ABC | Volume 110, Nº2, February 2018

Original Article Correlation between Very Short and Short-Term Blood Pressure Variability in Diabetic-Hypertensive and Healthy Subjects Karina R. Casali, 1 Beatriz D. Schaan, 2,4 Nicola Montano, 3 Daniela Massierer, 4 Flávio M.F. Neto, 5 Gabriela H. Teló, 2 Priscila S. Ledur, 2 Marilia Reinheimer, 5 Graciele Sbruzzi, 1 Miguel Gus 4 Instituto de Cardiologia - Fundação Universitária de Cardiologia (IC/FUC); 1 Universidade Federal do Rio Grande do Sul; 2 L. Sacco Hospital - University of Milan, Milan - Itália; 3 Divisão de Cardiologia - Hospital de Clínicas de Porto Alegre; 4 Serviço de Cardiologia - Hospital de Clinicas de Porto Alegre, Porto Alegre, RS – Brazil 5 Mailing Address: Gabriela Teló • Rua Ramiro Barcelos, 2400 - 2° andar sala 218. Postal Code 90035-003, Santana, Porto Alegre, RS - Brazil E-mail: gabitelo@hcpa.edu.br Manuscript received May 19, 2017, revised manuscript July 02, 2017, accepted July 21, 2017 DOI: 10.5935/abc.20180020 Abstract Background: Blood pressure (BP) variability can be evaluated by 24-hour ambulatory BP monitoring (24h-ABPM), but its concordance with results from finger BP measurement (FBPM) has not been established yet. Objective: The aim of this study was to compare parameters of short-term (24h-ABPM) with very short-term BP variability (FBPM) in healthy (C) and diabetic-hypertensive (DH) subjects. Methods: Cross-sectional study with 51 DH subjects and 12 C subjects who underwent 24h-ABPM [extracting time-rate, standard deviation (SD), coefficient of variation (CV)] and short-term beat-to-beat recording at rest and after standing-up maneuvers [FBPM, extracting BP and heart rate (HR) variability parameters in the frequency domain, autoregressive spectral analysis]. Spearman correlation coefficient was used to correlate BP and HR variability parameters obtained from both FBPM and 24h-ABPM (divided into daytime, nighttime, and total). Statistical significance was set at p < 0.05. Results: There was a circadian variation of BP levels in C and DH groups; systolic BP and time-rate were higher in DH subjects in all periods evaluated. In C subjects, high positive correlations were shown between time-rate index (24h-ABPM) and LF component of short-term variability (FBPM, total, R = 0.591, p = 0.043); standard deviation (24h-ABPM) with LF component BPV (FBPM, total, R = 0.608, p = 0.036), coefficient of variation (24h-ABPM) with total BPV (FBPM, daytime, -0.585, p = 0.046) and alpha index (FBPM, daytime, -0.592, p = 0.043), time rate (24h-ABPM) and delta LF/HF (FBPM, total, R = 0.636, p = 0.026; daytime R = 0,857, p < 0.001). Records obtained from DH showed weak positive correlations. Conclusions: Indices obtained from 24h-ABPM (total, daytime) reflect BP and HR variability evaluated by FBPM in healthy individuals. This does not apply for DH subjects. (Arq Bras Cardiol. 2018; 110(2):157-165) Keywords: Hypertension; DiabetesMellitus, Type 2; Autonomic Nervous System; Blood PressureMonitoring, Ambulatory. Introduction Blood pressure (BP) variability results from the interplay between external environmental stimuli, vascular system, and biological autonomic regulation of circulation. 1 Abnormalities in BP variability, evaluated by continuous intra-arterial ambulatory BP monitoring, are associated with poor outcomes in normotensive and hypertensive subjects. 2-4 Noninvasive methods such as finger BP measurement (FBPM) are good alternatives to invasive BP monitoring, as they are accurate non-invasive estimates of beat-to-beat radial BP, providing data that can estimate very short-term BP variability. 5,6 In addition, beat-to-beat records allow the extraction of information regarding heart rate (HR) variability that is directly related to cardiac autonomic control impairment 7,8 and associated with poor outcomes in both general 9 and diabetic populations. 10 However, due to practical and economic reasons, this method cannot be routinely used in the evaluation of outpatients. The development of noninvasive 24-hour ambulatory BP monitoring (24h-ABPM), with multiple readings throughout day and night, has made short-term BP variability estimate through several indices possible. 11 However, there are major differences between BP variability obtained from beat‑to‑beat records and that obtained by 24h-ABPM. Besides the duration of the series—very short- (FBPM) or short-term (24h-ABPM)—, BP series obtained by FBPM allows studying beat-to-beat variability, while 24h-ABPM series are sampled every 10‑15 minutes within 24 hours. 6 While non-invasive beat-to-beat methods allow detecting fast oscillations resulting from inter‑beat variations, it is inefficient to access very slow waves in short series; 24h-ABPM, in turn, detects slow variations only. 12,13 As both methods provide information about BP signals originating from the same cardiovascular system, a correlation between oscillatory components of overlapping bands obtained from FBPM and 24h-ABPM is expected; 157