ABC | Volume 113, Nº2, August 2019

Original Article Casonatto et al. Citrulline and post-exercise hypotension Arq Bras Cardiol. 2019; 113(2):218-228 The study protocol was registered in ClinicalTrials.gov (NCT03378596). Anthropometry The weight was measured by a digital anthropometric scale (Urano, OS 180A, Canoas, Brazil), with an accuracy of 0.1 kg and height was measured by a stadiometer with an accuracy of 0.1 cm, in accordance with the procedures described by Gordon et al. 23 The body mass index was defined as the body mass (kg) divided by the square of the body height. Office blood pressure (laboratorial phase) The office BP measurements were taken with an oscillometric device (Omron MX3 Plus, Bannockburn, USA) previously validated for clinical measures in adults. 24 Firstly, the participants remained seated (rest period) in a calm, quiet, and thermoneutral (22°-24°C) environment for 20 min. BP was measured three times during the rest period (at 10 min, 15 min, and 20 min). The rest BP value was considered as the average of these three measurements. Immediately following the sessions (exercise or control), BP was measured in a quiet environment for 60minutes. The BP measurements were taken according to the American Heart Association recommendations. 25 Ambulatorial blood pressure measures (ambulatorial phase) The ambulatorial BP measurements were taken with an oscillometric device (Dyna-MAPA – São Paulo, Brazil) attached on the left arm, always by the same investigator, in accordance with procedures described by the American Heart Association. 25 The participants received instructions to keep their arm extended during the measures. The device was calibrated by direct comparison with a mercury sphygmomanometer, by a trained technical person, in agreement with recommendations. 25 The monitor was set to register the systolic and diastolic BP and HR every 20 minutes during “daytime” (08:00 am to 11:00 pm) and every 30 minutes during “night-time” (11:00 pm to 08:00 am) to reduce sleep disturbances. The device screen was electronically blinded to avoid feedback. All participants were instructed to register and report their sleep time in a diary on the following day. The data were recorded in the device memory and then sent to a computer using specific software (Aplicação Dyna Mapa – Version 5.0.382.12) for analysis. The average of the valid readings was above 90% for all participants. Statistical analysis Assuming a standard deviation (SD) of 5 mmHg 26 for the systolic BP, an alpha of 5% and a desired statistical power of 80% for detecting a minimum difference of 7 mmHg, 26 7 subjects were required in each group. The Shapiro-Wilk test was employed to examine the normality of the data distribution. The data are presented in the text as mean and SD. Levene’s homogeneity test was applied before the ANOVA one-way. Turkey (if homogeneity was accepted) or Games-Howell (if homogeneity was not accepted) multiple comparisons were employed to examine differences between pairs of trials. Effect size from the paired t-test was calculated (d = mean/SD). The results of the periods (60 minutes, awake, asleep, and 24 hours) were obtained through the average of the respective periods. Statistical significance was defined as p < 0.05. To assess the inter-individual BP responsiveness, we first assessed the reproducibility of pre-exercise BP between “-10” and “0” minutes of rest BP measures (see Figure 1). For this analysis, the intraclass correlation coefficient (ICC, model 2,k), coefficient of variation (CV) and standard error measurements (SEM) were considered. The CV between pre-exercise BP measurements was calculated as follows: CV = 100x(2x(SD d /√ 2 )/(X 1 +X 2 ). SD d represents the SD of the differences between the 2 measurements, and X 1 and X 2 represent the 2 measurement means, respectively. SEM = SDx√(1-ICC), with SD representing the SD of the first pre-exercise BP measure. The SEM was used as a measure of variability, but primarily for the minimal detectable change (MDC) calculation. MDC, which is the minimal change necessary to provide confidence that the results are not a random variation or measurement error, was calculated as follows: MDC=z-score (95% CI) xSEMx√2. All above‑mentioned procedures were adopted according to Haley and Fragala-Pinkham 27 and Darter et al. 28 The participants were considered as “responder” if their BP reached a value equal to or greater than the MDC. The statistical analysis was generated using the SPSS, version 20, system for windows. Results The exercise intensity was reduced for three participants (15 [EC], 18 and 19 [EP]) because they reported RPE ≥ 15. The exercise intensity based on HR reserve varied between 51-59% during the time (9 ± 3 minutes) that they could not keep 60-70% HR reserve. The results of the variables related to the MDC calculation were: 0.915 (ICC), 5.08 (CV), 5.25 (SEM) and 2.37 (MDC) for systolic BP, and 0.846 (ICC), 5.64 (CV), 3.74 (SEM) and 1.69 (MDC) for diastolic BP. Table 2 presents the absolute and relative BP changes in the different experimental groups. A significant reduction in systolic BP was identified for the EC when compared with the CP in the first 60 minutes (relative changes) after the experimental session. The same lowering effects were observed over the course of 24 hours (absolute changes) and for the “awake” and “24 hour” periods, considering relative changes (Δ% Awake and Δ% 24 hours) compared to the CP. No significant differences were found for the CC and EP. Table 3 presents the effect sizes from the paired t-test (rest vs 60 min, awake, asleep, and 24 hours) for each experimental group. Considering systolic BP, the CP showed a significant effect for the asleep and 24 hours, the EP showed a significant effect for the awake and asleep, and the EC showed a significant effect for all periods (60 min, awake, asleep, and 24 hours). Additionally, considering diastolic BP, the CP presented a significant effect for the asleep, the CC presented a significant effect for the 60 min, asleep, and 24 hours, the EP presented a significant effect for the awake and asleep, and the EC presented a significant effect for the awake, asleep, and 24 hours. 220