Chest_Jan05_Initial and final exercise heart rate transients_influence of gender, aerobic fitness, and clinical status.pdf

7/28/2019 Chest_Jan05_Initial and final exercise heart rate transients_influence of gender, aerobic fitness, and clinical status.

1/12

DOI 10.1378/chest.127.1.3182005;127;318-327Chest

and Claudio Gil S. ArajoDjalma Rabelo Ricardo, Marcos Bezerra de Almeida, Barry A. FranklinFitness, and Clinical Status

: Influence of Gender, Aerobic*TransientsInitial and Final Exercise Heart Rate

http://chestjournal.chestpubs.org/content/127/1/318.full.html

services can be found online on the World Wide Web at:The online version of this article, along with updated information and

ISSN:0012-3692)http://chestjournal.chestpubs.org/site/misc/reprints.xhtml(

written permission of the copyright holder.this article or PDF may be reproduced or distributed without the priorDundee Road, Northbrook, IL 60062. All rights reserved. No part ofCopyright2005by the American College of Chest Physicians, 3300Physicians. It has been published monthly since 1935.

is the official journal of the American College of ChestChest

2005 American College of Chest Physiciansby guest on November 1, 2011chestjournal.chestpubs.orgDownloaded from
http://chestjournal.chestpubs.org/content/127/1/318.full.htmlhttp://chestjournal.chestpubs.org/content/127/1/318.full.htmlhttp://chestjournal.chestpubs.org/content/127/1/318.full.htmlhttp://chestjournal.chestpubs.org/site/misc/reprints.xhtmlhttp://chestjournal.chestpubs.org/site/misc/reprints.xhtmlhttp://chestjournal.chestpubs.org/http://chestjournal.chestpubs.org/http://chestjournal.chestpubs.org/http://chestjournal.chestpubs.org/http://chestjournal.chestpubs.org/site/misc/reprints.xhtmlhttp://chestjournal.chestpubs.org/content/127/1/318.full.html


2/12

Initial and Final Exercise Heart RateTransients*

Influence of Gender, Aerobic Fitness, and ClinicalStatus

Djalma Rabelo Ricardo, PhD; Marcos Bezerra de Almeida, PhD;Barry A. Franklin, PhD; and Claudio Gil S. Araujo, MD, PhD

Study objectives: To compare the independent and additive data provided by initial and finalheart rate (HR) exercise transients, and to analyze both according to gender, aerobic fitness,clinical status, and medication usage.

Design: Retrospective study.Setting: Exercise medicine clinic.Patients: A total of 544 subjects (363 men) with a mean ( SD) age of 50 14 years (age range,10 to 91 years), including asymptomatic and coronary artery disease patients.Measurements and results: HR transients were obtained from the following two exerciseprotocols: 4-s exercise test (4sET) followed by a maximal cardiopulmonary cycling exercise test(CPET). The initial HR transient was represented by the cardiac vagal index (CVI), which wasobtained by the 4sET, and the final transient (ie, HR recovery [HRR]) was determined by thefollowing equation: CPET maximal HR the 1-min postexercise HR. Transients were modestlyrelated (r 0.22; p < 0.001) when adjusted for age, aerobic fitness, clinical status, and negativechronotropic action drug usage. The transients were unrelated to gender (vs CVI, p 0.10; vsHRR, p 0.15). Subjects with a measured maximum oxygen uptake (VO2max) exceeding 100% ofthe predicted maximal aerobic power showed higher CVIs than those in less aerobically fit

subjects (VO2max < 50% subgroup, p 0.009; VO2max < 75% subgroup, p 0.034). Bothtransient results differed for asymptomatic and cardiac subjects (CVI, 1.32 0.02 vs 1.42 0.02,respectively [p 0.001]; HRR, 33 1 beats/min (bpm) vs 37 1 bpm, respectively [p 0.009]).Conclusions: The initial and final HR transients were modestly related, suggesting a potentiallycomplementary clinical role for both measurements in the assessment of autonomic function inpatients with coronary artery disease. Although both HR transients tended to behave similarlyunder the influence of several variables, the initial HR transient, measured during 4sET, wasmore likely to discriminate distinct subgroups compared with the final HR transient.

(CHEST 2005; 127:318327)

Key words: autonomic nervous system; exercise; 4-s exercise test; heart rate recovery; heart rate transient; vagal tone

Abbreviations: ANOVA analysis of variance; bpm beats per minute; CI confidence interval; CPET cardio-

pulmonary exercise test; CVI

cardiac vagal index; 4sET

4-s exercise test; HR

heart rate; HRR

heart raterecovery; METmetabolic equivalent; NCAD negative chronotropic action drug; V o2 oxygen uptake;V o2maxmaximal oxygen uptake

The heart rate (HR) response to dynamic exercisefollows a well-defined pattern, which is primarily

modulated by the autonomic nervous system.13

During the first few seconds of exercise, there is arapid HR increase, known as the initial transient,that is exclusively mediated by vagal inhibition,regardless of exercise intensity.4 As the exercisecontinues, there is increasing sympathetic activity,

proportional to the intensity of the exercise, whichprogressively accelerates the HR. Immediately afterexercise, a final transient represented by a decreas-ing HR response is observed. This is a result of vagalreactivation and a reduction in the sympatheticstimulation, with the latter contributing more effec-tively to the slow or late deceleration phase ofpostexercise HR.4

exercise and the heart

318 Exercise and the Heart

http://chestjournal.chestpubs.org/http://chestjournal.chestpubs.org/http://chestjournal.chestpubs.org/http://chestjournal.chestpubs.org/


3/12

While the physiologic basis of HR transients inexercise has been studied for some decades throughselective pharmacologic blockade of both branchesof the autonomic nervous system, it is only in thelast years that these HR responses have been linkedto clinical outcomes. Numerous studies have nowshown that adults with autonomic dysfunction, man-ifested as reduced cardiac vagal tone,59 have higherall-cause and cardiovascular mortality rates.5,1012

This impaired vagal-dependent cardioprotection canbe assessed through varied patterns of HR behaviorat rest5,11 and during the initial exercise transients(rest to exercise)4 and the final exercise transients(exercise to rest).1214 Exercise-related HR modula-tion indexes have been widely used to analyze car-diac autonomic function, and to provide a powerfuland independent prognostic indicator of mortality inadults.15

Since the pioneering studies undertaken at theCleveland Clinic,12,16,17 and later independently con-firmed by other authors,13,14,18 it has been demon-strated that a slower HR final transient, as measuredduring the first 1 or 2 min of the postexercise period,both during maximal exercise12 and submaximalexercise,19 significantly increases the relative risk ofdeath. However, these studies were not designedto investigate the possible underlying physiologicmechanisms (eg, parasympathetic dysfunction). In-terestingly, while the initial fast transient (ie, the first4 s) is exclusively vagus dependent, this is probablynot the case with the final transient, especially if alonger period of 1 min, which proved its prognosticvalue in clinical studies, is considered. As previouslydiscussed, during the final HR transient, an impor-tant and simultaneous participation of both branchesof the autonomic nervous system is expected.

To assess the relative contribution of vagal activity,Araujo et al20 proposed and validated a 4-s exercisetest (4sET) protocol, in which predominantly vagalactivity at rest is suddenly withdrawn by fast un-loaded cycling exercise. Although Araujo et al21 haverecently demonstrated the reliability of 4sET, itsprognostic value for cardiovascular and all-causemortality remains unknown. Thus, our major aims inthis study were as follows: (1) to compare the

independent and additive data provided by initialand final HR exercise transients; and (2) to analyzeboth HR transients according to gender, aerobicfitness, clinical status, and medication usage.

Materials and Methods

Sample

We reviewed our laboratory data from nonathletes who werevoluntarily submitted to a detailed medical and functional eval-uation between 1995 and 2003. From a total of 2,198 assess-ments, we were able to select 544 subjects (363 men; mean[ SD] age, 50 14 years; age range, 10 to 91 years) who hadperformed a 4sET followed by a maximal cardiopulmonaryexercise test (CPET) utilizing a ramp protocol,22 both on a cycleergometer. A single physician (C.A.) evaluated and supervised allexercise tests. Before the procedures, all subjects read and signeda specific informed consent form approved by the institutionalresearch committee. Nearly all subjects (98%) were white andrepresented a higher socioeconomic class.

Clinical Data

A functional evaluation was preceded by clinical anamnesis andphysical examination, a 12-lead supine resting ECG, and spirom-etry. Relevant clinical features as well as the regular use ofnegative chronotropic action drugs (NCADs), such as amioda-rone, -blockers, and calcium-channel blockers, were carefullyrecorded.

Exercise Protocols

4sET: The 4sET was pharmacologically validated20 for theisolated assessment of the integrity of cardiac vagal tone

through the analysis of HR initial transient (ie, the rest-to-exercise transition), as reported by our group20,23,24 andothers.25 Briefly, the 4sET consists of unloaded pedaling asfast as possible on a cycle ergometer (EC-1600; Cat Eye;Tokyo, Japan) from the fourth to the eighth seconds of a 12-smaximal inspiratory apnea. The subject remains seated on thecycle ergometer, and, after the HR stabilizes, the verbalcommands of four evaluators guide the actions to be sequen-tially performed at 4-s intervals, as follows: (1) a fast maximalinspiration, primarily through the mouth; (2) as fast as possiblepedaling; (3) sudden cessation of pedaling; and (4) expira-tion.20,21,26 A continuous recording of a single ECG lead,usually CC5 or CM5, for 35 s at 25 mm/s, beginning 5 s beforethe command for maximal inspiration was obtained with two

different ECG systems (Cardiolife TEC 7100; Nihon-Kohden;Tokyo, Japan; or Elite Ergo PC 3.2.1.5; Micromed; Braslia,Brazil).

To quantify cardiac vagal tone, the following ECG R-Rintervals (identified and measured with 10-ms resolution) wererecorded: the longest R-R interval (ie, either the interval obtainedimmediately before the onset of exercise or the first one after theonset of exercise); and the shortest R-R interval during the 4-sexercise (generally, the last one). Two maneuvers were typicallycarried out, with a dimensionless index (the greatest ratiobetween the aforementioned intervals obtained during both4sET maneuvers) being chosen to represent the cardiac vagalindex (CVI).

Previous studies have shown that the CVI magnitude is notdependent on the presence or absence of resistance to pedal

*From the Physical Education Graduate Program (Prof. RabeloRicardo, Prof. Bezerra de Almeida, and Dr. Araujo), Gama FilhoUniversity, Rio de Janeiro, Brazil; and William Beaumont Hos-pital (Dr. Franklin), Royal Oak, MI.Manuscript received August 23, 2004; revision accepted August24, 2004.Reproduction of this article is prohibited without written permis-sion from the American College of Chest Physicians (e-mail:[email protected]).Correspondence to: Claudio Gil Soares de Araujo, MD, PhD,Clnica de Medicina do Exerccio, 22031 070 Rua SiqueiraCampos, 93/101 Rio de Janeiro, RJ, Brazil; e-mail: cgaraujo@

iis.com.br

www.chestjournal.org CHEST / 127 / 1 / JANUARY, 2005 319



4/12

movement,4 on active or passive execution,26,27 on whether theexercise is undertaken with the lower or upper limbs,28 or even

without a ergometer usage, in the orthostatic position.29 More-over, within-day and between-day reliability have been shown tobe very high.21

Maximal CPET: The CPETs were conducted using a cycleergometer with direct collection and analysis of expired gases(VO2000; MedGraphics; St. Paul, MN), immediately after the4sET, according to an individualized ramp protocol, to achieve a

duration of between 8 and 12 min.22,30

The subjects were verballyencouraged to exercise to volitional fatigue (ie, exhaustion),regardless of the maximal HR attained.

For the 4sET and the CPET, both feet remained fixed to thepedals so as to enhance motor performance and to improvemechanical efficiency. The ECG was continuously monitored viaa single lead (CC5 or CM5) from pretest rest to at least 5 minpostexercise. Immediately after the test, subjects were helped offthe cycle ergometer and placed on an adjacent stretcher in thesupine position, generally within 20 to 40 s.

Procedures for HR Analysis: HR transients were obtained fromthe following two exercise protocols: 4sET; and CPET. On the4sET, the initial transient was represented by the CVI, aspreviously described. On CPET, the final transient (ie, the HR

recovery [HRR]) was determined by the following equation:maximal measured HR the 1-min postexercise HR. Both

values were obtained from averaging seven R-R interval dura-tions at respective times. For comparative purposes, the pre-dicted maximal HR was calculated according to the followingequation: 210 0.65 age (in years).31

The relationship between both HR transients expressed as CVIand HRR was initially quantified and adjusted according to thefollowing variables: age; aerobic fitness; clinical status; and

NCAD use.The values related to the initial and final HR transients thenwere compared according to the following six features: (1)gender-specific absolute aerobic fitness ( 5 metabolic equiva-lents [METs]; 5 to 7.9 METs; 8 to 11 METs; and 11 METs)32;(2) predicted aerobic fitness (expressed as a percentage, and forage and gender)33; (3) HRR (cutoff point, 23 beats/min [bpm])34;(4) the presence or absence of NCAD use; (5) chronotropicincompetence, defined as the inability to reach 85% of thepredicted maximal HR (220 age in years) during maximalexercise testing35; and (6) clinical status, distinguishing betweenpatients with coronary artery disease and asymptomatic subjects.Accordingly, we analyzed HR transients from several differentperspectives.

Table 1Characteristics of the Study Sample (n 544) With Specific Reference to Gender, Demographics,Cardiopulmonary Variables, Clinical Status, and Medications Usage*

Variables All (n 544) Men (n 363) Women (n 181) p Value 95% CI

DemographicsAge, yr 50 1 50 1 50 1 1.00 2.202.20

(1091) (1191) (1082)Weight, kg 77 1 83 0.9 66 1 0.001 14.9819.02

(28162) (48184) (28128)Height, cm 169 0.4 174 0.4 160 0.5 0.001 13.0714.93

(137195) (148195) (137178)Body mass index 26.8 0.22 27.4 0.3 25.6 0.4 0.001 1.092.50

(15.157.6) (18.457.6) (15.148.6)HR

CVI 1.37 0.01 1.37 0.01 1.37 0.02 1.00 0.030.03(1.012.62) (1.042.62) (1.012.20)

Predicted HR, bpm 177 0.38 177 0.5 178 0.6 0.087 2.150.14(151203) (151203) (157204)

Maximum HR, bpm 160 1 160 1 159 2 0.479 1.773.77(82207) (82207) (96198)

HRR, bpm 34 1 34 1 35 1 0.371 3.201.20(276) (276) (875)

Aerobic fitnessEstimated V o2, mL/(kgmin) 26.1 0.36 27.6 0.4 23.1 0.6 0.001 3.605.50

(9.255.5) (10.655.5) (9.251.5)V o2max, mL/(kgmin) 27 0.45 29.9 0.5 22.4 0.7 0.001 6.308.70

(4.861) (9.561.1) (4.856.2)METs 7.8 0.1 8.1 0.2 6.4 0.2 0.001 1.262.14

(1.417.5) (2.717.5) (1.416.1)V o2, % predicted 87.3 1.22 93.4 1.2 75.2 1.9 0.001 15.221.2

(20.0211.0) (30.6211) (20185.8)Clinical status

Asymptomatic 132 (24) 81 (23) 51 (28)Coronary artery disease 115 (21) 103 (28) 12 (7)Other diseases and mixed cases 297 (55) 179 (49) 118 (65)

NCADsNo 396 (73) 252 (69) 144 (80)Yes 148 (27) 111 (31) 37 (20)

*Values given as mean SEM (minimummaximum) or No. (%), unless otherwise indicated.Related to difference of means.




5/12

Statistical Analysis

To investigate between-group differences, t tests and one-wayand two-way analysis of variance (ANOVA) were used, followedby Bonferroni post hoc comparisons where appropriate. Bothlinear and stepwise regressions, and partial and Pearson product-moment correlations were used to assess the relationship be-tween variables. Partial correlations were adjusted according tothe aforementioned variables. The significance level was definedat p 0.05, and the 95% confidence intervals (CIs) were calcu-lated for demographic and cardiorespiratory variables. A statisti-cal software package (SPSS, version 11; SPSS; Chicago, IL) wasused for the statistical analyses.

Results

Descriptive statistics of the major variables arepresented in Table 1. CVI and HRR results wereonly modestly related, as shown in Figure 1. Thisassociation was even smaller after adjustments forage, aerobic capacity, clinical status, and NCAD use

(r

0.22; p

0.001), with age being the most influ-ential variable.Men and women with a lower measured maximal

aerobic power, expressed as METs, generally dem-onstrated slower initial and final HR transients asshown by CVI and HRR data (Fig 2). One-wayANOVA revealed differences (p 0.05) amongsome of the subsets that had been stratified accord-ing to aerobic fitness, more specifically at the ends ofthe continuum (ie, 5 vs 8 to 11 METs, 5 vs 11METs, and 5 to 7.9 vs 11 METs) for the femaleHR initial transient. Although statistical analysis

showed differences (p

0.036) for the final HR

transient, post hoc analysis failed to identity them.For male subjects, the transients differed in thesubsets for initial and final HR transients, except for 5 vs 5 to 7.9 METs, and for 8 to 11 vs 11 METs,respectively.

Age-adjusted two-way ANOVA showed that thetransients were unrelated to gender (CVI, p 0.10;HRR, p 0.15). When within-group analyses per-

formed according to aerobic fitness were carried out,differences between men and women in the cohortwith 5 METs for CVI (1.20 0.02 vs 1.30 0.02,respectively; p 0.01) and in the groups with 8METs for HRR (ie, 5 METs, p 0.001 [21 vs 32bpm, respectively]; and 5 to 7.9 METs, p 0.025[32 vs 36 bpm, respectively]) were observed. How-ever, there were nearly twice as many more womenthan men (67 vs 35, respectively) in the 5 METsgroup. Additionally, the associations adjusted for age,clinical status, and NCAD use between both tran-sients and directed measured maximal aerobic power

expressed as METs were r 0.19 (p 0.001) forthe initial transient and r 0.08 (p 0.056) for thefinal transient, when collective data from both sexeswere considered.

Subjects with 50% of age-predicted and gender-predicted maximal aerobic power tended to showslower initial and final HR transients (Fig 3). Al-though no statistically significant differences(p 0.052) for HRR were detected among thesubsets of subjects, such differences were noted forCVI in the groups with 50% of aerobic power vsthose with 100% of aerobic power (p 0.009),

and in the groups with 75% of aerobic power vsthose with 100% of aerobic power (p 0.034).The correlation coefficients between the HR tran-sients and the percentages of predicted maximalaerobic power, adjusted for age, clinical status, andmedication usage, were r 0.17 (p 0.001) forCVI, and r 0.09 (p 0.031) for HRR, which aresimilar to the values obtained for absolute values ofmeasured maximal oxygen uptake (V o2max), ex-pressed as METs. However, only 7% of the studysample (28 women and 8 men) had aerobic capacityvalues that were 50% of the age-predicted and

gender-predicted values.By using the percentage of predicted HR achieved

with CPET (reference value, 85%) to divide thesample according to the presence or absence ofchronotropic incompetence in NCAD nonusers,those with chronotropic incompetence demon-strated slower HR transients, which is especiallyevident in the initial transient (Table 2). Approxi-mately 17% of the subjects demonstrated an abnor-mal HR descent at 1 min after exercise (ie, 23bpm). The analysis of the initial transient of theseindividuals revealed significantly lower CVI valuesFigure 1. Nonadjusted relationship between CVI and HRR.




6/12

(mean CVI, 1.23 0.02 vs 1.40 0.01, respectively;p 0.001; 95% CI, 0.27 to 0.66).

Interestingly, in the 27% of the subjects who wereregularly receiving NCADs, their use did not sub-stantially influence the HR transients, despite thepresence of marginal statistically significant differ-ences (Table 3). When the sample was dividedaccording to clinical status, mean differences forboth transients were found between the coronaryartery disease group and the asymptomatic group(CVI, 1.32 0.02 vs 1.42 0.02, respectively[p 0.001; 95% CI, 0.16 to 0.04]; HRR, 33 1vs 37 1 bpm, respectively [p 0.009; 95% CI, 7to1]) [Fig 4]); however, those with coronary arterydisease were older by comparison (mean age, 58 1vs 44 1 years, respectively; p 0.001; 95% CI,7 to 11).

Discussion

Relationship Between Initial and Final Transients

Our results demonstrated a small, albeit signifi-cant, association among age-adjusted, medication-

adjusted, aerobic fitness-adjusted, and clinical status-adjusted initial and final HR transients, as shown bylinear and partial regression coefficients. These re-sults underlie the assumption that, although bothcoefficients are dependent, at least in part, on theautonomic nervous system,4 different physiologicmechanisms are responsible for the initial and finalHR transients. Thus, it seems that the informationobtained from both transients is complementary, asonly 4% of variation of one of them may account forthe variation in the other.

It has been pharmacologically demonstrated that

Figure 2. Measured aerobic fitness (V o2max) and initial HR transients (top) and final HR transients(bottom) for women and men.




7/12

the rapid initial HR transient is exclusively mediatedby vagal inhibition.20 On the other hand, the physi-ologic mechanisms involved in the final or immedi-ate postexercise transient are unclear. Imai et al36

demonstrated that the first 30 s of recovery areprimarily mediated by vagal reactivation, regardlessof age and exercise intensity in healthy adults, ath-letes, and individuals with heart failure. Neverthe-less, it seems that the adrenergic component reduc-tion plays a greater role in the subsequent HRdecrement.

Although the prognostic clinical usefulness of

HRR presents strong epidemiologic evidence underdifferent situations and methods, it would seem to bean oversimplification to assume that the HR descentduring the postexercise period, mainly when ob-served for 30 s, was primarily due to a parasym-pathetic activity recovery. The great diversity ofmethods used to investigate HRR after exercisemade the comparison of the various studies tenuousat best. Selected methodological differences were asfollows: (1) test characteristics (maximal12 vs sub-maximal19); (2) type of ergometer (cycle37 vs tread-mill14,38,39); (3) different recovery times (from 30 s36

to 5 min39); (4) recovery format (active12 vs pas-sive34); and (5) recovery position (standing,13 sit-ting,18 supine,38 or lateral decubitus,34), as well asother variations in protocol such as the time spent toadopt the resting position, respiratory rate, and tidalvolume,40 and whether the subject was allowed totalk or move during this period.

In order to achieve a generalized use in clinicalpractice, a given clinical or physiologic measurementmust be repeatable in the same individual under thesame controlled conditions. Recently, an analysis ofpreviously published studies41 on the methods ofquantitating the initial and final HR transients sug-gested that HRR measurements may be unreliable inmen and women (90 patients; age range, 30 to 80years) with or without symptoms. Conversely, theintraday and interday CVI calculations were highlyreliable in a large and heterogeneous sample (1,699patients; age range, 8 to 85 years),21 under a varietyof clinical statuses. Therefore, in contrast to CVI, thewidespread use of HRR measurements may presentan additional challenge (ie, low reliability), poten-tially compromising the clinical analysis and implica-tions.

Aerobic Fitness

The methodological strengths of the present studyinclude discrimination between genders and a directmeasurement of expired gases during maximal exer-cise, which was supervised and analyzed by the sameexperienced physician. Individuals with a lower

aerobic fitness, expressed as a measured value or asa percentage of the age-predicted and gender-predicted V o2max, also had slower initial and finalHR transients. Similar results had been reportedpreviously in epidemiologic studies that comparedaerobic fitness,32 HR transients,12,13 and cardiovas-cular mortality. Theoretically, because the slowerHR initial transient is due to vagal dysfunction, theseindividuals would be expected to demonstrate ahigher mortality rate, due to loss of vagal-dependentcardioprotection and increased vulnerability to ma-lignant ventricular arrhythmias.15

Figure 3. Aerobic fitness according to age-predicted and gen-der-predicted V o2max and initial HR transients (top) and finalHR transients (bottom).




8/12

Our data suggest that the initial HR transienttends to stratify the absolute or relative aerobicfitness better than HRR, regardless of gender. Incontrast to CVI, which showed a clear differentia-tion, the final HR transient basically divided the fourgroups in two (ie, 5 and 5 METs, or 50% and 50% of predicted oxygen uptake (V o2). This be-comes even clearer when aerobic fitness is analyzed

according to the percentage of the predictedV o2max, in which ANOVA failed to identify anydifference among the groups in the final HR tran-sient. Thus, the use of HRR as a marker of aerobicfitness seems to lack precision,42,43 since differentphysiologic mechanisms are involved. However, ourstudy sample included a relatively low prevalence ofsubjects with aerobic fitness of 50% of predicted.

Our data showed weak associations between aer-obic fitness, whether expressed in METs or inpercentages of predicted V o2max, and the initial andfinal HR transients. Although statistically significant,

the correlations were relatively low, which is inagreement with the findings of other studies.4 Incontrast, CVI performed better when compared withHRR, suggesting an inability to predict HR transientresponses from aerobic fitness and vice versa. This isrelevant because it is well-documented that aerobicfitness alone, whether measured directly or esti-mated, is an excellent predictor of cardiovascular andall-cause mortality.32,44 Thus, as has been previouslydemonstrated by Cole et al,12 the HRR (ie, the finaltransient) provides independent clinical prognosticinformation. It is possible that the initial HR tran-

sient, which was widely discussed in this study, alsomay provide a relevant, complementary prognosticindicator, based on its more objective dependence

on cardiac vagal integrity, high reliability, and mod-est association with the final transient.

Both HR transients seemed to be gender indepen-dent, which is in agreement with what we havepreviously described for CVI,45 at least in adults. Onthe other hand, there were differences in HRRamong boys and girls undergoing submaximal exer-cise protocols.46 Considering the present data for

categories of low aerobic fitness (ie,

8 METs and 75% predicted V o2), women showed higher andmore normal values for HR transients than did men.These provocative data thus suggest that womenwith low aerobic fitness tend to better preserve theirautonomic function and, accordingly, maintain ahigher degree of cardioprotection.47

Chronotropic Incompetence

A lower maximal HR during exercise testing isassociated with a greater risk of cardiovascular eventsand all-cause mortality.35 Chronotropic incompe-

tence may be a marker of autonomic modulation,reflecting the underlying cardiovascular disease. Ourresults suggest that individuals with blunted valuesfor initial and final HR transients often also hadconcomitant evidence of chronotropic incompe-tence, with an even higher discriminatory powerderived from the former.

HRR

Although most study subjects had a chronic clini-cal status that may have influenced autonomic activ-

ity, only 17% showed abnormal HRR responsesaccording to the cutoff criteria used (23 bpm). Thus,the use of absolute cutoffs, as originally proposed,

Table 3HR Transients According to NCAD Usage*

Variable Without NCAD (n 396) With NCAD (n 148) p Value 95% CI

Initial transient (CVI) 1.39 0.01 1.30 0.02 0.001 0.040.13(1.012.34) (1.042.62)

Final transient (HRR), bpm 35 1 32 1 0.008 0.825.57(276) (676)

*Values given as mean SEM (minimummaximum), unless otherwise indicated.

Table 2HR Transients According to Chronotropic Incompetence Without NCAD Utilization*

Variable 85% (n 54) 85% (n 342) p Value 95% CI

HR, % predicted 77 0.8 96 0.3 0.001 23.0514.95(5984) (85112)

Initial transient (CVI) 1.22 0.02 1.42 0.01 0.003 0.230.09(1.012.62) (1.042.34)

Final transient (HRR), bpm 25 1 37 1 0.062 24.620.62(253) (376)

*Values given as mean SEM (minimummaximum), unless otherwise indicated.




9/12

may not be an adequate discriminator.37 Alterna-tively, one might explore the use of a percentage ofmaximal exercise HR obtained at the first minute ofrecovery as an index of subsequent mortality. An-other clinically relevant analysis might be the quan-tification of HRR in relation to HR reserve (ie,maximum HR resting HR), that is, the percentageof the subjects HRR to which the HR is reduced at1 min after maximal exercise testing.

More recently, Raymond48 challenged the mech-anisms responsible for postexercise HRR, even con-

testing its prognostic value. Currently, it seems thatthe HRR requires additional study before definitivecriteria for its clinical use and interpretation can beadopted.

NCADs

Individuals receiving NCADs were more likely toexhibit slower HR transients. Because these drugsact primarily on the sympathetic nervous system,which is largely unrelated to the initial transient asstudied on 4sET,20 the differences found may reflectdisease-related autonomic impairment rather than adrug effect per se.

Clinical Status

As expected, individuals with known coronaryartery disease, compared with their healthy counter-parts, demonstrated a more blunted HR response forboth the initial and final HR transients. Once again,however, the initial transient, as assessed by 4sET,showed greater discriminatory power compared withHRR. The difference between the groups is proba-bly due to the fact that, in addition to being com-posed of older individuals,4951 the group of patientswith coronary artery disease had greater autonomicdysfunction, a feature that is inherent in their clinicalstatus.5,52,53

Limitations, Clinical Applications, and Future

ResearchThe present study has potentially confounding

variables, such as the severity and duration of thecardiovascular disorders presented, and the magni-tude and location of any residual ischemic myocar-dium or fibrotic areas. The results suggest that thephysiologic mechanisms underlying the initial andfinal HR transients are partially distinct, as measuredby 4sET and CPET. Moreover, the clinical prognos-tic value of cardiac vagal activity, as assessed by the4sET and how it compares with HRR, remainsunresolved.

Considering the limitations of the standardizationof postexercise HRR measurement, and the simplic-ity, potentially greater safety, and time required forthe performance and analysis of 4sET, it seems thatthe latter procedure should be additionally investi-gated and quantified. The simultaneous analysis ofboth HR transients may be complementary, aug-menting the clinically relevant and prognostic valueof the chronotropic response to exercise. Finally,methodologic improvements may be needed forfuture studies. These may include standardizing thetime that the subject takes to assume a supine resting

Figure 4. Final and initial HR transients according to clinicalstatus.




10/12

position after exercise and incorporating the conceptof HR reserve into the interpretation of the final HRtransient.

Conclusion

We found the following: (1) that the initial and

final HR transients are significantly, albeit modestly,related, suggesting a potentially complementary clin-ical role for both measurements in the assessment ofautonomic function and coronary artery disease; and(2) that both HR transients tended to behave simi-larly under the influence of several variables, al-though the initial HR transient, measured during4sET, was more likely to discriminate distinct sub-groups compared with the final HR transient.

References

1 Ekblom B, Astrand PO, Saltin B, et al. Effect of training oncirculatory response to exercise. J Appl Physiol 1968; 24:518528

2 Jose AD. Effect of combined sympathetic and parasympa-thetic blockade on heart rate and cardiac function in man.Am J Cardiol 1966; 18:476478

3 Maciel BC, Gallo L Jr, Marin Neto JA, et al. Autonomicnervous control of the heart rate during dynamic exercise innormal man. Clin Sci (Lond) 1986; 71:457460

4 Araujo CGS. Fast on and off heart rate transients atdifferent bicycle exercise levels. Int J Sports Med 1985;6:6873

5 Kleiger RE, Miller JP, Bigger JTJ, et al. Decreased heart ratevariability and its association with increased mortality after

acute myocardial infarction. Am J Cardiol 1987; 89:2562626 La Rovere MT, Bersano C, Gnemmi M, et al. Exercise-

induced increase in baroreflex sensitivity predicts improvedprognosis after myocardial infarction. Circulation 2002; 106:945949

7 Braith RW, Edwards DG. Neurohormonal abnormalities inheart failure: impact of exercise training. Congest Heart Fail2003; 9:7076

8 Robinson TG, Dawson SL, Eames PJ, et al. Cardiac barore-ceptor sensitivity predicts long-term outcome after acuteischemic stroke. Stroke 2003; 34:705712

9 Sevre K, Lefrandt JD, Nordby G, et al. Autonomic function inhypertensive and normotensive subjects: the importance ofgender. Hypertension 2001; 37:13511356

10 Huikuri HV, Makikallio TH, Airaksinen KEJ, et al. Measure-ment of heart rate variability: clinical tool or a research toy?J Am Coll Cardiol 1999; 34:18781883

11 La Rovere MT, Bigger JTJ, Marcus FI, et al. Baroreflexsensitivity and heart rate variability in prediction of totalcardiac mortality after myocardial infarction: ATRAMI (Au-tonomic Tone and Reflexes After Myocardial Infarction)Group. Lancet 1998; 351:487494

12 Cole CR, Blackstone EH, Pashkow FJ, et al. Heart raterecovery immediately after exercise as a predictor of mortal-ity. N Engl J Med 1999; 341:13511357

13 Desai M, De la Pena-Almaguer E, Mannting F. Abnormalheart rate recovery after exercise as a reflection of anabnormal chronotropic response. Am J Cardiol 2001; 87:11641169

14 Morshedi-Meibodi A, Larson MG, Levy D, et al. Heart raterecovery after treadmill exercise testing and risk of cardiovas-cular disease events (The Framingham Heart Study). Am JCardiol 2002; 90:848 852

15 Buch AN, Coote JH, Townend JN. Mortality, cardiac vagalcontrol and physical training: whats the link? Exp Physiol2002; 87:423435

16 Lauer MS, Okin PM, Larson MG, et al. Impaired heart rateresponse to graded exercise: prognostic implications of chro-

notropic incompetence in the Framingham Heart Study.Circulation 1996; 93:1520152617 Seshadri N, Gildea TR, McCarthy K, et al. Association of an

abnormal exercise heart rate recovery with pulmonary func-tion abnormalities. Chest 2004; 125:12861291

18 Nissinen SI, Makikallio TH, Seppanen T, et al. Heart raterecovery after exercise as a predictor of mortality amongsurvivors of acute myocardial infarction. Am J Cardiol 2003;91:711714

19 Cole CR, Foody JM, Blackstone EH, et al. Heart raterecovery after submaximal exercise testing as a predictor ofmortality in a cardiovascularly healthy cohort. Ann InternMed 2000; 132:552555

20 Araujo CGS, Nobrega ACL, Castro CLB. Heart rate re-sponses to deep breathing and 4-seconds of exercise beforeand after pharmacological blockade with atropine and pro-pranolol. Clin Auton Res 1992; 2:3540

21 Araujo CGS, Ricardo DR, Almeida MB. Intra and interdaysreliability of the 4-second exercise test. Rev Bras MedEsporte 2003; 9:299303

22 Buchfurher M, Hansen J, Robinson T, et al. Optimizing theexercise protocol for cardiopulmonary assessment. J ApplPhysiol 1983; 55:15581564

23 Nobrega ACL, Castro CLB, Araujo CGS. Relative roles of thesympathetic and parasympathetic systems in the 4-s exercisetest. Braz J Med Biol Res 1990; 23:12591262

24 Lazzoli JK, Soares PPS, Nobrega ACL, et al. Electrocardio-graphic criteria for vagotonia: validation with pharmacologicalparasympathetic blockade in healthy subjects. Int J Cardiol

2003; 87:23123625 Knopfli BH, Bar-Or O. Vagal activity and airway response to

ipratropium bromide before and after exercise in ambient andcold conditions in healthy cross-country runners. Clin J SportMed 1999; 9:170176

26 Nobrega ACL, Williamson JW, Araujo CGS, et al. Heart rateand blood pressure responses at the onset of dynamic exer-cise: effect of Valsalva manoeuvre. Eur J Appl Physiol 1994;68:336340

27 Nobrega ACL, Araujo CGS. Heart rate transient at the onsetof active and passive dynamic exercise. Med Sci Sports Exerc1993; 25:3741

28 Araujo CGS, Nobrega ACL, Castro CLB. Similarities be-tween fast initial heart rate response to arm and leg cycling

exercise [abstract]. J Cardiopulm Rehabil 1993; 13:34829 Almeida MB, Ricardo DR, Araujo CGS. Validation of the4-second exercise test in the orthostatic position. Arq BrasCardiol 2004; 83:160164

30 Myers J, Buchanan N, Smith D, et al. Individualized ramptreadmill: observations on a new protocol. Chest 1992; 101:236S241S

31 Lange-Andersen K, Shepard R, Denolin H, et al. Fundamen-tals of exercise testing. Geneva, Switzerland: World HealthOrganization, 1971

32 Myers J, Prakash M, Froelicher VF, et al. Exercise capacityand mortality among men referred for exercise testing.N Engl J Med 2002; 346:793801

33 Jones NL. Clinical exercise testing. 4th ed. Philadelphia, PA:WB Saunders Company, 1997




11/12

34 Watanabe J, Thamilarasan M, Blackstone EH, et al. Heartrate recovery immediately after treadmill exercise and left

ventricular systolic dysfunction as predictors of mortality: thecase for stress echocardiography. Circulation 2001; 104:19111916

35 Lauer MS, Francis GS, Okin PM, et al. Impaired chrono-tropic response to exercise stress testing as a predictor ofmortality. JAMA 1999; 281:524529

36 Imai K, Sato H, Hori M, et al. Vagally mediated heart rate

recovery after exercise is accelerated in athletes but bluntedin patients with chronic heart failure. J Am Coll Cardiol 1994;24:15291535

37 Gaibazzi N, Petrucci N, Ziacchi V. One-minute heart raterecovery after cycloergometer exercise testing as a predictorof mortality in a large cohort of exercise test candidates:substantial differences with the treadmill-derived parameter.Ital Heart J 2004; 5:183188

38 Shetler K, Marcus R, Froelicher VF, et al. Heart raterecovery: validation and methodological issues. J Am CollCardiol 2001; 38:19801987

39 Cheng YJ, Lauer MS, Earnest CP, et al. Heart rate recoveryfollowing maximal exercise testing as a predictor of cardio-

vascular disease and all-cause mortality in men with diabetes.Diabetes Care 2003; 26:20522057

40 Soares PPS, Nobrega ACL, Araujo CGS. Initial heart ratetransient to dynamic exercise performed at apnea: influenceof the preceding inspiratory flow. Arq Bras Cardiol 1994;63:287292

41 Yawn BP, Ammar KA, Thomas R, et al. Test-retest reproduc-ibility of heart rate recovery after treadmill exercise. Ann FamMed 2003; 1:236241

42 Bell GJ, Snydmiller GD, Davies DS, et al. Relationshipbetween aerobic fitness and metabolic recovery from inter-mittent exercise in endurance athletes. Can J Appl Physiol1997; 22:7885

43 McArdle WD, Pechar GS, Katch FI, et al. Percentile norms

for a valid step test in college women. Res Q 1973; 44:498500

44 Kavanagh T, Mertens DJ, Hamm LF, et al. Prediction oflong-tern prognosis in 12169 men referred for cardiac reha-bilitation. Circulation 2002; 106:666671

45 Araujo CGS, Nobrega ACL, Castro CLB. Vagal activity:effect of age, sex and physical pattern. Braz J Med Biol Res1989; 22:909911

46 Mahon AD, Anderson CS, Hipp MJ, et al. Heart rate

recovery from submaximal exercise in boys and girls. Med SciSports Exerc 2003; 35:20932097

47 Airaksinen KE, Ikaheimo MJ, Linnaluoto M, et al. Genderdifference in autonomic and hemodynamic reactions toabrupt coronary occlusion. J Am Coll Cardiol 1998; 31:301306

48 Raymond LW. Vagal pas de deux: heart-lung interplay inpostexercise heart rate recovery. Chest 2004; 125:11861191

49 Srinivasan K, Sucharita S, Vaz M. Effect of standing on shortterm heart rate variability across age. Clin Physiol FunctImaging 2002; 22:404 408

50 Wood RH, Hondzinski JM, Lee CM. Evidence of an associ-ation among age-related changes in physical, psychomotorand autonomic function. Age Ageing 2003; 32:415421

51 Bonnemeier H, Richardt G, Potratz J, et al. Circadian profileof cardiac autonomic nervous modulation in healthy subjects:differing effects of aging and gender on heart rate variability.J Cardiovasc Electrophysiol 2003; 14:791799

52 Huang P, Leu H, Chen J, et al. Usefulness of attenuated heartrate recovery immediately after exercise to predict endothe-lial dysfunction in patients with suspected coronary arterydisease. Am J Cardiol 2004; 93:1013

53 Liao D, Carnethon M, Evans GW, et al. Lower heart ratevariability is associated with the development of coronaryheart disease in individuals with diabetes. Diabetes 2002;51:35243531




12/12

DOI 10.1378/chest.127.1.318

2005;127; 318-327ChestClaudio Gil S. Arajo

Djalma Rabelo Ricardo, Marcos Bezerra de Almeida, Barry A. Franklin andAerobic Fitness, and Clinical Status

: Influence of Gender,*Initial and Final Exercise Heart Rate Transients

November 1, 2011This information is current as of

http://chestjournal.chestpubs.org/content/127/1/318.full.htmlUpdated Information and services can be found at:

Updated Information & Services

http://chestjournal.chestpubs.org/content/127/1/318.full.html#ref-list-1This article cites 51 articles, 22 of which can be accessed free at:

References

http://chestjournal.chestpubs.org/content/127/1/318.full.html#related-urls

This article has been cited by 1 HighWire-hosted articles:Cited Bys

http://www.chestpubs.org/site/misc/reprints.xhtmlfound online at:Information about reproducing this article in parts (figures, tables) or in its entirety can bePermissions & Licensing

http://www.chestpubs.org/site/misc/reprints.xhtmlInformation about ordering reprints can be found online:

Reprints

"Services" link to the right of the online article.Receive free e-mail alerts when new articles cite this article. To sign up, select the

Citation Alerts

PowerPoint slide format. See any online figure for directions.articles can be downloaded for teaching purposes inCHESTFigures that appear in

Images in PowerPoint format

b t N b 1 2011h tj l h t bD l d d f
http://chestjournal.chestpubs.org/content/127/1/318.full.htmlhttp://chestjournal.chestpubs.org/content/127/1/318.full.htmlhttp://chestjournal.chestpubs.org/content/127/1/318.full.htmlhttp://chestjournal.chestpubs.org/content/127/1/318.full.html#ref-list-1http://chestjournal.chestpubs.org/content/127/1/318.full.html#ref-list-1http://chestjournal.chestpubs.org/content/127/1/318.full.html#ref-list-1http://chestjournal.chestpubs.org/content/127/1/318.full.html#related-urlshttp://chestjournal.chestpubs.org/content/127/1/318.full.html#related-urlshttp://chestjournal.chestpubs.org/content/127/1/318.full.html#related-urlshttp://www.chestpubs.org/site/misc/reprints.xhtmlhttp://www.chestpubs.org/site/misc/reprints.xhtmlhttp://www.chestpubs.org/site/misc/reprints.xhtmlhttp://www.chestpubs.org/site/misc/reprints.xhtmlhttp://chestjournal.chestpubs.org/http://chestjournal.chestpubs.org/http://chestjournal.chestpubs.org/http://chestjournal.chestpubs.org/http://www.chestpubs.org/site/misc/reprints.xhtmlhttp://www.chestpubs.org/site/misc/reprints.xhtmlhttp://chestjournal.chestpubs.org/content/127/1/318.full.html#related-urlshttp://chestjournal.chestpubs.org/content/127/1/318.full.html#ref-list-1http://chestjournal.chestpubs.org/content/127/1/318.full.html

Documents

Chest_Jan05_Initial and final exercise heart rate transients_influence of gender, aerobic fitness, and clinical status.pdf