ABC | Volume 111, Nº1, July 2018

Original Article Bassi et al T2DM and HTN on HRV and cardiorespiratory status Arq Bras Cardiol. 2018; 111(1):64-72 Additionally, it is well established that exercise capacity, which is a strong predictor of cardiovascular and overall mortality, 12 is reduced in patients with T2DM compared with non-diabetic subjects 13 as well as hypertensive patients. 14 Although the causes of reduced exercise capacity in these populations are unknown, cardiac autonomic dysfunction may play an important role in the development of heart disease in diabetic patients leading to impaired exercise capacity. 15 Recently, new variables derived from the cardiopulmonary exercise test (CPET), such as circulatory power (CP) and ventilatory power (VP) have been used for the clinical evaluation of heart failure patients as important markers of exercise limitation 16 . These indices could provide a potentially valuable measure of cardiopulmonary function in the coexistence of TM2DM and HTN. Considering this knowledge gap, the primary objective of the present study was to assess the cardiac autonomic modulation in T2DM patients with and without HTN. The secondary objective was to verify if HRV indices are correlated with exercise capacity in these patients. We hypothesized that patients affected by DMT2 and HTN would have an altered cardiac autonomic control when compared with diabetics and that there would be a correlation between HRV indexes and exercise capacity. Methods Design The present investigation is a cross sectional study. Participants A total of 60 patients (mean age ± SD = 51 ± 7 years; 42 male and 18 female) diagnosed with T2DM, followed at the cardiovascular outpatient clinic of the Federal University of Sao Carlos (UFSCar), agreed to participate in the study. Patients were divided into two groups according to the presence or not of HTN: 1) DMT2 (n = 32; 20 males and 12 female) and 2) T2DM + HTN (n = 28; 20 males and 8 female). Duration of DMT2 and HTN was recorded, based on the date of diagnosis self-reported by patients. The experimental procedures were performed in the UFSCar Cardiopulmonary Physiotherapy Laboratory. Inclusion criteria for both groups consisted of age between 40 and 60 years and clinically diagnosed DMT2 – based on fast glycemia and hemoglobin A1c (HbA1c) values, according to current guidelines – currently under hypoglycemics and clinically stable for at least 6 months. All patients were sedentary (self-reported). In the DMT2 +HTN group, diabetic subjects had clinical diagnosis of HTN and were under hypoglycemic and antihypertensive therapy. Exclusion criteria consisted of a history consistent with coronary heart disease or other concomitant respiratory diseases. RR interval recording The RR intervals were recorded continuously using a Polar S810i telemetry system (Polar Electro Oy, Kempele, Finland) at a sampling rate of 500Hz, and these data were used to derive the HRV indices. Each subject rested for 10 minutes before the initiation of data collection to ensure HR stabilization. The RR interval signal was continuously recorded for 10 minutes, while the patient rested in supine position, breathing spontaneously. Participants were instructed not to speak unnecessarily during the evaluation to avoid HR signal interference. HRV analysis The RR interval signals were transferred to a microcomputer and reviewed by visual inspection by an independent examiner to verify the quality of the signals and detect any abnormalities. Segments which presented any abnormalities were discarded. The data were transferred to Kubios HRV analysis software (MATLAB, version 2 beta, Kuopio, Finland) and a stable and free of artifacts series of 256 sequential RR intervals was selected and analyzed. To analyze the tachograms, a multivariate approach was followed, which allows for a comprehensive assessment of the cardiac autonomic function. The nonlinear dynamic properties of HRV were analyzed by calculating approximate entropy (ApEn), 17 correlation dimension (CD) 18 and Poincaré plots 19 . ApEn quantifies the regularity of a time series and represents a simple index of the overall complexity and predictability of the signal. High ApEn values indicate high irregularity, while smaller values indicate a more regular signal. Thus, higher ApEn values reflect better health and function. 17 The CD index represents a measure of the dimensionality of the space occupied by the state vectors or the number of the degrees of freedom of a time series, also referred to as fractal dimension. A higher CD reflects more degrees of freedomof the cardiac sinoatrial node and, therefore, a greater range of possible adaptive responses to internal or external stimuli in an ever-changing environment. 20 Poincaré plots were built for each RR interval series and the following two descriptors were computed: (i) SD1 – the standard deviation measuring the dispersion of points perpendicular to the line-of-identity. This parameter is usually interpreted as a measure of short-term HRV, which is mainly influenced by respiratory sinus arrhythmia (parasympathetic modulation); and (ii) SD2 – the standard deviation measuring the dispersion of points along the identity line, which is interpreted as a measure of both short- and long-term overall HRV. Shannon entropy (SE) was computed to quantify the degree of complexity of the distribution of the signals samples. 21 A set of time domain HRV parameters were calculated, including: (i) mean and standard deviation of RR intervals (SD RR), in ms; (ii) square root of the mean squared differences of successive RR intervals (RMSSD), in ms; and (iii) geometrical parameters, including the integral of the RR interval histogram divided by the height of the histogram (RR tri index) and the baseline width of the histogram (TINN), in ms. A spectral analysis was performed on the tachograms, in order to calculate the signal spectral power in the frequency band between 0.03 Hz and 0.14 Hz (low-frequency [LF] band) and in the frequency band between 0.15 Hz and 0.4 Hz (high‑frequency [HF] band), both expressed in 65