Aging Process on Spectrally Determined Spontaneous
Baroreflex Sensitivity
A 5-Year Prospective Study
Jean-Pierre Fauvel, Catherine Cerutti, Ignace Mpio, Michel Ducher
Abstract—The interindividual age-related decrease in baroreflex sensitivity (BRS) was reported in many cross-sectional
studies. However, the long-term intraindividual decrease in BRS has never been confirmed by longitudinal studies. Data
obtained from a 5-year prospective study designed to assess the 5-year stress effects on blood pressure (BP) provided
the opportunity to assess longitudinal aging process on spectrally determined BRS (S-BRS) using the cross spectral
analysis. This analysis was carried out in 205 men aged between 18 and 50 years who had 2 valid beat to beat BP
recordings (Finapress) at a mean 5-year interval. At inclusion and at end of follow-up, S-BRS was significantly
correlated with age (r⫽⫺0.50, P⬍0.001, r⫽⫺0.33, P⬍0.001 respectively). Interestingly, the slopes and the intercepts
were not significantly different at a 5-year interval. This result is in favor of the good reproducibility of S-BRS. The
attenuation with age of S-BRS was calculated at 3.6% a year. This decrease was slightly higher than the one obtained
with the baseline data (2.3% per year). This longitudinal study provided, for the first time, an estimate of the slope of
the age-related physiological S-BRS decrease in a mid-aged healthy male population. Our findings reinforce the interest
of evaluating spontaneous BRS reported to predict hypertension and cardiovascular events in various populations.
(Hypertension. 2007;50:543-546.)
Key Words: mental stress 䡲 baroreflex 䡲 hypertension 䡲 blood pressure 䡲 reproducibility
B
aroreflex sensitivity (BRS) is a relevant index of the
autonomic control of blood pressure (BP). Historically,
BRS was determined using drug-induced BP and heart rate
(HR) opposite variations. More recently, noninvasive techniques of determining spontaneous BRS have been implemented,1–3 the spectral method being one of the most popular.
The spectral determination of BRS (S-BRS) is a reliable
measure of BRS because it has often been reported to be
correlated with the pharmacologically determined BRS.4 –7
S-BRS is easy to determine, and repeated measures are easy
to make. Aging process on S-BRS has already been reported
in several cross sectional studies.8 –12 Such studies only report
interindividual S-BRS variation with age. Thus, S-BRS has
consistently been reported to decrease with age. Physiological
aging in men was associated with a marked reduction in
baroreflex buffering. A differential aging process on baroreflex buffering BP and cardiovagal baroreflex was reported.10
The decrease in baroreflex buffering with aging was related
to an increase in basal sympathetic nerve activity and to a
reduction in systemic Alpha1-adrenergic vascular responsiveness whereas the decrease in cardiovagal baroreflex was
related to a decrease in parasympathetic pathways.10,13 However, to our knowledge, the intraindividual BRS alteration
with age has never been reported. For that purpose, we
analyzed the S-BRS 5-year aging process in a cohort of 205
healthy young men. This unique data set was extracted from
a 5-year follow-up study designed to assess stress effects on
BP.14 In our study, because the sensitivity of the BRS control
of HR was studied using the cross spectral analysis of SBP
and HR, only the cardiovagal baroreflex was analyzed.
Subjects and Methods
Subjects
The study was conducted on full-time employment subjects working
in a chemical company, aged between 18 and 55 years (37⫾10
years). Among the 473 eligible subjects, 370 volunteered to participate in the study: 330 healthy subjects were included if their dipstick
urinalysis were within normal limits and if they were free of any
current medication that could interfere with BP regulation; 319
subjects had a valid baseline Finapres recording. After inclusion in
1995, subjects were followed for outcome until 2001. None of the
subjects was lost for follow-up and each subject who was planned to
discontinue the study had a final evaluation for the primary end point
of the stress study. During the 5-year follow-up, 94 subjects had a
premature final visit. Main reasons for discontinuation during
follow-up were retirement (31 subjects), job transfer (22 subjects),
and medical reason (n⫽9). No Finapres recording was planned in
case of premature discontinuation. Among the 257 subjects who had
a final visit 5 years later, 5 subjects refused to have a Finapres
Received March 10, 2007; first decision March 30, 2007; revision accepted June 8, 2007.
From the Hôpital E. Herriot (J.-P.F., I.M., M.D.), Department of Nephrology, and INSERM ERI22 and Université Lyon 1 EA4173 (J.-P.F., C.C.),
Université Claude Bernard, Lyon, France.
Correspondence to J-P. Fauvel, Département de Néphrologie et Hypertension Artérielle, Hôpital E. Herriot, 69437 Lyon Cedex 03, France. E-mail
jean-pierre.fauvel@chu-lyon.fr
© 2007 American Heart Association, Inc.
Hypertension is available at http://hyper.ahajournals.org
DOI: 10.1161/HYPERTENSIONAHA.107.090811
543
544
Hypertension
September 2007
Characteristics of the 205 Study Subjects at Baseline and at
the End of Follow-Up
tion of the modulus of the transfer function between SBP and HR
variations in the mid-frequency band (0.07 to 0.15 Hz).
Parameter
Baseline
End of Follow-Up
P Values
Statistical Analysis
BMI, kg/m
24.9⫾3.4
25.6⫾5.7
⬍0.001
Alcohol, score
2.10⫾0.99
1.91⫾0.95
⬍0.05
SBP, mm Hg
127⫾9
131⫾13
⬍0.001
63⫾8
63⫾9
NS
12.4⫾5.6
10.0⫾5.7
⬍0.001
Data are expressed as mean⫾SD in the text and tables. Mean S-BRS
was computed for 6 clusters of age (ⱕ25, 26 to 30, 31 to 35, 36 to 40,
41 to 45, ⱖ46 years). Mean values of S-BRS in each cluster of age were
compared using an ANOVA followed by a post hoc Fischer test. The
normal distribution of each variable was tested using a KolmokorowSmirnoff test. The age-related decrease in S-BRS in each cluster of age
was compared in absolute values and in percentage. The paired Student
t test was used to compare inclusion to end of follow-up mean values.
Pearson coefficients of correlation were computed to assess time process
for each variable. All statistical analyses were performed using Statistica
6 software (StatSoft Inc). P⬍0.05 was considered statistically
significant.
2
HR, bpm
S-BRS, ms/mm Hg
Data are expressed as means⫾SD.
recording and 27 recordings were not valid for S-BRS assessment.
Because the population was mainly composed of men (20 women),
the presented data of baroreflex aging process were restricted to the
205 men who had 2 valid beat-to-beat BP recordings at a 5-year
interval (5.3⫾1.4 years).
Anthropometric variables (age, body mass index [BMI]) are
summarized in the Table. Alcohol consumption was classified into 5
levels after interview (⬍1 drink/week, ⬍1 drink/d, 1 to 2 drinks/d, 3
to 4 drinks/d, ⬎4 drinks/d). All patients gave their written informed
consent and the study protocol was approved by the independent
Lyons Ethics Committee.
Protocol of Work-Site BP Measurement
The study was conducted at work site, in a quiet room with a
controlled temperature at 20°C. Work-site systolic (SBP) and diastolic (DBP) BP were measured 3 times in a row with the use of a
mercury sphygmomanometer, after 5 minutes resting in a recline
position. The average of the last 2 SBP and DBP measurements was
taken into account. These measurements were made at the beginning
and end of follow-up.
Beat-to-beat BP was also recorded using a Finapres device (model
2300, Ohmeda). The cuff was wrapped around the forefinger of the
nondominant arm resting on a table, whose level was adjusted to obtain
a less than 5 mm Hg difference with the previously determined BP
(mercury sphygmomanometer). The equipped arm of the seated subject
was held in the same position throughout the procedure. After a
2-minute period of familiarization, the automatic calibration was
switched off. BP and HR were then recorded for 10 minutes at rest.
BP Signal Analysis
Signal acquisition and data processing to assess spontaneous BRS
was previously described.15 In brief, BP signals were sampled at a
rate of 100 Hz and data were stored for further processing. Data
processing was performed on a 4-minute recording. The spectral
determination of BRS was assessed on a 256 points completed to 512
points by zero padding. Cross-spectral analysis of SBP and HR was
performed. The zero padding was used to increase the frequency
resolution of the LF band. S-BRS was estimated through computa-
Results
Mean age of subjects was 36⫾10 years at baseline. Subjects
whose baseline BP was greater than 140/90 mm Hg represented
11.2% of the population. The Table summarizes the mean
characteristics of the study population at baseline. SBP and HR
individual values were significantly correlated 5 years later
(r⫽0.61, P⬍0.001; r⫽0.59, P⬍0.001, respectively). S-BRS had
a normal distribution (d⫽0.072, NS). S-BRS individual values
were negatively correlated with age at inclusion and end of
follow-up (r⫽⫺0.50, P⬍0.001, Figure 1A; r⫽⫺0.33,
P⬍0.001, Figure 1B). After a 5-year interval, mean S-BRS was
significantly lower (Table). S-BRS at baseline was significantly
correlated with S-BRS at the end of follow-up (r⫽0.45,
P⬍0.001; Figure 1C). The 5-year decrease in S-BRS was
significant in clusters of ages (20 to 25, 26 to 30, 36 to 40, 41 to
45 years; Figure 2). However, the age-related decrease in SBRS,
neither in absolute values (Figure 3) nor in percentage (data not
shown), were significantly different between clusters of age. The
age related decrease in S-BRS was significantly correlated with
the baseline S-BRS (r⫽⫺0.51; P⬍0.001). BRS was slightly
correlated with BMI at baseline (r⫽⫺0.18, P⬎0.05) but not at
the end of follow-up (r⫽⫺0.09; NS). Five-year changes in
S-BRS were not significantly linked with those in BMI
(r⫽⫺0.14; NS).
Discussion
Cross-sectional studies suggested that BRS decreased with age
when comparing the BRS of subjects of various ages.8 –12 Our
longitudinal study is the first to confirm the decrease in S-BRS
Figure 1. Correlations between age and S-BRS at inclusion (A) and 5 years later (B). Correlation between S-BRS at inclusion and 5
years later (C). S-BRS indicates spectral baroreflex sensitivity.
Fauvel et al
Aging Process on Baroreflex
545
Figure 3. Box plot representation of the 5-year S-BRS alteration
by cluster of age. S-BRS indicates spectral baroreflex sensitivity.
Figure 2. S-BRS at inclusion and 5 years later in each cluster of
age. *P⬍0.05 at a 5-year interval. S-BRS indicates spectral
baroreflex sensitivity.
over time in men. Furthermore, between 20 and 50 years, the
5-year decrease in S-BRS is constant and not age-related.
We implemented a longitudinal study designed to assess the
stress effects on 5-year BP alteration in a large sample of
subjects.14 Individual BP stress reactivity and individual impact
of job strain were tested for BP alteration. Stress BP reactivity
was measured with Finapres device at a 5-year interval between
inclusion and end of follow-up. This data set provided a unique
opportunity to assess aging process on S-BRS.
BRS can be determined in the frequency domain by the
computation of cross-spectral analysis but also in the time
domain using the sequence method.3 The determination of
BRS in the temporal domain using the sequence method had
consistently been reported to be strongly correlated with
spectral BRS determination.15–18 As in our experience the
sequence method appeared to be slightly less reproducible at
short term (1 week), we now only determine BRS in the
frequency domain. Recently, Johnston et al also reported
lower coefficients of variation and reproducibility coefficients if BRS were determined with the spectral method
instead of the sequence method.19
As already reported in cross-sectional studies, S-BRS
among subjects aged between 18 and 50 years decreased with
age. In young adults, Davy et al20 showed that baroreflex
sympathetic pathway was preserved with age. Baroreflex
control of sympathetic nerve activity was not altered whereas
baroreflex control of parasympathetic nerve activity was
attenuated by aging in humans.13 Moreover Laitinen et al12
confirmed these results in a larger sample of subjects (n⫽63,
aged 23 to 77 years) using both the sequence and the
cross-spectral determination of BRS for Finapres recordings.
Such a result is important to consider because a baroreflex
impairment that is the consequence of the decrease of the
vagal autonomic function is considered as an independent
marker of mortality.21–22 We also found such an aging process
on BRS when analyzing baseline data. Five years later,
S-BRS was also significantly correlated with age. Interestingly, the slopes and the intercepts were not significantly
different at a 5-year interval, which is in favor of the good
reproducibility of the spectral determination of BRS. The
significant test-retest correlation obtained in a large number
of subjects at a 5-year interval showed that S-BRS has a
predictable time evolution. For instance, at a 5-year interval
the test-retest correlation coefficient of S-BRS (r⫽0.45,
P⬍0.001) was within the range of the SBP one (r⫽0.61,
P⬍0.001). Time evolution of S-BRS is thus as predictable as
the SBP one. Because S-BRS is partly derived from SBP
determination, any estimate of its time evolution is limited by
the time evolution of the BP signal. Because S-BRS is
reproducible in the short term15,23–25 and has a predictable
evolution in the long term, S-BRS can be considered an
intrinsic characteristic of each subject and therefore used in
research studies and clinical practice.
In cross-sectional studies, a large variability in interindividual BRS was consistently observed. Our study is the first
one to analyze the aging process on BRS by means of a
prospective design. A longitudinal design provided a better
control of BRS variability, each subject being his own
control. The slope of the age-related decrease in S-BRS
(⫺0.45 ms/mm Hg/year) was higher to the one obtained with
the cross-sectional data (⫺0.28 ms/mm Hg/year). In our
longitudinal study, the decrease in S-BRS represented 3.6%
per year. Although not significant, Figure 3 suggests that
S-BRS decrease with age could be attenuated after 45 years.
However, Umetani et al26 reported a linear decrease in
autonomic control from 10 to 99 years. A longitudinal study
including older subjects would be of major interest.
Limitations
Although controversial, cardiovagal baroreflex gain has been
reported to be significantly lower in women compared with
men.27 In our study, it was not possible to analyze gender
effects. Thus, our results should be restricted to measuring
S-BRS decrease between 18 and 50 years in a male population. In our study, physical activity was not recorded. Physical training increases BRS and might decrease cardiovascular risk in a given subject. The physiological decrease in BRS
is unclear. Major hypotheses include aging process on the
intima/media ratio of the carotid and aging process on the
speed of nerve conduction. The age-related decrease in
S-BRS could also be a mixture of cell aging, reduced physical
546
Hypertension
September 2007
activity, and BMI increase. As our cohort study did not
control these factors, it was not possible to differentiate
relative influences of each potential mechanism.
Perspectives
The BRS that is easily determined using a 5-minute rest BP
recording should be more frequently monitored. A normal
BRS is an index of the integrity of the autonomic nervous
system. It is an intrinsic characteristic of each subject which
might be considered as an integrator of many physiopathological events. BRS is increased by physical activity and
conversely altered in diabetes and hypertension. A study
whose aim would be to evaluate its prognostic value in
healthy populations to predict hypertension should be undertaken. BRS has also been associated with metabolic risk
factors and proposed to be an ideal index of cardiovascular
risk in hypertension.28 Furthermore, the decrease in the vagal
component of BRS may also be considered as an independent
risk factor of cardiovascular events as recently reported in
chronic renal failure.29 The contribution of BRS to treatment
effects could be considered in clinical trial. For instance, in
secondary prevention, endarterectomy of the internal carotid
artery showed the increase of BRS and its improving was
correlated to cardiovascular events at a 5-year interval.30
This longitudinal study confirms the age-related decrease
in BRS already reported in cross sectional studies. In men
aged 18 to 50 years, the decrease in BRS was evaluated at
3.6% a year. Our findings reinforce the interest of evaluating
spontaneous BRS known to predict hypertension31 and cardiovascular events in various populations.21,22
Acknowledgments
The authors thank Christiane Bertheux-Beaufrere for translation support. The authors also thank all of the subjects who volunteered for
the study.
Disclosures
None.
References
1. Robbe HW, Mulder LJ, Ruddel H, Langewitz WA, Veldman JB, Mulder
G. Assessment of baroreceptor reflex sensitivity by means of spectral
analysis. Hypertension. 1987;10:538 –543.
2. Pagani M, Somers V, Furlan R, Dell’Orto S, Conway J, Baselli G, Cerutti
S, Sleight P, Malliani A. Changes in autonomic regulation induced by
physical training in mild hypertension. Hypertension. 1988;12:600 – 610.
3. Bertinieri G, di Rienzo M, Cavallazzi A, Ferrari AU, Pedotti A, Mancia
G. A new approach to analysis of the arterial baroreflex. J Hypertens.
1985;3:S79 – 81.
4. Colombo R, Mazzuero G, Spinatonda G, Lanfranchi P, Giannuzzi P,
Ponikowski P, Coats AJ, Minuco G. Comparison between spectral analysis and the phenylephrine method for the assessment of baroreflex
sensitivity in chronic heart failure. Clin Sci (Lond). 1999;97:503–513.
5. Pitzalis MV, Mastropasqua F, Passantino A, Massari F, Ligurgo L, Forleo
C, Balducci C, Lombardi F, Rizzon P. Comparison between noninvasive
indices of baroreceptor sensitivity and the phenylephrine method in postmyocardial infarction patients. Circulation. 1998;97:1362–1367.
6. Watkins LL, Grossman P, Sherwood A. Noninvasive assessment of
baroreflex control in borderline hypertension: comparison with the phenylephrine method. Hypertension. 1996;28:238 –243.
7. Parlow J, Viale JP, Annat G, Hughson R, Quintin L. Spontaneous
cardiac baroreflex in humans: comparison with drug-induced
responses. Hypertension. 1995;25:1058 –1068.
8. Ebert TJ, Morgan BJ, Barney JA, Denahan T, Smith JJ. Effects of aging
on baroreflex regulation of sympathetic activity in humans. Am J Physiol.
1992;263:H798 –H803.
9. Gribbin B, Pickering TG, Sleight P, Peto R. Effect of age and high blood
pressure on baroreflex sensitivity in man. Circ Res. 1971;29:424 – 431.
10. Jones PP, Christou DD, Jordan J, Seals DR. Baroreflex buffering is
reduced with age in healthy men. Circulation. 2003;107:1770 –1774.
11. Fluckiger L, Boivin JM, Quilliot D, Jeandel C, Zannad F. Differential
effects of aging on heart rate variability and blood pressure variability.
J Gerontol A Biol Sci Med Sci. 1999;54:B219 –B224.
12. Laitinen T, Hartikainen J, Vanninen E, Niskanen L, Geelen G, Lansimies
E. Age and gender dependency of baroreflex sensitivity in healthy
subjects. J Appl Physiol. 1998;84:576 –583.
13. Matsukawa T, Sugiyama Y, Mano T Age-related changes in baroreflex
control of heart rate and sympathetic nerve activity in healthy humans.
J Auton Nerv Syst. 1996, 12;60:209 –212.
14. Fauvel JP, M’Pio I, Quelin P, Rigaud JP, Laville M, Ducher M. Neither
perceived job stress nor individual cardiovascular reactivity predict high
blood pressure. Hypertension. 2003;42:1112–1116.
15. Ducher M, Fauvel JP, Gustin MP, Cerutti C, Najem R, Cuisinaud G,
Laville M, Pozet N, Paultre CZ. A new non-invasive statistical method to
assess the spontaneous cardiac baroreflex in humans. Clin Sci (Colch).
1995;88:651– 655.
16. Laude D, Elghozi JL, Girard A, Bellard E, Bouhaddi M, Castiglioni P,
Cerutti C, Cividjian A, Di Rienzo M, Fortrat JO, Janssen B, Karemaker
JM, Leftheriotis G, Parati G, Persson PB, Porta A, Quintin L, Regnard J,
Rudiger H, Stauss HM. Comparison of various techniques used to
estimate spontaneous baroreflex sensitivity (the EuroBaVar study). Am J
Physiol Regul Integr Comp Physiol. 2004;286:R226 –R231.
17. Westerhof BE, Gisolf J, Stok WJ, Wesseling KH, Karemaker JM. Timedomain cross-correlation baroreflex sensitivity: performance on the
EUROBAVAR data set. J Hypertens. 2004;22:1371–1380.
18. Porta A, Baselli G, Rimoldi O, Malliani A, Pagani M. Assessing baroreflex
gain from spontaneous variability in conscious dogs: role of causality and
respiration. Am J Physiol Heart Circ Physiol. 2000;279:H2558–H2567.
19. Johnson P, Shore A, Potter J, Panerai R, James M. Baroreflex sensitivity
measured by spectral and sequence analysis in cerebrovascular disease:
methodological considerations. Clin Auton Res. 2006;16:270 –275.
20. Davy KP, Tanaka H, Andros EA, Gerber JG, Seals DR. Influence of age
on arterial baroreflex inhibition of sympathetic nerve activity in healthy
adult humans. Am J Physiol. 1998;275:H1768 –H1772.
21. Osterziel KJ, Hanlein D, Willenbrock R, Eichhorn C, Luft F, Dietz R.
Baroreflex sensitivity and cardiovascular mortality in patients with mild
to moderate heart failure. Br Heart J. 1995;73:517–522.
22. La Rovere MT, Bigger TJ, Marcus FI, Mortara A, Schwartz PJ; for the
ATRAMI (Autonomic Tone and Reflexes After Myocardial Infarction)
investigators. Baroreflex sensitivity and heart rate variability in prediction of
total cardiac mortality after myocardial infarction. Lancet. 1998;351:
478–484.
23. Iellamo F, Legramante JM, Raimondi G, Castrucci F, Massaro M, Peruzzi
G. Evaluation of reproducibility of spontaneous baroreflex sensitivity at
rest and during laboratory tests. J Hypertens. 1996;14:1099 –1104.
24. Davies LC, Francis D, Jurak P, Kara T, Piepoli M, Coats AJ. Reproducibility
of methods for assessing baroreflex sensitivity in normal controls and in
patients with chronic heart failure. Clin Sci (Lond). 1999;97:515–522.
25. Herpin D, Ragot S. Mid- and long-term reproducibility of noninvasive
measurements of spontaneous arterial baroreflex sensitivity in healthy
volunteers. Am J Hypertens. 1997;10:790 –797.
26. Umetani K, Singer DH, McCraty R, Atkinson M. Twenty-four hour time
domain heart rate variability and heart rate: relations to age and gender
over nine decades. J Am Coll Cardiol. 1998;31:593– 601.
27. Beske SD, Alvarez GE, Ballard TP, Davy KP. Gender difference in cardiovagal baroreflex gain in humans. J Appl Physiol. 2001;91:2088–2092.
28. Lantelme P, Khettab F, Custaud MA, Rial MO, Joanny C, Gharib C,
Milon H. Spontaneous baroreflex sensitivity: toward an ideal index of
cardiovascular risk in hypertension? J Hypertens. 2002;20:935–944.
29. Johansson M, Gao SA, Friberg P, Annerstedt M, Carlstrom J, Ivarsson T,
Jensen G, Ljungman S, Mathillas O, Nielsen FD, Strombom U.
Baroreflex effectiveness index and baroreflex sensitivity predict all-cause
mortality and sudden death in hypertensive patients with chronic renal
failure. J Hypertens. 2007;25:163–168.
30. Hirschl M, Kundi M, Blazek G. Five-year follow-up of patients after
thromboendarterectomy of the internal carotid artery: relevance of baroreceptor sensitivity. Stroke. 1996;27:1167–1172.
31. Ducher M, Cerutti C, Chatellier G, Fauvel JP. Is high job strain associated
with hypertension genesis? Am J Hypertens. 2006;19:694 –700.