using SPSS® software version 20. The statistical significance was p ≤ 0.05. Results Of the 25 volunteers analyzed, one normotensive volunteer did not meet the acceptability criteria for the analysis of the cardiac and peripheral autonomic modulation, and one normotensive volunteer and two prehypertensive patients did not attend to the analysis of arterial baroreflex function. Table 1 shows the demographic and clinical characteristics of the groups evaluated. In addition to laboratory tests for glycemia, total cholesterol and triglycerides (Table 1), 13 normotensive volunteers and nine prehypertensive subjects measured serum creatinine levels (0.85 ± 0.21 and 0.94 ± 0.21 mg/dL, respectively), p = 0.350), and nine normotensive and seven prehypertensive patients measured serum uric acid levels (4.09 ± 1.55 and 4.84 ± 1.12 mg/dl, respectively, p = 296). No differences were observed between groups in any of the laboratory variables analyzed. Analysis of vascular function, measured by forearm vascular conductance during resting and reactive hyperemia, is shown in Figure 1. Vascular conductance increased during hyperemia in both the normotensive (p<0.01) and prehypertensive (p<0.01). In addition, although the prehypertensive group presented greater forearm vascular conductance both at rest (p = 0.05) and at the peak of reactive hyperemia (p = 0.04), this difference between groups tends to be more pronounced during the reactive hyperemia maneuver (interaction effect: p = 0.05). Indices of cardiac autonomic modulation were similar between the groups (Table 2). However, in the peripheral autonomic modulation, greater variability (VarianceSBP) and higher VLFSBP and LFSBP spectral components were observed in prehypertensive patients compared to normotensive patients (Table 2). Additionally, we observed a lower gain of baroreflex control in prehypertensive patients (LFSBP-iRR gain), but similar LFSBP-iRR delay time between groups (Figure 2). Table 3 shows the central frequency, phase shift and coherence of the LF component of the SBP-iRR relationship, as well as the central frequency and coherence of the LF and HF components of the relationship between respiratory activity and the iRR. 168