A study of presbycardia, with gender differences favoring ageing women

A study of presbycardia, with gender differences favoring ageing women

David F. Goldspink a,⁎, Keith P. George a, Paul D. Chantler a, Richard E. Clements a,Lisa Sharp a, Gary Hodges a, Claire Stephenson a, Thomas P. Reilly a, Ashish Patwala b,

Tamas Szakmany c, Lip-Bun Tan d, N. Timothy Cable a

a Research Institute for Sports and Exercise Sciences, Liverpool John Moores University, Liverpool, L3 2EF, UKb Cardiology, The Cardiothoracic Centre in Liverpool, Thomas Street, Liverpool, L14 3PE, UK

c Department of Critical Care, Whiston Hospital, St Helens and Knowsley Hospital NHS Trust, UKd Academic Unit of Molecular Cardiovascular Medicine, University of Leeds, LS1 3EX, UK

Received 1 April 2008; accepted 28 June 2008Available online 20 August 2008

Abstract

Background: The impact of ageing on the human cardiovascular system has been the subject of several studies in recent years, but withinsufficient emphasis on defining sex-specific differences. To rectify this, gender-specific differences in structure and function in the humancardiovascular system were studied in a European population during natural ageing.Methods: Cardiac power output (CPO) was measured and integrated with changes in left ventricular (LV) mass, diastolic, systolic and limbblood flow, blood pressure and exercise capacity in 93 health-screened men and 122 women, aged 20 to 75 years.Results: Correlatingwith a 21% loss of LVmass,maximumcardiac pumping (i.e. CPOmax= Q̇max xMAPmax) and reserve (CR = CPOmax −CPOrest)capacities decreased 20–25%with age in male hearts. In contrast, CPOmax, CR and LVmasswere all preserved in ageing women.Maximum cardiacoutput (Q̇max; 26–32%), peak forearm blood flow (FBFpeak; 61%) and exercise capacity (40–50%) all decreased, but more so in men than women. Incontrast, systemic vascular resistance (68–75%) and mean arterial pressure (MAPmax; 14–26%) increased in both sexes.

CPOrest decreased 27% in men, but was unchanged in women, despite lower early:late diastolic filling (48–51%), Q̇rest (19–23%) andFBFrest (56%) in both sexes.Conclusions: Understanding sex-specific differences in cardiovascular ageing is important for public health and biomedical research, givenincreasingly larger older populations and the need to prevent and treat cardiovascular disease.© 2008 Elsevier Ireland Ltd. All rights reserved.

Keywords: Healthy ageing; Gender differences; Overall cardiac function; Central and peripheral blood flow; LV mass; Exercise capacity

1. Introduction

Dorland's Medical Dictionary defines Presbycardia as“impaired cardiac function attributed to the ageing process,occurring in association with recognizable changes of senes-cence in the body and in the absence of convincing evidenceof other forms of heart disease”. The impact of ageing onhumans has been the subject of several studies in recent years

because of the biological, clinical and economic importance ofburgeoning ‘ageing populations’ in developed countries.Because the incidence of coronary heart and cardiovasculardisease increases with age, and these are the most commoncauses of morbidity and mortality in the western world, thecardiovascular system has been a major focus of attention [1].

Many studies on human ageing exhibit a wide range offactors influencing biological variables, such as differences inethnicity, activity, diet, socio-economic status, sex, hormonaland health status. Equally varied has been the techniques usedto detect age-related changes within the cardiovascular sys-tem. Such factors probably explain why after so many studiesdiscrepancies still exist. These include a wide variation in theperceived impact of ageing on heart size and its LV mass in

International Journal of Cardiology 137 (2009) 236–245www.elsevier.com/locate/ijcard

⁎ Corresponding author. Research Unit for Human Development andAgeing, Research Institute for Sports and Exercise Sciences, Liverpool JohnMoores University, 15-21 Webster Street, Liverpool, L3 2ET, UK. Tel.: +44151 231 4328/4554; fax: +44 151 231 4353.

E-mail address: [email protected] (D.F. Goldspink).

0167-5273/$ - see front matter © 2008 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.ijcard.2008.06.086

mailto:[email protected]

http://dx.doi.org/10.1016/j.ijcard.2008.06.086

both men and women [2–4]. Similarly, discrepancies exist asto the impact of ageing on blood flow generation at rest andpeak exercise [5–7].

While age remains a major risk factor in both sexes, someage-related physiological adaptations within the cardiovascu-lar system of healthy individuals, and women in particularly,remain unclear. Hence, we sought to identify sex-specificdifferences, and potential spatial and temporal variations with-in the cardiovascular system, with ageing by conducting astudy inwhich several relatedmeasurements weremade on thesame rigorously health-screened men and women. Left ven-tricular mass, cardiac power output as a measure of overallcardiac function [8], central and peripheral blood flow, bloodpressure and exercise capacity were measured in men andwomen between the ages of 20 and 75 years, thereby coveringmost of the adult life span.

1.1. Our hypothesis

Sex-specific differences emerge within the cardiovascularsystem with ageing.

2. Materials and methods

This study was approved by the Human Ethics Commit-tee of Liverpool John Moores University and all participantsgave their written informed consent.

2.1. Participants

Approximately 500 volunteers were recruited in Liver-pool in the North West of England, with the aim of studyingapproximately 100 of each sex. Strict health criteria wereadopted before admission into the study, with all participantsundergoing extensive medical history/lifestyle question-naires followed by a maximal treadmill exercise stress test,with concomitant ECG. Only those who were free from overtcardio-respiratory and neuromuscular problems, were nor-motensive, had a BMI b30 kg/m2, were neither cigarettesmokers nor taking medications were investigated further.This amounted to 93 sedentary, but healthy men and 122women, ranging from 20 to 75 years. All participants re-frained from consuming food (3 h), caffeine (6 h), alcohol(12 h) and strenuous exercise (24 h) before measurementswere made on each individual over 2 days.

2.2. Body composition

LBM (i.e. fat- and bone-free masses) were determinedby DEXA (Hologic Inc, Horizon Park, Levensesteenweg,Belgium).

2.3. Echocardiography

While in a supine resting position participants underwent2D-echocardiographic assessment (Acuson Cypress Echo

System, Acuson Corporation, CA). Using standard echocar-diographic methods [9], the LV mass of each subject wasdetermined (CV=7.1%).

2.4. Cardio-respiratory functions

2.4.1. Stage 1. Oxygen consumption during maximal exercisestress test

Each subject's heart rate was determined from a 12-leadECG, which was monitored throughout a treadmill exercisestress test designed to measure their V̇O2max [8,10]. Oxygenconsumption (V̇O2), CO2 production (V̇CO2), end tidal par-tial pressure of CO2 (PETCO2), tidal ventilation (V̇E) and res-piratory rate were measured using the Medgraphics CPX-Dsystem (Medgraphics Corporation, St Paul, Minnesota,USA). V̇O2max was considered to have been achieved whena plateau in V̇O2 had been reached despite additional in-creases in workload, a HR greater than 95% of the subject'sage-predicted maximal value and a RER N1.1 (CV forV̇O2max=3.5%).

2.4.2. Stage 2. Measurements of CPOMAP was calculated [11] from brachial artery blood pres-

sure measured by manual auscultation at rest and at two-minute intervals up to, and including, maximum exercise(CV for MAPrest=5.1% and MAPmax=3.7%). Duplicatemeasurements of cardiac output (Q̇) were made by rebreath-ing either 10% CO2 at rest [12], or 4% CO2 at maximumexercise [9], from a gas mixture contained within twice thesubject's tidal volume [13], with CV for Q̇rest =4.2% andQ̇max=2.3%. CPOrest (CV=2.2%) and CPOmax (CV=4.5%)were derived from these measurements of MAP and Q̇.

2.5. Calculations

CPO (Watts) = (Q̇×MAP)×2.22×10− 3 [8].CR=CPOmax−CPOrest.a-v diff (ml O2/dl blood)= (V̇O2max

/Q̇max) / 10.SVR (dyne sec cm5)= (MAP/Q̇)×80.

2.6. Forearm blood flow

Resting FBF was measured by strain gauge plethysmo-graphy [14] following 30 min of supine rest at 22 °C. Venousocclusion cuffs were placed around the wrist (to occludeblood flow to the hand) and upper arm, with the latter rapidlyinflated to 50 mm Hg every 15 s (i.e. 10 s of inflationfollowed by 5 s of deflation) to impede venous return but notarterial influx. The resultant increase in forearm volume wasmeasured by mercury-in-silastic strain gauges positionedaround the widest portion of the forearm. Peak FBF wasdetermined after 5 min of arterial occlusion (200 mm Hg) ofthe forearm. After removal of occlusion, reactive hypaeremicblood flowwas measured (as above) every 15 s for 2 min, withpeak limb blood flow typically occurring within 15–30 s.Mean values were obtained from triplicate measurements of

237D.F. Goldspink et al. / International Journal of Cardiology 137 (2009) 236–245

FBFrest (CV=5.7%) and FBFpeak (CV=4.8%), with FVCcalculated from FBF/MAP.

2.7. Multivariate allometric scaling

LVmass, CPO and V̇O2max are known to be size-dependent.Where appropriate these variables were scaled allometrically toLBM using the equation y=axb, where the exponent b is theslope of the log–log plot and a is derived from the antilog of theintercept on the y-axis [10]. Power-function ratios (e.g. LVmass/LBMb) were derived for body size and body compositionindependent variables.

2.8. Statistical analyses

All data are presented as means±SEM. Normally dis-tributed data were analyzed using parametric tests, linearregression analyses performed to determine the associationamong variables. Pearson's product-moment correlation co-efficients were used to indicate the magnitude and directionof relations between variables. The slopes of the regressionlines for men and women were compared using analysis ofcovariance, which included an interaction term. Significantdifferences were expressed as P≤0.05.

3. Results

3.1. Demographics

A total of 122 women (mean 47±14 years) and 93 men(mean 43±18 years), ranging from 20 to 75 years, wereintensively studied after being rigorously screened and foundfree of any known cardiovascular diseases and defined assedentary [15].

Table 1LV mass, cardiac power and peripheral blood flow±allometric scaling toLBM in 20 year-old men and women.

Men Women

Body mass (kg) 83.0±1.3 68.2±1.0*BMI (kg/m2) 26.7±0.4 25.9±0.3Surface area (m2) 2.0±0.01 1.7±0.01*LV mass (g) 196±39 169±31*LV mass (g/kg LBM0.56) 20.2±3.1 20.7±3.4CPOrest (W) 1.0±0.01 0.8±0.01**CPOrest (W/kg LBM0.64) 0.08±0.01 0.07±0.01CPOmax (W) 5.18±0.1 4.07±0.1**CPOmax (W/kg LBM0.73) 0.29±0.01 0.26±0.01CR (W/kg LBM0.76) 0.21±0.01 0.19±0.01FBFrest (ml/100 ml tissue/min) 3.79±0.9 3.45±0.9FBFpeak (ml/100 ml tissue/min) 51.0±1.4 43.9±1.0*

*Pb0.05; **Pb0.01.

Fig. 1. Left ventricular mass and blood flow in men and women at rest.Changes in LV mass (A), early to late (E:A ratio) diastolic filling (B) andLVEF (C) for men (blue open symbols) and women (red solid symbols).

238 D.F. Goldspink et al. / International Journal of Cardiology 137 (2009) 236–245

3.2. Body composition

Body mass, BMI and surface area were all greater inmen than women, but none changed significantly over the55 years covered (Table 1). However, significant(Pb0.05) alterations in body composition occurred, withmen and women gaining 29% and 31% more fat whilelosing 9% and 11% LBM, respectively. Such age-relatedchanges in body composition need to be accounted forwhen measuring cardiac function and morphology, asshown below.

3.3. LV morphology and CPO

Although averaged, absolute values of LV mass, CPOmax

and CPOmax were greater in the larger men, these gender

differences disappeared after allometric scaling for LBM(Table 1). This demonstrates the importance of appropriatenormalization of data when investigating gender differ-ences and clearly indicates that, relative to body size andcomposition, the pumping capacity of the heart isequivalent in men and women. Nonetheless, significantsex-specific differences emerged with ageing.

Between 20 and 75 years LV mass decreased (Pb0.001)by 21% in men (Fig. 1A), but increased non-significantly(P = 0.19) by 13% inwomen. Consistent with these changes inmorphology, CPOmax (20%) and CR (25%) decreased(Pb0.0001) in men (Fig. 2A and C), but not in women(Fig. 2B and D), whether allometrically scaled to LBM (datanot shown) or not. This lowering of power output from themale heart equated to 1.6Wover 55 years, but only 0.06 W inwomen (P=0.8).

Fig. 2. Effects of ageing on cardiac functional capacity in men and women. CPO at rest (open symbols) and maximal exercise (solid symbols) are shown for men(A) and women (B), as well as CR for both men (C) and women (D).



3.4. Central and peripheral haemodynamics

3.4.1. Measured at maximal exerciseWhen considering the heart's maximum generation of

blood flow it was apparent that Q̇max decreased significantly(Pb0.0001) in both sexes across the 55 years. This washowever more pronounced in men (32%; Fig. 3A) thanwomen (26%; Fig. 3B), because significant (Pb0.05) reduc-tions in both SVmax (13%; from 120 to 104 ml) and HRmax

(23%, from 195 to 150 beats min−1) were observed in men,while in women only HRmax decreased (19%; from 190 to154 beats min−1) significantly (Pb0.0001).

The higher absolute values of Q̇max in men versus women(Fig. 3A and B) were matched by similarly higher rates ofpeak FBF (Table 1). However, with ageing similar overallpercent reductions in both FBFpeak (60%) and FVCpeak

(62%) were observed (Pb0.05) in both sexes (Fig. 4).According to the best regression fit, these peripheral changesoccurred more steeply between 20 and 40 years (approxi-mately 60% of total), with slower rates of decline thereafter(Fig. 4).

In contrast to changes in blood flow, MAPmax increasedby 14% in men (Fig. 3C) and 26% in women (Fig. 3D)across the 55 years. The greater increase in MAPmax inwomen corresponded with a marginally steeper age-relatedincrease in SVRmax (Pb0.001), i.e. 75% versus 65% in men(Fig. 3E and F). The increase in male MAPmax arose from a53% increase in DBPmax (Pb0.001) with no change inSBPmax (195±2.0 mm Hg), while the steeper rise in femaleMAPmax involved increases (Pb0.05) in both SBPmax (9%;from 181 to 197 mm Hg) and DBPmax (63%; from 58 to

95 mm Hg). So, while more pronounced decreases in max-imal blood flow were found in ageing men, the reverse wastrue with increases in blood pressure, these being greater inwomen.

3.4.2. Measured at restWe next investigated whether any such age-related, gen-

der differences in cardiac function were apparent at rest.CPOrest decreased by 27% (Pb0.001; Fig. 2A) in men, butremained unchanged in women (Fig. 2B).

In terms of central blood flow, similar (Pb0.001) de-creases (48–51%) in LV passive:active diastolic filling wereobserved in both sexes (Fig. 1B). LVEFs were all normal,and while marginally higher in women (67±7%) than men(64±6%) did not change with age (Fig. 1C). Althoughresting heart rates remained unchanged at 67±2 beats min−1,stroke volumes decreased more in men (38%) than women(19%), leading to a more marked reduction in Q̇rest for men(41%; Fig. 3A) than women (23%; Fig. 3B). Although limbblood flow remained higher in men than women (Table 1),this fell by similar extents (FBFrest by 56% and FVCrest by54%) in both sexes.

With ageing significant (Pb0.01) increases in MAPrestoccurred in both men (10%; Fig. 3C) and women (16%;Fig. 3D), along with age-related increases in SVRrest (Fig. 3Eand F).

3.5. Aerobic exercise capacity

Whilst health screening and determining CPOmax, V̇O2max

was measured in each participant. This enabled us to establish

Fig. 4. Peak forearm blood flow in men and women, with age. Changes in peak FBF (A) and FVCmax (B) for men (blue open symbols) and women (red closedsymbols).

Fig. 3. Age-related changes in blood flow and pressure generation in men and women. Resting (open symbols) and maximal (solid symbols) values of Q̇, MAPand SVR are presented for men (A, C and E) and women (B, D and E).


that all our participants were essentially sedentary [15],thereby eliminating activity status as a potential confoundinginfluence from our study.

Although the gap between the higher absolute values ofV̇O2max for men versus women (Fig. 5A) narrowed whenratio corrected for differences in body mass (Fig. 5B), thisremained significantly (Pb0.05) different. Although it iscommon practice to normalize data in this way, thephysiological variable with the strongest correlation toexercise capacity, cardiac morphology and function isLBM [10]. Also, women possess proportionately lessLBM. Allowance for this is crucial if meaningful genderdifferences are to be established. All apparent genderdifferences in exercise capacity (Fig. 5A and B) disappearedwhen allometrically scaled to LBM (Fig. 5C).

Exercise capacity was significantly (Pb0.01) lower in theoldest men (50%) and women (40%), compared with theiryoungest counterparts (Fig. 5). Differences in a-vO2diff werecalculated; these suggesting that muscles of older men (24%)and women (17%) extract less oxygen than 20 year-olds(Fig. 5D).

4. Discussion

On average, women in Western societies have superiorlife expectancies than men, although the reasons for this arenot clear. Presbycardia may be a major contributing factor,and for this reason we conducted this important epidemio-logical survey into ageing-related structural and functionalchanges in the cardiovascular system in a population of healthy

Fig. 5. Exercise capacity in men and women, with age. V̇O2max, in men (blue open symbols) and women (red solid symbols) are presented before (A), and afterallowing for differences in body mass (B), or scaling to LBM (C). Differences in a-vO2diff (D).


individuals in Liverpool (home city of the Beatles), England.Our measurements show that there were no significantdifferences in vascular ageing between men and women, butsignificantly greater cardiac ageing in men than women, evenafter correcting for body composition and size. This observa-tion confirms that men are significantly more susceptible topresbycardia than women. Whether this impacts on lifeexpectancy or not cannot be determined because this studyexcluded, a priori, any contribution of cardiovascular diseaseson our measurements of ageing-related changes in cardiovas-cular morphology and function.

Our observations indicate that if measurements of cardiacmorphology, function and exercise capacity are not correctedfor differences in body composition, and LBM in particular,sex-specific differences may be erroneously assumed (Table 1,Fig. 5A–C). Allometric scaling to LBM is mathematicallymore precise [10], whilst recognizing that women possessproportionately less skeletal muscle and that this tissue con-sumes most of the oxygen during exercise.

The advantage of measuring overall cardiac function asCPO is demonstrated here as it is an all encompassing mea-sure, taking into account all age-related changes in central(e.g. contractility) and/or peripheral (e.g. loading conditions)factors likely to influence overall cardiac performance. Assuch CPO provides the best available measure of overallorgan function [16] and has proved to be one of the mostpowerful predictors of prognosis in heart failure patients;superior to Q̇ which only reflects cardiac flow generation perse [17–21]. Because of the close relationship betweenCPOmax and prognosis in cardiac dysfunction [17–21], itseemed reasonable to use this approach to investigate wheth-er the effects of cardiac ageing are gender specific, or not.The use of CPO has enabled us to gain the following im-portant insight into presbycardia:

i. that the hydraulic power delivered by the heart to thebody is commensurate with the size and compositionof the body, because when normalized appropriatelythe maximal pumping and reserve capacities of theheart are essentially the same in men and women(Table 1),

ii. nonetheless, sex-specific differences exist in responseto natural ageing, and because these are unaffected bybody size/composition as shown after corrections, theymust represent real biological phenomena,

iii. that the heart diverts less energy to blood flow gen-eration and more towards generating higher bloodpressure to overcome greater vascular resistances inolder individuals (Fig. 3), as reflected by decreases inlimb blood flow (Fig. 4).

iv. that the decline of overall cardiac function in mencontributes to their steeper diminution of exercise ca-pacity, compared with women (Fig. 5) with age.

The significant age-related decreases (13–25%) in SVmax,CPOmax and CR (Fig. 2A and C) in men correlated well with

a 21% loss of LV mass (Fig. 1A), whether measured by Echoor MRI [2,4]. In contrast, over the same age range the femaleheart demonstrates resilience to ageing, with no reductions inLV mass (Fig. 1A [2–4]), SVmax, CPOmax or CR (Fig. 2Band D). There are two possible mechanistic explanations forthese sex-specific differences:

First, over the adult life span while men progressivelylose billions (30–35%) of their ventricular myocytes, womenpreserve their full complement [3]. By delaying myocytesenescence and stimulating anti-apoptotic signals, oestrogen[22] and higher local and systemic levels of IGF-1 [23]protect the female heart and vasculature by lowering theirrate of myocyte death, which in both women and femalerodents is 2- to 3-fold lower than in the equivalent malehearts [24,25]. With the recent ground breaking discovery ofresident stem and progenitor cells in the hearts of humansand animals, myocyte renewal is now accepted as importantin myocyte homeostasis [26]. Sex-related differences in theregulation of both the myocyte death and renewal pathways,may ultimately provide a definitive explanation for this fas-cinating gender difference in myocyte attrition, which willclearly impact on LV mass (Fig. 4) and contractility, espe-cially in men during ageing.

Second, as a consequence of changes in ventricular andarterial compliance [27] and increased SVR (Fig. 4C–F),systolic and pulse pressures increase, with the pulse wavetraveling faster to coincide with late systole in older indi-viduals, thereby increasing the pulsatile afterload and work-load on their LV [28]. Since this is usually more pronouncedin older women than men [28], this haemodynamic effect,along with the preservation of more ventricular myocytes,could be critical in preserving both LV mass (Fig. 1A) andCPOmax (Fig. 2C and D) in ageing women, as opposed to thesignificant losses of myocytes and cardiac function in thehearts of older men [3].

Like CPO, LVEF is measured as an index of cardiacfunction. However LVEF fails to correlate with the sex-specific differences observed here, for while resting LVEFswere slightly higher in women than men [29], unlike CPO,LVEF did not change in either sex with age (Fig. 1C). This isconsistent with the long-known observation that LVEF doesnot correlate with exercise capacity or V̇O2max[30]. Strokevolumes however were influenced by age, but less so in olderwomen than in men, thereby explaining the smaller reduc-tions in female Q̇rest and Q̇max (Fig. 3A and B). As well asagreeing with some [6,7], but not all [5] previous studies thisobserved lowering of central blood flow (Fig. 3A and B)with age was in keeping with reductions in limb blood flow(Fig. 4) and exercise capacities (Fig. 5); all more pronouncedin men than women. However according to best fit regres-sion lines, unlike the linear age-related decline in Q̇max,FBFpeak and FVCpeak appeared to decline more steeplybetween 20 and 40 years, and to similar extents in both sexes(Fig. 4). Similar changes in peripheral blood flow have alsobeen described in the leg [31,32]. While active vasodilationstill occurs during exercise, older men and women possess


less vasodilatory capacity (Fig. 4) [32]. Presumably this islinked to an age-related diminution of physical activity lev-els, with modulation of the regulatory impact of local musclemetabolites and shear forces on endothelial-dependent andendothelial-independent vasodilatory mechanisms. Suchreductions in peripheral blood flow, particularly in sedentaryolder people, increase the risk of thrombosis, atherosclerosisetc. The loss of oestrogens, as both a vasodilatory and vaso-protective agent, may also contribute to these changes inperipheral blood flow (Fig. 4) [32,33] and the increasedpostmenopausal incidence of cardiovascular disease andcardiac events [33]. However, we were not able to detect anyspecific deflections in blood flow relating to the menopause(Fig. 4).

The magnitude of the changes in VO2max (Fig. 5) isconsistent with earlier observations in cross-sectional [31]and longitudinal [34,35] studies, except that many previousstudies studied only one sex and normalized data to bodymass and not LBM. Along with the changes in cardiacmaximum pumping and reserve capacities, the decline inVO2max could clearly influence quality of life and functionalindependence.

4.1. Limitations of the present study

Although desirable, a longitudinal study of similar sizecovering an age span of 55 years was not realistically possible.Even in the few longitudinal studies examining age-relatedchanges in VO2max, let alone cardiovascular function, criticalcompromises have had to be made, e.g. either using fewparticipants (n=5) [34] or by including participants who,although classified as healthy, were current or former cigarettesmokers [35]; an exclusion criterion in our study. Other dif-ficulties in conducting longitudinal studies [35] are demon-strated by the fact that the decline in VO2max accelerates withincreasing age, and so far the predicted changes in VO2max perdecade have been calculated from measurements made over amedian interval of only 7.9 years.

Since our study was cross-sectional, we employed strictentry criteria to minimise biological variations in our com-munity-based participants to dissociate, as far as possible,any confounding effects of either disease or activity status onthe ageing process, something seldom done previously. Wewere not however able to control for differences in dietaryintake.

Individual investigators have their own preferences formeasuring cardiac output (Q̇) non-invasively. Some considerthe rebreathing of an inert gas (e.g. C2H2 or N2O) to be moreaccurate than CO2, as used here [8]. However, the automatedequilibrium CO2 method at rest [12] and the exponentialCO2 method at maximum exercise [13,36] have recentlybeen shown to give very similar measurements of Q̇ asrebreathing N2O [37]. All of these non-invasive approachesare more convenient, and safer with exercise, than using the‘gold standard’ invasive methods of the Direct Fick, dye- orthermo-dilution techniques, which themselves are called into

question if applied during maximum exercise [38]. In therelatively few studies that have directly compared theseinvasive and non-invasive techniques on healthy humans atrest and during exercise, similar ranges of variation have beenobserved [38]. Any perceived inherent weaknesses in ouranalytical techniques (e.g. using Echo and notMRI to examinecardiac morphology, and rebreathing CO2 rather than an inertgas to measure Q̇) were consistently applied to each and everyone of our male and female participants, using the same trainedinvestigators throughout. Hence, any intrinsic technical orhuman variability was applied equally throughout the entirestudy. As the gender-specific, age-related differences for LVmasses and CPO were already highly statistically significant,any technical improvements would only have increased theselevels further, but without affecting the overall outcome of thestudy. Also, bymaking all our cardiovascularmeasurements onthe same cohort of participants our data could be integratedmore effectively than by themore usual approach of comparingdata sets from smaller, and often incompatible, studies.

5. Conclusion

In this epidemiological survey on a representative popula-tion of our European society we have characterised the impactof natural ageing on the cardiovascular system. Many of thechanges, e.g. V̇O2max, SV, Q̇, FBF andCPO probably representphysiological adaptations to reduced activity levels, and ageingappears to be a continuous process between 20 and 75 years asno specific age of onset was detected. A significant, novelfinding is the sex-specific susceptibility to cardiac ageing, withwomen showing greater resilience than men to the loss ofcardiac myocytes, LV mass and overall organ function. Evenafter correcting for differences in body composition, as well assize betweenmen andwomen, sex-related disparities in cardiacmorphology and function persisted with ageing. Based onthese discoveries, the question why women's hearts are betterprotected from ageing requires more intensive biomedicalresearch, especially in view of the steady global rise inlife expectancy and its enormous medical and economicconsequences.

Acknowledgements

We are grateful to the British Heart Foundation and DunhillMedical Trust for the financial support of this research.

The authors of this manuscript have certified that theycomply with the Principles of Ethical Publishing in theInternational Journal of Cardiology [39].

References

[1] Lakatta EG. Age-associated cardiovascular changes in health: impacton cardiovascular disease in older persons. Heart Fail Rev2002;7:29–49.

[2] Dannenberg AL, Levy D, Garrison RJ. Impact of age on echocardio-graphic left ventricular mass in a healthy population (The FraminghamStudy). Am J Cardiol 1989;64:1066–8.


[3] Olivetti G, Giordano G, Corradi D, et al. Gender differences and ageing:effects on the human heart. J Am Coll Cardiol 1995;26:1068–79.

[4] Hees PS, Fleg JL, Lakatta LG, Shapiro EP. Left ventricularremodelling with age in normal men versus women: novel insightsusing three-dimensional magnetic resonance imageing. Am J Cardiol2002;90:1231–6.

[5] Lakatta EG. Cardiovascular regulatory mechanisms in advanced age.Physiol Rev 1993;73:413–53.

[6] Ogawa T, Spina J, Martin WH, et al. Effects of age, sex, andphysical training on cardiovascular responses to exercise. Circulation1992;86:494–503.

[7] Fleg JL, O'Connor C, Gerstenblith G, et al. Impact of age on thecardiovascular response to dynamic upright exercise in healthy menand women. J Appl Physiol 1995;78:890–900.

[8] Cooke GA, Marshall P, al-Timman JK, et al. Physiological cardiacreserve: development of a non-invasive method and first estimates inman. Heart 1998;79:289–94.

[9] Devereux RB, Alonso DR, Lutas EM. Echocardiographic assessmentof left ventricular hypertrophy: comparison to necropsy findings. Am JCardiol 1986;57:450–8.

[10] Chantler P, Clements RE, Sharp L, George KP, Tan LB, Goldspink DF.The influence of body size on measurements of overall cardiacfunction. Am J Physiol 2005;289(5):H2059–65.

[11] Meaney E, Alva F, Moguel R, Meaney A, Alva J, Webel R. Formulaand nomogram for the sphygmomanometric calculation of the meanarterial pressure. Heart 2000;84:64.

[12] Collier CD. Determination of mixed venous CO2 tensions by re-breathing. J Appl Physiol 1956;9:25–9.

[13] Defares JG. Determination of PvCO2 from the exponential CO2 riseduring rebreathing. J Appl Physiol 1958;13:159–64.

[14] Whitney RJ. The measurement of volume changes in human limbs.J Physiol 1953;121:1–27.

[15] American College of Sports Medicine. American College of SportsMedicine's guidelines for exercise testing and prescription. London:Lippincott Williams and Wilkins; 2000.

[16] Nicholls DP, Riley MS. Measuring cardiac power output — the acidtest. Eur Heart J 2001;22:1368–70.

[17] Tan LB. Cardiac pumping capability and prognosis in heart failure.Lancet 1986:1360–3 ii.

[18] Williams SG, Cooke GA, Wright DJ, et al. Peak cardiac power output:a powerful prognostic indicator in chronic heart failure. Eur Heart J2001;22:1496–503.

[19] Cotter G, Williams SG, Tan BL. The role of cardiac power in heartfailure. Curr Opin Cardiol 2003;18:215–22.

[20] Fincke R, Hochman JS, Lowe AM, et al. Cardiac power is the strongesthemodynamic correlate of mortality in cardiogenic shock: a reportfrom the SHOCK trial registry. J Am Coll Cardiol 2004;44(2):340–8.

[21] Mendoza DD, Cooper HA, Panza JA. Cardiac output predicts mortalityacross a broad spectrum of patients with acute cardiac disease. AmHeart J 2007;153:366–70.

[22] Patten RD, Pourati I, Aronovitz MJ, et al. 17 beta-estradiol reducescardiomyocyte apoptosis in vivo and in vitro via activation of phospho-inositide-3 kinase/Akt signaling. Circ Res 2004;95:692–9.

[23] Leri A, Kajstura J, Li B, et al. Cardiomyocyte ageing is gender-dependent: the local IGF-1-IGF-1R system.HeartDis 2000;2(2):108–15.

[24] Guerra S, Leri A, Wang X, et al. Myocyte death in the failing humanheart is gender dependent. Circ Res 1999;85:856–66.

[25] Mallat Z, Fornes P, Costagliola R, et al. Age and gender effectson cardiomyocyte apoptosis in the normal human heart. J Gerontol2001;56A:M719–23.

[26] Torella D, Ellison GM, Mendez-Ferrer S, Ibanez B, Nadal-Ginard B.Resident human cardiac stem cells: role in cardiac cellular homeostasisand potential for myocardial regeneration. Nature Clin PractCardiovasc Med 2006;3(1):S1–6.

[27] Najjar SS, Schulman SP, Gerstenblith G, et al. Age and gender affectventricular–vascular coupling during aerobic exercise. J AmColl Cardiol2004;44(3):611–7.

[28] Franklin S, Gustin W, Wong ND, et al. Hemodynamic patterns of age-related changes in blood pressure; The Framingham Heart Study.Circulation 1997;96:308–15.

[29] ChungAK, Das SR, LeonardD, et al.Women have higher left ventricularejection fractions than men independent of differences in left ventricularvolume: the Dallas Heart Study. Circulation 2006;113:1597–604.

[30] Cohn JN, Johnson GR, Shabetai R, et al. Ejection fraction, peak exerciseoxygen consumption, cardiothoracic ratio, ventricular arrhythmias, andplasma norepinephrine as determinants of prognosis in heart failure. TheV-HeFT VA Cooperative Studies Group. Circulation 1993;87:VI5–16.

[31] Dinenno FA, Jones PP, Seals DR, Tanaka H. Limb blood flow andvascular conductance are reduced with age in healthy humans: relationto elevations in sympathetic nerve activity and declines in oxygendemand. Circulation 1999;100:164–70.

[32] Proctor DN, Koch DW, Newcomer SC, et al, Leuenberger UA.Impaired leg vasodilation during dynamic exercise in healthy olderwomen. J Appl Physiol 2003;95:1963–70.

[33] Oparil S, Arthur C. Corcoran Memorial Lecture. Hormones andvasoprotection. Hypertension 1999;33:170–6.

[34] McGuire DK, Levine B, Williamson J, et al. A 30 year follow-up of theDallas bed rest and training study, I. Effects of age on thecardiovascular response to exercise. Circulation 2001;104:1350–7.

[35] Fleg JL, Morrell CH, Bos AG, et al. Accelerated longitudinal decline ofaerobic capacity in healthy older adults. Circulation 2005;112:674–82.

[36] Vanhees L, Defoor J, Schepers D, Brusselle S, Reybrouck T, Fagard R.Comparison of cardiac output measured by two automated methods ofCO2 rebreathing. Med Sci Sports Exerc 2000;32:1028–34.

[37] Jakovljevic DG, Nunan D, Donovan G, et al. Comparison of cardiacoutput determined by different rebreathing methods at rest and peakexercise. Eur J Appl Physiol 2008;102:593–9.

[38] Warburton DER, Haykowsky MJF, Quinney HA, Humen DP, Teo KK.Reliability and validity of measures of cardiac output during incrementalto maximal aerobic exercise. Sports Med 1999;27:23–41.

[39] Coats AJ. Ethical authorship and publishing. Int J Cardiol2009;131:149–50.


A study of presbycardia, with gender differences favoring ageing women

Documents