Abstract
Subclinical organ damage represents an intermediate stage in the continuum of vascular disease and a determinant of overall cardiovascular risk. We investigated the associations of pulse wave velocity (PWV), ambulatory arterial stiffness index (AASI), and office pulse pressure (PP) with several target organ damages (TODs) in newly diagnosed and never-treated patients with essential hypertension with respect to their dipping profile.
One hundred sixty-eight hypertensive patients with recently diagnosed and never-treated stage I–II essential hypertension were evaluated with respect to the relationship of PWV, AASI, and office PP with TOD including microalbumin (MAU) levels, cognitive function, intima-media thickness (IMT), coronary flow reserve (CFR), left ventricular mass (LVM), left ventricular filling pressures, diastolic dysfunction, and left atrium (LA) enlargement.
Simultaneous estimation of AASI, PWV, and office PP independently associated with the following: (i) CFR (P < 0.01), 24-h urine albumin excretion rates (P < 0.05), left ventricular diastolic dysfunction (P < 0.01), and LA enlargement (P < 0.01) in never-treated hypertensive patients; (ii) CFR (P < 0.05), IMT (P < 0.01), left ventricular diastolic dysfunction (P < 0.05), and LA enlargement (P < 0.05) in dippers; and (iii) CFR (P < 0.05) and LA enlargement (P < 0.01) in nondippers. Nonindependent relationships revealed between (i) AASI and left ventricular filling pressures and (ii) PWV and cognitive dysfunction in never-treated hypertensive patients.
The simultaneous estimation of three noninvasive indexes of arterial stiffness leads to valuable information regarding their association with TOD including CFR, MAU levels, IMT, left ventricular diastolic dysfunction, and LA enlargement in never-treated hypertensive patients regarding their dipping status.
Signs of subclinical organ involvement, otherwise called target organ damage (TOD), should be investigated thoroughly in hypertensive patients. Subclinical organ damage represents an intermediate stage in the continuum of vascular disease and a determinant of overall cardiovascular risk. TODs include left ventricular hypertrophy, left ventricular diastolic dysfunction, left atrium (LA) enlargement, impaired coronary flow reserve (CFR), increased intima-media thickness (IMT), elevated levels of microalbuminuria, and cognitive impairment.1 Ambulatory blood pressure (BP) measurement may reveal nondipper hypertensive patients. Nondipping status, defined as a reduced nocturnal decline in BP, is another noninvasive determinant of cardiovascular events.2
Arterial stiffness represents an independent predictor for all-cause and cardiovascular mortality and morbidity in patients with essential hypertension, diabetes, or end-stage renal disease. Arterial stiffness can noninvasively be estimated by measuring carotid–femoral pulse wave velocity (PWV), that is, the velocity of the pulse wave to travel a given distance between two sites of the arterial system.3,4 Ambulatory arterial stiffness index (AASI) is a recently proposed indicator of arterial stiffness that is derived from ambulatory BP recordings.5 Although AASI has only limited reproducibility, it improves risk stratification.6 AASI has been shown to strongly correlate with several measures of arterial stiffness including PWV and augmentation index, and to provide prognostic information on cardiovascular mortality and TOD.7 On the other hand, large clinical trials have shown that office pulse pressure (PP), defined as systolic BP (SBP) minus diastolic BP (DBP), is a strong predictor of coronary events in patients with essential hypertension.8
An association between increased indexes of arterial stiffness and left ventricular diastolic dysfunction or elevated left ventricular filling pressures has been shown in patients with coronary artery disease, diabetes, obesity, and hypertension.9,10,11,12 PWV has been related with several TODs in patients with newly diagnosed essential hypertension.13,14,15 In untreated hypertensives, as well as in patients with type 1 diabetes, AASI has been shown to relate with microalbuminuria and carotid stiffness index.16 PP may predict LA size at a very early stage of hypertension disease in nondipper patients.17 In the present study, we aimed to investigate any existing associations of PWV, AASI, and office PP with a number of TODs in newly diagnosed and never-treated patients with essential hypertension with respect to their in-dipping profile.
Methods
Study population. We studied 178 hypertensive, consecutive, nondiabetic patients (mean age 53 ± 12 years, 89 men) visiting the outpatient clinic of our department between January 2006 and December 2008 with recently diagnosed and never-treated stage I–II essential hypertension.1 Ten patients were excluded from the study because of the incomplete data. Patients reported that their BP (SBP, DBP, or both) was found elevated either by medical personnel during a routine annual checkup or by themselves incidentally. Patients were subjected to the following examinations within 2 weeks from the baseline visit: three office BP measurements in each one of the three subsequent visits in the hypertension outpatient clinic; 24-h ambulatory BP monitoring (ABPM); blood and urine sampling for routine blood chemistry and urine examination; standard 12-lead electrocardiogram; transthoracic echocardiogram; carotid–femoral PWV measurement; microalbumin (MAU) levels in 24-h urine collection; and Mini Mental State Examination (MMSE) as a screening test for global cognitive impairment. Informed consent had been obtained during the initial visit of the study that was approved by the ethical committee of the hospital.
Patients with white-coat hypertension, secondary hypertension, congestive heart failure, previous myocardial infarction, cardiomyopathy, stroke, cardiac valve diseases, history of coronary artery bypass grafting, atrial fibrillation, renal insufficiency, overt proteinuria, diabetes mellitus, and those patients under medication for noncardiovascular diseases were excluded from the study. All patients underwent a noninvasive test (cardiopulmonary exercise test) to exclude myocardial ischemia. None of the patients was on treatment with statin or cardioactive medications, and none of the female patients was on hormone replacement treatment.
Details regarding methodology have been described in previous studies.13,14,17
Summarizing, morning office BP was measured approximately in the same morning hour of the day, by the same cardiologist with a mercury sphygmomanometer (first and fifth phases of Korotkoff sounds taken as SBP and DBP, respectively) after the patients had rested for a period of 5–10min in the sitting position. Three measurements were taken at 1-min intervals, and the average was used to define clinic SBP and DBP. ABPM was carried out on the nondominant arm using the valid recorder Spacelab 90207 (General Electric Health Care, Berlin, Germany). Dipping status was defined as the BP, both SBP and DBP, decrease >10% at night.18 The ABPM device was set to obtain BP readings at 15-min intervals during the day (07:00–23:00) and at 20-min intervals during the night (23:00–07:00). All patients had >75% of successful readings. We estimate AASI using all the 24-h ambulatory BP recordings of each subject, DBP was plotted against SBP. We defined AASI as one minus this regression slope. The stiffer the arterial tree, the closer the regression slope and AASI are to 0 and 1, respectively.6 As AASI depends on nocturnal BP fall, an interval of 30min between nocturnal readings was considered necessary for AASI calculation.
Left ventricular diastolic function was assessed by measuring peak E and A waves of transmitral flow (E/A ratio). Tissue Doppler imaging was performed in order to measure peak Ea and Aa waves and subsequently Ea/Aa ratio at the endocardial portion of the basal site of the lateral wall on the apical four-chamber view of the left ventricle. E/Ea ratio represented an index of left ventricular filling pressures. Left ventricular mass (LVM) and left ventricular mass index (LVMI: LVM/BSA) were measured in all patients using the Devereux formula according to the Penn Convention Protocol.19 LA enlargement was defined as LA diameter ≥39mm.20 IMT was calculated using B-mode ultrasound imaging. The ratio of VTI-Vd (time integral of peak diastolic coronary blood flow velocity) after adenosine infusion (140g/kg/min) to VTI-Vd at baseline during color-guided Doppler echocardiography was used to access CFR. Impaired CFR was defined as <2.5m/s. Arterial stiffness was estimated by PWV measurement using Complior SP (Artech Medical, Pantin, France), a computerized device that permits automatic calculation of PWV. Time delay between the recorded carotid and femoral arterial waves was recorded, whereas distance separating the transducers was superficially measured resulting in PWV calculation as the average of at least 10 cardiac cycles.21 MAU was analyzed by nephelometry (immunochemical assay, BN Prospec; Dade Behring, Marburg, Germany). The normal level of urine albumin excretion is <30mg/24h. Folstein's MMSE was used as a measure of global cognitive dysfunction. Patients were tested during a morning visit in a silenced room by an experienced neurologist. A full score on the MMSE is 30, higher scores indicate higher function, whereas cognitive impairment was defined as a score of ≤27 and cognitive dysfunction as a score of <24 (ref. 22).
Statistical analysis. All variables are expressed as mean ± s.d., or median and interquartile range regarding their normal distribution or not, respectively. Variables were tested by the Kolmogorov–Smirnov test to assess the normality of distribution. All variables except MMSE, IMT, and MAU were normally distributed. The latter ones (MMSE, IMT, and MAU) were analyzed after transformation into ranks. Both unpaired Student's t-test and χ2-test were used in order to compare numeric or not differences within groups. Simple linear regression was used to identify the relations between variables. Multiple linear regression analysis using stepwise procedure was performed in order to explore multiple linear relations between arterial stiffness indexes (AASI, PWV, and 24-h PP) and TOD (MMSE, MAU, IMT, CFR, E/A, E/Ea, LAH). Age, sex, body mass index, atherosclerotic factors (clinic or 24-h BP, smoking, dyslipidemia) were forced into the regression analysis model as independent variables, to take into account any possible relationship with the examined dependent variables. In order to avoid collinearity between office and ambulatory parameters, SBP, mean BP, and PP (model A) as well as ambulatory SBP and ambulatory PP (model B) were entered in the multivariate model separately. Associations are presented by means of correlation coefficient (B) and P value. The level of statistical significance was determined as a P value <0.05. Statistical analysis was performed on a SPSS 13 version (SPSS, Chicago, IL).
Results
Study population characteristics
Total population was divided into two groups according to dipping profile: dippers (n = 80, mean age 51 ± 12 years), and nondippers (n = 88, mean age 56 ± 11 years). Demographic, clinical, and echocardiographic characteristics of all hypertensive patients, dippers, and nondippers are listed in Table 1.
Determinants of arterial stiffness indexes
AASI was related with the following: (i) 24-h PP, body surface area (BSA), 24-h urine albumin excretion rates, and E/Ea ratio (all hypertensives), (ii) IMT (dippers), and (iii) 24-h PP, ABPM dia, and smoking habit (nondippers).
PWV was related with the following: (i) office PP, BP sys, 24-h PP, ABPM sys, age, CFR, and E/A ratio (all hypertensives, dippers, nondippers); (ii) BP mean and LA divided by height (LAH) (all hypertensives, dippers); (iii) BMI and MMSE (all hypertensives, nondippers); (iv) 24-h urine albumin excretion rates and LAH (dippers); (v) E/Ea ratio (all hypertensive patients).
Office PP was related with the following: (i) PWV, 24-h PP, BP sys, BP mean, ABPM sys, heart rate, age, CFR (all hypertensives, dippers, nondippers); (ii) BSA, E/A ratio, and LAH (all hypertensives, nondippers) in all hypertensive patients; (iii) smoking, IMT, and E/Ea ratio (dippers). All the data mentioned above are presented in Supplementary Table S1 online.
Independent association of TODs to arterial stiffness indexes
By multivariate linear regression analysis in regard to all hypertensives, AASI was independently associated with 24-h urine albumin excretion rates (model A). PWV was independently associated with CFR, E/A, LAH (model A and model B). Office PP was independently associated with LAH (model B) (Table 2).
Multiple regression analysis of independent associations between arterial stiffness indexes (AASI, PWV, PP) and target organ damage in hypertensive patients regarding their dipping status
Multiple regression analysis of independent associations between arterial stiffness indexes (AASI, PWV, PP) and target organ damage in hypertensive patients regarding their dipping status
In dippers, AASI was independently associated with IMT (model A and model B). PWV was independently associated with CFR, LAH (model A) and CFR, E/A ratio (model B). Office PP was independently associated with IMT (model A and model B) (Figure 1).
Independent relationships of arterial stiffness indexes (a, AASI; b, office PP; c, PWV) with target organ damages (IMT, CFR, LAH, E/A) in dipper untreated hypertensive patients. AASI, ambulatory arterial stiffness index; CFR, coronary flow reserve; E/A, transmitral E wave divided by A wave; IMT, intima-media thickness; LAH, left atrium divided by height; PP, pulse pressure; PWV, pulse wave velocity.
In nondippers, PWV was independently associated with CFR (model A and model B). Office PP was independently associated with LAH (model B).
Discussion
There is increasing interest in developing noninvasive methods that can easily be applied in clinical practice to measure arterial compliance that in turn plays an important role in the pathogenesis of cardiovascular events. In the present study, we demonstrated that each one of the noninvasive arterial stiffness indexes studied (AASI, PWV, office PP) was independently associated with one or more TODs in never-treated patients with essential hypertension with respect to their dipping status.
ABPM is used more and more in clinical practice for the diagnosis and management of hypertension.1 It has been recently demonstrated that nondipping is a marker of TOD and a predictor for cardiac and cerebrovascular risk in hypertensive and normotensive subjects. Although dipping prevents damage to brain, kidneys, heart, and blood vessels, nondipping is associated with increased morbidity and mortality.23
A novel method for estimating arterial stiffness, named AASI, based on ABPM was recently proposed.5 Strong evidence exists from cross-sectional analysis and prospective cohort studies that AASI predicts cardiovascular risk within the normotensive and hypertensive range.7,24 AASI is associated with signs of subclinical TOD in patients with primary hypertension. Untreated hypertensive patients with microalbuminuria, carotid abnormalities, or left ventricular hypertrophy showed higher ambulatory arterial stiffness index as compared with those without it. A strong correlation was found between AASI and microalbuminuria. The greater the arterial stiffness the higher the prevalence and degree of renal damage. In a randomly recruited European population, the AASI was a strong predictor of stroke, beyond traditional cardiovascular risk factors, including the mean arterial pressure and PP.16,24,25 AASI is strongly dependent on the degree of nocturnal BP fall in hypertensive patients as it reflects the dynamic relation between DBP and SBP through the whole day and is increased in nondipping independent of BP.26,27 In our study, a positive relationship between AASI and 24-h urine albumin excretion rates as well as left ventricular filling pressures was revealed. Moreover, AASI independently associated with 24-h urine albumin excretion rates in all hypertensive patients. When dipping status was taken into account, we found that AASI was independently associated with intima-media thickness in dipper hypertensive patients. However, no correlation between AASI and TOD was found in nondippers.
In several studies, a relationship between PWV and diastolic dysfunction, CFR, carotid intima-media thickness, cognitive dysfunction, and microalbuminuria has been revealed in patients with newly diagnosed essential hypertension.13,14,15 In our study, those results were confirmed, whereas an independent relationship of PWV with CFR, left ventricular diastolic dysfunction, and LA enlargement was revealed. When dipping status was taken into account, only CFR was associated by PWV in nondippers, whereas PWV associated with CFR, left ventricular diastolic dysfunction, and LA enlargement in dippers. It seems that in patients with increased 24-h hypertensive burden and subsequent augmented cardiovascular risk, nondippers, a different TOD pattern compared to dippers is observed.
Office PP, as an indicator of arterial stiffness, is a strong predictor of coronary events in patients with essential hypertension.8 Furthermore, both the degree of nocturnal BP fall and the LA size are positively associated with increased PP.17,23,28 In our study, it was shown that in untreated hypertensives, either dippers or not dippers, PP was not independently related with CFR. It seems that comparing these two indexes of arterial stiffness (PWV and PP), only PWV has the power, at least in our study, to independently associate with impaired CFR in untreated hypertensive patients, independently of their dipping status.
In the Dublin Outcome Study, comparing different indexes of arterial stiffness, it was found that AASI was a stronger predictor of stroke mortality than PP and a better predictor of cardiovascular mortality among normotensive than among hypertensive subjects, whereas PP was a superior prognostic marker in hypertensive patients.7 Furthermore, Hansen et al. showed that AASI, but not PWV, predicted stroke, whereas the opposite was true for the composite of all cardiovascular events in a general population.29
In our study, an excellent correlation was revealed between PWV and office PP in all three studied groups. However, it is interesting that we did not manage to find any correlation between AASI and PWV or office PP in either group of never-treated patients studied (all hypertensives, dippers, nondippers). It is known that Li et al. found a significant relationship between AASI and carotid to femoral PWV in a group of normotensive Chinese volunteers.5 Schillaci et al. also reported that AASI is only weakly related to a widely used index of aortic stiffness, such as PWV, in a group of untreated hypertensive patients. This relationship was not independent and it was importantly affected by other factors.27 Finally, it was found that the association of AASI and brachial PP was restricted to dippers in a population that was evaluated for kidney donation.8 It seems that further studies are needed in order to explore the relationship between AASI and PWV or office PP. AASI might not be a precise index of arterial stiffness, but rather a BP component just predicting cardiovascular outcome.24 Although both ambulatory PP and AASI may be regarded as surrogates of arterial stiffness, these parameters have different physiological meanings than the arterial stiffness itself and should be considered at best as surrogates of other direct measures of arterial stiffness, such as aortic PWV, which remains the gold standard in order to measure arterial stiffness.30
In conclusion, the simultaneous estimation of three noninvasive indexes of arterial stiffness leads to valuable information regarding their association with TOD including CFR (PWV), 24-h urine albumin excretion rates (AASI), intima-media thickness (AASI, PP), left ventricular diastolic dysfunction (PWV, PP), and LA enlargement (PWV, PP) in never-treated hypertensive patients as well as in dippers and nondippers. We propose that all three tests (AASI, PWV, and office PP) might be used to screen newly diagnosed and untreated hypertensive patients, as we do in our hypertension clinic, because the amount of information regarding TOD significantly increases.
Study limitations
One of the limitations of this cross-sectional study is the relatively low number of patients in each group. Because hypertension has a high prevalence in population, a greater number of patients should be necessary in order to generalize the results of this study in untreated hypertensives, treated hypertensives with medications that affect arterial stiffness calculation, or patients with severe hypertension.
Additionally, no cause–effect relationship between AASI, PP and PWV (as arterial stiffness indexes) and TOD could be really concluded as our study has a cross-sectional design.
Finally, due to the increased number of multiple regression models performed, some of the significant statistical findings may occur due to chance.
Disclosure
The authors declared no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
