Comparison of the Cardiopulmonary Responseto Exercise in Recipients of Dual Sensor DDDRPacemakers Versus a Healthy Control GroupERIC PAGE,* PASCAL DEFAYE,† JEAN-LUC BONNET,‡ CATHERINE DURAND,*and AMEL AMBLARD‡From *UCP.X and the †Department of Cardiology, University Hospital Grenoble, and the ‡ELA Medical,Clinical Research Department, Le Plessis-Robinson, France

PAGE, E., ET AL.: Comparison of the Cardiopulmonary Response to Exercise in Recipients of Dual SensorDDDR Pacemakers Versus a Healthy Control Group. The authors previously have shown in healthy sub-jects that age related loss of muscular strength did not alter the chronotropic response during treadmillexercise, whether with sudden onset of effort, as in the chronotropic assessment exercise protocol (CAEP)or more gradual effort as in the Harbor exercise protocol. This study was performed to verify that in patientssuffering from chronotropic insufficiency, and in absence of other effort-limiting disorders, “physiologic”pacing enables a cardiorespiratory response comparable to that of age-matched healthy subjects. Further-more, the aim of the study was to confirm that the response of a new dual sensor-based pacing system wasproperly adapted to the metabolic demand, whether during CAEP or during Harbor test, by subjecting pa-tients to both protocols. All study participants were able to undergo treadmill exercise testing, had normalcardiopulmonary function tests at rest, and no cardiac, muscular, or pulmonary disease. A healthy group(control) included 16 subjects (mean age 70.4 ± 3.9 years), and the test group (pacemaker [PM]) included9 subjects (mean age 67.1 ± 10.8 years) permanently paced for isolated chronotropic insufficiency witha dual sensor pacing system. All subjects underwent CAEP and Harbor tests with measurements of gasexchange, 24 hours apart, in randomized order. All subjects reached an appropriate level of exercise, asexpressed by mean lactate plasma concentrations, which were slightly higher in the control than the PMgroup during CAEP (4.9 ± 1.9 vs 3.7 ± 1.9 mmol/L, NS) and Harbor (5.3 ± 1.9 vs 3.8 ± 1.8 mmol/L, P <

0.05) tests. No statistical difference was observed in VO2 and VE at peak exercise between the two groupsduring either test. In the PM group, heart rate at peak exercise and metabolic reserve slope were higherduring the CAEP than the Harbor protocol. These two measurements were significantly lower than in thecontrol group. The PM group also had lower plasma lactate concentrations and dyspnea/fatigue scores.The Harbor test seems less suitable than the CAEP test to study the chronotropic response of pacemakerswith dual sensors during exercise. A high performance of the new dual sensor-based pulse generator wasconfirmed in this physically fit patient population, whose peak heart rate was considerably higher thanin other similar studies. (PACE 2003; 26[Pt. II]:239–243)

chronotropic insufficiency, dual sensor pacing, heart rate adaptation, Harbor test,chronotropic assessment exercise protocol

IntroductionAging is associated with an inescapable loss

of muscular strength in healthy, sedentary indi-viduals. This process becomes conspicuous pastage 70, as a result of muscular fibrosis, which canonly be mitigated by nearly daily physical exer-cise. When evaluated in the laboratory, this lossof muscular strength becomes particularly notice-able when exercise is performed on a bicycle. Ag-ing, sedentary subjects are unable to pedal at aprescribed rate beyond a certain workload before

Address for reprints: Eric Page, M.D., UCP.X, 45 MarieReynoard Ave., 38100 Grenoble, France. Fax: (33)0476338469;e-mail: ericpagecardio@wanadoo.fr

they reach their maximal heart rate (HR). From thatstandpoint, treadmill testing is better adapted toan elderly population.1,2 In a previous study theauthors have shown that in a group of healthysubjects whose mean age was 70 years, loss ofmuscular strength did not alter the chronotropicresponse during treadmill exercise, whether withsudden onset of effort, as in the chronotropic as-sessment exercise protocol (CAEP) or more grad-ual effort (Harbor protocol).3 This confirmed thereliable adaptation to a mixed 70-year-old popu-lation, representing to the majority of paced pa-tients, of the mathematical model of Wilkoff, whodeveloped the CAEP protocol in a male population<50 years old.4

In 1998, an experimental study of theTalent DR pacemaker (PM) (Ela Medical,

PACE, Vol. 26 January 2003, Part II 239

PAGE, ET AL.

Montrouge, France) showed that an antero-posterior accelerometer-based activity sensorcoupled with a physiological minute ventilation(VE) sensor yielded an excellent correlationbetween simulated and sinus rate response curvesin patients with preserved exercise inducedchronotropic response.5 The clinical validity ofthese results was confirmed by comparing the re-sults obtained in the healthy control subjects of thepresent study, with those observed in age-matchedpatients, studied under identical methodologicalconditions, who were permanently paced forchronotropic insufficiency.6

This study was performed to verify that, inpatients suffering from chronotropic insufficiencyduring exercise, and in absence of other effort-limiting disorders and left ventricular dysfunc-tion, “physiologic” pacing enables a cardiorespi-ratory response comparable to that of age-matchedhealthy subjects. Furthermore, the aim of thisstudy was to confirm that the pacing system re-sponse was properly adapted to the metabolic de-mand, whether during CAEP or during Harbortest, by subjecting the patients to both exerciseprotocols.

Patients and MethodsThe protocol of this single center study was

reviewed and approved by the National EthicalReview Committee, and all patients granted theirwritten, informed consent to participate in thestudy.

Patient Population

Inclusion criteria was (1) age ≥ 60 and<80 years; (2) ability to undergo treadmill exercisetesting; (3) normal cardiopulmonary function testsat rest, including echocardiogram; (4) successfulperformance of a bicycle exercise test confirmingthe absence of cardiac (coronary disease, arrhyth-mias, congestive failure), muscular, or pulmonarydisease; and (5) body mass index <25. Two groupsof patients were studied. The previously describedhealthy control group3 included eight men andeight women (mean age 70.4 ± 3.9 years, range61–78 years). The group of PM recipients includedsix men and three women, whose mean age was67.1 ± 10.8 years (range 60–74 years). All TalentDR PMs were implanted at least 1 month beforethe study for treatment of isolated chronotropicinsufficiency.6

The Talent DR is a dual chamber, rate respon-sive PM that combines a VE sensor with an an-teroposterior body acceleration sensor. The pacingsystem uses the sensor best adapted to ongoing ex-ercise conditions. The accelerometer contributed

to a faster response of the chronotropic functionat the onset and at the end of exercise whereasduring effort, VE, measured cycle-by-cycle, pro-vides a rate response proportional to the level ofactivity.

Exercise Testing

Each study participant underwent two exer-cise tests, 24 hours apart, at the same time ofday to minimize the effect of circadian physio-logical variations. The order of tests was set by arandomization table. Both treadmill exercise testsincluded measurements of gas exchange (CPX,Medical Graphics, St Paul, MN, USA). The char-acteristics of the standard CAEP test used for theevaluation of rate responsive PMs are listed inTable I. At each stage of exercise, the speed andslope are increased, resulting in an exponentialincrease in workload. The Harbor test offers threedifferent exercise schedules (Table II), which tai-lor the test according to the predicted patientexercise capacity.7 Following a warm-up stageof 3 minutes, the increase in workload is ap-plied linearly minute-by-minute. Since the goal ofthis test is to measure maximal exercise capacity,patients are incited to reach maximal clinical (fa-tigue, dyspnea) and metabolic (respiratory quo-tient >1.15) criteria. In the PM group, the PM wasprogrammed in the DDDR mode with dual sen-sor rate responsiveness, a lower pacing rate (LPR)set at 60 beats/min, and a maximum pacing rateclose to the maximal predicted heart rate (MPHR)(220 beats/min - patient age).

Table I.

Characteristics of CAEP Test

Stage Duration (min) Speed (MPH) Grade % METs

0 2 1 0 1.51 2 1 2 22 2 1.5 3 2.83 2 2 4 3.64 2 2.5 5 4.65 2 3 6 5.86 2 3.5 8 7.57 2 4 10 9.68 2 5 10 12.19 2 6 10 14.3

10 2 7 10 16.511 2 7 10 19

MET values for each stage were derived form published data(Reference 1) CAEP = chronotropic assessment exerciseprotocol; MET = metabolic equivalents; MPH = miles per hour.

240 January 2003, Part II PACE, Vol. 26

EXERCISE RESPONSE OF DUAL SENSOR DDDR PACEMAKER

Table II.

Harbor Test: Three Schedules Were Applied According to Each Patient’s Exercise Capacity

Light Schedule Intermediate Schedule Vigorous Schedule

Duration Speed Grade Duration Speed Grade Duration Speed GradeStage (min) (MPHR) (%) (min) (MPHR) (%) (min) (MPHR) (%)

0 3 1 0 3 1 0 3 1 01 1 1.2 1 1 1.4 1 1 1.6 22 1 1.4 2 1 1.8 2 1 2.2 43 1 1.6 3 1 2.2 3 1 2.8 64 1 1.8 4 1 2.6 4 1 3.4 85 1 2 6 1 3 6 1 4 106 1 2.2 8 1 3.4 8 1 4.6 147 1 2.2 10 1 3.4 10 1 5.2 148 1 2.2 12 1 3.4 12 1 5.8 14

MPH = miles per hour.

Data Collection

During each test, oxygen uptake (VO2), carbondioxide production (CO2), HR, and VE were mea-sured cycle-by-cycle, and recorded at each stage ofexercise and at peak effort. The ventilatory thresh-old was defined as the VO2 at which VE/VO2 rosewithout a concomitant increase in VE/VCO2.8 Twominutes after the end of exercise, blood was ob-tained from the earlobe for measurement of lactatelevels. At the end of each test, following a 5-minuteperiod of recovery, the patient’s perception of levelof effort reached and degree of dyspnea was gradedon a scale from 1 to 6 (Table III).

Calculated Parameters

Cardiac reserve (%HRR) was defined as thepercentage of HR variation between the pro-grammed resting rate and the MPHR in the controlgroup, and between the LPR and the maximal pro-grammed pacing rate in the PM group. The mea-

Table III.

Effort and Dyspnea Perception Score Graded by EachPatient at Peak Exercise

Score Perceived Effort Dyspnea

1 Very easy Minimally dyspneic2 Easy Somewhat dyspneic3 Relatively easy Moderately dyspneic4 Somewhat strenuous Dyspneic5 Strenuous Very dyspneic6 Exhaustive Severely dyspneic

sured metabolic reserve (Measured%MR) was de-fined as the percentage of variation in measuredVO2 between resting state and peak effort. Fromthese measurements, the slope of the linear rela-tionship between %HRR versus Measured %MR(Measured Reserve Slope = MS) was calculatedfor each patient and each exercise test. By mathe-matical extrapolation, the mean MS for each groupand each test was calculated at a null intercept.

At peak exercise, the percent VO2 reached wasdetermined with respect to the maximal theoret-ical VO2 (%MaxVO2), defined by the method ofWasserman et al.7 Likewise, percent functional ca-pacity at peak exercise was defined with respect tothe MPHR (%MaxHR).

Data Management and Statistical Analyses

All data are presented as means ± SD. Stu-dent’s paired t-tests were used to compare CAEPversus Harbor tests in a same group. Student’s non-paired t-tests were used to compare the two groupsin each exercise test. A linear regression analysisfor submaximal VO2 comparison during each testand each protocol was performed. MS was com-pared with the ideal slope 1.

ResultsIn the control group the mean MPHR was

150 ± 4 versus 153 ± 11 beats/min in the PM group(NS). The maximal programmed pacing rate was152 ± 12 beats/min and, as expected, was not sta-tistically significant compared to MPHR. The “in-termediate” exercise schedule was completed byeight patients in the control group and five patientsin the PM group, and the “vigorous” schedule wascompleted by eight patients in the control group

PACE, Vol. 26 January 2003, Part II 241

PAGE, ET AL.

Table IV.

Control Group

CAEP Harbor P

Exercise duration 14.7 ± 2.9 9.3 ± 1.5 <0.001(min)

HR at peak 154 ± 11 151 ± 10 NS(beats/min)

%MaxHR (%) 103 ± 7 101 ± 7 NSVO2 at AT (mL/kg/ 18.8 ± 6.1 18.4 ± 4.4 NS

per minute)VO2 at peak (mL/ 28.7 ± 6.4 28.4 ± 5.5 NS

kg/per minute)%Max VO2 123 ± 25 122 ± 18 NSVE at peak (L/min) 72 ± 20 68 ± 20 NSLactates (mmol/L) 4.9 ± 1.9 5.3 ± 1.9 NSEPS (1–6) 2.5 ± 1.4 2.8 ± 0.9 NSMS 1.1 ± 0.16 1.1 ± 0.18 NS

HR = heart rate; AT = anaerobic threshold; VE = ventilation;MS = metabolic slope; EPS = effort perception score.

and four patients in the PM Group. No patient ineither group was limited to the “light” exerciseschedule.

Tables IV and V, respectively, summarize thedata collected in the control group and in the PMgroup. All study participants reached an appropri-ate level of exercise, as expressed by mean lactateplasma concentrations, which were slightly higherin the control than the PM group during CAEP(4.9 ± 1.9 vs 3.7 ± 1.9 mmol/L, NS) and Harbor

Table V.

PM group

CAEP Harbor P

Exercise duration (min) 15.2 ± 3.3+ 9.9 ± 1.3+ <0.001HR at peak (beats/min) 144 ± 12* 138 ± 13** <0.01%MaxHR (%) 93 ± 6** 91 ± 7** <0.01VO2 at AT (mL/kg/per min) 17.7 ± 6.1+ 17.4 ± 8.5+ NSVO2 at peak (mL/kg/per min) 28.8 ± 10.7+ 29.1 ± 10.5+ NS%MaxVO2 (%) 135 ± 46* 137 ± 46* NSVE at peak (L/min) 73 ± 24+ 72 ± 28+ NSLactates (mmol/L) 3.7 ± 1.9+ 3.8 ± 1.8* NSEPS (1–6) 1.4 ± 0.7* 1.3 ± 0.5*** NSMS 0.84 ± 0.13 0.77 ± 0.15 <0.05

Comparisons vs control group: +NS; *P = 0.05; **P, = 0.01; ***P = 0.001.HR = heart rate; AT = anaerobic threshold; VE = ventilation; MS = metabolic slope; EPS = effortperception score.

(5.3 ± 1.9 vs 3.8 ± 1.8 mmol/L, P < 0.05) tests. Nostatistical difference was observed in VO2 and VEat peak exercise between the two groups duringeither test.

In the PM group, HR at peak exercise andthe metabolic reserve slope were higher duringthe CAEP than the Harbor protocol. These twomeasurements were significantly lower than inthe control group. The PM group also had lowerplasma lactate concentrations and dyspnea/fatiguescores.

DiscussionHR adaptation by rate responsive pacing

with dual sensors in patients suffering fromchronotropic insufficiency during exercise nor-malizes aerobic capacity.9 In comparison with agroup of healthy subjects, maximal O2 consump-tion, imposed workload, and maximal VE weresimilar in the paced patient population of thepresent study. The results were independent of theexercise protocol, whether the workload increasedexponentially, as in the CAEP test, or linearly, as inthe Harbor test. In contrast, maximal HR reachedin the PM group was lower than in the controlgroup in both tests. This may have been due toa lesser effort accomplished by patients, as sug-gested by lower lactate concentrations and dys-pnea/fatigue scores at peak exercise. This lessereffort was certainly attributable to some reserva-tion on the part of the medical team to challengepatients to reach maximal levels of exercise, aswas done systematically in healthy subjects. Con-sequently, the mean MPHR was limited to 90%,as opposed to 106% in the control group, andthe metabolic slope was distinctly shallower in

242 January 2003, Part II PACE, Vol. 26

EXERCISE RESPONSE OF DUAL SENSOR DDDR PACEMAKER

the control group, particularly during the Harbortest (0.84 ± 0.13 during CAEP vs 0.77 ± 0.15 dur-ing Harbor). The present results with the CAEPtest are, nevertheless concordant with those ofother published reports, which used the metabolicslope to measure the chronotropic response ofrate responsive PMs. This method was applied byKay,10 in 1998, in ten PM recipients with vari-ous sensors, including VE (n = 3), dP/dT (n = 4),and activity (n = 4). Though the mean curve valuewas 0.81 ± 0.25, a wide interindividual disper-sion of the results was observed, ranging between0.30 and 1.19. In a more homogeneous group of31 patients paced with the same accelerometer +VE dual sensor combination, Freedman et al.11

have recently reported a mean metabolic slopevalue of 0.92 ± 0.25, considerably closer to the nor-mal predicted value of 1. However, in 7 of the 31patients, the measured value was 2 SD below thatpredicted theoretical value. The authors empha-sized the importance of reaching peak exercise andobserved that the simulation of submaximal exer-cise by excluding the last stage lowered the slopeto <0.8, falling to nearly 0.6 when the last twostages were omitted. These results are concordantwith those observed in the current study, wheretwo patients who reached a peak HR equivalentto their MPHR had the steepest metabolic slopes(0.90 and 0.99 during CAEP), whereas two patientswith the shallowest slopes (0.67 and 0.65) hadonly reached 93% and 84% of their MPHR respec-

tively, when the test was terminated because offatigue. In all cases, the authors were able to verifythat the increase in HR was not limited by deviceprogramming.

Various studies of PMs with combined activ-ity + VE sensors have confirmed the predominantcontribution of the VE sensor during intense ef-fort.12−14 In this study, rate responsiveness offeredby the Talent DR 213 pulse generator was mostlybased on the VE sensor during effort, while theaccelerometer contributed to a faster response atthe onset and end of the tests. Consequently, thehigher VE was at peak exercise, the closer waspeak HR to the maximal programmed pacing rate.In the PM group, VE was similar at peak exer-cise during both tests. However, as mentioned ear-lier, mean peak HR reached in the PM group was6 beats/min lower during the Harbor than dur-ing the CAEP test. As opposed to the authors’ ini-tial hypothesis, the Harbor test seems less suitablethan the CAEP test to study the chronotropic re-sponse of PMs with dual sensors during exercise.The 1-minute duration of each stage versus 2 min-utes during CAEP is probably too short, resultingin a delayed response by the sensing system. Fi-nally, this study confirmed the high performanceof the Talent DR pulse generator in this physicallyfit patient population whose peak HR was con-siderably higher than in other studies, which hadenrolled patients with distinctly greater physicallimitations.

References1. Williams MA. Exercise testing and training in the elderly cardiac

patient. Champaign, IL, Human Kinetics Publishers, 1994, pp. 11–23.

2. Guire DK, Levine BD, Williamson JW, et al. A 30-year follow-up ofthe Dallas Bedrest and Training Study. Effect of age on cardiovascu-lar adaptation to exercise training. Circulation 2001; 4:1358–1366.

3. Page E, Bonnet JL, Durand C. Comparison of metabolic expenditureduring CAEP versus a test adapted to aerobic capacity (Harbor test)in elderly healthy individuals. PACE 2000; 23:1772–1777.

4. Wilkoff BL, Corey J, Blackburn G. A mathematical model of thecardiac chronotropic response to exercise. J Electrophysiol 1989;3:176–180.

5. Bonnet JL, Geroux L, Cazeau S. French Talent DR Pacemaker Inves-tigators. Evaluation of a dual sensor rate responsive pacing systembased on a new concept. PACE 1998; 21:2198–2203.

6. Camm AJ, Fei L. Chronotropic incompetence – Part II: Clinical im-plications. Clin Cardiol 1996; 19:503–508.

7. Wasserman K, Hansen JE, Sue DY, et al. Protocols for exercisetesting. In K. Wasserman et al. (ed): Principles of Exercise Test-ing and Interpretation. Philadelphia, Lea & Febiger, 1994, pp. 58–71.

8. Dickstein K, Barvik S, Aarsland T, et al. A comparison of method-ologies in detection of the anaerobic threshold. Circulation 1990;81(Suppl. 1):II38–II48.

9. Ellestad MH. Chronotropic incompetence. The implications ofheart rate response to exercise (compensatory parasympathetic hy-peractivity?) Circulation 1996; 93:1485–1487.

10. Kay GN. Quantitation of chronotropic response: Comparison ofmethods for rate-modulating permanent pacemakers. J Am CollCardiol 1992; 20:1533–1541.

11. Freedman RA, Hopper DL, Mah J, et al. Assessment of pacemakerchronotropic response: Implementation of the Wilkoff mathemati-cal model. PACE 2001; 24:1748–1754.

12. Alt E, Combs W, Fotuhi P, et al. Initial clinical experience with anew dual sensor SSIR pacemaker controlled by body activity andminute ventilation. PACE 1995; 18:1487–1495.

13. Ovsyshcher I, Guldal M, Karaoguz R, et al. Evaluation of a newrate adaptive ventricular pacemaker controlled by double sensors.PACE 1995; 18:386–390.

14. Lascault G, Pansard Y, Scholl JM, et al. Dual chamber rate respon-sive pacing and chronotropic insufficiency. Comparison of doubleand respiratory sensors. Arch Mal Cœur 2001; 94:190–195.

PACE, Vol. 26 January 2003, Part II 243