CHF SEPTEMBER/OCTOBER 2000EXERCISE AND ABNORMAL VENTILATION IN CHF 243

Patients with chronic heart failure frequently reportshortness of breath during daily activities as their primary symptom. In recent years, many efforts havebeen made by researchers to explain the mechanisms that underlie the characteristic heightened ventilatory response to activity in patients with chronic heart failure. The degree to which the ventilatory response toexercise is heightened parallels the severity of the disease,and measuring the ventilatory gas exchange response toexercise can help quantify the patient’s response to therapy. Prior to the 1990s, patients with chronic heartfailure were generally discouraged from participating inprograms of exercise training. However, in the lastdecade, studies have demonstrated that exercise trainingis quite safe for these patients, and a multitude of bene-fits have been reported. Among the benefits of trainingare improvements in the abnormal ventilatory responseto exercise. Although many mechanisms could potentiallyexplain this response, it appears most likely that this improvement after training is due to a reduction in lactate accumulation and an attenuation of the heightenedmuscle receptor reflex response that occurs in chronic heartfailure. This article reviews the mechanisms of dyspnea inchronic heart failure, along with recent studies assessingthe effects of training on abnormal ventilatory responses toexercise in these patients. (CHF. 2000;6:243–249,255)©2000 by CHF, Inc.

Jonathan Myers, PhDFrom the Cardiology Division, Palo Alto Veterans Administration Health Care System, and Stanford University, Palo Alto, CA

Address for correspondence reprint/requests:Jonathan Myers, PhD, Palo Alto VA Health Care System, Cardiology Division, 111C, 3801 Miranda Avenue, Palo Alto, CA 94304Manuscript received May 5, 2000; accepted May 23, 2000

A hallmark feature of patients with chronic heartfailure (CHF) is exercise intolerance characterizedby fatigue, shortness of breath, or both. The degreeof exercise intolerance is important to characterizein patients with CHF, since it has implications formorbidity, disability, and prognosis, and is oftenthe reason a patient seeks medical attention. As, ex-ercise intolerance is an important clinical feature inthese patients, therapeutic interventions are largelyaimed at improving this symptom. Once thought tobe contraindicated in patients with CHF, exercisetraining is now commonly employed in these pa-tients, and a wide range of benefits, including im-proved exercise capacity, has been demonstratedover the last decade.1,2

On the surface, the explanation for exercise in-tolerance in CHF appears simple. The heart’s im-paired pumping capacity delivers inadequateamounts of oxygen to the working muscle, leadingto inadequate energy supply, lactate accumulation,shortness of breath, and fatigue. However, the exer-cise response in CHF is far more complex thancommonly thought just 10 years ago. An abundanceof evidence has accumulated to suggest that cardiacpump function is poorly related to exercise capacityand symptom generation in CHF. Rather, it is gen-erally thought that factors in the peripheral muscu-lature, and not central hemodynamics, largelydetermine both shortness of breath and fatigue.These peripheral factors include abnormalities inendothelial function, vasodilatory capacity, distrib-ution of cardiac output, heightened chemo- and er-goreceptor sensitivity, skeletal muscle histology,and oxidative enzyme activity.3–5

A great deal of effort has also recently been direct-ed toward abnormalities of the ventilatory response toexercise in CHF. Minute ventilation (VE) is markedlyhigher among patients with CHF than in normal sub-jects at any given level of exercise.3–6 This has beenexpressed as a higher VE at matched exercise work-loads, a higher VE at any given oxygen uptake, and ahigher VE for any given level of CO2 production(VE/VCO2 slope).7–9 The latter measurement is fre-quently used as an index of ventilatory drive, and has

Effects of Exercise Training on AbnormalVentilatory Responses to Exercise in Patients with Chronic Heart Failure

CHF SEPTEMBER/OCTOBER 2000244 EXERCISE AND ABNORMAL VENTILATION IN CHF

been shown to reflect the severity of CHF7 and prog-nosis.9 Any attempt to explain the exercise limitationin CHF must also include an explanation for the in-creased ventilatory response to exercise. Mechanismsthat have been put forth to explain this response in-clude increased intrapulmonary pressures, elevatedphysiologic dead space, early metabolic acidosis, al-tered ventilatory control, abnormal breathing pat-terns, including rapid, shallow respiration, andheightened sensitivity of receptors in the muscle.3–7,10

Despite a great deal of research directed toward thisarea, a complete explanation of the mechanisms(s)underlying these abnormal ventilatory responses toexercise in patients with CHF is lacking.

Mechanism of Abnormal Exercise VentilationIn attempting to explain dyspnea in CHF, it is inter-esting to note that pulmonary function per se doesnot limit exercise in these patients. A large breathingreserve at peak exercise has been repeatedly ob-served; this is the ratio of maximal exercise VE tomaximal voluntary ventilation at rest, and is alsotermed the VE/MVV ratio or dyspnea index. In otherwords, as occurs in a normal individual, 20%–40% ofthe CHF patient’s ventilatory capacity remains, evenat maximal levels of exertion. In addition, a decreasein arterial oxygen saturation, another marker ofsymptomatic pulmonary disease, rarely occurs inCHF. Thus, although some patients with CHF clearlyhave abnormal pulmonary hemodynamics, pul-monary function generally does not limit exercise.

Pulmonary Wedge Pressure. Elevated pulmonarywedge pressure (PCW) is common in patients withCHF and is thought to decrease lung compliance11–13

and to stimulate pulmonary juxtacapillary receptors,which stimulate ventilation.12,14,15 Traditionally, eleva-tions in PCW have therefore been thought to be themain cause of hyperventilation and consequent exer-tional dyspnea. Several investigators have attemptedto quantify the role of PCW as a cause of dyspnea dur-ing exercise in patients with CHF. Interestingly, manyrecent studies have found exercise capacity to be relat-ed to PCW at rest8,16,17 but not during exercise.16,17

Moreover, acute pharmacologically-induced reduc-tions in pulmonary wedge pressure do not significant-ly reduce the ventilatory response to exercise.16 Thesefindings suggest that hyperventilation during exercisein patients with CHF is not, in fact, related to increas-es in PCW. Although pulmonary hypertension is fre-quently seen in patients with heart failure, exercisedoes not appear to be limited by ventilatory manifesta-tions of elevated pulmonary pressures.

Estimated Ventilatory Dead Space to TidalVolume Ratio. The ventilatory requirement for agiven level of work also depends on the ineffectivefraction of tidal volume (VT), that is, the ventilatorydead space (Vd). This ratio, Vd/VT, is frequently ele-vated in patients with CHF, suggesting a ventila-tion-perfusion mismatch, and is one mechanismthat underlies the hyperventilatory response to ex-ercise in these patients. Higher than normal Vd/VTvalues during exercise have been attributed primar-ily to increases in physiologic dead space, which arelargely due to a reduction in pulmonary blood flowvia reduced cardiac output.6–8,18,19 The elevatedVd/VT makes breathing inefficient because greaterventilation is required to maintain arterial oxygen.

Early Lactate Accumulation. A striking feature ofthe exercise response in the patient with CHF isearly lactate accumulation. Numerous investigatorshave associated early lactate accumulation in theblood with metabolic acidosis, hyperventilation, andexercise intolerance in patients with CHF.3,4,8,18,20

Lactate that accumulates in the blood during exer-cise must be buffered to maintain physiologic pH,and the bicarbonate buffering process yields a sec-ondary source of CO2, which stimulates ventilation.

Chemo- and Ergoreceptor Sensitivity. A recentparadigm suggests the presence of a specific ventila-tory signal arising from the exercising muscle, whichis abnormally enhanced in CHF.5,10,21,22 These sig-nals appear to contribute to the abnormal hemody-namic, autonomic, and ventilatory responses toexercise that characterize CHF. Afferent fibers pres-ent in the skeletal muscle (ergoreceptors) are sensi-tive to metabolic changes related to muscular work.These receptors appear to mediate circulatory adap-tations occurring in the early stages of exercise, arestimulated by metabolic acidosis, and are partiallyresponsible for sympathetic vasoconstriction.22 Chuaand colleagues23 have also recently demonstratedthat hypoxic chemosensitivity is increased in CHF,and that this heightened chemosensitivity is correlat-ed with the VE/VCO2 slope. The results of these en-hanced ergoreflex and chemoreceptor responses arehyperventilation and heightened sympathetic out-flow, which causes an increase in peripheral resis-tance and thus a decrease in muscle perfusion.

Effect of Exercise Training onAbnormal Ventilation in CHFWhile exercise training programs for patients withCHF were discouraged prior to the 1990s due to con-cerns about safety and benefits, such programs have

CHF SEPTEMBER/OCTOBER 2000EXERCISE AND ABNORMAL VENTILATION IN CHF 245

now become an accepted part of the therapeutic regi-men for CHF. From the evidence gathered in themultitude of studies published during the last decade,the benefits of exercise training in patients with CHFappear to be due primarily to adaptations in the pe-ripheral musculature rather than the heart itself.2,24,25

These studies have focused on exercise tolerance,skeletal muscle oxidative capacity, histologic changesin the muscle, endothelial function, remodeling ofthe myocardium after an infarction, and quality oflife.2 Training also appears to favorably modify theventilatory response to exercise in CHF, althoughsuch data are comparatively sparse and the mecha-nism underlying this response has not been fullycharacterized. A partial normalization of exerciseventilation in CHF represents an important therapeu-tic achievement, since dyspnea is such an importantcause of the exercise limitation and morbidity. Be-cause an appreciation for these abnormal ventilatoryresponses to exercise and the benefits of training inCHF are relatively recent, only five studies havespecifically addressed the influence of training on ab-normal ventilatory responses to exercise in these pa-tients.26–30 All of these studies have demonstrated thattraining reduces VE at a matched8,22,26–28 submaximalworkload, but only two have attempted to explore themechanisms involved.

In a controlled trial of exercise training in patientswith CHF, Coats and colleagues26 assessed the slopeof the relation between minute ventilation and therate of carbon dioxide production (VE/VCO2)throughout exercise, an index which has been shownto parallel the severity of CHF.6,7,9 The VE/VCO2slope was reduced after 8 weeks of training(0.038±0.04 to 0.034±0.03, p<0.05). In addition, VEwas reduced at matched submaximal workloads (25and 50 watts) after training (Fig. 1). These changeswere associated with a reduction in the sensation ofbreathlessness and the ease of performing daily activ-ities, according to Likert symptom scores. Similarly,Sullivan et al27 trained 12 CHF patients for 4–6months and reported a reduction in the ventilatoryresponse at submaximal levels throughout a progres-sive exercise test. During a symptom-limited, sub-maximal, constant work rate protocol, ventilation andCO2 production were reduced at matched work rates(representing a mean of 79±11% of maximal capaci-ty), even though the oxygen cost of the work (VO2)was unchanged. Endurance time on this submaximalprotocol increased by nearly 50%.

Davey and colleagues28 assessed the effects of amodest home-based training program (20-minutesessions 5 days/week for 8 weeks) on the ventilatoryresponse to exercise in 22 patients with New YorkHeart Association Class I or II heart failure. After

the training program, small but significant increasesin work capacity (10% and 9% increases in peakwatts and peak VO2, respectively) were observed.However, submaximal endurance improved by aconsiderable 33%. Both VCO2 and VE were reducedat matched submaximal workloads. As in the previ-ous studies, the VE/VCO2 slope decreased aftertraining. Exercise tolerance in these patients was re-lated to various measures of ventilatory efficiency.

Piepoli et al22 assessed the potential effects of ex-ercise training on muscle afferents (peripheral mus-cle, or ergoreflex, contributions to ventilation) inpatients with CHF. The ergoreflex contribution wasquantified by a post-handgrip regional circulatoryocclusion technique. After 6 weeks of a forearmtraining protocol, the ergoreflex contribution to ex-ercise VE was reduced by 58% (p<0.05 vs. controls).These unique findings not only appear to confirmthe presence of a muscle “ergoreflex” stimulus toventilation, but they also suggest that alterations inthese muscle receptors play an important role in ex-ercise intolerance in CHF. Previous studies havedemonstrated that isolated training of the forearmresults in greater submaximal endurance and sub-stantially corrects the impaired oxidative capacity ofthe skeletal muscle in CHF.29,30 The study by Piepoliet al22 suggests that in addition, exercise trainingmay reduce the heightened ergoreflex activity thatcontributes to excessive dyspnea and fatigue in CHF.

Improvement in abnormal ventilation after train-ing might also involve hemodynamic changes (re-duced pulmonary pressures or improved ventilation-perfusion matching), metabolic changes reflected by areduction in lactate accumulation, a change in venti-latory control, or a change in the ventilatory patternthat makes breathing more efficient. Accurate evalua-tion of these mechanisms would require the simulta-

Figure 1. Minute ventilation at rest and at 25 watts, 50watts, and peak exercise before (open boxes) and after(shaded boxes) exercise training in patients with CHF. *p<0.05 Before vs. after training. From Coats et al.26

Min

ute

V

enti

lati

on

(I/

min

)

Rest 25W 50W Peak

Exercise Load

CHF SEPTEMBER/OCTOBER 2000246 EXERCISE AND ABNORMAL VENTILATION IN CHF

neous measurement of ventilatory gas exchange, in-vasive intrapulmonary pressures, cardiac output, andarterial blood gases. We recently performed a studyin which theses variables were acquired simultaneous-ly during upright exercise before and after a 2-monthhigh intensity residential training program in pa-tients with reduced ventricular function after a my-ocardial infarction.8

In the residential training program, the patientsperformed two 1-hour walking sessions daily, andfour 45-minute sessions of individually prescribedcycling weekly. This resulted in 29% and 39% in-creases in peak VO2 and VO2 at the lactate thresh-old, respectively, whereas no differences wereobserved among control subjects. Although no dif-ferences were observed within or between groups inresting ejection fraction, maximal cardiac outputincreased by 1.7 l/minute in the trained group,which contrasts with the results of several studies ofCHF showing either no change after training25,31,32

or a more modest increase in cardiac output, on theorder of 1.0 l/minute.26,27 Figure 2 illustrates the ef-fects of training on oxygen uptake, heart rate, per-ceived exertion, and lactate responses throughoutexercise. Figure 3 presents the effects of training onVE, VE/VO2, respiratory rate, and tidal volume. Figure 4 shows the effect of the training program onthe slope of the relationship between VE and VCO2.Figure 5 illustrates the relationships between restingand maximal exercise pulmonary wedge pressuresand peak oxygen uptake. Figure 6 presents the rela-tionships between maximal exercise VE/VCO2 and

maximal values for cardiac output, PCW, Vd/VT, andmean pulmonary artery pressure prior to beginningthe training program. These results are discussedbelow, in the context of the mechanisms underlyingexertional dyspnea in CHF.

Baseline Ventilatory Response to Exercise. Atbaseline, arterial PCO2 remained normal through-out exercise (falling from 34.0±3.8 to 30.9±4.2 mmHg), and VE and VCO2 were tightly coupled, sug-gesting that neural and chemoreceptor control ofVE were normal. VE/VCO2 was characteristicallyshifted upward (Fig. 4), which reflects a heightenedVd/VT.7 The heightened ventilatory response to ex-ercise is also evidenced by an elevated VE/VO2, par-ticularly on the initial test (Fig. 3).

Ventilation-Perfusion Mismatching. Ventilation-perfusion mismatching has been identified as a majorfactor contributing to an increase in physiologic deadspace (Vd/Vt) in CHF.3,4,7,18,19 Underlying this re-sponse is reduced pulmonary perfusion caused by anattenuated cardiac response to exercise.18 The inverseassociation between the ventilatory response to exer-cise (VE/VCO2) and maximal cardiac output in ourstudy (Fig. 6) suggests that decreased pulmonary per-fusion did in fact underlie an increased physiologicdead space by increasing ventilation-perfusion mis-matching. A better matching of ventilation and perfu-sion, and thus a reduction in Vd/VT, could, in theory,be accomplished by either an improvement in cardiacoutput or a reduction in physiologic dead space. We

Figure 2. Influence of exercise training on oxygen uptake,perceived exertion, heart rate, and lactate at matchedramp work rates during exercise. Open squares representbaseline values and closed circles denote post-training. A main effect (p<0.01) was observed for all parameters except oxygen uptake.

Figure 3. Influence of training on ventilatory responses toexercise at matched ramp work rates during exercise. Opensquares indicate baseline values and closed squares indi-cate post-training. A significant reduction (p<0.01) wasobserved for VE/VO2 (the relationship between minute ven-tilation and the oxygen cost).

CHF SEPTEMBER/OCTOBER 2000EXERCISE AND ABNORMAL VENTILATION IN CHF 247

hypothesized that an increase in maximal cardiacoutput with training observed in some previous stud-ies26,27 would parallel a reduction in physiologic deadspace. Although a substantial increase in maximalcardiac output occurred in the trained group, we didnot observe a significant reduction in the ratio ofdead space to tidal volume, making improved ventila-tion-perfusion matching an unlikely explanation forthe improved ventilatory response to exercise.

Efficiency of Ventilation. The significant reduc-tions in VE/VO2 (Fig. 3) and the reduction in theslope of the relationship between VE and VCO2(Fig. 4) confirm a beneficial effect of exercise train-ing on the efficiency of ventilation during exercise.These measures are well recognized indices of ven-tilatory efficiency and have been used as markers ofthe severity of CHF.6,7,9

Breathing Pattern. Rapid, shallow respiration ischaracteristic of the response to exercise in patientswith CHF,6,33,18 and it has been suggested that thisbreathing pattern reflects an effort on the part ofthe patient to reduce the work of breathing.12 Thisresponse has been expressed as the ratio of tidalvolume to respiratory rate.6,33 Initially, our patientsdemonstrated a ratio of tidal volume to breathingrate which was similar to that observed in previousstudies among patients with CHF,6,33 and traininghad only a slight effect on this relationship. Theoverall reduction in the ventilatory requirement forwork (VE/VO2) that we observed after training wasattributable mainly to a lower respiratory rate,whereas tidal volume remained relatively un-changed (Fig. 3). Training had only minimal effectson the breathing pattern.

Metabolic Acidosis. Previous studies have demon-strated that exercise training causes a delay in the lac-tate or ventilatory threshold in normal subjects34,35

and in patients who have sustained a myocardial in-farction.36,37 A delay in the threshold of hyperventila-tion would be particularly important in patients withCHF, because dyspnea is such an important limitationto their daily activities.38 Whether training causes a re-duction in lactate production or an increase in the rateof lactate clearance has been debated, but clearlytraining delays lactate accumulation during exercise.39

Our data confirm the work of others6,40 with respect toboth a delay in the lactate threshold (39% increase inVO2 at the lactate threshold) and an overall reductionin the slope of the relationship between blood lactateand work rate throughout exercise (Fig. 2). Becauseventilation is an important component of perceived ef-fort,41 it follows that a marked reduction in perceivedexertion was observed throughout exercise after train-ing (Fig. 2).

Pulmonary Pressures. Relatively weak relation-ships were observed between resting and exercisePCW, peak VO2, and the ventilatory requirement forexercise (Fig. 5), confirming previous assertions thatstimulation of pulmonary receptors does not underlieabnormal ventilatory responses to exercise. The influ-ence of exercise training on pulmonary vascular pres-sures has been disputed. Some studies havedemonstrated increases in left ventricular filling pres-sures during exercise after training in this popula-tion,31,32 whereas others have shown decreases32,42 or

Figure 4. The slope of the relationship between minuteventilation and VCO2 before (open circles) and after(closed circles) training in the exercise group (0.33 beforevs. 0.27 after, p<0.01).

Figure 5. The relationships between resting and maximal ex-ercise pulmonary wedge pressures and peak oxygen uptake among patients in both groups before randomization.

CHF SEPTEMBER/OCTOBER 2000248 EXERCISE AND ABNORMAL VENTILATION IN CHF

no change.27,31,43 We did not observe any changes inPCW, pulmonary artery pressure, or pulmonary vas-cular resistance in either the trained or controlgroups, and the relationships between pulmonarypressures, exercise capacity, and ventilation were sim-ilar before and after the study period in both groups.

Mechanism of Reduced Ventilatory Response to ExerciseOne of the most important factors governing exer-cise tolerance in CHF is the level of ventilation thatcan be sustained. Exercise training has clearly beenshown to be one intervention that reduces the venti-latory response to exercise and contributes to an in-crease in exercise tolerance. Because few studieshave taken a mechanistic approach to this question, acomplete explanation for the reduction in the venti-latory response to exercise requires further study.The study by Piepoli and coworkers22 focused onmuscle afferent receptor effects, e.g., those in the pe-ripheral musculature, whereas our study8 addressedcentral adaptations, e.g., pulmonary pressurechanges, ventilation-perfusion matching, breathingpatterns, and ventilatory control.

According to the alveolar gas equation, ventila-tion is governed by three factors: 1) the PaCO2 setpoint, i.e., the level at which PaCO2 is regulated(ventilatory control); 2) the ratio of ventilation toperfusion (Vd/VT); and 3) CO2 production, orVCO2. In our study, ventilation and VCO2 weretightly coupled, and PaCO2 remained stable during

exercise, as others have observed in patients withCHF.18,44 These relationships did not change withtraining, suggesting that altered ventilatory controldoes not explain the reduced ventilatory responseto exercise in these patients. However, a growingbody of evidence suggests an important role of pe-ripheral muscle receptors in stimulating ventila-tion.5,10,21,22 This, along with the findings of Piepoliet al22 demonstrating diminution of these signalsafter localized forearm training, suggest an alterna-tive, peripheral contribution to ventilation, whichadapts to training. Whether this adaptation occursin response to more standardized aerobic trainingprograms in these patients is unknown.

We hypothesized that if exercise training wouldincrease maximal cardiac output, as others had re-cently reported in CHF,26,27 an improvement inVd/VT would occur via enhanced pulmonary bloodflow. Although we observed a trend for a reductionin maximal Vd/VT, the insignificant change in thisresponse suggests that this mechanism does notcontribute in a major way to the reduction in theventilatory response to exercise.

The third factor in the alveolar gas equation,CO2 production, might be reduced after trainingthrough a reduction in lactate production, an in-crease in lactate removal, or both. The reduction inblood lactate levels after training has been consid-erable.2,8,24,40 This is likely related to an improve-ment in skeletal muscle metabolism; patients withCHF are known to exhibit abnormalities in mito-chondrial volume and oxidative enzymes, and train-ing has been demonstrated to partially reversethese abnormalities.2,25,45 The mediation of exerciseventilation with training by a reduction in bloodlactate has also been demonstrated in normals46

and patients with COPD.47 Lastly, a reduced tidalvolume could also increase ventilation by raisingthe Vd/Vt; anatomic dead space, although consid-ered to be fixed in absolute terms, will be increasedif a given tidal volume is ventilated more frequently(i.e., a high ratio of tidal volume to respiratoryrate). In our study, training caused only a slight de-crease in the ratio of maximal tidal volume to respi-ratory rate. Although the reduced ventilationthroughout exercise was mediated primarily by areduction in respiratory rate (Fig. 3), the overallbreathing pattern does not appear to be altered sig-nificantly by training.

SummaryDyspnea on exertion remains the hallmark symptomof patients with CHF, and heightened ventilation isrelated to the severity of heart failure. Most treat-

Figure 6. The relationships between maximal exerciseVE/VCO2 and maximal values for cardiac output, pul-monary capillary wedge pressure (PCW), Vd/Vt, and meanpulmonary artery pressure (PA) in both groups at baseline.VE/VCO2=relationship between minute ventilation and therate of carbon dioxide production; Vd/VT=ratio of ventila-tory dead space to tidal volume.

CHF SEPTEMBER/OCTOBER 2000EXERCISE AND ABNORMAL VENTILATION IN CHF 249

ment strategies are directed toward improving fa-tigue or mortality, but in recent years increasing at-tention has focused on ameliorating the heightenedventilatory response to exercise. Exercise trainingappears to partially normalize the ventilatory re-sponse to exercise primarily via two mechanisms: 1)reduced lactate accumulation; and 2) a reduction inthe exaggerated ergoreflex response that character-izes CHF. Recent studies have greatly enhanced ourunderstanding of the cardiopulmonary response toexercise in CHF, and a growing body of evidencesuggests that training can improve abnormalities ofventilation in response to exercise in these patients.

REFERENCES1 Agency for Health Care Policy and Research. Clinical

Practice Guidelines for Cardiac Rehabilitation, 1995.2 Piepoli MF, Flaiter M, Coats AJS. Overview of studies of

exercise training in chronic heart failure: The need for aprospective randomized multicenter European trial . Eur Heart J. 1998;19:830–841.

3 Sullivan MJ, Hawthorne, M. Exercise intolerance in patientswith chronic heart failure. Prog Cardiovasc Dis. 1995;38:1–22.

4 Myers J, Froelicher VF. Hemodynamic determinants of exercise capacity in chronic heart failure. Ann Intern Med.1991;115:377–386.

5 Harrington D, Coats AJS. Mechanisms of exercise intolerance incongestive heart failure. Curr Opin Cardiol. 1997;12:224–232.

6 Myers I, Salleh A, Buchanan N, et al. Ventilatory mechanismsof exercise intolerance in chronic heart failure. Am Heart J.1992;124:7–10.

7 Wada O, Asanoi H, Miyagi K, et al. Importance of abnormallung perfusion in excessive exercise ventilation in chronicheart failure. Am Heart J. 1992;125:790–798.

8 Myers J, Dziekan G, Goebbels U, et al. Influence of high in-tensity training on the ventilatory response to exercise in pa-tients with reduced ventricular function. Med Sci Sports Exerc.1999;31:929–937.

9 Chua TP, Ponikowski P, Harrington D, et al. Clinical correlatesand prognostic significance of the ventilatory response to exercisein chronic heart failure. J Am Coll Cardiol. 1997;29:1585–1590.

10 Clark AL, Piepoli M, Coats AJS. Skeletal muscle and the control of ventilation on exercise: Evidence for metabolic receptors. Eur J Clin Invest. 1995;25:299–305.

11 Gazetopolous N, Davies H, Oliver C, et al. Ventilation andhemodynamics in heart disease. Br Heart J. 1966;28:11–16.

12 Ingram RH, McFadden ER. Respiratory changes during exercise in patients with pulmonary venous hypertension.Prog Cardiovasc Dis. 1976;19:109–115.

13 Parker GW, Gorlin R. Immediate post-exercise vital capacity:A measure of increased pulmonary capillary pressure. Am J Med Sci. 1969;257:365–370.

14 Reed JW, Ablett M, Cotes JE. Ventilatory response to exer-cise and to carbon dioxide in mitral stenosis before and after valvulotomy: Causes of tachypnea. Clin Sci Molec Med.1978;54:9–16.

15 Paintal AS. Mechanism of stimulation of type J pulmonary receptors. J Physiol (Lond). 1969;55:439–445.

16 Fink L, Wilson J, Ferraro N. Exercise ventilation and pul-monary artery wedge pressure in chronic stable congestiveheart failure. Am J Cardiol. 1986;5:249–253.

17 Szlachcic J , Massie BN, Kramer BL, et al. Correlates andprognostic implication of exercise capacity in chronic conges-tive heart failure. Am J Cardiol. 1985;55:1037–1042.

18 Sullivan MJ, Higginbotham MB, Cobb FR. Increased exerciseventilation in patients with chronic heart failure: Intact venti-latory control despite hemodynamic and pulmonary abnor-malities. Circulation. 1988;77:552–559.

19 Buller NP, Poole-Wilson PA. Mechanism of the increasedventilatory response to exercise in patients with chronic heartfailure. Br Heart J. 1990;63:281–283.

20 Sullivan MJ, Cobb FR. The anaerobic threshold in chronicheart failure: Relation to blood lactate, ventilatory basis, re-producibility, and response to exercise training. Circulation.1990;81(suppl II):II47–II58.

21 Clark AL, Poole-Wilson PA. Exercise limitation in chronicheart failure: Central role of the periphery. J Am Coll Cardiol.1996;28:1092–1102.

22 Piepoli M, Clark AL, Volterrani M. Contribution of muscleafferents to the hemodynamic, autonomic, and ventilatory responses to exercise in patients with chronic heart failure.Effects of physical training. Circulation. 1996;93:940–952.

23 Chua TP, Clark AL, Amadi A, et al. Relation betweenchemosensitivity and the ventilatory response to exercise inchronic heart failure. J Am Coll Cardiol. 1996;27:650–657.

24 Dubach P, Myers J, Dziekan G, et al. Effect of exercise train-ing on myocardial remodeling in patients with reduced ven-tricular function following myocardial infarction: Applicationof MRI. Circulation. 1997;95:2060–2067.

25 Hambrecht R, Niebauer I, Fiehn E, et al. Physical training inpatients with stable chronic heart failure: Effects on cardiorespi-ratory fitness and ultrastructural abnormalities of leg muscles.J Am Coll Cardiol. 1995;25:1239–1249.

26 Coats AJS, Adamopoulos S, Radaelli A, et al. Controlled trialof physical training in chronic heart failure. Exercise perfor-mance, hemodynamics, ventilation and autonomic function.Circulation. 1992;85:2119–2131.

27 Sullivan MI, Higginbotham MB, Cobb FR. Exercise trainingin patients with severe left ventricular dysfunction. Hemody-namic and metabolic effects. Circulation. 1988;78:506–515.

28 Davey P, Meyer T, Coats A, et al. Ventilation in chronic heartfailure: Effects of physical training. Br Heart J. 1992;68:473–477.

29 Stratton JR, Dunn JF, Adamopoulos S, et al. Training partiallyreverses skeletal muscle metabolic abnormalities during exer-cise in heart failure. J Appl Physiol. 1994;76:1575–1582.

30 Minotti JR, Johnson EC, Hudson TL, et al. Skeletal muscleresponse to exercise training in congestive heart failure. J Clin Invest. 1990;86:751–758.

31 Jette M, Helter R, Landry F, et al. Randomized 4-week exer-cise program in patients with impaired left ventricular func-tion. Circulation. 1991;84:1561–1567.

32 Scalvini S, Marangoni S, Volterrani M, et al. Physical rehabili-tation in coronary patients who have suffered from episodesof cardiac failure. Cardiology. 1992;80:417–423.

33 Clark AL, Chua TP, Coats AJS. Anatomical dead space, venti-latory pattern, and exercise capacity in chronic heart failure.Br Heart J. 1995;74:377–380.

34 Davis JA, Frank MH, Whipp BI, et al. Anaerobic threshold al-terations caused by endurance training in middle-aged men.J Appl Physiol. 1979;46:1039–1045.

35 Ready AE, Quinney HA. Alterations in anaerobic thresholdas the result of endurance training and detraining. Med SciSports Exerc. 1982;14:292–297.

36 Hambrecht R, Niebauer J, Marburger C, et al. Various intensitiesof leisure time physical activity in patients with coronary artery dis-ease: Effects on cardiorespiratory fitness and progression of coro-nary atherosclerotic lesions. J Am Coll Cardiol. 1993;22:468–477.

37 Sullivan M, Ahnve S, Froelicher VF, et al. The influence ofexercise training on the ventilatory threshold of patients withcoronary heart disease. Am Heart J. 1985;109:458–463.

38 Oka RK, Stotts NA, Dae MW, et al. Daily physical activity lev-els in congestive heart failure. Am J Cardiol. 1993;71:921–925.

39 Myers J, Ashley E. Dangerous curves: A perspective on exercise,lactate, and the anaerobic threshold. Chest. 1997;111:787–795.

40 Sullivan MJ, Higginbotham MB, Cobb FR. Exercise trainingin patients with chronic heart failure delays ventilatory anaer-obic threshold and improves submaximal exercise perfor-mance. Circulation. 1989;79:324–329.

41 Robertson RI, Noble BJ. Perception of physical exertion:Methods, mediators, and applications. Exerc Sport Sci Rev.1997;25:407–452.

CHF SEPTEMBER/OCTOBER 2000250 EXERCISE AND ABNORMAL VENTILATION IN CHF

42 Tavazzi L, Ignone G. Short-term hemodynamic evolutionand late follow-up of post-infarct patients with left ventriculardysfunction undergoing a physical training program. EurHeart J. 1991;12:657–665.

43 Dubach P, Myers J, Dziekan C, et al. Effect of high intensityexercise training on central hemodynamic responses to exer-cise in men with reduced left ventricular function. J Am CollCardiol. 1997;29:1591–1598.

44 Weber KT, Kinasewitz CT, Janicki JS, et al. Oxygen utiliza-tion and ventilation during exercise in patients with chroniccardiac failure. Circulation. 1982;65:1213–1223.

45 Hambrecht R, Fiehn E, Yu I, et al. Effects of endurancetraining on mitochondrial ultrastructure and fiber typedistribution in skeletal muscle of patients with stablechronic heart failure. J Am Coll Cardiol. 1997;29:1067–1073.

46 Casaburi R, Storer TW, Wasserman K. Mediation of reducedventilatory response to exercise after endurance training. J Appl Physiol. 1987 63:1533–1538.

47 Casaburi R. Mechanisms of the reduced ventilatory require-ment as a result of exercise training. Eur Respir Rev.1995;25:42–46.