Journal of Human Hypertension (2003) 17, 153–158 & 2003 Nature Publishing Group All rights reserved 0950-9240/03 $25.00 www.nature.com/jhh ORIGINAL ARTICLE Diastolic blood pressure is an important determinant of augmentation index and pulse wave velocity in young, healthy males JNu¨ rnberger1, S Dammer1, A Opazo Saez2, T Philipp1 and RF Scha¨fers1 1Department of Nephrology, University of Essen, Essen, Germany; 2Department of Cellular and Integrative Physiology, Indiana University School of Medicine, Indianapolis, IN, USA Pulse wave velocity (PWV) and augmentation index are P ¼ 0.0002) and total peripheral resistance (r ¼ 0.266, widely used measures of arterial stiffness. The purpose P ¼ 0.0204). After multiple regression analysis, augmen- of this study was to evaluate the role of blood pressure tation index remained significantly correlated only to as a determinant of both indices independent of DBP (b ¼ 0.347, P ¼ 0.0051). Using simple regression potentially confounding factors including gender, age analysis, PWV was correlated to age (r ¼ 0.304, and cardiovascular disorders. A total of 77 young, P ¼ 0.0067), systolic blood pressure (r ¼ 0.280, healthy subjects were investigated under resting condi- P ¼ 0.0129). DBP (r ¼ 0.455, Po0.0001), MAP (r ¼ 0.446, tions. Augmentation index was derived by pulse wave Po0.0001) and heart rate (r ¼ 0.348, P ¼ 0.0018). After analysis using carotid applanation tonometry. PWV was multiple regression analysis, PWV remained correlated determined from pressure tracing over the carotid and only to age (b ¼ 0.218, P ¼ 0.0422) and DBP (b ¼ 0.4105, femoral artery. The relations between stiffness markers P ¼ 0.0316). In summary, DBP is an important determi- and haemodynamic parameters were analysed by sim- nant of augmentation index and PWV in young, healthy ple (r) and multiple (b) regression analysis. Using simple males. Further studies are needed to characterize the regression analysis, augmentation index was correlated impact of blood pressure on arterial stiffness in other to age (r ¼ 0.292, P ¼ 0.0105), diastolic blood pressure populations including females and older subjects. (DBP, r ¼ 0.483, Po0.0001), mean arterial blood pressure Journal of Human Hypertension (2003) 17, 153–158. (MAP, r ¼ 0.381, P ¼ 0.0007), pulse pressure (r ¼À0.414, doi:10.1038/sj.jhh.1001526 Keywords: augmentation index; pulse wave velocity; diastolic blood pressure; arterial stiffness Introduction ness have been developed such as characteristic impedance,8,9 PWV constitutes one of the most Stiffening of the arterial wall is an inevitable process 1 widely used surrogates of arterial stiffness. that occurs during ageing and is one major Recently, a new technique, pulse wave analysis, mechanism responsible for morbidity and mortality 2,3 has been developed that offers a simple way to in cardiovascular disease. Thus, measuring arter- measure arterial function.1 Pulse pressure waves are ial stiffness has become a widely used tool for recorded at a single conveniently exposed artery, investigating the function of large arteries in such as the carotid artery, and used to calculate aug- epidemiological and clinical studies. Arterial stiff- mentation index, a measure that reflects the degree ness has been traditionally assessed by measuring 4,5 to which central arterial pressure is enhanced by pulse wave velocity (PWV), which has been wave reflection of the pulse wave.1 This new shown to predict cardiovascular mortality in pa- technique is reproducible, noninvasive and increas- tients with essential hypertension and chronic renal 10,11 6,7 ingly used in clinical studies. failure. Although other measures of arterial stiff- A number of nonpathological factors influence augmentation index and PWV. It has been consis- tently reported that age,1,12 height,13,14 heart rate15,16 Correspondence: Dr R.F. Scha¨fers, Division of Nephrology, School and gender12,14 influence augmentation index, of Medicine, University of Essen, Hufelandstrasse 55, 45122 whereas PWV appears to be a more robust stiffness Essen, Germany. measure influenced by fewer factors including age1,5 E-mail: [email protected] 17 Received 16 August 2002; revised 6 November 2002; accepted 13 and heart rate. However, despite the wide use November 2002 of both augmentation index and PWV, there are Arterial stiffness and diastolic blood pressure JNu¨rnberger et al 154 controversial results in the literature regarding the electrocardiogram and routine laboratory screening. relation between both stiffness measures and blood Investigations were carried out in the Department of pressure, the classical haemodynamic parameter. Nephrology at the University of Essen, Germany. In some studies, augmentation index was found to The study was approved by the Ethics Committee of be correlated to diastolic blood pressure (DBP) but the University of Essen Medical School and was in not systolic blood pressure (SBP).18,19 Other studies accordance with the principles laid down in the reported that augmentation index was correlated to Declaration of Helsinki. both SBP and DBP20 or only to SBP.21,22 Interestingly, Measurements were performed in a quiet, tem- Cameron et al 21 found in a study population of 262 perature-controlled (241C) room. On the study day, hypertensive subjects that augmentation index was volunteers reported to the laboratory at 7–8 a.m. correlated to DBP only in males and that this after an overnight fast to eliminate postprandial relation was lost when augmentation index was haemodynamic changes. During the investigation, normalised for age. There are also inconsistent subjects remained in the supine position. Each sub- results in studies only measuring mean blood ject was instrumented with a blood pressure cuff pressure. London et al 23 found augmentation index and tape electrodes to measure transthoracic im- to be correlated to mean arterial pressure in patients pedance. After a period of 30 min of complete rest, with end-stage renal failure. In contrast, Wilkinson haemodynamic measurements were performed. All et al 24 did not find a correlation between augmenta- measurements were taken by the same investigator. tion index and mean arterial blood pressure in subjects with diabetes mellitus type 1. There is also conflicting evidence in the literature regarding Anthropometric and haemodynamic parameters the relation between PWV and blood pressure. Some studies showed that PWV was only correlated Height and weight were measured and body mass to SBP.22,25 Others found that PWV was correlated index (BMI) was calculated as weight to height to both SBP and DBP18 or was correlated to SBP in squared. Blood pressure (mmHg) was measured both men and women, but was correlated to DBP with a standard mercury sphygmomanometer with only in women.26 the disappearance of Korotkow’s sound defined as The controversial results reported in the literature DBP. Pulse pressure (PP) was calculated by sub- may be attributed to the large variability in the stracting DBP from SBP. Mean arterial pressure population studied, such as the broad age range, (MAP) was defined as DBP plus one-third of PP. gender distribution and the presence of different Heart rate (HR) (bpm) was calculated from the RR- diseases or drug regimens. Additionally, age-related interval of the electrocardiogram, which was re- changes in the vasculature including a widening corded using a Siemens-Cardirexs multichannel ink of pulse pressure,27 changes in heart rate, blood jet recorder (Siemens Medizintechnik, Erlangen, pressure and vascular resistance28 may further Germany) as previously described.29,30 complicate the evaluation of stiffness determinants. We used impedance cardiography to measure These controversial results suggest that it is neces- central haemodynamics. Impedance cardiography sary to identify the determinants of arterial stiffness agrees closely with measurements obtained by markers in the absence of the potential contribution Doppler echocardiography or thermodilution,31 of age, gender or disease-related changes of the and is acceptable for clinical use particularly in vasculature. studies investigating young subjects free from any The purpose of the present study was to evaluate cardiovascular disease.32 In our laboratory, we have the role of blood pressure as a determinant of determined that the variability of stroke volume augmentation index and PWV in a homogeneous measured by impedance cardiography is less than population of young, healthy males. We also 5.5% (expressed as coefficient of variation).30 Stroke measured central haemodynamics including stroke volume (SV) (ml) was measured by impedance volume, cardiac output and total peripheral resis- cardiography using the standard approach with tance. We found that DBP but not SBP is strongly circular tape electrodes and graphical signal analy- correlated to both augmentation index and PWV. sis according to Kubicek’s equation.33 A ‘Kardio- Dynagraph’ was used to record changes in trans- thoracic impedance (Heinz Diefenbach Elektrome- Materials and methods dizin, Frankfurt, Germany). Cardiac output (CO) À1 Study population and protocol (1  min ) was calculated as CO ¼ HR  SV/1000. Total peripheral resistance (TPR) (dyne  s  cmÀ5) The study population included 77 young (23–35 was calculated as MAP divided by CO. years old), male volunteers who gave informed, Pulse transmission time (ms) was determined written consent. Only male subjects were selected noninvasively from pressure tracings over the because augmentation index has been shown to