Left ventricular remodeling and arterial afterload in older women with uncontrolled and controlled hypertension

Abstract

Objective

The prevalence of hypertension increases with advancing age in women. Blood pressure control is more difficult to achieve in older women and despite well-controlled blood pressure, the cardiovascular mortality remains high. However, the underlying mechanisms are not understood.

Methods

19 women with uncontrolled hypertension on drug treatment (70±2 [SE] yrs, ambulatory awake blood pressure; 152±2/84±2 mmHg), 19 with controlled hypertension (68±1 yrs, 128±2/71±2 mmHg), and 31 healthy normotensive women (68±1 yrs, 127±1/73±1 mmHg) were recruited. Participants were weaned from antihypertensive drugs and underwent 3 weeks of run-in prior to cardiac-vascular assessments. Left ventricular morphology was evaluated with cardiac MRI. Arterial load and vascular stiffness were measured via ultrasound and applanation tonometry.

Results

Left ventricular mass normalized by body surface area was not different between hypertension groups (uncontrolled vs. controlled: 50.0±1.7 vs. 51.8±2.3 g/m²), but it was lower in the normotensive group (41.7±0.9 g/m², one-way ANOVA P=0.004). Likewise, central pulse wave velocity was not different between hypertension groups (11.5±0.6 vs 11.1±0.5 m/s), lower in the normotensive group (9.1±0.3m/s, one-way ANOVA P=0.0001). Total peripheral resistance was greater in uncontrolled HTN compared to normotensive group (2051±323 vs. 1719±380 dyn•sec•cm⁻⁵), while controlled HTN group (1925±527 dyn•sec•cm⁻⁵) was not different to either groups.

Conclusion

Regardless of current blood pressure control, hypertensive older women exhibited increased cardiac mass and arterial stiffness compared to normotensives. Future large-scale longitudinal studies are warranted to directly investigate the mechanisms for the high cardiovascular mortality among older hypertensive women with well-controlled blood pressure.

INTRODUCTION

During early and middle adulthood, men have a greater prevalence of hypertension (HTN) compared to women, but the prevalence and severity of HTN increase distinctly in women after menopause^1–3. Indeed, the prevalence of HTN increases twice after menopause, independent of other traditional risk factors^{4, 5}. Furthermore, after six decades of their lives, the prevalence of HTN-related cardiovascular complications is greater in older women compared with age-matched men and these are the leading cause of death in women^{6, 7}. Despite remarkable advances in pharmacotherapy, blood pressure (BP) control remains difficult to achieve particularly in older women. The Framingham Heart Study revealed a sex difference in BP control rates reporting an age-associated gradual decrease in BP control rates was more distinctive in women than in men.^8–11.

More importantly, it is noteworthy to mention that clinical focus is overwhelmingly on BP control of hypertensive individuals. Although many epidemiological reports have shown the BP control with antihypertensive medications profoundly reduce the risk of CVD in the HTN populations, several prospective observational studies revealed that the cardiovascular mortality rate remains high in HTN women despite adequate BP control^12–14. These counterintuitive clinical outcomes suggest that factors associated with the prevention of cardiovascular events in this population may be vastly complex and beyond BP control. However, little information is available from studies in humans on the physiological mechanisms involved.

The heart and vasculature are the main components of the human cardiovascular system, and interplay between cardiac and arterial factors determines the efficiency of the circulatory system¹⁵. The arterial afterload provides the main resistance to the left ventricle (LV) to generate the flow to the circulatory system and influences the magnitude of energy transmitted from the heart to peripheral circulation. Under conditions of chronically increased afterload, the LV is exposed to constant mechanical stress leading to a LV hypertrophy. Previous findings indicated a close association between LV hypertrophy and cardiovascular outcomes in not only the general population but also hypertensive patients^16–18 and there was positive relationship between LV mass and the severity of HTN^{19, 20}. LV mass predicts future risk of cardiovascular events in both men and women but a larger relative risk associated with LV hypertrophy was found in women than in men^{19, 21}. Clinical trials have shown that regression of LV mass with antihypertensive drug treatment leads to a decrease in cardiovascular events^{22, 23}. In addition, arterial load is closely associated with properties of large conduit arteries and the LV as a pump. Arterial load, composed of resistive and pulsatile components, is an important determinant of myocardial oxygen consumption and a key stimulus for myocardial hypertrophy^{24, 25}. It has been demonstrated that the age-related stiffening of large arteries is more pronounced in women compared with their male-counterparts^26–28. The stiffer large arteries likely contributed to the greater prevalence of hypertension and its associated cardiovascular complications in older women^26–28. To our knowledge, however, control of BP by antihypertensive drugs and its impact on unfavorable LV remodeling in older women remains grossly understudied, and comparative contributions of arterial load to LV in hypertensive older women is unknown.

Therefore, the purposes of this cross-sectional study was to compare the LV remodeling and arterial properties in hypertensive older women with uncontrolled and controlled HNT, as well as in older normotensive women. We hypothesized that current blood pressure control by antihypertensive agents would not normalize cardiac morphology and arterial afterload in older women with hypertension.

METHODS

Participants

We studied 19 uncontrolled hypertensive, 19 controlled hypertensive and 31 normotensive older women. Uncontrolled HTN is defined as ambulatory awake systolic pressure ≥ 140 mmHg and/or diastolic pressure ≥ 90 mmHg under stable antihypertensive medication regimens and controlled HTN is systolic pressure lower than 140 mmHg and/or diastolic pressure lower than 90 mmHg on drug treatment. Volunteers with ambulatory awake systolic pressure ≥ 160 mmHg and/or ≥100 mmHg were excluded for safety reasons. We recruited participants who were between 60-85 years old, free of overt cardiovascular and other clinical diseases, diabetes mellitus (fasting glucose ≥126 mg/dl or 2-hour oral glucose tolerance test; glucose level ≥200 mg/dl) and chronic kidney disease (serum creatinine >1.5 mg/dl), assessed by medical history, physical examination, resting ECG, urinalysis and blood chemistries. All participants were postmenopausal, defined as the absence of menses for at least 2 years and all were not on hormone therapy for at least 6 months prior to study participation.

After screening, hypertensive women were weaned progressively from their drugs (wash-out period) to eliminate the confounding influence of various antihypertensive medications, which was followed by a 3-week run-in. During the run-in period, all participants were required to maintain a healthy lifestyle, according to the Eight Report of the Joint National Committee standard guidelines²⁹. Twenty-four hour ambulatory BP measurements (ABPM) (Oscar 2; SunTech Medical Instruments) were performed at the screening visit (before drug wash-out) and immediately after the end of the run-in period to assure the effect of antihypertensive agents were eliminated before data collection. We believe that 3-week run-in with no drug treatment would not dramatically change arterial properties and LV morphology, since cardiovascular remodeling requires at least 4 months to occur³⁰.

The study was approved by the Institutional Review Board of the University of Texas Southwestern Medical Center and Texas Health Presbyterian Hospital Dallas, and conformed to the ethical standards of the Declaration of Helsinki. Prior to obtaining written informed consent, the purpose, nature and risk of the study procedures were explained to the participants and questions were addressed.

Cardiac magnetic resonance imaging

To evaluate the geometric cardiac remodeling we used magnetic resonance imaging (MRI) which is the recommended method for clinical trials investigating LV mass regression³¹. Cardiac MRI examinations were performed using a 1.5 T clinical MRI scanner (Gyroscan Intera, Philips). Short axis slices of the heart images were obtained in 6 mm thickness with 4 mm gap. Measurements included LV mass, end-diastolic (LVEDV) and end-systolic volumes (LVESV), and mass-to-volume ratio (LVR) by an experienced investigator who was blinded to participant characteristics. Obtained images were analyzed using the semiautomatic software tool of QMass V.7.2 (Medis Medical Imaging Systems, the Netherlands). Left ventricular wall thickness (LVWT) was measured on short axis images at the midventricular level (papillary muscle level) in the anterior, inferior, lateral and septal region. Cardiac MRI measurements were normalized by body surface area (BSA) and expressed as LV mass index (LVMI), LVEDV index (LVEDVI), LVESV index (LVESVI) and LVWT index (LVWTI). All measurements were assessed at end-diastole, which was identified as the frame with the largest LV volume. The typical error of the within-individual variability in the manual planning of cardiac MRI was 1.6% for LVMI and 1.3% for LVEDVI, and the between-individual typical error was 4.6% for LVMI and 4.1% for LVEDVI in our laboratory³².

Pulse wave velocity

We used carotid-to-femoral (central PWV) and carotid-to-radial pulse wave velocity (peripheral PWV), the gold standard measurement of arterial stiffness, as indexes of central and peripheral arterial stiffness using the SphygmoCor MM3 system (AtCor Medical). central PWV was determined by recording pressure pulse waves at the right carotid and femoral arteries using a high-fidelity micromanometer (Millar Instruments, Houston, TX) and calculating the distance between the recording sites divided by the time delay between the carotid and femoral pulse waves. peripheral PWV was determined from the right carotid and radial waveforms. The distance was measured along the body surface using a non-stretchable tape. The straight distance from the carotid recording site to the suprasternal notch was subtracted from the suprasternal notch to the femoral recording site in the calculation of central PWV. The distance between the suprasternal notch and the carotid site added to suprasternal notch to the radial site in the calculation of peripheral PWV. At least 3 measurements were repeated, and the average of the 3 highest quality recordings were used for analysis.

Wave reflection and left ventricle afterload

Arterial wave reflection was assessed using the SphygmoCor system. Radial artery pressure waveforms were recorded by applanation tonometry with a high fidelity micromanometer. A generalized transfer function, which has been validated both intra-arterially and noninvasively, was used to generate the aortic pressure waveform. The aortic pressure waveform provided the measurements of aortic BPs, central aortic pulse pressure (CPP), augmented pressure (AP; the difference between the first and second systolic shoulders of the aortic systolic BP), augmentation index (AIx), and AIx adjusted for a heart rate of 75 beats/min (AIx @ 75). LV wasted pressure energy (LVEw)^{33, 34} is a component of extra myocardial oxygen requirement that is a result of early systolic wave reflection. LVEw was estimated as [1.333 × (AP × Δt_r) × π/4] with 1.333 as the conversion factor for mmHg/s to dynes s/cm and Δt_r is the systolic duration of the reflected wave. At least 3 measurements were repeated, and the average of the 3 highest quality recordings, established as an in-device operation quality index of >90% were used for analysis.

Cardiac output and arterial afterload

Brachial cuff BP was measured by electrosphygmomanometry (model 4240; suntech Medical Instruments) with a microphone placed over the right brachial artery to detect Korotkoff sounds. Heart rate was recorded from lead II of the electrocardiogram. At least 20 min post instrumentation, cardiac output was measured by of the modified acetylene rebreathing method^{35, 36}. Cardiac output and supine BPs were measured in 5 min interval, at least three times and the average value was recorded. All these variables were normalized by BSA. Total peripheral resistance (TPR) was determined as 80 × mean BP divided by cardiac output. Effective arterial elastance (Ea) was defined as brachial SBP × 0.9 divided by stroke volume. Total arterial compliance (TAC) was calculated as stroke volume divided by pulse pressure.

Arterial images

Image of the right common carotid artery was collected using an ultrasound/Doppler system (iE33, Philips) equipped with a high-resolution transducer (11 MHz) positioned at 2 cm proximal to the carotid bifurcation with simultaneous ECG recordings. Diameters were measured using image-analysis software (QLAB, Philips). Systolic and diastolic areas were averaged over 4 continuous beats. The carotid pressure waveforms were obtained with a tonometry transducer and the signal was calibrated by the brachial diastolic and mean arterial pressure measured by arm cuff. Carotid arterial compliance, a measure of arterial buffering capacity³⁷, was calculated as: C=[(D1-D0)/D0]/[2(P1-P0)]π(D0)², where D1 and D0 are the maximal and minimal carotid diameter, and P1 and P0 are the highest and lowest aortic pressures. Carotid arterial distensibility, a measure of arterial elastic properties³⁸, was calculated as: [2(D1-D0)/D0]/(P1-P0).

Study procedures

All measurements were performed in the morning ≥ 2 hours after light meal, in a quiet and environmentally controlled laboratory with an ambient temperature of ≈ 25°C (day 1). Participants abstained from alcohol, caffeine, medication use and strenuous physical activity for at least 24 hours prior to data collection. At least 20 minutes after supine resting, cardiac output measurements were performed followed by arterial stiffness measurements. Central (carotid) and peripheral (brachial) BP were obtained using tonometry and arm cuff at the brachial artery, and central and peripheral PWV were assessed. Ultrasonography on the common carotid artery was performed for the assessment of carotid artery compliance. Cardiac MRI (for the assessment of cardiac morphology) was performed at the University of Texas Southwestern Medical Center at Dallas on a separate day (day 2). The MRI testing occurred at a similar time of the day as day 1 to eliminate possible diurnal variations.

While all participants (19 uncontrolled, 19 controlled, and 31 normotensives) successfully completed the other assessments during the procedure, not all but most of participants (16 uncontrolled HTN, 15 controlled HTN, and 29 normotensive) completed cardiac MRI measurement.

Statistical analysis

Values are expressed as means ± SEM. Statistical analysis was performed using SPSS Statistics (Version 23, Chicago IL) and statistical significance was set at P< 0.05. To examine differences among three groups, one-way ANOVA was used. The sample size calculation for a one-way ANOVA has been made based on the preliminary data obtained from our laboratory. We assumed that the minimum difference of LV mass index to be detected was 5 g/m² and that the estimated standard deviation was 4.8 g/m². We would need n=19 per group to detect the difference among the groups with power = 0.80 and alpha=0.05. To examine the interaction between group (uncontrolled vs. controlled vs. normotensive) and ABPM (screening vs. run-in), a two-way (2×2) mixed factorial ANOVA with repeated measures was used. Post hoc multiple comparisons were adjusted using the Bonferroni correction. To examine relations of interest, Pearson product-moment correlation coefficients were used.

RESULTS

Participant characteristics are presented in Table 1. Age, height, and weight were not different among the groups. Also, there were no group differences in body mass index and BSA. The percentage of antilipidemic and antihypothyroidic medication use was similar among the groups. Table 2 contains the percentage of BP medication use in uncontrolled and controlled HTN women, which shows similar results between the groups. Four participantes in the uncontrolled HTN group and 2 in the controlled HTN group were excluded during the wash out and/or run-in period.

Table 1.

Participant characteristics

	Normotensives	Uncontrolled HTN	Controlled HTN	P Value
n	31	19	19
Age, years (range)	68±1 (60–82)	70±1 (60–84)	68±2 (60–79)	0.6
Height, cm	162.5±1.1	163.1±1.1	159.3±1.1	0.07
Weight, kg	71.0±2.2	74.4±2.4	72.8±2.9	0.6
BMI, kg/m2	26.8±0.7	28.1±1.0	28.7±1.1	0.3
BSA, m2	1.78±0.03	1.83±0.03	1.79±0.04	0.6
Medication
Antilipidemic, n	5 (26%)	6 (32%)	4 (21%)
Antihypothyroidic, n	3 (10%)	3 (16%)	2 (11%)

Data are mean±SE. P value; one way ANOVA, HTN: hypertension; BMI: body mass index; BSA: body surface area

Table 2.

Antihypertensive drug use

	Uncontrolled HTN	Controlled HTN
Beta Blockers, n	4 (21%)	4 (21%)
Diuretics, n	7 (37%)	8 (42%)
ACE inhibitors, n	3 (16%)	4 (21%)
ARBs, n	5 (32%)	4 (21%)
CCBs, n	3 (16%)	3 (16%)
Combination Treatment, n	3 (16%)	4 (21%)

HTN: hypertension; ACE: angiotensin-converting-enzyme; ARBs: angiotensin receptor blockers; CCBs: calcium channel blockers.

Ambulatory blood pressure

ABPM data are presented in Table 3. At the screening visit, ambulatory awake systolic and diastolic BP were similar between controlled HTN and normotensive women but greater in uncontrolled HTN women compared to both groups (P=0.001). After the wash-out and 3-week run-in period, both ambulatory awake systolic and diastolic BP were elevated in controlled HTN women (P<0.05) but remained the same in uncontrolled HTN women (P=0.5). Post run-in ABPM results were greater in the uncontrolled and controlled HTN groups compared to the normotensive group. Asleep ABPM results were similar to those of awake measurements.

Table 3.

Ambulatory BP at the screening visit and immediately after run-in

	Normotensives		Controlled HTN		Uncontrolled HTN
	Screening	Run-in	Screening	Run-in	Screening	Run-in
Awake SBP (mmHg)	127±1†	124±1Ψ	128±2†	150±2*	152±2	154±2
Awake DBP (mmHg)	73±2†	72±2Ψ	71±2†	83±2*	84±2	85±2
Sleep SBP (mmHg)	112±2†	110±2Ψ	115±3†	133±3*	134±3	138±3
Sleep DBP (mmHg)	62±2†	60±2Ψ	61±2†	72±2*	70±2	72±2

Data are mean±SE; HTN: hypertension; SBP: systolic blood pressure; DBP: diastolic blood pressure.

^*P≤0.01 vs. screening within the same group;

^ΨP≤0.01 vs. uncontrolled and controlled HTN;

^†P≤0.01 vs. uncontrolled HTN.

Cardiac MRI measurements

The uncontrolled and controlled HTN groups showed greater LVMI compared to the normotensive group (Figure 1-A). There was no difference in LVMI between the uncontrolled and controlled groups. Similarly, LVWTI was greater in uncontrolled and controlled HTN women compared to normotensive women (Figure 1-B). LVEDVI and LVESVI were not different among three groups (Table 5). LVR was higher in uncontrolled and controlled HTN women than in normotensive women.

FIGURE 1.

Cardiac morphology in older women with uncontrolled and controlled hypertension (HTN), as well as in normotensive women. Panel A: Comparison of Left Ventricular Mass Index (LVMI), Panel B: Comparison of Left Ventricular Wall Thickness Index (LVWTI). Values are mean±SE.

Table 5.

Cardiac magnetic resonance imaging measurements

	Normotensives	Uncontrolled HTN	Controlled HTN	P Value
LVEDVI (ml/m2)	60.1±1.7	63.2±2.8	63.7±2.2	0.4
LVESVI (ml/m2)	16.2±0.8	13.9±1.3	14.8±0.9	0.2
LVR (g/ml)	0.71±0.02*†	0.81±0.03	0.81±0.02	0.002

Data are mean±SE; P value; one way ANOVA, LVEDVI: left ventricular end diastolic volume index; LVESVI: left ventricular end systolic volume index; LVR: left ventricular mass to volume ratio

^*P=0.01 vs. uncontrolled HTN and

^†P=0.007 vs. controlled HTN

Arterial stiffness, central blood pressure and wave reflection

Central PWV was greater in the uncontrolled and controlled HTN groups compared with the normotensive group, while there was no difference between the HTN groups (Table 4). There was no difference between uncontrolled and controlled HTN in carotid compliance and distensibility (Table 4). The uncontrolled HTN group showed significantly lower carotid compliance and distensibility compared to normotensive group (post hoc; P≤0.01), while it was not significantly different to controlled HTN group. There was no difference in carotid systolic and diastolic diameters among groups (Table 4).

Table 4.

Arterial compliance and diameters, central pressure, and wave reflection

	Normotensives	Uncontrolled HTN	Controlled HTN	P Value
Central PWV (m/s)	9.1±0.3*†	11.5±0.6	11.1±0.5	0.0001
Compliance (mm2/mmHg)	0.15±0.02*	0.08±0.01	0.12±0.02	0.03
Distensibility (10−3/kPa)	18.1±1.3*	12.4±1.0	14.6±0.8	0.01
Carotid systolic diameter (cm)	6.41±0.13	6.76±0.14	6.62±0.22	0.3
Carotid diastolic diameter (cm)	6.12±0.12	6.54±0.26	6.36±0.22	0.3
CPP (mmHg)	44±2*†	59±3	52±3	0.0001
AP (mmHg)	12±1*	18±1	14±2	0.003
AIx (%)	25.1±1.7*	33.4±2.4	25.5±2.6	0.02
AIx@75 (%)	21.1±1.7*	28.4±2.0	20.4±2.2*	0.02

Data are mean±SE; P value; one way ANOVA, HTN: hypertension; CPP: central pulse pressure; PWV: pulse wave velocity; AP: augmented pressure; AIx: augmentation index; AIx@75: augmentation index at heart rate 75

^*P<0.05 vs. uncontrolled HTN;

^†P<0.05 vs. controlled HTN.

Central pulse pressure was greater in the HTN groups compared to the normotensive group. AP was the greatest in the uncontrolled HTN and the lowest in normotensive women (ANOVA, P=0.003, Table 4). AIx and AIx @ 75 were greater in the uncontrolled HTN group compared to the controlled HTN and normotensive groups (P≤0.05, Table 4).

Cardiac and Arterial afterload

LVEw was greater in uncontrolled HTN compared to controlled HTN and normotensive women (Figure 2-A). LVEw was not different between controlled HTN and normotensive women. There was a significant positive correlation between LVEw and LVMI in uncontrolled and controlled HTN women (r=0.65, P=0.02 and r=0.58, P=0.03, respectively) but not in normotensive women (r =−0.01, P=0.9).

FIGURE 2.

Cardiac and arterial afterload in older women with uncontrolled and controlled hypertension (HTN), as well as in normotensive women. Panel A: Comparison of cardiac afterload; left ventricle wasted pressure energy (LVEw). Panel B: Comparison of arterial afterload; total peripheral resistance (TPR). Panel C: Comparison of arterial afterload; total arterial compliance (TAC). D: Comparison of arterial afterload; effective arterial elastance (Ea). Values are mean±SE

TPR was the greatest in uncontrolled HTN women and lowest in normotensive women (Figure 2-B). There was no difference in TAC between uncontrolled and controlled HTN groups and TAC was significantly lower in the uncontrolled group compared to normotensive group (Figure 2-C). Ea index showed similar results as those of TAC (Figure 2-D).

Hemodynamics parameters

Hemodynamic parameters are presented in Table 6. Systolic and diastolic BP were greater in HTN groups compared with normotensive group. Heart rate, Qc index, and SV index were not different among groups.

Table 6.

Hemodynamic parameters

	Normotensives	Uncontrolled HTN	Controlled HTN	P Value
Systolic BP (mmHg)	109±2*†	137±2	131±2	0.0001
Diastolic BP (mmHg)	61±1*†	77±1	75±1	0.0001
MAP (mmHg)	77±1*†	97±1	93±1	0.001
Heart Rate (beats/min)	68±7	70±8	69±9	0.9
Qc index (l/min per m2)	2.07±0.07	1.96±0.06	2.13±0.08	0.3
SV index (ml/m2)	30.6±1.0	28.5±0.9	31.4±1.6	0.3

Data are mean±SE; P value; one way ANOVA, HTN: hypertension; BP: blood pressure; MAP: mean arterial pressure; SV: stroke volume Ea: effective arterial elastance

^*P≤0.001 vs. uncontrolled HTN,

^†P≤0.001 vs. controlled HTN

DISCUSSION

This is the first study to examine the differences in cardiac remodeling and arterial afterload in uncontrolled and controlled hypertensive older women. Our novel data showed that current BP control with antihypertensive drugs was not associated with favorable cardiac geometric remodeling and vascular destiffening in hypertensive older women. LV enlargement and vascular stiffness in this population seems to be independent of BP control status in our findings.

Cardiac geometric remodeling in hypertensive older women

Although meta-analysis studies indicated that stable antihypertensive drug treatment is associated with LV hypertrophy regression^39–41, hypertensive older women exhibited a greater LVMI and LVWTI compared to normotensive women in the current investigation regardless of current BP control status. It is possible that the discrepancy is mainly due to different study populations (e.g. mixed genders and wide age range of participants in meta-analysis versus only older women in our study). Previous cross-sectional studies demonstrated a clear sex-related difference in the LV enlargement^42–44 and older women appear to respond more sensitively to the HTN and obesity than similarly aged men, as those factors are related to detrimental LV changes^{45, 46}. Despite antihypertensive treatment, LV enlargement is more persistent in postmenopausal women^{47, 48}, as our data consistently indicated, and it may partially explain the greater risk for cardiovascular events in older hypertensive population even with well-controlled BP.

The meta-analysis studies suggest that angiotensin-converting-enzyme inhibitors, angiotensin II receptor antagonists, and calcium channel blockers are superior to beta blockers and diuretics in the regression of LV hypertrophy^39–41. The observed LV enlargement in the current study, independent of BP control, partially can be explained by the fact that about 60% of the participants were treated with beta blockers and diuretics. Also, there might be other factors, mainly sex hormone deficiency related, modulating the LV geometric remodeling exclusively in older hypertensive women, which warrant further investigations.

Arterial afterload and compliance in hypertensive older women

Another important finding is arterial afterload and carotid artery compliance were not different between the HTN groups. No group differences were observed between uncontrolled and controlled HTN women in arterial afterload, described by TPR and TAC⁴⁹. In addition, Ea, the lumped index of effective resistive and pulsatile afterload^{50, 51}, was not different between uncontrolled and controlled HTN women. Central arterial properties, carotid artery compliance and distensibility, were comparable with arterial afterload components. However, it is noteworthy that controlled HTN women showed no significant difference in aforementioned parameters compared to normotensive women, either. Recently, it was suggested that among hypertensive individuals, who had long-term controlled BP appeared to have delayed progression of arterial stiffness with age compared with those who persistently had uncontrolled BP⁵². Women with controlled HTN showed modestly lower, but not statistically significant, arterial afterload in the current study and further investigations are necessary to clarify the association between arterial afterload factors and BP control rate in hypertensive older women.

LV wasted pressure energy in hypertensive older women

We examined for the first time the LVEw in hypertensive older women and its association with LV enlargement. Prolonged exposure to elevated LV afterload has been indicated as the main contributor of LV hypertrophy development in HTN^{53, 54}. LVEw was significantly lower in controlled HTN compared to uncontrolled HTN. LV afterload is consist of steady resistive (arterial afterload) and pulsatile components, and the pulsatile component is dependent on central arterial properties and peripheral wave reflections⁵⁵. Previous studies found that pulsatile components of LV afterload, characterized by AP and wave reflection, is associated with LV hypertrophy and the impact of wave reflection on LV afterload is determined by the interaction between amplitude of the AP and duration of ejection^{56, 57}. LVEw was introduced as the parameter resulting from both AP amplitude and duration of wave reflection⁵⁸. In our data, increased LVEw was correlated with LVMI in both uncontrolled and controlled HTN and it is consistent with the previous finding that LVEw contributes to LV remodeling in the hypertensive population⁵⁸. Based on the correlation between LVEw and LVMI in hypertensive women, the relation between two factors might be causative but further prospective cohort study is required to establish the definite causal relationship.

Study limitations

Our study has several limitations. First of all, because of the nature of the cross-sectional design with a relatively small sample size, it is not appropriate to develop further interpretation regarding the causal relationship between HTN and cardiac-arterial alterations. However, in the light of evidences we found, it is possible to set a hypothesis for future large scale longitudinal study investigating the association between LV enlargement and HTN control. Secondly, we did not document the time since the diagnosis of hypertension and/or onset of antihypertensive treatment. The time that exposed to the pressure overload by high BP is a critical determinant in cardiac adaptation and the duration of antihypertensive treatment is closely associated with regression of hypertension induced cardiac and vascular alterations. However, the primary focus of the current study was to evaluate the characteristics of cardiac-vascular components in hypertensive older women with uncontrolled and controlled BP. Future studies need to examine the time dependent impact of various hypertensive drugs on cardiac-vascular alterations. Third, we did not collect the information regarding the age of menopause transition and the past history/duration of hormone therapy. There is growing awareness that menopause and hormone therapy play an important role in cardiovascular physiology in older women. However, all participants did not use hormonal supplement at least for 6 months prior to participating in the study. lastly, our participants were older hypertensive women, free of major cardiovascular, pulmonary, neurological and/or renal diseases, but a part of them were on a stable drug regimen, besides antihypertensive, for controlling hypothyroidism or elevated blood lipids. Although this may have influenced our results, the percentage of medication use was similar among groups; therefore, this factor should not influence the observed group differences. It is common for older women to be on antilipidemic and antihypothyroidism medications, which enhances the generalizability of our findings to the general older female population.

Conclusion

A key concept emerging from current investigation is HTN older women with antihypertensive drug treatment exhibited LV enlargement compared with normotensive counterparts independent of current BP control status. Adequate BP control was not correspondingly associated with favorable characteristics of cardiac and arterial function. Our observations possibly set up hypotheses explaining why the risk of cardiovascular mortality remains high in older hypertensive women regardless of BP control status. Large sample size prospective studies are warrant for better understanding of the relationships among LV enlargement, arterial load and BP control. It is an important question for the rational choice of antihypertensive treatment to achieve “beyond BP control”, specifically targeted to improve residual cardiovascular risk factors. There may be opportunity for strategies to further reduce cardiovascular morbidity, which could correlate with changes in LV remodeling.