ABC | Volume 110, Nº4, April 2018

Original Article Silva et al Postural hypotension: spectral analysis Arq Bras Cardiol. 2018; 110(4):303-311 infer from their results that autonomic system and baroreflex dysfunction are associated with OH. Harrington et al. 11 evaluated baroreflex sensitivity by digital plethysmography in the elderly – 75 normotensive and 64 hypertensive patients, without medications. The authors reported reduced baroreflex, with impairment of HF component. Kawaguchi et al. 12 shown a decrease in the LF/HF ratio and decreased cerebral perfusion measured by infrared spectroscopy in a group of 80 elderly subjects after passive standing. A more recent study on hypertensive patients, 27 18 with OH and 64 without OH (mean age of 74.2 years), demonstrated that, despite lower systolic volume in those with OH, no significant differences in change in the LF/HF ratio after orthostatism were found, which is in agreement with our results. Barantke et al. 13 demonstrated a decline in all components of HRV with age, and an association between LF and baroreflex sensitivity in orthostatic position. This evidence that LF reflects baroreflex function rather than sympathetic innervation measured by 6-[18F] fluorodopamine has also been demonstrated by other authors. 28,29 Consequently, our findings suggest the hypothesis that the lack of increase in HR and LF with orthostatismmay be related to baroreflex dysfuntion, 27,30 which predisposes to OH. Blood pressure may also influence the prevalence of OH. Studies on hypertensive elderly patients demonstrated higher prevalence of OH in those with higher BP levels. Gangavati et al. 31 followed 722 elderly patients and found a prevalence of OH of 19% in participants with uncontrolled hypertension (BP ≥ 140/90 mmHg) and of 5% in those with controlled hypertension (BP < 140/90 mmHg). Mean age was 78 years in both groups. Valbusa et al. 32 reported similar findings but with different OH prevalence – the authors evaluated 994 patients with mean age of 88 years; the prevalence of OH was 13% in hypertensive patients with BP ≤ 140/90 mmHg and 23% in those with BP > 140 mmHg. In the present study, no difference between the groups was found in baseline BP in the supine position or in the prevalence of hypertension. Regarding the association of OH with the use of medications, in a study with 189 patients aged 75 or older with OH, the prevalence of OH was of 35%, 58%, 60% and 65% in those patients using none, one, two, three or more medications, respectively. Although the study included medications other than anti-hypertensive agents, hydrochlorothiazide was associated with higher prevalence of OH (65%). 33 Analysis of a cohort of 3,775 women aged between 60 and 80 years demonstrated that the use of three or more anti-hypertensive agents had a 2.2 greater chance of developing OH in comparison with patients taking no medications. 34 In the present study, although the use of ACE inhibitors was significantly higher in patients with OH, this drug was not a predictor of this condition, which may be explained by its role on autonomic modulation. 22 Diabetes mellitus may also result in autonomic dysfunction. 4 In our study, its prevalence was 17.1% in the population, with no difference between the groups and, thereby, had no influence on the results. As previously reported, clinical manifestations of OH that may lead to falls, fractures, presyncope and syncope cause functional impairment in the elderly, which is known as frailty syndrome. 3-5 In the current study, previous symptoms including dizziness, presyncope and syncope were more frequent in patients in the case group and, according to the literature, these symptoms may be associated with frailty syndrome and lower BP values after orthostatism. 4,35 Data in the literature on the association between frailty and risk for cardiovascular disease are scarce. A study on 1,622 elderly men aged between 71 and 92 years showed an association between frailty and increased risk factors, including waist circumference, lipid profile and SAH, despite similar prevalence of these factors between frail and non-frail elderly persons. Cardiovascular risk scores were not calculated, but this association was independent of established cardiovascular disease. 36 In the present study, patients were assessed for cardiovascular risk using the Framingham 14 and PROCAM scores, 15 with no difference between the groups. It is worth mentioning that 75 years is the age limit for the use of these scores. Limitations The main limitations of this study were the number of patients and the fact that they were assessed only once, which made the evaluationof reproducibility of results impossible. The use of digital plethysmography for measurement of BP levels in orthostatic position would enable the early detection of OH. Besides, we did not evaluate the very low frequency (VLF) component of HRV, associated with renin-angiotensin-aldosterone system, thermoregulation and peripheral vasomotor tone. Conclusions In the study population, lower LF and HR in the supine position was found in patients with OH, regardless of gender, BP in supine position and use of beta-blockers. HR in the supine position was an independent predictor for OH with an odds ratio of 3.23 for values lower than 61 bpm. Author contributions Conception and design of the research and Statistical analysis: Silva RMFL; Acquisition of data and Analysis and interpretation of the data: Silva RMFL, Miranda CES, Barbosa MT, Bicalho MAC; Writing of the manuscript: Silva RMFL, Miranda CES; Critical revision of the manuscript for intellectual content: Silva RMFL, Miranda CES, Barbosa MT. Potential Conflict of Interest No potential conflict of interest relevant to this article was reported. Sources of Funding There were no external funding sources for this study. Study Association This article is part of the thesis of master submitted by Carlos Eduardo de Souza Miranda, fromUniversidade Federal de Minas Gerais. 309