Arterial Pressure, Heart Rate, and Cerebral Hemodynamics across the Adult Life Span

Abstract

Age-related alterations in systemic and cerebral hemodynamics are not well understood. The purpose of this study is to characterize age-related alterations in beat-to-beat oscillations in arterial blood pressure (BP), heart rate (HR), cerebral blood flow (CBF), cardiac baroreflex sensitivity (BRS) and dynamic cerebral autoregulation (dCA) across the adult life span. We studied 136 healthy adults aged 21–80 years (60% women). Beat-to-beat BP, HR and CBF velocity were measured at rest and during sit-stand maneuvers to mimic effects of postural changes on BP and CBF. Transfer function analysis was used to assess BRS and dCA. Carotid-femoral pulse wave velocity (cfPWV) was measured to assess central arterial stiffness. Advanced aging was associated with elevated cfPWV, systolic and pulse BP, cerebrovascular resistance and CBF pulsatility, but reduced mean CBF velocity. Compared to the young and middle-aged, older adults had lower beat-to-beat BP, HR and CBF variability in the low frequency ranges (LF) at rest, but higher BP and CBF variability during sit-stand maneuvers. BRS was reduced, while dCA gain was elevated at rest in older adults. Multiple linear regression analysis indicated that systolic BP variability is correlated positively with cfPWV independent of HR variability. In conclusion, advanced aging is associated with elevated pulsatility in BP and CBF, reduced beat-to-beat LF oscillations in BP, HR and CBF, and impaired BRS and dCA at rest. The augmented BP and CBF variability in older adults during sit-stand maneuvers indicate diminished cardiovascular regulatory capability and increased hemodynamic stress on the cerebral circulation with aging.

Introduction

Advanced aging is associated with increased risks of cardio- and cerebrovascular disease which may be attributed at least in part to alterations in systemic and cerebral hemodynamics.¹ Central arterial stiffening, a hallmark of vascular aging, is related to increases in systolic and pulsatile blood pressure (BP) and cerebral blood flow (CBF) pulsatility.^{2, 3} Excess increase in arterial pulsatility may cause cerebrovascular remodeling, endothelial dysfunction, increases in cerebrovascular resistance and reduction in CBF.³ Aging is also associated with reductions in beat-to-beat heart rate (HR) variability, which has been linked to cardiovascular and brain health.^{4, 5} However, the relationship between age and BP variability is controversial^5–7 and little is known about CBF variability.

BP and CBF regulation in the time scales of several seconds to minutes are related primarily to the baroreflex and dynamic cerebral autoregulation (dCA), respectively.⁸ The baroreflex regulates BP by monitoring the distortions of baroreceptors located in the central elastic arteries and provides feedback control of HR and systemic vascular resistance via the autonomic nervous system to reduce BP variability.⁹ On the other hand, dCA maintains a relatively constant CBF in face of BP fluctuations which may be enhanced due to baroreflex dysfunction.¹⁰ Aging appears to differentially affect these two regulatory mechanisms, such that cardiac baroreflex function declines with age^{9, 11}; while no effects of age on dCA were observed.^{11, 12}

Previous studies suggest that baroreflex and dCA may interact in the regulation of CBF in face of acute changes in arterial pressure.¹³ For example, Tzeng et al. observed an inverse correlation between cardiac baroreflex sensitivity (BRS) and dCA metrics in healthy young adults suggesting the existence of compensatory functionalities between these regulatory mechanisms.¹³ However, others did not find such a relationship in older adults.^{11, 14} In these studies, the sample size was small and no middle-aged subjects were included. Therefore, further studies are warranted to elucidate the effects of aging on the relationship between baroreflex function and dCA.

Further understanding of age-related alterations in baroreflex function and dCA may provide new insights into cardiovascular and brain health in older adults. In addition, it is important to reveal potential sex differences in baroreflex function and CBF regulation.^{3, 15, 16} In this regard, impairment of baroreflex function and dCA may lead to dizziness, syncope, or fall in older adults associated with postural change induced hypotension in daily life.¹⁷ In previous studies, we and others have employed squat-stand or sit-stand maneuvers to mimic effects of postural changes on BP and CBF regulation.^{11, 18} Assessments of baroreflex function and dCA under these conditions not only provide practical and clinical implications, but also may improve the reliability of these assessments because of enhanced BP and CBF variability.¹⁸

The purposes of the present study were: 1) to characterize age as well as sex-related alterations in beat-to-beat oscillations in BP, HR, and CBF at rest and during sit-stand maneuvers across the adult life span, and 2) to examine cardiac baroreflex function and dCA under these conditions. We hypothesized that beat-to-beat BP and CBF variability both are reduced with advanced aging at rest, but are enhanced during sit-stand maneuvers reflecting diminished cardiovascular and cerebrovascular regulatory capability in older men and women.

Methods

Participants

One hundred and thirty-six healthy participants aged between 21 to 80 years (81 women; 83% Caucasian, 10% African American, 7% Asian) were recruited through flyers and newspaper advertisements from the Dallas-Fort Worth metroplex. Exclusion criteria included the presence of ischemic or structural heart disease screened by 12-lead electrocardiogram and echocardiography, office BP > 140/90mmHg confirmed by ambulatory blood pressure monitoring, carotid artery atherosclerotic plaque or stenosis with > 50% occlusion imaged by ultrasound, diabetes screened by the presence of symptoms, use of antidiabetic drugs, or fasting blood glucose > 126mg/dL, body mass index (BMI) > 40 kg/m², smoking, active alcohol or drug abuse, brain damage or trauma, and the presence or history of cerebrovascular (e.g., stroke), neurologic, psychiatric, or inflammatory diseases. Pregnant or breast-feeding women and individuals who participated in regular moderate to high intensity aerobic exercise training (> 30 minutes per session, 3 times per week over the past 2 years) were also excluded.

This study was approved by the Institutional Review Board of the University of Texas Southwestern Medical Center and Texas Health Presbyterian Hospital Dallas and was performed in accordance with the guidelines of the Declaration of Helsinki and Belmont Report. A written informed consent was signed by all subjects for study participation.

Study protocol

All data were collected in an environmentally controlled laboratory with an ambient temperature of ~ 22 °C. Subjects abstained from caffeinated beverages, alcohol, and vigorous exercise at least 24 hours before the study. After resting in the quiet, seated position for ⩾15 minutes, a 5-minute baseline data was collected during spontaneous breathing. Next, a repeated sit-stand maneuver was performed for 5 minutes with a duty cycle of a 10-second sit and a 10-second stand. An investigator verbally coached subject during the maneuver. This protocol was designed to induce clinically relevant postural hemodynamic oscillations specifically at 0.05 Hz and to improve coherence between the measured hemodynamic variables.¹⁸ The 0.05 Hz was chosen because of dCA is likely to be effective at low frequencies and cardiac baroreflex function also can be assessed at this frequency.^{18, 19}

Measurements

HR was measured by a 3-lead electrocardiogram (Hewlett-Packard, Andover, MA). Arterial BP was recorded from the left middle finger at the participant’s heart level using a photoplethysmography (Finapres Medical Systems, Amsterdam, The Netherlands). This method can reliably assess beat-to-beat changes in BP that is correlated well with intra-arterial recordings.²⁰ Intermittent arterial pressure was measured in the right arm by an ECG-gated electrosphygmomanometer (Tango+, Suntech Medical, Morrisville, NC). The brachial BP recording was used to corroborate the finger measurement to ensure that a finger cuff was properly placed. End-tidal CO₂ (ETCO₂) was monitored with a nasal cannula using a capnograph (Capnogard, Novamatrix). All data were simultaneously recorded at a sampling frequency ≥ 250 Hz and stored for offline analysis. Baseline carotid-femoral pulse wave velocity (cfPWV) was measured using applanation tonometry (SphygmoCor 8.0, AtCor Medical) to assess central arterial stiffness.

CBF velocity (CBFV) was measured in the middle cerebral artery (MCA) using a transcranial Doppler (Multi-Dop X2, Compumedics/DWL, Singen, Germany). A 2-MHz Doppler probe was placed over the temporal window using a headgear (Spencer Technologies, Northborough, MA) and fixated at a constant angle and depth where the optimal CBFV signal was obtained according to a standard procedure.²¹ Transcranial Doppler has a high temporal resolution (>100 ms) and allows noninvasive and repeatable estimates of changes in CBF on a beat-to-beat basis.

Data analysis

Preprocessing

Hemodynamic data were analyzed by AcqKnowledge (BIOPAC Systems, Goleta, CA, USA) and DADiSP (Newton, MA, USA) software according to the procedures described in detail elsewhere.¹⁰ First, artifacts were visually identified and cleaned from the data segment used for spectral and transfer function analysis. Second, all hemodynamic data were averaged for each cardiac cycle to obtain beat-to-beat mean values.²² Normalized CBFV (CBFV%) relative to the mean value of the 5-minute data segment was calculated for spectral and transfer function analyses.^{14, 22} Third, the time series of beat-to-beat BP, HR, R-R interval, and CBFV% were linearly interpolated, resampled at 2 Hz to obtain equidistant time series, and detrended by a 3rd order polynomial curve fitting. Finally, these data were subdivided into 256-point segments with 50% overlap, and a Hanning window was applied for spectral and transfer function analysis. Cerebrovascular resistance index (CVRi) was calculated by dividing mean arterial pressure (MAP) by mean CBFV. Pulsatility index (PI) was calculated as [(systolic CBFV-diastolic CBFV)/mean CBFV].²³

Systemic Oscillations and Baroreflex Function

Spectral powers of beat-to-beat HR, R-R interval, systolic BP (SBP), and diastolic BP (DBP) variability were calculated from the low frequency range (LF: 0.05–0.15 Hz) at baseline and a point frequency of 0.05 Hz during sit-stand maneuvers.^{14, 18, 24} The baroreflex sensitivity (BRS) was determined by the transfer function gain between SBP and R-R interval in LF at baseline and at a point frequency of 0.05 Hz during sit-stand maneuvers.^{14, 18} The frequency band of 0.05–0.15 Hz was selected a priori based on the literature that showed that transfer function coherence of the baroreflex function is low below 0.05 Hz, which may compromise the validity of BRS estimation.²⁵ Also, spectral power and transfer function estimates at frequencies higher than 0.15 Hz may be influenced by respiration.²⁴

Cerebral Oscillations and Cerebral Autoregulation

Spectral powers of beat-to-beat MAP and CBFV% variability as well as their transfer function estimates were calculated from very low (VLF: 0.02–0.07 Hz), low (LF: 0.07–0.20 Hz), and high (HF: 0.20–0.35 Hz) frequency ranges at baseline and from a point frequency of 0.05 Hz during sit-stand maneuvers as reported in previous studies.^{14, 22} The explanation of transfer function parameters has been described previously.¹⁰ In brief, transfer function gain quantifies the magnitude relationship between the input and output signals (i.e. SBP and R-R interval for BRS, MAP and CBFV% for dCA respectively) whereas the phase provides a measure of their temporal displacement. Coherence function provides a measure of linear correlation between the signals and ranges from 0 to 1.

Statistical Analysis

Subjects were divided into three age groups: young (21–44 years), middle age (45–64 years) and old (65–80 years). Two-way analysis of variance (ANOVA) was used to examine the main and interaction effects of age and sex on subject characteristics and steady-state hemodynamics.¹⁵ Chi-squared test was used to test group difference in categorical variables. Due to non-normal distributions of transfer function and power spectral variables, two-way ANOVA with age and sex as factors were performed after rank transformation. For age-significant variables, the post-hoc test with a Bonferroni correction was used to make pairwise group comparisons. The analysis of covariance with body mass index (BMI) as a covariable was performed to reveal its potential effects on the age-related differences in BRS and dCA after rank transformation.¹⁵ Spearman’s (non-parametric) correlation analysis was used to assess simple correlations between hemodynamic variables. Multiple linear regression analysis was used to assess the relative contributions of cfPWV, HR, or R-R interval variability to SBP variability after rank transformation. Statistical significance was defined by 2-tailed tests with P<0.05. All data were reported as mean ± standard deviation unless otherwise stated. Data were analyzed with SPSS 20.0 (SPSS, Inc, Chicago, IL).

Results

Subject Characteristics and Steady-State Hemodynamics

Table S1 summarizes subject characteristics and steady-state hemodynamics for each age group. The brachial SBP and pulse pressure increased with age, while DBP first increased in the middle age and then decreased in the old. The old group showed reductions in systolic, diastolic, and mean CBFV when compared to the young group, while PI, CVRi, and cfPWV were elevated. Old adults also showed lower ETCO₂ relative to the young. Women had higher mean CBFV, but lower brachial BP and CVRi than men (Table S1).

BP, HR Variability, and BRS

Spectral powers of beat-to-beat HR, R-R interval, SBP and DBP variability in the LF range all decreased progressively with age at rest (Table 1 and Figure 1). However, SBP variability during sit-stand maneuvers was significantly higher in the old than in the young, while both HR and R-R spectral powers remained lower in the old (Table 1 and Figure 1). Multiple linear regression analysis indicated that about 34% and 37% of SBP spectral power (variability) can be explained by HR variability and cfPWV under resting conditions and during sit-stand maneuvers, respectively; and cfPWV contribution to SBP variability was higher during sit-stand maneuvers than that at rest (Table S2). Similar results were obtained with either HR or R-R interval variability.

Table 1.

Power spectral analysis of beat-to-beat blood pressure, heart rate, and cerebral blood flow velocity variability

Variable		Young		Middle age		Old		P-value
		Men	Women	Men	Women	Men	Women	Age group	Sex	Interaction
Systemichemodynamics
LF	HR, bpm2	6.47 ± 6.69	3.98 ± 2.48	2.37 ± 1.67	1.76 ± 1.77 *	1.22 ± 2.2	0.62 ± 0.58 *†	<0.001	0.017	0.636
	R-R Interval, ms2	1153 ± 1608	580 ± 752	396 ± 479	254 ± 266 *	231 ± 382	91 ± 104 *†	<0.001	0.014	0.402
	SBP, mmHg2	9.71 ± 5.05	7.14 ± 5.85	8.53 ± 3.88	7.05 ± 4.3	5.6 ± 4.67	4.66 ± 2.83 *†	0.001	0.034	0.438
	DBP, mmHg2	4.32 ± 2.77	3.72 ± 2.76	3.19 ± 1.8	2.11 ± 1.58 *	1.42 ± 1.09	1.06 ± 0.71 *†	<0.001	0.018	0.315
Sit-Stand	HR, bpm2/Hz	3473 ± 2956	3503 ± 5379	1605 ± 1503	730 ± 618 *	1110 ± 1413	1025 ± 1919 *	<0.001	0.012	0.463
	R-R Interval, ms2/Hz	289817 ± 224809	276781 ± 375300	183063 ± 173898	77332 ± 83811 *	115973 ± 95869	56764 ± 68151 *	<0.001	0.002	0.323
	SBP, mmHg2/Hz	4511 ± 5073	3896 ± 4380	7601 ± 5723	4998 ± 5569	11848 ± 11311	7474 ± 6352 *	0.001	0.054	0.329
	DBP, mmHg2/Hz	1450 ± 1399	1463 ± 1214	2056 ± 983	1119 ± 1005	2314 ± 1735	1592 ± 1267	0.225	0.016	0.055
Cerebral hemodynamics
VLF	MAP, mmHg2	4.82 ± 3.37	3.22 ± 1.62	6.9 ± 3.95	4.86 ± 3.03	5.85 ± 5.14	4.4 ± 5.12	0.057	0.005	0.905
	CBFV%, %2	19.82 ± 19.01	9.6 ± 5.87	22.79 ± 15.27	13.14 ± 8.07	17.77 ± 13.02	15.07 ± 13.94	0.138	0.001	0.562
LF	MAP, mmHg2	3.43 ± 1.75	3.81 ± 2.37	2.87 ± 1.59	2.27 ± 1.61 *	2.32 ± 3.08	1.88 ± 2.27 *†	<0.001	0.459	0.501
	CBFV%, %2	10.86 ± 7.59	11.69 ± 7.52	8.63 ± 5.53	8.06 ± 4.91 *	9.65 ± 10.23	9.25 ± 9.96*	0.014	0.745	0.745
HF	MAP, mmHg2	0.38 ± 0.42	0.58 ± 0.45	0.53 ± 0.45	0.51 ± 0.49	0.4 ± 0.33	0.36 ± 0.47	0.071	0.957	0.058
	CBFV%, %2	1.86 ± 1.49	2.32 ± 1.65	1.93 ± 1.48	2.65 ± 2.11	2.48 ± 2.42	2.38 ± 1.67	0.751	0.195	0.702
Sit-Stand	MAP, mmHg2/Hz	2235 ± 2416	2145 ± 1882	3418 ± 1722	2052 ± 1971	4859 ± 3720	3399 ± 2708*	0.003	0.043	0.108
	CBFV%, %2/Hz	6598 ± 5353	4062 ± 3071	9593 ± 6078	5471 ± 4475	13148 ± 7758	7957 ± 6048*†	<0.001	<0.001	0.728

Data are mean ± standard deviation. Bold values represent P<0.05.

^*vs. young;

^†vs. middle age.

Young, 21–44 years; Middle age, 45–64 years; Old, 65–80 years; HR, heart rate; SBP, systolic blood pressure; DBP, diastolic blood pressure; MAP, mean arterial pressure; CBFV%, normalized mean cerebral blood flow velocity; VLF, very low frequency; LF, low frequency; HF, high frequency. Sit-stand maneuvers were performed at 0.05 Hz and its point frequency data are reported.

Figure 1.

Group-averaged power spectral density of R-R interval (A& B), SBP (C&D) and DBP (E&F) oscillations during baseline (rest) and sit-stand maneuvers. Solid lines represent the young group; dashed lines represent the middle-aged group; and dotted lines represent the old group. SBP, systolic blood pressure; DBP, diastolic blood pressure.

Age was associated with a substantial reduction of BRS either at rest or during sit-stand maneuvers (Table 2 and Figure 2). There was also an age-related increase in BRS phase (more negative values) during sit-stand maneuvers (Table 2 and Figure S1). The coherence between SBP and R-R interval at rest was lower in the old group than in the young and middle-aged groups, but similar during sit-stand maneuvers. Analysis of covariance did not reveal effects of BMI on age-related differences in BP, HR, R-R variability and BRS.

Table 2.

Transfer function analysis of cardiac baroreflex function and dynamic cerebral autoregulation

Variable		Young		Middle age		Old		P-value
		Men	Women	Men	Women	Men	Women	Age group	Sex	Interaction
Cardiac baroreflex function
LF	Gain, ms/mmHg	8.87 ± 4.26	7.84 ± 3.67	5.98 ± 2.34	4.87 ± 2.49 *	4.17 ± 2.34	3.41 ± 2.07 *†	<0.001	0.052	0.713
	Phase, rad	−0.80 ± 0.21	−0.94 ± 0.40	−0.94 ± 0.28	−1.19 ± 0.38 *	−0.95 ± 0.31	−1.08 ± 0.40	0.006	<0.001	0.796
	Coherence	0.74 ± 0.09	0.64 ± 0.13	0.70 ± 0.09	0.60 ± 0.12	0.60 ± 0.20	0.50 ± 0.16 *†	<0.001	<0.001	0.987
Sit-Stand	Gain, ms/mmHg	9.02 ± 4.57	8.61 ± 3.16	5.01 ± 2.83	4.22 ± 2.00 *	3.44 ± 1.99	2.64 ± 1.33 *†	<0.001	0.186	0.498
	Phase, rad	−0.56 ± 0.28	−0.72 ± 0.48	−0.61 ± 0.24	−0.93 ± 0.36 *	−0.87 ± 0.20	−1.04 ± 0.42 *	<0.001	0.001	0.169
	Coherence	0.90 ± 0.15	0.90 ± 0.13	0.95 ± 0.06	0.87 ± 0.15	0.93 ± 0.13	0.91 ± 0.16	0.838	0.044	0.559
Dynamic cerebral autoregulation
VLF	Gain, %/mmHg	1.57 ± 0.60	1.19 ± 0.45	1.44 ± 0.36	1.30 ± 0.41	1.54 ± 0.50	1.36 ± 0.55	0.744	0.002	0.636
	Phase, rad	0.82 ± 0.52	0.71 ± 0.45	0.88 ± 0.47	0.85 ± 0.45	0.90 ± 0.46	0.84 ± 0.40	0.537	0.482	0.914
	Coherence	0.60 ± 0.16	0.50 ± 0.16	0.66 ± 0.17	0.59 ± 0.16	0.68 ± 0.15	0.53 ± 0.18	0.072	<0.001	0.662
LF	Gain, %/mmHg	1.64 ± 0.47	1.61 ± 0.36	1.64 ± 0.46	1.74 ± 0.36	2.01 ± 0.73	2.07 ± 0.57 *†	0.001	0.414	0.590
	Phase, rad	0.60 ± 0.26	0.52 ± 0.16	0.64 ± 0.29	0.61 ± 0.20	0.52 ± 0.21	0.53 ± 0.22	0.067	0.287	0.501
	Coherence	0.69 ± 0.15	0.74 ± 0.13	0.65 ± 0.14	0.67 ± 0.13	0.62 ± 0.17	0.62 ± 0.20	0.054	0.321	0.772
HF	Gain, %/mmHg	2.00 ± 0.52	1.85 ± 0.54	1.79 ± 0.61	2.10 ± 0.47	2.15 ± 0.59	2.56 ± 0.76 *†	0.001	0.139	0.117
	Phase, rad	0.20 ± 0.26	0.16 ± 0.14	0.17 ± 0.24	0.16 ± 0.25	0.19 ± 0.22	0.14 ± 0.21	0.717	0.516	0.902
	Coherence	0.67 ± 0.17	0.69 ± 0.17	0.65 ± 0.16	0.67 ± 0.15	0.61 ± 0.15	0.60 ± 0.20	0.095	0.790	0.973
Sit-Stand	Gain, %/mmHg	1.94 ± 0.55	1.49 ± 0.41	1.65 ± 0.35	1.84 ± 0.77	1.73 ± 0.63	1.54 ± 0.47	0.531	0.054	0.180
	Phase, rad	0.73 ± 0.28	0.81 ± 0.27	0.81 ± 0.20	0.91 ± 0.26	0.81 ± 0.30	0.93 ± 0.30	0.212	0.013	0.947
	Coherence	0.93 ± 0.08	0.95 ± 0.05	0.98 ± 0.02	0.92 ± 0.11	0.94 ± 0.19	0.96 ± 0.11 *	0.021	0.158	0.162

Data are mean ± standard deviation. Bold values represent P<0.05.

^*vs. young;

^†vs. middle age.

Young, 21–44 years; Middle age, 45–64 years; Old, 65–80 years; VLF, very low frequency; LF, low frequency; HF, high frequency. Sit-stand maneuvers were performed at 0.05 Hz and its point frequency data are reported.

Figure 2.

Group-averaged transfer function gain of BRS (A&B) and dCA (C&D) during baseline (rest) and sit-stand maneuvers. Solid lines represent the young group; dashed lines represent the middle-aged group; and dotted lines represent the old group. BRS, baroreflex sensitivity; dCA, dynamic cerebral autoregulation.

Women had significant lower LF HR, R-R interval, SBP and DBP variability at rest and lower HR, R-R interval and DBP variability during sit-stand maneuvers than men (Table 1). Women also had lower cardiac baroreflex phase (more negative values) and coherence than men at rest and during sit-stand maneuvers (Table 2).

CBFV Variability and dCA

Spectral powers of beat-to-beat MAP and CBFV% variability in the LF range decreased with age under rest, but similar in the VLF and HF ranges (Table 1 and Figure 3). In contrast, spectral powers of MAP and CBFV% during sit-stand maneuvers were augmented in the old compared to the young.

Figure 3.

Group-averaged power spectral density of MAP (A&B) and CBFV% (C&D) oscillations during baseline (rest) and sit-stand maneuvers. Solid lines represent the young group; dashed lines represent the middle-aged group; and dotted lines represent the old group. MAP, mean arterial pressure; CBFV%, normalized mean cerebral blood flow velocity.

Both LF and HF dCA gains were higher in the old than in the young and middle-aged at rest (Table 2 and Figure 2). However, dCA gains were not different during sit-stand maneuvers among groups. No age-related differences in dCA phase were observed during sit-stand maneuvers while the coherence was slightly elevated in the old compared with the young (Figure S2). Analysis of covariance did not reveal effects of BMI on age-related differences in MAP and CBFV% variability and dCA.

Women had lower MAP and CBFV% variability in VLF at rest and during sit-stand maneuvers than men (Table 1). Women also had lower VLF dCA gain and coherence at rest, and higher phase during sit-stand maneuvers than men (Table 2).

Correlation between BRS and dCA

The associations between BRS and dCA gain in the LF ranges at rest and during sit-stand maneuvers are presented in Figure S3. BRS was correlated positively with dCA gain in the young, but not in the middle-aged (baseline: R²=0.016, P=0.383; sit-stand maneuvers: R²=0.013, P=0.429) and the old (baseline: R²=0.007, P=0.672; sit-stand maneuvers: R²=0.050, P=0.268).

Discussion

This is the first study to characterize age-related differences in beat-to-beat oscillations in BP, HR, CBF, cardiac BRS and dCA under resting conditions and during sit-stand maneuvers across the adult life span. The major findings are fourfold. First, we found that beat-to-beat HR, R-R, SBP, DBP variability and cardiac BRS in the LF range all were reduced at rest in the older adults relative to young subjects. Cardiac BRS was reduced also during sit-stand maneuvers while SBP variability was augmented significantly in the old adults. Second, we found that MAP and CBFV variability in the LF range were reduced also at rest in the older adults while dCA gain was elevated. However, dCA gain remained unchanged despite the augmented MAP and CBFV variability during sit-stand maneuvers in the older adults. Third, we found that women had lower HR, R-R interval and DBP variability in the LF range and lower MAP and CBFV% variability in the VLF range at rest and during sit-stand maneuvers than men. In addition, although no sex differences in cardiac BRS was observed, women had lower dCA gain and coherence in the VLF range at rest, and a higher phase during sit-stand maneuvers than men, suggesting a better dCA. Fourth, BRS assessed in the LF ranges was correlated positively with dCA gain in the young, but not in the middle-aged and old subjects. Collectively, these findings demonstrate the presence of age-related reductions in BP, HR, CBF variability in the LF ranges and impaired cardiac BRS and dCA in older adults at rest. Furthermore, augmented BP and CBF variability during sit-stand maneuvers in older adults indicate diminished cardiovascular regulatory capability and increased hemodynamic stress on the cerebral circulation with aging.

Steady-State Systemic and Cerebral Hemodynamics

The observation of higher cfPWV, CVRi, and PI of CBFV in older adults in this study is consistent with previous reports.^{3, 15, 23} The age-related reduction in cerebral capillary density and deformation of the microvascular wall might be the morphological basis of the increases in cerebrovascular resistance.²⁶ In addition, endothelial dysfunction and vessel wall smooth muscle degeneration may result in impaired vasodilatory responses and sustained cerebral vasoconstriction with age.²⁷ Age-related increase in CBFV pulsatility may be explained by a downstream increase in cerebral microvascular resistance as well as the upper-steam elevation of pulse pressure.²⁸ With advancing age, progressive stiffening of central elastic arteries increases PWV, which in turn may cause premature arterial wave reflection and augmentation of central pulse pressure while decreasing diastolic BP.³ Consistent with these age-related cardiovascular changes, we observed that age-related reduction in diastolic CBFV (i.e. 29% reduction in the old when compared with the young) was greater than the reduction of systolic CBFV (12%), which resulted in augmented pulsatility. In addition, reduced pressure wave reflections from the downstream cerebrovascular bed due to an increased transmission also may contribute to the augmented CBF pulsatility in the MCA in older adults.³ Clinically, excess increases in arterial BP and CBF pulsatility may lead to cerebrovascular remodeling and compensatory increases in vascular resistance, which over the time may cause a damage to the cerebral microcirculation and increase the risk of cerebral small vessel disease.²⁹

BP, HR, and CBFV Variability

Advanced aging is associated with substantial decreases in HR variability,⁴ which most likely is related to the alterations in autonomic neural activity. Zhang et al. demonstrated that autonomic blockade markedly reduced HR variability in young adults.²⁴ Aging is associated with significant reductions in parasympathetic, but increases in sympathetic neural activity, which may lead to reductions in HR variability in older adults.³⁰ In addition, reduction in cardiac BRS with aging may reduce HR variability.⁹

Aging effects on beat-to-beat BP variability are inconclusive. Previous studies showed that BP variability was increased,⁷ decreased,^{5, 6} or remained unchanged in older adults.⁶ In the present study, older adults showed diminished SBP and DBP variability in the LF range at rest, while SBP variability was augmented significantly during sit-stand maneuvers when compared to young adults. Furthermore, we found that SBP variability was correlated positively with cfPWV independent of the contributions from HR or R-R variability. These findings suggest that baroreflex-mediated BP regulation during postural changes is diminished with advanced age via either stiffening of the barosensory arteries or a diminished windkessel effect of the central arteries. This diminished capability of baroreflex function and/or arterial stiffening may explain why aging increases BP variability in daily living, as measured by ambulatory BP monitoring, and the risk of brain and other end-organ damage.³¹

Age-related differences in beat-to-beat CBFV variability across the adult lifespan have not been studied extensively. Previous studies compared CBFV variability between young and older adults under resting conditions showed reduced CBFV variability in the LF range in older adults consistent with the findings of the present study.¹¹ However, the augmented CBFV variability during sit-stand maneuvers in older subjects observed in the present study was not reported in previous studies.¹¹ This discrepancy may be related to the relatively small sample size used in these studies. We used a normalized unit to measure changes in CBFV (%) which is based on the consideration that differences in baseline CBFV values and/or the insonated MCA diameters may confound the comparisons of CBFV changes among individual subjects.⁸ Of note, normalized CBFV (%) has been shown to correlate better with CBF volumetric changes.³²

Interestingly, although dCA transfer function gains in the LF and HF ranges were increased in older adults under resting conditions (suggesting an impaired autoregulation with age), it remained unchanged during sit-stand maneuvers associated with the augmented BP and CBFV variability. Therefore, even though dCA may be preserved with advanced aging during postural change induced BP oscillations, augmented CBFV variability suggests that intermittent brain hypoperfusion may occur under these conditions, which in turn may lead to symptoms of dizziness, syncope, or fall in older adults .¹⁷

Baroreflex and dCA Relationship

Age-related decline in cardiac BRS is consistent with previous findings.⁹ However, in contrary to the notion that age does not alter dCA,⁸ we observed that dCA transfer function gain in the LF and HF ranges was increased in older adults at rest. This discrepancy cannot be explained simply by the differences in the methods used for dCA assessment because no differences dCA transfer function gain in the LF range between the young and old adults have been reported using the similar methods.¹¹ Thus, further studies with large sample size are warranted to understand the effects of aging on dCA. The underlying mechanism(s) of the frequency-dependent effects of aging on dCA observed in this study is unknown and may be related to the aging effects on autonomic neural activity.²⁰

Furthermore, consistent with previous studies, we observed that cardiac BRS was correlated positively with dCA transfer function gain in the young, but not in the middle-aged and older adults under resting conditions.^{11, 13, 14} The new finding is that this relationship was also presented during sit-stand maneuvers. A higher cardiac BRS may lead to a better control of BP, thus a less demand for dCA to maintain CBF constant which may explain the observed BRS and dCA relationship in the young subjects. However, the underlying mechanisms for age-related interactions between BP and CBF regulation cannot be determined in this study.

Sex differences

In a previous study, we reported that women had higher volumetric CBF measured with PC-MRI and lower cerebrovascular resistance than men across the adult life span.¹⁵ Consistently, we observed that women had higher mean CBFV and lower CVRi than men in the present study. These findings together suggest that the observed sex differences in CBF may be attributed mainly to the differences in CBF velocity rather than the differences in cerebral blood vessel diameters between women and men. The new findings that women had lower HR, SBP and DBP variability in the LF ranges and lower MAP and CBFV variability in the VLF ranges at rest and during sit-stand maneuvers than men may be attributed to the observed sex differences in the sympathetic and/or cardiac BRS as well as autonomic neural control of dynamic changes in BP and CBF.^{16, 33} Interestingly, women also had lower dCA gain and coherence in the VLF range at rest, and higher phase during sit-stand maneuvers than men, suggesting a better autoregulation.¹⁰

Strengths and Limitations

The major strength of this study is an integrated examination of both systemic and cerebral hemodynamics, which provides a comprehensive view of the effects of aging on the both systems. Furthermore, the study population was stringently screened to exclude overt cardio- and cerebrovascular risk factors or diseases which may confound the aging effect.

There are several limitations that should be discussed. First, a cross-section nature of the present study precludes the understanding of the causal relationships between age and hemodynamic variables. Second, the use of linear transfer function method may limit its capability to quantify the nonlinear characteristics of systemic and cerebral hemodynamics. However, the average levels of coherence function higher than 0.5 for both BRS and dCA assessments at rest and during sit-stand maneuvers ensure the reliability of our findings. Third, cardiac BRS estimated in the present study represents closed-loop properties of the baroreflex function which may not be extrapolated to its open-loop characteristics.³⁴ However, baroreflex functions normally under closed-loop conditions which is more relevant to BP control in daily life. Fourth, changes in CBFV reflect changes in volumetric CBF only if the MCA diameter remains relatively constant which has been demonstrated under moderate changes in BP and arterial CO₂ concentrations.^35–37 Thus, beat-to-beat changes in CBFV observed in this study most likely reflect beat-to-beat changes in CBF. Finally, only healthy subjects were included in this study, whether the presence of cardiovascular or cerebrovascular diseases alters baroreflex and/or dCA associated with aging remains to be determined.

Conclusions

Advanced aging is associated with significant alterations in both the systemic and cerebral hemodynamics such as elevations in central arterial stiffness, SBP and pulse pressure, reductions in mean and diastolic CBFV, increases in CBFV pulsatility and cerebrovascular resistance. In addition, aging is associated with altered cardio- and cerebrovascular regulatory mechanisms, such as cardiac BRS and dCA. Although beat-to-beat BP and CBFV variability in the LF ranges were markedly reduced in older adults at rest, they were augmented significantly during sit-stand maneuvers. Overall, these findings support the viewpoint that age-related alterations in cerebral hemodynamics are related closely to those of the systemic circulation. The augmented BP and CBFV variability in older adults during sit-stand maneuvers indicate the presence of diminished cardiovascular regulatory capability and increased hemodynamic stress on the cerebral circulation with aging, which may lead to higher risks of cardio- and cerebrovascular disease.

Perspectives.

Advanced aging is associated with increased risks of cardio- and cerebrovascular disease, which may be attributed at least in part to the altered systemic and cerebral hemodynamics. The findings from this study provide an integrated view of effects of aging on both the systems and demonstrated the presence of diminished cardiovascular regulatory capability and increased hemodynamic stress on the cerebral circulation during sit-stand maneuvers in older adults. These findings provide new knowledge for our understanding of age-related alterations in cerebral hemodynamics and its relation to the systemic circulation, and may have implications for the understanding of age-related increases in cerebrovascular disease.

Novelty and Significance.

What Is New?

Aging is associated with reductions in beat-to-beat blood pressure (BP) and cerebral blood flow (CBF) variability in the low frequency ranges at rest. However, postural changes augment BP and CBF variability to a greater extent in the old than young subjects indicating the presence of diminished cardiovascular regulatory capability and increased hemodynamic stress on the cerebral circulation with advanced aging.

What Is Relevant?

Understanding the aging effects on BP and CBF regulations may provide new insight into cardiovascular and brain health in older adults.

Summary

Aging is associated a phenotype of increases in systolic and pulse BP, CBF pulsatility and cerebrovascular resistance, and reductions in beat-to-beat BP and CBF variability in the low frequency ranges under resting conditions. The augmented BP and CBF variability in older adults during sit-stand maneuvers indicate the presence of diminished cardiovascular regulatory capability and increased hemodynamic stress on the cerebral circulation with aging.