Respiratory Medicine (2010) 104, 1557e1565

ava i lab le at www.sc ienced i rec t . com

journa l homepage : www.e lsev ier . com/ loca te / rmed

Effects of interval exercise training on respiratorydrive in patients with chronic heart failure

Athanasios Tasoulis a, Ourania Papazachou a, Stavros Dimopoulos a,Vasiliki Gerovasili a, Eleftherios Karatzanos a, Theodoros Kyprianou a,Stavros Drakos b, Maria Anastasiou-Nana b, Charis Roussos a,Serafim Nanas a,*

a Pulmonary & Critical Care Medicine Department, Cardiopulmonary Exercise Testing and Rehabilitation Laboratory,“Evgenidio” Hospital, National & Kapodistrian University of Athens, 20, Papadiamantopoulou St, Athens 11528, GreecebThird Clinic of Cardiology, NKUA, Athens, Greece

Received 25 October 2009; accepted 16 March 2010

KEYWORDSVentilatory pattern;Mouth occlusionpressure;Inspiratory pressure;Respiratory drive;Interval training;Chronic heart failure

* Corresponding author. Tel.: þ30 21E-mail address: a-icu@med.uoa.gr

0954-6111/$ - see front matter ª 201doi:10.1016/j.rmed.2010.03.009

Summary

Background: Patients with chronic heart failure (CHF) suffer from ventilatory abnormalities.This study examined the effects of interval exercise training on the respiratory drive in CHFpatients.Methods: Forty-six clinically stable CHF patients (38 males/8 women, mean age Z 53 � 11years) participated in an exercise rehabilitation program (ERP) 3 times/week, for 12 weeksby interval training modality with or without the addition of resistance training. All patientsunderwent symptom-limited cardiopulmonary exercise testing (CPET), and measurements ofmouth occlusion pressure at 100 ms (P0.1) and maximum inspiratory muscle strength (PImax)before and after ERP. Respiratory drive was estimated by mouth occlusion pressure P0.1 andP0.1/PImax ratio at rest, and the ventilatory pattern by resting mean inspiratory flow (VT/TI)and by VT/TI at identical CPET workloads, before and after ERP. We also studied a controlnon exercising group of 11 patients (8 men and 3 women).Results: P0.1 at rest decreased from 3.04� 1.52 to 2.62� 0.9 cmH2O (pZ 0.015), P0.1/PImax % atrest from 4.56 � 3.73 to 3.69 � 2.03 (p Z 0.006), resting VT/T I from 0.44 � 0.10 to0.41 � 0.10 l/s (pZ 0.014), and VT/TI at identical work rate from 2.13 � 0.59 to 1.93 � 0.58 l/s(p Z 0.001) after ERP. VO2 at peak exercise increased from 16.3 � 4.8 to 18.5 � 5.3 ml/kg/min(p < 0.001) in the exercise group. No improvement was noted in the control group.

0 7201918/6973036448; fax: þ30 210 7244941.(S. Nanas).

0 Published by Elsevier Ltd.

1558 A. Tasoulis et al.

Conclusions: ERP by interval training improves the respiratory drive and ventilatory pattern at restand during exercise in CHF patients.ª 2010 Published by Elsevier Ltd.

Introduction

Dyspnea and fatigue are the main disease manifestations,which limit the physical activity and lower the quality of lifeof patients suffering from chronic heart failure (CHF).1e3

Several central or peripheral pathophysiological mecha-nisms, such as low cardiac output, pulmonary hypertension,4

increased pulmonary capillary wedge pressure during exer-cise, and peripheral myopathy,5 contribute to exerciseintolerance in CHF. Respiratory abnormalities, includingrestrictive and obstructive breathing patterns,6e8 decreasedlung compliance,9 decreased diffusion capacity,10 hyper-ventilation,11 ventilationeperfusion mismatch,12 bronchialhyper-responsiveness,13 respiratory muscle weakness14,15

and decreased inspiratory capacity16 have also beenobserved in these patients. The respiratory drive in patientssuffering from CHF is significantly higher than in normalsubjects, indicating the exertion of a greater respiratoryeffort, associated with an increased dyspnea score.17

Exercise by continuous aerobic training has been describedto improve functional capacity, quality of life and clinicaloutcomesofCHFpatients.18e20Recentdata showthat intervalaerobic training can be a valid alternativemodality of trainingthat improves maximal oxygen uptake even in CHF patientswith severely impaired cardiovascular function.21e23 Recentstudies have also suggested that resistance training incombination with aerobic training induces beneficial effectsin aerobic capacity, muscle strength and peripheral endo-thelial function, with neuro-modulation and anti-inflammatory effects in these patients.24,25

Our hypothesis was that interval aerobic training aloneor with the addition of resistance training would reverserespiratory drive abnormalities in CHF patients.

We therefore, prospectively examined the effects ofa 12-week, hospital-based, supervised exercise rehabilita-tion program (ERP) on CHF patients’ respiratory drive.

Materials and methods

Patients of our heart failure clinic presenting with stable CHF,NYHA functional class � III, left ventricular ejection fraction(LVEF) less than 50%were referred to our rehabilitation centerin order to participate in a non hospital based rehabilitationprogram. The patients had no need for hospitalization andtheirmedical treatmentwasoptimizedand stable forat least 2months including beta-blockers, angiotensin convertingenzyme inhibitors, diuretics, nitrates, spironolactone, amio-darone and digoxin. The diagnosis of CHFwas based on clinicalevaluation and laboratory testing, using non-invasive andinvasive procedures that included electrocardiogram, chestroentgenogram, echocardiogram, right heart catheterization,radionuclide ventriculography, coronary angiography andmyocardial biopsy, as clinically indicated. Patients wereexcluded if symptom-limitedcardiopulmonaryexercise testing(CPET) performed according to the American Thoracic

Society/American College of Chest Physicians Statement onCPETwas contraindicated26 and if they presented with severeCOPD or severe valvular disease. 69 participated in finalrandomization. 23 were dropped out after randomization dueto hospitalization for respiratory infections (3 patients), heartfailure decompensation (2 patients) and inability due topersonal reasons to follow the time table of the program ororthopedic related problems (18 patients).

The baseline characteristics of the 46 (38 men and 8women) CHF patients are listed in Table 1. A history of type IIdiabetes mellitus treated with oral anti-diabetics was eli-cited in 3 patients. The study was designed as a randomizedcontrolled parallel two-group study. The patients enrolled inthe study were randomly assigned, according to VO2 peak(cut-off 16 ml/kg/min) and age (cut-off 50 years old), tointerval aerobic or combined interval aerobic plus strengthexercise training program, so that the two groups werestratified, and attended a three-month rehabilitationprogram. The primary end point of the study was to investi-gate the effect of exercise training on respiratory drive andbreathing pattern. The study also aimed at revealing thedifferences, if such exist, between the two exercise modes.The investigator performing the CPET and respiratory drivemeasurements was not aware of the patients’ specificassignment to the aerobic or combined group.

All patients underwent maximal symptom-limited cardio-pulmonary exercise testing (CPET), before the beginning andafter completion of the program. No significant differencesexisted among the groups in respect to age, gender, NYHAclass, CHF etiology, LVEF or medical treatment.

Agroupof11patients (8menand3women)presentingwithstable CHF, matched for age (53 � 11 years), BMI (26 � 4 kg/m2), and exercise capacity was also included in the study asa control group that did not participate in the rehabilitationprogram. The control group was a non randomized, nonparallel group, that consisted of patients that presented withstable CHF, NYHA functional class � III, left ventricular ejec-tion fraction (LVEF) less than 50% with no need for hospitali-zation and optimal medical treatment, who were referred toour rehabilitation center and were selected prospectively.The diagnosis of CHF was based on clinical evaluation andlaboratory testing as in the exercise group. These patients didnot wish to participate in the rehabilitation program eventhough theywere encouraged to do so, aswere the patients inthe exercise group. They were given standard instructions fortheir daily activity, encouraged to walk 3e5 times per weekfor 20e30 min and they had the same clinical observation asthe patients randomized to our study.

This study was approved by our Institutional BioethicsCommittee, and written informed consent was obtainedfrom all patients.

Exercise prescription

Exercise CHF group underwent aerobic exercise training,with or without strength training on electromagnetically

Table 1 Baseline characteristics of 46 study patients.

All patients Aerobic group Combined group

Age, years 53 � 11 53 � 11 53 � 12Men/women 38/8 19/2 19/6Body mass index, kg/m2 27.2 � 4.6 27.0 � 4.7 27.4 � 4.6New York heart association functional class

I 13 6 7II 26 13 13III 7 2 5

Left ventricular ejection fraction, % 34.9 � 10.5 34.1 � 11 35.6 � 10.3Underlying heart diseaseCardiomyopathy

Dilated 27 11 16Ischemic 16 8 8Valvular 3 2 1

Drug regimensAngiotensin-converting enzyme inhibitors 37 17 20b-Adrenergic blockers 41 19 22Diuretics 42 20 22Spironolactone 26 12 14Digoxin 14 6 8Nitrates 2 2 e

Amiodarone 16 7 9

Values are means � SD or numbers of observations.

Respiratory drive and interval training in CHF 1559

braked cycle ergometers (Cateye Ergociser, EC1600;Tokyo, Japan) 3 times weekly, for 12 weeks. Any missedsessions were added to the end of the program, so thatthe 36 sessions were completed. Patients assigned toaerobic training (n Z 21) exercised at a 50% of baselineSteep Ramp Test (SRT) (which was performed ona different day from the CPET) alternating 30 s of exercisewith 60 s of rest for 40 min per session.22 The trainingintensity was adjusted every 6 sessions according to a newSRT. The patients assigned to combined training (n Z 25)exercised at a 50% of baseline SRT, alternating 30 s ofexercise with 60 s of rest for 20 min per session and alsoattended 20 min strength training, so that the total timeof exercise was the same in both groups. Strength trainingincluded training of quadriceps, at 55e65% of 2 resistantmaximum (2RM), training of ham strings, at 55e65% of 2RMminus 1 kg, training of the biceps and muscles of theshoulder zone, at an intensity level defined by the weightsthe patient could lift up to 10 repetitions. Each musclegroup was trained at each limb separately with a pause of30 s between the groups. The exercise sessions weresupervised by physicians, exercise physiologists andphysical therapists, including continuous monitoring ofthe electrocardiogram and peripheral oxygen saturation,regular blood pressure measurements, and Borg scalescore assessment.

Cardiopulmonary exercise testing

All patients performed a symptom-limited, ramp-incre-mental CPET on an electromagnetically braked Ergoline 800cycle ergometer (SensorMedics, Anaheim, CA) before andafter the completion of the program. In order to reach testduration of 8e12 min, the individual increments in workload

were calculated, using the equation published by Hansenet al.27 Gas exchange was measured with the patientbreathing through a low resistance valve, with the noseclamped, using amodelVmax 229D instrument (SensorMedics)calibrated with a known gas mixture before each test.Breath-by-breath oxygen uptake (VO2), carbon dioxideoutput (VCO2) and ventilation (VE) were measured with theon-line system. All measurements were recorded for 2min atrest, during unloaded pedaling for 3 min before exercise,throughout exercise, and during the first 5 min of recovery.Baseline VO2 was the average of measurements made at restfor 2 min. Peripheral O2 saturation was monitored continu-ously by pulse oxymetry. Heart rate and rhythm were moni-tored by a MAX 1, 12-lead electrocardiographic system(Marquette Electronics, Milwaukee, WI) and blood pressurewas measured every 2 min with a mercury sphygmomanom-eter. All patients were verbally encouraged to exercise tointolerable leg fatigue or dyspnea.

Cardiopulmonary measurements

The gas exchange measurements were used to calculateVO2 at peak exercise (VO2p) and at the anaerobic threshold,and the ventilatory equivalent of VCO2 (VE/VCO2) slopebetween exercise onset and anaerobic threshold. Peak VO2

was calculated as the average of measurements made for20 s immediately before the end of exercise. The anaerobicthreshold was determined using the V-slope technique,28

and the result was confirmed graphically from a plot ofventilatory equivalent for oxygen (VE/VO2) and carbondioxide (VE/VCO2) against time. The ventilatory response toexercise was calculated by linear regression as the slope ofVE vs. VCO2 from the beginning of exercise to anaerobicthreshold, when the relationship is linear.29 In order to

1560 A. Tasoulis et al.

examine VO2 kinetics during the early recovery phase, thefirst-degree slope of VO2 for the first minute of recovery(VO2/t-slope) was calculated.

29 Peak work rate was definedas the highest workload achieved and maintained for 30 s,at a �50 rpm pedaling rate.

Pulmonary function tests

SpirometryAfter familiarization of the patient with the laboratoryenvironment, measurements of forced vital capacity (FVC)and forced expiratory volume at 1 s (FEV1) were madein the sitting position with a closed-circuit spirometer(SensorMedics) as recommended by the American ThoracicSociety. The measurements were made before andrepeated after ERP.30

Maximal inspiratory pressureIn all patients, the maximal inspiratory pressure (PImax) wasmeasured at rest, using a Vmax 229 system for pulmonaryand metabolic tests (SensorMedics), by a method similar tothat described by Black and Hyatt.31 For the measurementof PImax, each patient was instructed to exhale to theresidual volume, and then produce a maximal inspiratoryeffort through a closed mouthpiece. The highest peakpressure reached during 2 (out of at least 3) attempts thatdiffered by <5% was averaged and retained for analysis. Alltests were performed before and after ERP.

Respiratory driveMeasurements of the mouth occlusion pressure (P0.1) weremade 100 ms after the onset of inspiration at rest. P0.1 wasthe inspiratory pressure developed 100 ms after closure ofa valve at the level of functional residual capacity, at rest(Vmax 229, SensorMedics). After 4 calm breaths with constantfunctional residual capacity (FRC) the level of FRC wasdetected. Then the valve was randomly closed, to minimizethe likelihood of a deliberate respiratory effort exerted bythe patient against the valve. The measurement wasrepeated 6e8 times, and P0.1 was the average of 4 measure-ments that differed by <5%.32,33 P0.1 was expressed as an

Table 2 Respiratory drive and breathing pattern indices before

Exercise rehabilitation p

At rest

Before After

P0.1 (cmH2O) 3.04 � 1.52 2.62 �P0.1/PImax % 4.56 � 3.73 3.69 �Mean inspiratory flow (l/s) 0.44 � 0.10 0.41 �Respiratory cycle (%) 43.9 � 6.1 45.9 �Breathing frequency (breath/min) 19.5 � 4.4 19.6 �Tidal volume (l) 0.60 � 0.11 0.57 �Minute ventilation (l/min) 11.4 � 2.8 11.1 �Partial pressure of end-tidal carbondioxide (mmHg)

34.3 � 4.5 35.4 �

Values are means � SD.P0.1: mouth occlusion pressure, 100 ms after the onset of inspiration, P

absolute value (cmH2O), and as a percentage of PImax (P0.1/PImax%), to normalize its measurement for individual differ-ences in inspiratory muscle strength. Mean inspiratory flow(VT/TI) and end-tidal carbon dioxide (PETCO2), VE, VT,breathing frequency (fb), inspiratory time (TI), total respi-ratory time (TTot), and respiratory cycle (TI/TTot), werecalculated at rest and breath-by-breath during exercise.

Statistical analysis

All continuous variables are presented as means � SD. Forbaseline characteristics, means of continuous variableswere compared by paired Student’s t-test, and categoricalvariables by chi-square test. Cardiopulmonary measure-ments made at baseline and after the completion of theexercise program or at follow up for the control group werecompared by paired-samples t-test or Wilcoxon’s signed-rank test, as appropriate. Comparison between groupswas made by repeated measures analysis. The level ofsignificance was set at <0.05.

Results

After completion of the ERP, mean P0.1, P0.1/PImax, and VT/TIat rest, were lower than before ERP (Table 2, Figs. 1a and b,2a). In contrast, a significant increase in TI/TTot wasobserved after ERP (Table 2). There were no statisticallysignificant differences after ERP in mean fb, VT, VE, andPETCO2 at rest (Table 2). We compared, in each patient, theindices of ventilatory pattern at peak workload before ERPwith those measured at the same workload after ERP. MeanVT/TI, fb, VE and TI/TTot decreased and PETCO2 increased(Figs. 2b, 3). VT did not change significantly (Table 2). VO2 atpeak exercise increased (16.3 � 4.8 vs. 18.5 � 5.3 ml/kg/min, p < 0.001 and 1.31 � 0.43 vs. 1.49 � 0.51 L/min,p < 0.001 for the whole exercise group, 16.4 � 4.1 vs.17.8 � 4.6 ml/kg/min, p < 0.006 and 1.34 � 0.36 vs.1.43 � 0.34 L/min, p Z 0.022 for aerobic group, 16.2 � 5.3vs. 19.1 � 5.8, p < 0.001 and 1.28 � 0.49 vs. 1.54 � 0.62 L/min, p< 0.001 for combined group). Maximumworkload alsoincreased (101 � 37 vs. 119 � 41 W, p < 0.001 for the whole

and after exercise rehabilitation program for the whole group.

rogram

At identical work rate

p Before After p

0.9 0.015 e e e

2.03 0.006 e e e

0.10 0.014 2.13 � 0.59 1.93 � 0.58 0.0017.3 0.008 47.8 � 3.4 46.3 � 2.6 0.0025.0 ns 38.4 � 8.2 33.9 � 6.9 <0.0010.11 0.058 1.59 � 0.39 1.58 � 0.38 ns2.7 ns 61.1 � 17.2 53,6 � 16.3 <0.0015.2 ns 32.2 � 5.5 36.2 � 6.7 <0.001

0.1/PImax %: mouth occlusion pressure/max inspiratory pressure %.

Figure 1 a. Error bar of P0.1 at rest before and after ERP in 46patients with CHF. b. Error bar of P0.1/PImax at rest before andafter ERP in 46 patients with CHF.

Figure 2 a. Error bar of VT/TI at rest before and after ERP in46 patients with CHF. b. Error bar of VT/TI at identical work-load before and after ERP in 46 patients with CHF.

Figure 3 Error bar of PETCO2 at identical workload beforeand after ERP in 46 patients with CHF.

Respiratory drive and interval training in CHF 1561

group, 104� 28 vs. 117� 29 W for aerobic group, p< 0.001,99 � 43 vs. 120 � 48 W for combined group, p < 0.001) aswell as VO2/t-slope (0.48 � 0.24 vs. 0.6 � 0.27 L min�1

min�1, p < 0.001 for the whole group, 0.49 � 0.24 vs.0.57 � 0.25 L min�1 min�1, p Z 0.024 for aerobic group,0.46 � 0.23 vs. 0.63 � 0.30 L min�1 min�1, p < 0.001), whileVE/VCO2 slope (31.5 � 6.0 vs. 30.3 � 5.5, p Z 0.14)decreased but not statistically significantly. FEV1, FEV1%

predicted, FVC, FVC % predicted and FEV1/FVC, did notchange significantly after ERP. No differences were notedregarding respiratory drive parameters and cardiopulmo-nary parameters between the two groups of training. Bothimproved P0.1, P0.1/PImax, VT/TI at rest, TI/TTot at rest andVT/TI, fb, VE, PETCO2 at identical work rate as well as VO2 atpeak, VO2 at anaerobic threshold, VO2/t-slope, maximumworkload and VE/VCO2 slope following the ERP (Tables 3 and4). Our patients show no difference in their weight and insystolic and diastolic blood pressure at rest and at peakexercise before and after the ERP.

There was no statistically significant difference atbaseline and 3 months after regarding respiratory driveand cardiopulmonary parameters in the control group(Table 5). Between the exercise group and the control

1562 A. Tasoulis et al.

group there was a statistical significant difference in VO2

peak and P0.1/PImax at rest before and after 3 months offollow up (Table 5).

Discussion

This prospective study has shown that both types oftraining, aerobic interval and interval plus strength trainingimproved respiratory drive and ventilatory pattern ofpatients suffering from CHF, at rest and during exercise. Toour knowledge this is a novel study that evaluates theeffects of interval and combined exercise training onrespiratory drive in CHF patients.

The respiratory centers are stimulated by centralrespiratory pacer cells, central and peripheral chemore-ceptors, upper airway receptors, other areas of the brain,pulmonary tension receptors, J receptors, and volitionalpathways. The signals are integrated to a single signal tothe respiratory muscles.34 The respiratory drive is a surro-gate measure of this neural output, which controls theventilatory pattern.17 Respiratory drive, expressed as P0.1,reflects both the neural output to, and the mechanicalresponse from, the inspiratory muscles.32 Its activation hasbeen studied in normal subjects and in patients sufferingfrom CHF. While it is used as an index of neural output tothe inspiratory muscles in normal subjects, absolute valuesof P0.1 might underestimate the effective neural drive inpatients with inspiratory muscle weakness. Consequently,the P0.1/PImax ratio is used as an index of central respiratoryoutput, normalized for inspiratory muscle strength.Previous studies have observed a higher respiratory drive,reflected by resting P0.1, P0.1/PImax and VT/TI ratio, inpatients with CHF compared to healthy controls, althoughVT was similar at baseline.17,35 Our study has shownimprovement in several respiratory drive parameters afterexercise in contrast with a previous study suggesting onlyimprovement in inspiratory muscle performance atcomparable submaximal workloads, probably due todifferent exercise modality of training and the smallnumber of patients in that study.36

Table 3 Respiratory drive and breathing pattern indices beforgroup of patients.

Exercise rehabilitation p

At rest

Before After

P0.1 (cmH2O) 3.04 � 1.54 2.64 �P0.1/PImax % 4.27 � 3.0 3.33 �Mean inspiratory flow (l/s) 0.43 � 0.08 0.39 �Respiratory cycle (%) 46.3 � 6.9 47.3 �Breathing frequency (breath/min) 18.1 � 2.9 19.5 �Tidal volume (l) 0.64 � 0.11 0.58 �Minute ventilation (l/min) 11.7 � 2.5 11.1 �Partial pressure of end-tidal carbondioxide (mmHg)

34.7 � 4.7 35.1 �

Values are means � SD.P0.1: mouth occlusion pressure, 100 ms after the onset of inspiration, P

Exercise training improves several manifestations ofCHF, including neuro-hormonal, muscular and ventilatoryabnormalities, microcirculation and also increases exercisecapacity significantly.19,20,37,38 Our observations suggestthat ERP activated pathophysiological pathways, whichsignificantly lowered an abnormally high baseline respira-tory drive. These changes caused significant decreases inVT/TI, VE and fb at equivalent workloads. No difference wasobserved in VT, indicating that the decrease in VE wasentirely attributable to a decrease in fb.

An excessive activation of the ergoreceptors has beenreported in patients presenting with CHF, along withcachexia, autonomic abnormalities consistent with sympa-thetic activation and vagal withdrawal, hemodynamicabnormalities, and elevated plasma catecholamine concen-trations.39,40 Therefore, the heightened sympathetic, vaso-constrictor, and ventilatory drive characteristic of CHFmightbe partially explained by an exaggerated contribution ofergoreflex during exercise. Physical training increases theexercise capacity and partially reverses the abnormalresponse to exercise by attenuating this reflex excitation.41

With ERP, the responses to the enhanced ergoreflex ofpatients in CHF were brought to levels closer to those ofnormal subjects. Although the mechanisms by which ERPimproves the pathophysiology of CHF are incompletelyunderstood, it is known to partially reverse early muscleacidosis and muscle atrophy, lower cytokines levels, andimprove peripheral hemodynamic function. Improvingmuscle function can attenuate the stimulus to the ergore-flex, restoring a physiological balance.40 These pathophysi-ological mechanisms might explain our observations.

Also neurohumoral activation, including activation of thesympathetic nervous system, characterizes CHF and patientswith the greatest sympathetic activation have the poorestsurvival. Muscle sympathetic nerve activity (MSNA) is provedto be greater in heart failure patients compared with normalcontrols, and exercise training results in dramatic reductionsin directly recorded resting sympathetic nerve activity.42 Inpatientswith CHF and SleepApnea regular exercise improvesOSA severity and sleep architecture and reduces MSNA. It isproposed that improvement in carotid chemoreflex control is

e and after exercise rehabilitation program for the aerobic

rogram

At identical work rate

p Before After p

0.9 0.057 e e e

1.76 0.062 e e e

0.08 ns 2.2 � 0.56 2.02 � 0.55 0.0066.3 ns 48.3 � 3.4 46.9 � 2.3 0.0663.6 ns 38.1 � 7.8 35.1 � 8.1 0.0040.12 0.011 1.67 � 0.28 1.62 � 0.30 ns2.3 ns 63.2 � 15.0 56.7 � 15.7 0.035.8 ns 32.0 � 5.8 35.0 � 6.3 0.16

0.1/PImax %: mouth occlusion pressure/max inspiratory pressure %.

Table 5 Cardiopulmonary exercise testing and respiratory drive indices before and after 3 months of follow up of bothexercise training and control group.

Exercise group Control group

Before After Before After p (Betweengroups)

Respiratory drive parameters at restP0.1 (cmH2O) 3.04 � 1.52 2.62 � 0.9** 2.62 � 0.82 2.71 � 0.9 nsP0.1/PImax % 4.56 � 3.73 3.69 � 2.03** 3.1 � 1.5 3.3 � 2.1 0.045Mean inspiratory flow (l/s) 0.44 � 0.10 0.41 � 0.10** 0.46 � 0.13 0.48 � 0.17 0.09Respiratory cycle (%) 43.9 � 6.1 45.9 � 7.3** 45.4 � 5.4 44.4 � 6.9 nsBreathing frequency(breath/min)

19.5 � 4.4 19.6 � 5.0 20.9 � 3.5 20.4 � 3.9 ns

Tidal volume (l) 0.60 � 0.11 0.57 � 0.11* 0.62 � 0.14 0.6 � 0.12 nsMinute ventilation (l/min) 11.4 � 2.8 11.1 � 2.7 12.3 � 2.6 12.5 � 3.2 nsPartial pressure of end-tidalcarbon dioxide (mmHg)

34.3 � 4.5 35.4 � 5.2 31.8 � 4.5 31.6 � 3.7 ns

Cardiopulmonary exercise testing parametersVO2 at peak exercise(ml/kg/min)

16.3 � 4.8 18.5 � 5.3* 15.8 � 5.9 16.4 � 6.5 0.05

VE/VCO2slope 31.5 � 6.0 30.3 � 5.5 34.7 � 9.5 35.3 � 8.3 nsMean inspiratory flow (l/s) atpeak

2.13 � 0.59 2.4 � 0.7* 2.07 � 0.8 2.12 � 0.74 0.069

Respiratory cycle (%)at peak 47.8 � 3.4 48.2 � 3.0 44.7 � 3.0 46.3 � 2.8 nsBreathing frequency (breath/min) at peak

38.4 � 8.2 40.6 � 6.8** 34.6 � 8.2 35.0 � 6.0 ns

Tidal volume (l) at peak 1.59 � 0.39 1.73 � 0.43* 1.6 � 0.56 1.64 � 0.54 0.077Minute ventilation (l/min) atpeak

61.1 � 17.2 70.5 � 21.6* 55.8 � 23.1 57.7 � 22.4 0.08

Partial pressure of end-tidalcarbon dioxide (mmHg) atpeak

32.2 � 5.5 32.1 � 6.2 31.7 � 5.9 31.9 � 5.1 ns

Values are means � SD.P0.1: mouth occlusion pressure, 100 ms after the onset of inspiration, P0.1/PImax %: mouth occlusion pressure/max inspiratory pressure %.Comparison within group analysis. *p < 0.001, **p < 0.05.

Table 4 Respiratory drive and breathing pattern indices before and after exercise rehabilitation program for the combinedgroup of patients.

Exercise rehabilitation program

At rest At identical work rate

Before After p Before After p

P0.1 (cmH2O) 3.04 � 1.53 2.59 � 0.92 ns e e e

P0.1/PImax % 4.84 � 4.37 3.98 � 2.21 0.099 e e e

Mean inspiratory flow (l/s) 0.45 � 0.12 0.41 � 0.12 0.052 2.06 � 0.61 1.86 � 0.6 0.031Respiratory cycle (%) 41.8 � 4.5 44.7 � 8.0 0.011 47.4 � 3.4 45.8 � 2.7 0.011Breathing frequency(breath/min)

20.7 � 5.1 19.6 � 6.0 ns 38.6 � 8.6 32.9 � 5.6 0.003

Tidal volume (l) 0.57 � 0.11 0.57 � 0.12 ns 1.53 � 0.46 1.55 � 0.44 nsMinute ventilation (l/min) 11.3 � 3.2 11.1 � 3.0 ns 59.4 � 18.9 51.0 � 16.6 0.006Partial pressure of end-tidalcarbon dioxide (mmHg)

34.0 � 4.3 35.6 � 4.6 0.071 32.3 � 5.3 37.3 � 6.9 0.001

Values are means � SD.P0.1: mouth occlusion pressure, 100 ms after the onset of inspiration, P0.1/PImax %: mouth occlusion pressure/max inspiratory pressure %.

Respiratory drive and interval training in CHF 1563

1564 A. Tasoulis et al.

implicated in this exercise training benefit. Since chemo-reflex control integrates in the central nervous system (CNS),it is reasonable to conclude that exercise leads to changes inthe CNS that may affect respiratory drive.43

Another significant finding was the improvement ofPETCO2 after ERP which is an indirect index of ventilatorefficiency. Furthermore, the ventilatory response to exer-cise as assessed by VE/VCO2 slope, a significant prognosisindicator that also reflects interstitial pulmonary edemademonstrated a trend to improve.44

In patients with Chronic Obstructive Pulmonary Disease(COPD) the degree of exertional dyspnea correlates withresting respiratory drive and at similar mechanical load andsimilar levels of respiratory muscle dysfunction, dyspneamay result from an individual’s response of the centralmotor output.45 These data support the concept that exer-tional dyspnea is modulated by afferent output from centralreceptors. In patients with COPD the beneficial effect ofexercise training in increasing aerobic capacity, neuromus-cular coupling of the ventilatory pump, yielding significantadaptations in the muscles and contributing to the relief ofboth exertional dyspnea is well established.46,47

Although there was no statistical difference betweenthe two types of exercise training it seems that combinedtraining improves respiratory cycle and PETCO2 at rest andat identical work rate more than interval training alonesuggesting that the greater peripheral muscle stimulationmight induce beneficial effects in peripheral chemorecep-tors activity, and endothelial function in CHF patients;however the small number of patients enrolled in this studylimits the comparison of the two types of training.

Clinical implications

P0.1, P0.1/PImax and VT/TI seem to be among the few vari-ables measurable at rest, which improved after ERP. It alsoseems that interval exercise training alone or with theaddition of resistance training not only improves exercisecapacity but also partially reverses respiratory driveabnormalities during exercise in chronic heart failurepatients. Our study also implies that exercise trainingmodulates chemosensitivity and minimizes the output ofthese central receptors potentially benefiting patients withchronic heart failure. The reliability of these parameters asa means of evaluating therapeutic interventions in thesepatients merits further investigation.

Study limitations

Comparison of the two groups of training did not demon-strate differences regarding respiratory drive and cardio-pulmonary parameters due to the small number ofparticipants. The power analysis for the comparison of thetwo training groups showed that the study was underpow-ered. Definite conclusions for the possible different effectsof the two types of training on respiratory drive cannot bededuced by the present study. A limitation of the study isalso that we did not use a specific questionnaire in order tomeasure and monitor the patients’ daily activity.

Another possible criticism of the present study is thecontroversial use of P0.1 to measure central drive in

patients with CHF and coexisting COPD as P0.1 is influencedby hyperinflation and by resistive and elastic loads, factorsthat lead to P0.1 underestimation.33

In conclusion, a 3-month supervised ERP by intervalexercise training alone or with the addition of resistancetraining improves the respiratory drive and the ventilatorypattern at rest and during exercise in CHF patients.

Acknowledgements

This work was supported by the Special Account forResearch Grants of the National and Kapodistrian Universityof Athens, Greece and Thorax Foundation.

Conflict of interest

None declared.

References

1. McParland C, Krishnan B, Wang Y, Gallagher CG. Inspiratorymuscle weakness and dyspnea in chronic heart failure. Am RevRespir Dis 1992;146:467e72.

2. Lipkin DP, Canepa-Anson R, Stephens MR, Poole-Wilson PA.Factors determining symptoms in heart failure: comparison offast and slow exercise tests. Br Heart J 1986;55:439e45.

3. Hamilton AL, Killian KJ, Summers E, Jones NL. Muscle strength,symptom intensity and exercise capacity in patients withcardiorespiratory disorders. Am J Respir Crit Care Med 1995;152:2021e31.

4. Butler J, Chomsky DB, Wilson JR. Pulmonary hypertension andexercise intolerance in patients with heart failure. J Am CollCardiol 1999;34(6):1802e6.

5. Clark AL, Poole-Wilson PA, Coats AJS. Exercise limitation inchronic heart failure: central role of the periphery. J Am CollCardiol 1996;28(5):1092e102.

6. Ries AL, Gregoratos G, Friedman PJ, Clausen JL. Pulmonaryfunction tests in the detection of left heart failure: correlationwith pulmonary artery wedge pressure. Respiration 1986;49:241e50.

7. Guazzi M, Agostoni P, Matturri M, Pontone G, Guazzi MD.Pulmonary function, cardiac function, and exercise capacity ina follow-up of patients with congestive heart failure treatedwith carvedilol. Am Heart J 1999;138:460e7.

8. Ravenscraft SA, Gross CR, Kubo SH, et al. Pulmonary func-tion after successful heart transplantation. Chest 1993;103:54e8.

9. Evans SA, Watson L, Cowley AJ, Johnston ID, Kinnear WJ. Staticlung compliance in chronic heart failure: relation with dysp-noea and exercise capacity. Thorax 1995;50:245e8.

10. Puri S, Baker BL, Dutka DP, Oakley CM, Hughes MB, Cleland JG.Reduced alveolarecapillary membrane diffusing capacity inchronic heart failure. Its pathophysiological relevance andrelationship to exercise performance. Circulation 1995;91:2769e74.

11. Clark AL, Volterrani M, Swan JW, Coats AJ. The increasedventilatory response to exercise in chronic heart failure:relation to pulmonary pathology. Heart 1997;77:138e46.

12. Clark AL, Volterrani M, Swan JW, Coats AJ. Ventilationeperfusionmatching in chronic heart failure. Int J Cardiol 1995;48:259e70.

13. Cabanes LR, Weber SN, Matran R, et al. Bronchial hyper-responsiveness to methacholine in patients with impaired leftventricular function. N Engl J Med 1989 May 18;320(20):1317e22.

Respiratory drive and interval training in CHF 1565

14. Mancini DM, Henson D, LaManca J, Levine S. Respiratory musclefunction and dyspnea in patients with chronic congestive heartfailure. Circulation 1992;86:909e18.

15. Nanas S, Nanas J, Kassiotis C, et al. Respiratory musclesperformance is related to oxygen kinetics during maximalexercise and early recovery in patients with congestive heartfailure. Circulation 1999;100:503e8.

16. Papazachou O, Anastasiou-Nana M, Sakellariou D, et al.Pulmonary function at peak exercise in patients with chronicheart failure. Int J Cardiol 2007;118:28e35.

17. Torchio R, Gulotta C, Greco-Lucchina P, et al. Closing capacityand gas exchange in chronic heart failure. Chest 2006;129:1330e6.

18. ExTraMATCH Collaborative. Exercise training meta-analysis inpatients with chronic heart failure (ExTraMATCH). BMJ 2004;328:189e90.

19. Belardinelli R, Georgiou D, Cianci G, Purcaro A. Randomized,controlled trial of long-term moderate exercise training inchronic heart failure. Effects on functional capacity, quality oflife and clinical outcome. Circulation 1999;99:1173e82.

20. Hambrecht R, Fiehn E, Weigl C, et al. Regular physical exercisecorrects endothelial dysfunction and improves exercisecapacity in patients with chronic heart failure. Circulation1998;98(24):2709e15.

21. Gerovasili V, Drakos S, Kravari M, et al. Physical exerciseimproves the peripheral microcirculation of patients withchronic heart failure. J CardiopulmRehabil Prev 2009 NoveDec;29(6):385e91.

22. Meyer K, Samek L, Schwaibold M, et al. Interval training inpatients with severe chronic heart failure: analysis andrecommendations for exercise procedures. Med Sci SportsExerc 1997 Mar;29(3):306e12.

23. Wisløff U, Støylen A, Loennechen JP, et al. Superior cardio-vascular effect of aerobic interval training versus moderatecontinuous training in heart failure patients: a randomizedstudy. Circulation 2007;115(24):3086e94.

24. Maiorana A, O’Driscoll G, Cheetham C, et al. Combined aerobicand resistance exercise training improves functional capacityand strength in CHF. J Appl Physiol 2000;88(5):1565e70.

25. Beckers P, Denollet J, Possemiers N, Wuyts F, Vrints C,Conraads V. Combined endurance-resistance training vs.endurance training in patients with chronic heart failure:a prospective randomized study. Eur Heart J 2008;29:1858e66.

26. American Thoracic Society/American College of Chest Physi-cians. ATS/ACCP. Statement on cardiopulmonary exercisetesting. Am J Respir Crit Care Med 2003;167:211e77.

27. Hansen JE, Sue DY, Wasserman K. Predicted values for clinicalexercise testing. Am Rev Respir Dis 1984;129(Suppl.):S49eS55.

28. Beaver WL, Wasserman K, Whipp BJ. A new method fordetecting anaerobic threshold by gas exchange. J Appl Physiol1986;60:2020e7.

29. Nanas S, Terrovitis J, Charitos C, et al. Ventilatory response toexercise and kinetics of oxygen recovery are similar in cardiactransplant recipients and patients with mild chronic heartfailure. J Heart Lung Transplant 2004;23:1154e9.

30. American Thoracic Society. Standardization of spirometry 1995update. Am Rev Respir Dis 1995;152:1107e36.

31. Black LF, Hyatt RE. Maximal respiratory pressures: normalvalues and relationship to age and sex. Am Rev Respir Dis 1969;99:696e702.

32. Whitelaw WA, Derenne JP, Milic-Emili J. Occlusion pressure asa measure of respiratory center output in conscious man.Respir Physiol 1975 Mar;23(2):181e99.

33. Whitelaw WA, Derenne JP. Airway occlusion pressure. J ApplPhysiol; 1993:1475e83.

34. Burton MD, Kazemi H. Neurotransmitters in central respiratorycontrol. Respir Physiol 2000;122(2e3):111e21.

35. Ambrosino N, Opasich C, Crotti P. Breathing pattern, ventila-tory drive and respiratory muscle strength in patients withchronic heart failure. Eur Respir J 1994;7:17e22.

36. Vibarel N, Hayot M, Ledermann B, Messner Pellen P,Ramonatxo M, Prefaut C. Effect of aerobic exercise training oninspiratory muscle performance and dyspnoea in patients withchronic heart failure. Eur J Heart Fail 2002;4:745e51.

37. Roditis P, Dimopoulos S, Sakellariou D, et al. The effects ofexercise training on the kinetics of oxygen uptake in patientswith chronic heart failure. Eur J Cardiovasc Prev Rehabil 2007Apr;14(2):304e11.

38. Dimopoulos S, Anastasiou-Nana M, Sakellariou D, et al. Effectsof exercise rehabilitation program on heart rate recovery inpatients with chronic heart failure. Eur J Cardiovasc PrevRehabil 2006 Feb;13(1):67e73.

39. Ponikowski P, Piepoli M, Chua TP, et al. The impact of cachexiaon cardiorespiratory reflex control in chronic heart failure.Eur Heart J 1999;20:1667e75.

40. Ponikowski P, Chua TP, Francis D, et al. Muscle ergoreceptoroveractivity reflects deterioration in clinical status andcardiorespiratory reflex control in chronic heart failure.Circulation 2001;104:2324e30.

41. Piepoli M, Clark A, Volterrani M, Adamopoulos S, Sleight P,Coats AJ. Contribution of muscle afferents to the hemody-namic, autonomic, and ventilatory responses to exercise inpatients with chronic heart failure: effects of physical training.Circulation 1996;93:940e52.

42. Roveda F, Middlekauff H, Rondon M, et al. The effects ofexercise training on sympathetic neural activation in advancedheart failure a randomized controlled trial. J Am Coll Cardiol2003;42(5).

43. Ueno L, Drager L, Rodrigues A, et al. Effects of exercisetraining in patients with chronic heart failure and sleep apnea.Sleep 2009;32(5).

44. Nanas S, Nanas J, Sakellariou D, et al. VE/VCO2 slope isassociated with abnormal resting haemodynamics and isa predictor of long-term survival in chronic heart failure. Eur JHeart Fail 2006;8:420e7.

45. Marin J, Montes de Oca M, Rassulo J, Celli B. Ventilatory driveat rest and perception of exertional dyspnea in severe COPD.Chest 1999;115:1293e300.

46. Gigliotti F, Coli C, Bianchi R, et al. Exercise training improvesexertional dyspnea in patients with COPD. Evidence of the roleof mechanical factors. Chest 2003;123:1794e802.

47. Vogiatzis I, Terzis G, Nanas S, et al. Skeletal muscle adapta-tions to interval training in patients with advanced COPD.Chest 2005;128:3838e45.