|Home | About | Journals | Submit | Contact Us | Français|
Proximal aortic diameter, including aortic root (AoR) diameter, has been inversely related to pulse pressure (PP) in cross-sectional studies. So, investigators have hypothesized that a smaller AoR diameter may increase risk of developing hypertension. Prospective studies are lacking to test this hypothesis.
We measured AoR diameter in 3195 Framingham Study participants (mean age 49 years, 57% women; 8460 person-examinations) free from hypertension and prior cardiovascular disease who underwent routine echocardiography. We related AoR to hypertension incidence and blood pressure (BP) progression (increment of ≥1 category, as defined by the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure).
On follow-up (median 4 years), 1267 individuals (15%, 661 women) developed hypertension and 2978 participants experienced BP progression (35%, 1588 women). In logistic regression models adjusted for age, sex, and height, AoR was positively associated with hypertension incidence (odds ratio [OR], 1.15; 95% confidence interval [CI], 1.08–1.23) and BP progression (OR, 1.09; 95% CI, 1.04–1.14) on follow-up. However, adjustment for other factors known to influence BP tracking (baseline systolic and diastolic blood pressure, smoking, diabetes, and weight) rendered these relations statistically non-significant (OR, 1.03; 95% CI, 0.96–1.11 for hypertension incidence; OR, 1.03; 95% CI, 0.97–1.08 for BP progression).
In our large community-based sample of non-hypertensive individuals, AoR diameter was not associated with hypertension incidence or BP progression prospectively after adjustment for potential confounders. Our prospective study does not support the notion that a smaller AoR predisposes to hypertension.
High blood pressure (BP) is a major risk factor for cardiovascular disease (CVD), estimated to account for 7.1 million deaths per year worldwide.1 Globally, over 60% of cerebrovascular disease and almost 50% of coronary heart disease has been attributed to suboptimal BP.1 Recent guidelines have put emphasis on the importance of the prevention and reduction of the community burden of hypertension, given its public health importance and potential as a modifiable vascular risk factor.2 Therefore, it is important to elucidate the pathophysiological mechanisms underlying hypertension, and to discover new approaches for identifying individuals prone to hypertension development.
Given the public health importance of hypertension, considerable attention has been focused recently on the complex relations of the aortic root (AoR) diameter and hypertension risk. On one hand, dilation of the AoR diameter has been noted in individuals with hypertension, although the prevalence of such dilation in recent studies is low.3 A larger AoR is recognized also as a cross-sectional correlate for the presence of cardiac and extra-cardiac target organ damage,4, 5 and a predictor of incident CVD longitudinally.6 The aforementioned studies would seem to suggest that a larger AoR is a marker of higher vascular risk. Yet, on the other hand, a parallel set of cross-sectional studies have repeatedly demonstrated an inverse association between AoR diameter and BP measurements, especially pulse pressure (PP).3–10 These latter observations have generated the hypothesis that a smaller AoR may be a marker of higher blood pressure (BP) and possibly of future hypertension risk.11, 12 However, this hypothesis has been challenged,13 and there is an ongoing debate over whether proximal aortic diameter, including that of the root, is implicated in the pathogenesis of hypertension.11, 14, 15
The availability of longitudinal BP measurements as well as echocardiographic AoR dimensions in the Framingham Heart Study provides a unique opportunity to test the hypothesis that AoR diameter predicts the development of hypertension. Thus, we investigated the associations of AoR dimensions and longitudinal BP tracking (including the incidence of hypertension) in a large, community-based sample of men and women free of hypertension and prior CVD at baseline.
The selection criteria and design of the Framingham Offspring Study have been detailed previously.16 Approximately every four years, Offspring cohort participants undergo routine medical history, physical examination including BP measurement, anthropometry, and laboratory assessment of CVD risk factors at the Heart Study. At several examinations participants also underwent routine transthoracic echocardiography using a standardized protocol (see below).
Participants who underwent routine echocardiographic measurement of the AoR diameter at examination cycles 2 (1979–83), 4 (1987–91), 5 (1991–95) or 6 (1995–98) (referred to as baseline examinations of the present investigation) were eligible if they also attended the subsequent examination cycle (exam 3, 5, 6, and 7, respectively; referred to as follow-up examinations) approximately four years later (Figure). This sampling design was chosen to maximize the use of the wealth of echocardiographic and BP data, to test our study hypothesis with adequate statistical power to discern modest associations. Thus, the same individual could contribute multiple observations to the present investigation if he/she attended more than one examination (and the corresponding follow-up examination). Of the 4107 unique individuals who attended two consecutive examinations and had aortic root measurement available, 184 were excluded due to prevalent CVD (coronary heart disease, cerebrovascular disease, intermittent claudication, or congestive heart failure) and further 725 due to prevalent hypertension defined as a systolic BP ≥140 mm Hg or a diastolic BP ≥90 mm Hg, or use of anti-hypertensive treatment. Three individuals had missing height information resulting in a final sample of 3195 unique individuals contributing 8460 person-examinations to the final study sample (mean age, 49 years; 57% women). The Institutional Review Board at Boston University Medical Center approved the study, and all participants gave written informed consent.
All study participants underwent routine transthoracic echocardiography using a standardized protocol. The echocardiographic equipment used for image acquisition varied across the four baseline examinations: Hoffrel 201 ultrasound receiver (and Aerotech transducer) at examination cycle 2; Hewlett Packard (model 77020AC) ultrasound machine at examinations 4 and 5; a Sonos 1000 Hewlett-Packard machine at examination 6. The AoR diameter was measured from the M-mode tracings in accordance with the American Society of Echocardiography guidelines using a leading-edge-to-leading-edge measurement of the maximal distance between the anterior aortic root wall and the posterior aortic root wall at end diastole.17 The reproducibility of AoR measurements was systematically assessed at the sixth examination and was excellent.18
We examined the associations of baseline AoR diameter to the incidence of two blood pressure outcomes on follow-up (median 4 years): 1) incidence of hypertension, defined as a systolic BP ≥140 mm Hg, diastolic BP ≥90 mm Hg, or use of anti-hypertensive medication;2 2) blood pressure progression, defined as an increase in BP category on follow-up (according to the sixth report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC VI).19 For this purpose, non-hypertensive participants were allocated to a BP category at baseline (systolic BP <120 mm Hg and diastolic BP <80 mm Hg; systolic BP 120–129 mm Hg or diastolic BP 80–84 mm Hg; or systolic BP 130–139 mm Hg or diastolic BP 85–89 mm Hg).19 An increase of BP category on follow-up in an individual was defined as an increment of at least one category or development of hypertension.
Because it is conceivable that a smaller AoR diameter may be associated with very modest increments in BP on follow-up that may not be reflected by either the development of hypertension or BP progression, we also evaluated changes in systolic and diastolic BP, and pulse pressure on follow-up in additional analyses.
We used pooled repeated observations in multivariable logistic regression analyses to relate AoR diameter at baseline to hypertension incidence and BP progression on follow-up in models adjusted for age, sex and height; and models adjusted for age, sex, height, weight, baseline systolic and diastolic blood pressure, smoking, and diabetes.
We examined effect modification by testing the statistical significance of the two-way interaction term between age, sex, obesity, systolic BP (above versus below the median) and AoR diameter separately for the outcomes of hypertension incidence.
We performed additional analyses examining the relations of AoR diameter to longitudinal changes in systolic and diastolic BP, and pulse pressure analyzed as continuous variables; general estimating equations were used to account for repeated observations on some individuals and the algorithm described in Levy et al. was employed to adjust for antihypertensive treatment on follow-up.20, 21 These models were adjusted for the covariates in the fully adjusted model.
In secondary analyses, we examined associations of AoR diameter at baseline to hypertension incidence and BP progression on follow-up after additional adjustment for fractional shortening, stroke volume calculated using the Teicholz formula,22 total and high-density lipoprotein cholesterol, or mean arterial pressure, in separate models. Also, we performed analyses without adjustment for baseline BP. In further secondary analyses, we related AoR diameter at baseline to: incidence of isolated systolic hypertension; changes in systolic and diastolic BP after exclusion of individuals who were started on antihypertensive medication during follow-up; and incidence of BP outcomes over eight years of follow-up (instead of four years).
We had statistical power of 80% to detect an odds ratio (OR) effect of 1.12 (per SD increment in AoR diameter) and 90% power for an OR of 1.14 (at an alpha of 0.05). All analyses were performed using SAS 9.1 (SAS Institute, Cary, NC) and a two-sided P-value of <0.05 was considered significant.
The characteristics of our study sample are shown in Table 1. On follow-up (median, 4 years), 1267 individuals (15%, 661 women) developed hypertension and 2978 participants experienced BP progression (35%, 1588 women). There was a positive and statistically significant association of AoR diameter with hypertension incidence on follow-up in analyses adjusted for age, sex and height (Table 2, left panel). If adjusting for a set of standard CVD risk factors, the association was attenuated altogether. None of the interaction terms evaluated reached statistical significance (P>0.1 for all), suggesting that the association of AoR diameter with incident hypertension was not modified by age, sex, level of systolic BP or obesity.
Similarly, there was a highly significant association between AoR diameter and BP progression on follow-up in age-, sex- and height-adjusted analyses (Table 2, right panel). Again, when adjusting for additional CVD risk factors, the association was attenuated altogether.
In additional analyses, AoR diameter was not related to longitudinal changes in systolic (mean change 3.0 mm Hg, SD 11.9), diastolic BP (mean change 0.6 mm Hg, SD 7.9), or pulse pressure (mean change 2.4 mm Hg, SD 10.0) evaluated as continuous outcomes (P=0.33, 0.97, and 0.48, respectively).
In secondary analyses, there was a borderline significant association of AoR with hypertension incidence in models that did not adjust for baseline BP (OR, 1.07; 95% confidence interval, 0.99–1.14; P=0.052). All other secondary analyses (with additional adjustment for fractional shortening, stroke volume, total and high-density lipoprotein cholesterol, or mean arterial pressure; associations of AoR diameter with isolated systolic hypertension incidence, changes in systolic and diastolic BP after exclusion of individuals who were started on antihypertensive medication during follow-up, and analyses based on eight years of follow-up) were statistically non-significant (P>0.1 for all).
Prior studies have established the cross-sectional associations between AoR diameter and different BP measures in individuals from the general population,6, 7, 10 in hypertensives,3–5 and in individuals with Marfan syndrome.9 Several of these studies have found stronger positive associations of AoR diameter to diastolic BP,5, 6 or pulse pressure,4, 7 than to systolic BP. Consistently, they have shown direct relations of AoR diameter with diastolic BP,3, 5, 6, 10 and inverse relations with pulse pressure.4, 7, 10 In some of these studies, the relations were adjusted for height and weight,5–7, 10 which together with age and sex have been shown to explain the major part of the inter-individual variation in AoR diameter.10 The inverse relation of AoR diameter with pulse pressure was confirmed in a recent study in hypertensives.23 In that cross-sectional study, elevated pulse pressure was mainly attributable to decreased AoR diameter, measured above the sinotubular ridge, and increased wall stiffness. The authors speculated that inter-individual differences in AoR diameter, in addition to age-related tissue alterations in the aortic wall, may contribute to the pathogenesis of increased pulse pressure and isolated systolic hypertension. However, this hypothesis is controversial and has been debated much over the past few years.11, 13, 14 It has been suggested that only a prospective study could clarify the potential role (if any) of a smaller AoR in the development of hypertension.15 To our knowledge, there are no previous studies examining the longitudinal relation between AoR diameter and BP.
In our large, community-based sample of men and women free from hypertension and prior CVD at baseline, AoR diameter was related positively to incident hypertension and progression of BP on follow-up in minimally-adjusted analyses, but not after additional adjustment for standard CVD risk factors. There was a borderline significant positive association of AoR with hypertension incidence in multivariable models that did not adjust for baseline BP. However, the OR were slightly above 1, indicating an association in the opposite direction to the hypothesis we tested, i.e. that a smaller AoR predisposes to development of hypertension. It should be noted that it is conceivable that AoR diameter influences BP earlier in the life course, and any impact of AoR diameter on BP in our middle-aged sample may have already occurred, Studies of the relations of AoR and BP tracking in younger samples are needed to investigate this possibility.
The strengths of our study include the large community-based sample, the direct measurement of AoR diameter, and the longitudinal design which allowed us to evaluate temporal relations of AoR diameter and BP. There were several limitations of our study. Our sample consisted of middle-aged, white individuals of European decent, limiting the generalizability of our findings to other age groups and ethnicities. Also, the variability of BP measurements could have introduced errors in the stratification of participants into BP categories. Additionally, the change in blood pressure in our sample was small, potentially limiting our power to detect a relation between change in diameter and change in blood pressure. Aortic diameter was measured at the root at the level of the sinuses of Valsalva, which are geometrically complex, potentially limiting the precisions of our diameter measurements. Also, M-mode AoR diameter measurements can be less accurate and may underestimate true AoR diameter (as measured from 2-dimensional images), which also should be considered as a limitation of the present study. The sinuses are substantially larger than the adjacent tubular segment of aorta, just above the sinotubular junction. Since characteristic impedance of the aorta has a strong inverse relation to diameter (power of -2.5),24 it is sensitive to measurements of AoR diameter and the level of measurement. Thus, small discrepancies between AoR diameters at the level of the sinuses as compared to the tubular proximal aorta may be amplified, which may result in missing potentially important physiological effects of variations in aortic diameter measured at a slightly higher level above the root. Indeed, there is evidence to suggest that the proximal aortic root diameter (as measured routinely) is less associated with BP than the ascending aortic diameter.25
The present study fulfills an important gap in our understanding of the complex relations of AoR to BP. Prospectively, AoR diameter measured at the level of the sinuses of Valsalva was not associated with hypertension incidence or BP progression over time after adjustment for potential confounders in our young-to-middle aged large community-based sample of individuals free of hypertension and prior CVD at baseline. Our study does not support a causal role for AoR diameter in the pathophysiology of hypertension development in this age group. As noted above, we measured aortic diameter at the root, and it is conceivable that proximal ascending aortic dimensions may play a more important role in influencing the propensity for future hypertension. Additional studies that measure ascending aortic dimensions at multiple points and also simultaneously assess proximal aortic compliance may help elucidate the relations of aortic properties (structure and function) to the development of an elevated pulse pressure and hypertension.
SOURCES OF FUNDING
This work was by the Swedish Society of Medicine and the Swedish Heart-Lung Foundation (Dr. Ingelsson), and the NIH/NHLBI Contract No. N01-HC-25195, 6R01-NS 17950 (Dr. Benjamin), HL080124 and 2K24HL4334 (Dr. Vasan).
Dr. Mitchell is owner of Cardiovascular Engineering Inc, a company that designs and manufactures devices that measure vascular stiffness. The company uses these devices in clinical trials that evaluate the effects of diseases and interventions on vascular stiffness.