doi: 10.1152/japplphysiol.00566.2006102:781-786, 2007. First published 26 October 2006;J Appl Physiol

T. Brink-Elfegoun, L. Kaijser, T. Gustafsson and B. Ekblomnervous system governorMaximal oxygen uptake is not limited by a central

You might find this additional info useful...

27 articles, 12 of which you can access for free at: This article citeshttp://jap.physiology.org/content/102/2/781.full#ref-list-1

15 other HighWire-hosted articles: This article has been cited by http://jap.physiology.org/content/102/2/781#cited-by

including high resolution figures, can be found at: Updated information and serviceshttp://jap.physiology.org/content/102/2/781.full

can be found at: Journal of Applied Physiology about Additional material and informationhttp://www.the-aps.org/publications/jappl

This information is current as of January 8, 2013.

http://www.the-aps.org/. 2007 the American Physiological Society. ISSN: 8750-7587, ESSN: 1522-1601. Visit our website at year (monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyrightphysiology, especially those papers emphasizing adaptive and integrative mechanisms. It is published 12 times a

publishes original papers that deal with diverse area of research in appliedJournal of Applied Physiology

at Southampton Solent Univ on January 8, 2013

http://jap.physiology.org/D

ownloaded from

Maximal oxygen uptake is not limited by a central nervous system governorT. Brink-Elfegoun,1,2 L. Kaijser,3 T. Gustafsson,3 and B. Ekblom1,21Department of Physiology and Pharmacology, Karolinska Institute, Stockholm; 2strand Laboratory ofWork Physiology and the Swedish School of Sport and Health Sciences, Stockholm; and 3Departmentof Laboratory Medicine, Section of Clinical Physiology, Karolinska Institute, Stockholm, SwedenSubmitted 19 May 2006; accepted in final form 17 October 2006

Brink-Elfegoun T, Kaijser L, Gustafsson T, Ekblom B. Maxi-mal oxygen uptake is not limited by a central nervous system gover-nor. J Appl Physiol 102: 781786, 2007. First published October 26,2006; doi:10.1152/japplphysiol.00566.2006.We tested the hypoth-esis that the work of the heart was not a limiting factor in theattainment of maximal oxygen uptake (V O2 max). We measured car-diac output (Q ) and blood pressures (BP) during exercise at twodifferent rates of maximal work to estimate the work of the heartthrough calculation of the rate-pressure product, as a part of theongoing discussion regarding factors limiting V O2 max. Eight well-trained men (age 24.4 2.8 yr, weight 81.3 7.8 kg, and V O2 max59.1 2.0 ml min1 kg1) performed two maximal combined armand leg exercises, differing 10% in watts, with average duration oftime to exhaustion of 4 min 50 s and 3 min 40 s, respectively. Therewere no differences between work rates in measured V O2 max, maxi-mal Q , and peak heart rate between work rates (0.02 l/min, 0.3 l/min,and 0.8 beats/min, respectively), but the systolic, diastolic, and cal-culated mean BP were significantly higher (19, 5, and 10 mmHg,respectively) in the higher than in the lower maximal work rate. Theproducts of heart rate times systolic or mean BP and Q times systolicor mean BP were significantly higher (3,715, 1,780, 569, and 1,780,respectively) during the higher than the lower work rate. Differencesin these four products indicate a higher mechanical work of the hearton higher than lower maximal work rate. Therefore, this study doesnot support the theory, which states that the work of the heart, andconsequently V O2 max, during maximal exercise is hindered by acommand from the central nervous system aiming at protecting theheart from being ischemic.

central governor; error of the method; maximal exercise; oxygenuptake

MAXIMAL AEROBIC POWER (V O2 max) is of vital importance forboth physical performance and health in general. That is whyquestions regarding this important physiological parameterhave been researched since 1923, when Hill and Luptonshowed that the oxygen uptake did not continue to increasewith increasing rate of work (17). Despite more than 80 yearsof scientific investigation of V O2 max, there is still no clearconsensus regarding the fundamental question: What sets theupper limit for V O2 max during exercise with a large musclemass included?

Over the years, several theories have been proposed aslimitation(s) factors to V O2 max, e.g., skeletal muscle oxidativecapacity, mainly total mitochondrial volume (peripheral limi-tation) (27), vasodilatation in relation to cardiac output (Q ) (7),symmorphosis theory where there is no single limiting factorfor V O2 max (28, 32), partly revised to different links in the

oxygen transport system chain during severe dynamic exerciseas resistances in Ohms law (8, 9).

According to our view, this theory of a harmonized oxygentransport system chain during maximal exercise is in line withthe one in which the volume of oxygen transported from theleft ventricle to the periphery [Q times oxygen content ofarterial blood (CaO2 )] is the basic determinant for V O2 max (11,12). In this thinking, changes in different factors within theoxygen transport system chain, such as variations in hemoglo-bin concentration, oxygen content of inspired air, arterialoxygen saturation, or type of exercise, may only modify theV O2 max obtained under optimal conditions.

However, another theory has been introduced into this dis-cussion (2224). The basis for this theory is that the centralnervous system (a central governor) controls the circulationduring severe exercise. No experimental data are presented,which could have supported the theory, but the author statesthat there is sufficient evidence for the opinion that the centralnervous system moderates the central circulation during max-imal exercise. By doing so, the circulation is primarily hin-dered to utilize its whole capacity to protect the heart musclefrom becoming ischemic. Thus, according to the central gov-ernor theory, V O2 max is only a consequence of the amount ofwork that the heart is allowed to perform and not vice versa,i.e., that the total volume of oxygen transported from the heartto the periphery has reached its maximum during maximalexercise with large muscle groups.

This central governor theory has been both supported (25)and rejected (5, 31). Since no consensus has been obtained andno earlier study has specifically addressed this question, theaim of the present investigation was to study the centralcirculation during different maximal work rates. It is acknowl-edged that the double product of blood pressure (BP) times Qand BP times heart rate (HR) reflects the myocardial oxygenuptake (MV O2) (20, 21). Our aim was therefore to measure BPand Q at two maximal work rates, differing in a range of1015% in watts (W). The work time had to be long enoughto create V O2 max in both tests.

The hypothesis for this study was that, during maximal andsupramaximal exercise with the use of large muscle groups inwell trained subjects, a higher rate of work would create ahigher BP than the lower one; at the same time, the HR, Q , andV O2 max would be equal in the two maximal tests. Therefore,the heart during the higher maximal work rate should haveproduced more mechanical work than during the lower one.Thus, if this is the case, the work of the heart would not havebeen hindered by the central nervous system at the lowermaximal work rate.

Address for reprint requests and other correspondence: B. Ekblom, strandLaboratory of Work Physiology, Swedish School of Sport and Health Sci-ences, Box 5626, S-114 86 Stockholm, Sweden (e-mail: bjorn.ekblom@gih.se).

The costs of publication of this article were defrayed in part by the paymentof page charges. The article must therefore be hereby marked advertisementin accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

J Appl Physiol 102: 781786, 2007.First published October 26, 2006; doi:10.1152/japplphysiol.00566.2006.

8750-7587/07 $8.00 Copyright 2007 the American Physiological Societyhttp://www. jap.org 781

at Southampton Solent Univ on January 8, 2013

http://jap.physiology.org/D

ownloaded from

METHODS

Subjects. Eight trained, nonsmoking male subjects, age 24.4 2.8yr, height 184.3 8.3 cm, weight 81.3 7.8 kg, and V O2 max 59.1 2.0 ml min1 kg1, volunteered to participate in this study. To avoidany training effect as a consequence of the study, the subjects had tobe fairly well trained, since the study included several preexperimen-tal maximal exercises to secure peak values of V O2 on severalmaximal rates of work, and two maximal exercises had to be carriedout within 2 h in the main experiment. Therefore, the subjects wereselected according to the following criteria: 1) a V O2 max of 55ml min1 kg1, 2) training history of 5 yr of endurance training,and 3) current training frequency of 4 times/wk. The protocol wasexplained to the subjects before they had given their written, informedconsent to participate in this study. The study was approved by theRegional Ethics Committee of Stockholm, Sweden.

All experimental tests were performed within a 6-wk period. Thesubjects performed the tests at about the same time of the day (2 h).There was 48 h between individual maximal tests, except in themain experiment, during which only 1 h elapsed between two maxi-mal tests. The subjects were asked to eat a small meal no less than 4 hbefore each test, to avoid caffeine-enriched beverages during this 4-hperiod, and not to be on medication during the test period.

All tests were carried out using a combined arm and leg ergometersystem, which consisted of two mechanically braked cycle ergometers(Monark 839E, Monark, Vansbro, Sweden). One of these was used asan ordinary leg cycle ergometer, whereas the other was placed on analuminium frame and used as an arm ergometer, with handgripsmounted on the cranks. The subjects were seated behind the armergometer, with the height of the saddle adjusted so that the arms,when extended, were just below heart level.

The ratio between arm and leg work in all tests was 20:80% to25:75% of the total rate of work. The cadence was set to 80 rpm. Thecadence figures, one for legs and the other for arms, were available ontwo computer screens. If the subject lost his cadence within reason-able limits, the Monark 839E system corrected the braking force sothat the work intensity set for the test would be maintained.

An important part in this study was to establish a clear leveling offof V O2 vs. rate of work to create a situation with different demands onthe heart while V O2 max was the same. This was done through anextensive pretest program in which several submaximal and maximalcombined arm and leg exercises were carried out. From the results ofthese tests, an individual leveling-off configuration of V O2, in relationto increasing rate of work, could be established in all subjects, andfrom this figure the lowest rate of work (L) that elicited V O2 max waschosen. The L work rate was used in the main test with cathetersindwelled, and the data obtained were compared with data from an10% higher maximal rate of work (H). V O2 max was reached ac-cording to strand and Rodahl (3a) when 1) total work time was 5min, 2) leveling off of V O2 vs. rate of work with V O2 on the highestwork rate being within 150 ml/min from previous highest obtainedvalue in the test (29), 3) subjective rate of perceived exertion of 16,and 4) and blood lactate concentrations above 8 mM.

We had to perform the L and H maximal tests on the same day.However, no difference was observed in V O2 values between the twomaximal tests during the pretests. Furthermore, there were no differ-ences between obtained values of V O2 and HR for the L rate of workwith and without catheters indwelled. We also used well-trainedsubjects who were familiar with repeated maximal exercises duringnormal training sessions. Thus there are no reasons to believe that thetest procedure during the main test influenced the conclusions of thisstudy (see DISCUSSION).

V O2 was measured continuously using an ergospirometry system(AMIS 2001, Innovision, Odense, Denmark) based on the mixedexpired method with an inspiratory flowmeter. Averaged values ofV O2 data of 10-s averages were available on the computer screen. Forthe gas analyzer calibration, high precision gases (16.00 0.04% O2

and 4.00 0.1% CO2, Air Liquide, Kungsangen, Sweden) andnormal air were used. Before each test, the ambient conditions weremeasured, and gas analyzers and inspiring flowmeter were calibrated.The calibration of the flowmeter was performed with a 3.0-liter airsyringe (Hans Rudolph) at a low, medium, and high flow velocity.This systems accuracy has been validated against the Douglas bagtechnique (18). HR was continuously measured using a HR monitor(model Polar S610, Polar Electro) integrated with the AMIS 2001.

A thin catheter was introduced in the brachial artery for samplingof blood and measurement of intra-arterial BP. A balloon-tippedcatheter was inserted percutaneously via an antecubital vein andadvanced under fluoroscopic control to a branch of the pulmonaryartery for sampling of blood. The hemoglobin concentration and bloodO2 saturation was measured by an automatic spectrophotometricmethod (ABL 520 Radiometer) radiometer in blood drawn from thebrachial and pulmonary artery. Q was determined by the Ficksprinciple, based on V O2 and the difference in O2 content in arterial(arteria brachialis) and mixed venous (arteria pulmonalis) blood.

Determinations of blood lactate concentrations were done on sam-ples of blood (20 l) obtained on fingertip blood and analyzed onBiosen 5140 (EKF, Diagnostic, Magdeburg, Germany). Calibration ofthe blood lactate analyzer was performed before each test and checkedby using a lactate standard of 12 mM. Calibration results within 0.1mM were accepted. The voltage was checked by a control solution of4.86.4 mM. RPE was evaluated using the 620 RPE scale, accord-ing to Borg (6), applied on central (dyspnoea, tachycardia) and local(active muscle) fatigue (10).

Main test. The preparation and procedure for each test was thesame for every subject. In the morning of the experimental day, thecatheters were inserted. After arrival to the exercise test room, saddleheight was adjusted, 11 ECG electrodes were applied, and the firstsubmaximal rate of work (200 W) was carried out for 4 min. V O2 andHR were measured continuously with the AMIS 2001 and the ECG.At the end of the submaximal work period, BP was recorded fromthe brachial artery. Thereafter blood samples for determination ofCaO2 and mixed venous O2 concentration were simultaneously ob-tained from the brachial and pulmonary artery catheters, respectively.After a short rest and 2 min of warm-up at 50% of final work rate,the L maximal rate of work was performed. The subjects worked untilsubjective exhaustion or until they could not keep the cadence. V O2and HR were measured continuously. At 2 min from start of the Lmaximal rate of work, BP in the brachial artery was recorded for thefirst time and then at least another three times. On each occasion, atleast 25 BP curves were recorded, and average values were calcu-lated. After the first BP recording, blood samples for determination ofCaO2 and mixed venous O2 concentration were simultaneously drawnfrom the brachial and pulmonary artery catheters, respectively. Atleast three pairs of blood samples were obtained during the L maximaltest.

After the subject had rested for 1 h, resting values of V O2, HR, andBP were recorded, and blood samples were drawn from the twocatheters for determination of Q while sitting on the cycle ergometer.Thereafter, using the same experimental protocol as in the firstexercise session, the 200-Wsubmaximal and the H maximal exercisewere carried out. However, since the H maximal rate of work washigher, the work time became shorter. Therefore, after the firstrecording of BP and Q at 2 min, only another 1 or 2 values of Q andBP could be obtained until the subject was exhausted (Fig. 1).

Statistical analysis. Statistical analyses were performed using Con-fidence Interval Analysis 2.0.0 (Trevor Bryant) and Statistica 7.1(StatSoft, Tulsa, OK). Changes between values were analyzed with95% confidence intervals (CI). A difference was regarded as a changewhen the zero value was not included in 95% CI for the difference.

Shapiro-Wilks W-test was applied to examine the normality in thedistribution of data. To detect differences between the L and the Htrials, paired t-tests were used. Significance of differences was deter-mined at P 0.05, and data are reported as means SD, unless

782 CARDIAC PERFORMANCE DURING MAXIMAL EXERCISE

J Appl Physiol VOL 102 FEBRUARY 2007 www.jap.org

at Southampton Solent Univ on January 8, 2013

http://jap.physiology.org/D

ownloaded from

otherwise stated. Mean BP was calculated from diastolic BP 1/3 ofthe difference between systolic and diastolic BP.

RESULTS

The differences in V O2 max and peak HR in the L combinedarm and leg work between the two pretests were 0.06 l/min and1.8 beats/min (P 0.03 and P 0.22, respectively). Betweenthe two pretests and the L trial, the differences in V O2 max andpeak HR were 0.01 and 0.04 l/min and 2.8 and 1.0 beats/min(P 0.95 and 0.81 and P 0.11 and 0.53, respectively).Between the two pretests and the H trial, the differences inV O2 max and peak HR were 0.01 and 0.07 l/min and 3.5 and 1.8beats/min (P 0.97 and 0.74 and P 0.04 and 0.46,respectively). Average peak blood lactate concentration afterthe final pretest was 13 2 mM.

All results and relevant statistics for the main experiment arepresented in Table 1 and Figs. 2 and 3.

The rates of work (mean, SD, range) of the two maximaltests were L 387 60 (320510) and H 431 73 (340565)W. Corresponding times to exhaustion were L 4 min 50 s (4min, 00 s to 6 min, 50 s) and H 3 min 40 s (3 min, 00 s to 5min, 10 s). Regarding the V O2 max, Q , and HR values in the Land H tests, the 95% CI of the differences included a zero. TheH maximal work load generated a higher systolic, diastolic,and calculated mean arterial BP than the L maximal work. Inthese BP measurements, the 95% CI for the differences did notinclude zero.

Since the HR was the same during the H and L maximalexercise, the average product of HR times systolic and HRtimes mean BP was 12% and 8% higher in the H than in the L

Fig. 1. Graph showing the experimental protocol within the region of times for the blood pressure (BP) and cardiac flow (Q ) and the maximal exercise intensitieswith their respective O2 uptake (V O2) values for the low (L) and high (H) maximal exercise. Time (min) on x-axis and V O2 (l/min) on y-axis (n 8).

Table 1. Gas exchange and BP data during submaximal and maximal exercise with two work rates

Submaximal L Submaximal H Maximal L Maximal H

Rates of work, W 200 200 38760 43173*V O2, l/min 2.820.09 2.840.09 4.580.61 4.600.58HR, beats/min 1409 1489* 1905 1915V E, l/min 715 715 16720 18321*SV, ml 13914 13218 13114a 13212Q, l/min 19.41.5 19.52.9 24.73.2a 25.03.8Systolic BP, mmHg 14714 14112 16023 17932*Mean BP, mmHg 1015 1037 10811 11715*Diastolic BP, mmHg 779 736 827 878HR systolic BP, beats min1 mmHg 20,4811,841 20,8381,962 30,4164,327 34,1315,987*HR mean BP, beats min1 mmHg 15,1211,610 15,2531,373* 20,5092,208 22,2892,894*Q systolic BP, l min1 mmHg 2,865448 2,766576 4,058551 4,628662*Q mean BP, l min1 mmHg 2,113327 2,018362 2,734311 3,005431*

Values are means SD (n 8). V O2, oxygen uptake; HR, heart rate; V E, minute ventilation; SV, stroke volume; Q, cardiac output; BP, blood pressure; L,low; H, high. *P0.05 H trial compared with the L combined arm and leg trial. (n 7); see DISCUSSION.

783CARDIAC PERFORMANCE DURING MAXIMAL EXERCISE

J Appl Physiol VOL 102 FEBRUARY 2007 www.jap.org

at Southampton Solent Univ on January 8, 2013

http://jap.physiology.org/D

ownloaded from

maximal exercise, respectively. Since Q was the same duringthe H and L maximal exercise (24.7 and 25.0 l/min, respec-tively), the increase in Q times systolic and Q times mean BPduring the H compared with the L maximal exercise was 14%and 10%, respectively. In all these increases, the 95% CI forthe differences did not include zero.

For values obtained during the two submaximal exercises,the 95% CI for the differences included zero only in HRmeasurement.

DISCUSSION

The aim of the present investigation was to study the centralcirculation during maximal exercise of two different magni-tudes and to analyze the obtained data in relation to the centralgovernor theory (2225). This theory states that the circulationand skeletal muscle activity during severe exercise are con-trolled by the central nervous system, primarily to protect theheart muscle from becoming ischemic, and that V O2 max is onlya consequence of the amount of work that the heart is allowed

to perform and not vice versa. This theory has been forwardedwithout any experimental scientific support. Since the earlyworks of Hill and others, many theories on the limitation(s) toV O2 max have come forward, but no one has been definitivelyand indisputably proven and accepted by all exercise physiol-ogists. According to our view, V O2 max is mainly a result of theamount of oxygen offered to the periphery from the leftventricle (Q CaO2 ) during maximal exercise. This is sup-ported by the results of previous experiments (2, 11, 12, 26).

In the present study, the double product for HR times bothsystolic and mean BP and Q times both systolic and mean BPwas higher in the H than the L maximal exercise, whereas theV O2 max was the same. According to Kitamura et al. (20) andNelson et al. (21), the double product for HR times BP flow isa satisfactory predictor of MV O2 and coronary blood flow innormal young subjects exercising. They observed a closecorrelation of HR times BP with MV O2 and coronary bloodflow. This would suppose that the heart is able to perform moremechanical work during the H than the L maximal exercise.

Fig. 2. Line of identity figures with L on x-axisand H on y-axis for maximal V O2 (V O2 max; l/min;A), heart rate (HR; beats/min; B), Q (l/min; C),systolic BP (mmHg; D), diastolic BP (mmHg; E),and calculated mean BP (mmHg; F). P 0.05between the H trial compared with the L combinedarm and leg trial.

784 CARDIAC PERFORMANCE DURING MAXIMAL EXERCISE

J Appl Physiol VOL 102 FEBRUARY 2007 www.jap.org

at Southampton Solent Univ on January 8, 2013

http://jap.physiology.org/D

ownloaded from

This can also be confirmed by a recent study where 35 maleand 16 female well-trained distance runners had identicalV O2 max on a treadmill between a maximal and a supramaximalexercise (30% above incremental V O2 max) (16).

This indicates that the limitation of the central circulationduring the L maximal exercise must be discussed in otherterms, i.e., in relation to flow limitations rather than a centralnervous inhibition as it was suggested in the central governortheory. As mentioned in the introduction and above, whendifferent relevant aspects regarding this topic are discussed,i.e., in (2, 4, 14, 15, 19, 26, 30), the volume of oxygentransported from the heart to the periphery seems to be themajor determinant for maximal oxygen transport and utiliza-tion during severe prolonged dynamic exercise with largemuscle groups involved. The results of the present studysupport this theory.

The higher product of Q or HR times BP at the H comparedwith the L maximal work rate indicates that the energetic costof the work of the heart is increased during the H maximalwork. The 812% increase in pressure-rate product in thisstudy should increase myocardial oxygen consumption byabout equal amount (21). Also, the 9% higher pulmonaryventilation on the H maximal work rate costs more oxygen (1,3). Thus, since V O2 and Q at the H and the L maximal worksessions were equal, it can be speculated that part of theaverage 1 min 10 s shorter time to exhaustion at the H maximalwork could be attributed to the increased oxygen cost of theheart and the ventilation, since, evidently, less volume ofoxygen is available for the working skeletal muscles. The orderof the L and H tests was not randomized for methodologicalreasons. We had to perform the L trial before the H trial;

otherwise, we would never have known what caused the effectson either the L or the H trial. A lower BP during the L trial afterthe H trial could have been interpreted as a consequence of theH trial. With this order, we have the reversed situation. Ifanything, the higher BP during the H trial could only have beena consequence from the H work.

In a recent review article, Vella and Robergs discusseddifferent stroke volume (SV) responses to incremental exercise(33). In a great number of articles published, SV continues toincrease from a medium submaximal exercise level up tomaximal exercise, whereas in an about equal amount of studiesSV remains unchanged during the same part of the exercisespectrum. Even a decrease in SV at peak exercise has beenreported. In the present study, the SV was 108 ml while sittingon the ergometer. During the two submaximal exercises, SVincreased to 139 and 132 ml, respectively, but remained un-changed during further exercise or to 131 and 132 ml at the Hand L experiments, respectively. Thus, in this study, SVincreased from rest to submaximal exercise but was unchangedfrom 6065% to 100% of V O2 max in both the L and H partof the study, which is in line with the comment forwarded inrelation to the Vella and Robergs article (9a). Early literatureindicates an error in a single determination of Q of 5% (13).

The variation in measurement of Q during submaximalexercise was 0.06 l/min (limit of agreement 3.11 to 3.22).The variation during maximal exercise, using the last values ofthe L and H tests, was 0.32 l/min (limit of agreement 2.81 to3.45). During the L trial, one of the subjects had his pulmonaryartery catheter removed from its location due to technicalreasons. This gave us very limited data during this measure-ment, and we decided to exclude this subjects Q and SV data

Fig. 3. Line of identity figures with L onx-axis and H on y-axis for the calculatedproducts of Q systolic BP (A), Q mean BP(B), HR systolic BP (C), and HR mean BP(D). P 0.05 between the H trial comparedwith the L combined arm and leg trial.

785CARDIAC PERFORMANCE DURING MAXIMAL EXERCISE

J Appl Physiol VOL 102 FEBRUARY 2007 www.jap.org

at Southampton Solent Univ on January 8, 2013

http://jap.physiology.org/D

ownloaded from

from this calculation (see Table 1; n 7). Using the last twomeasurements done during the last minutes of both the 7 L and8 H maximal exercises, when V O2 max was stable, the variationwas 0.12 l/min (limit of agreement 2.21 to 2.46). Thesefigures indicate that the variation in measurement of Q with thedirect Fick method is fairly small, which is important in thediscussion of factors limiting V O2 max. We used only well-trained subjects in this study, but there are no reasons tobelieve that the basic results and conclusions from this studycannot be applied to other types of healthy adult subjects suchas female and untrained people. However, in patients withreduced cardiovascular function or in elderly people, changesin functions and structures may very well be of such impor-tance that the conclusion from this study may not apply tothem.

In conclusion, the H maximal work generated a highersystolic, diastolic, and calculated mean arterial BP than the Lmaximal work rates, whereas V O2 max and Q were the same forthe two different maximal work rates. The heart was able toperform more mechanical work during the H than the Lmaximal exercise. This study suggests that the limitation of thecentral circulation during maximal work must be discussed inother terms rather than a central nervous inhibition, as sug-gested in the central governor theory.

ACKNOWLEDGMENTSWe thank the subjects who participated in this investigation and made this

study possible. We also thank Gunilla Hedin, Monica Borggren, and Marg-aretha Stolperud for excellent technical assistance.

GRANTSThis research was financially supported by the Swedish School of Sport and

Health Sciences, Stockholm, Sweden, and the Swedish Research Council.

REFERENCES

1. Aaron EA, Seow KC, Johnson BD, Dempsey JA. Oxygen cost ofexercise hyperpnea: implications for performance. J Appl Physiol 72:18181825, 1992.

2. Andersen P, Saltin B. Maximal perfusion of skeletal muscle in man.J Physiol 366: 233249, 1985.

3. Anholm JD, Johnson RL, Ramanathan M. Changes in cardiac outputduring sustained maximal ventilation in humans. J Appl Physiol 63:181187, 1987.

3a.strand PO, Rodahl K. Textbook of Work Physiology New York:McGraw-Hill, 1986. p. 283.

4. Bassett DR Jr, Howley ET. Maximal oxygen uptake: classical versuscontemporary viewpoints. Med Sci Sports Exerc 29: 591603, 1997.

5. Bergh U, Ekblom B, Astrand PO. Maximal oxygen uptake classicalversus contemporary viewpoints. Med Sci Sports Exerc 32: 8588,2000.

6. Borg G, Linderholm H. Perceived exertion and pulse rate during gradedexercise in various age groups. Acta Med Scand 472: 194206, 1967.

7. Clausen JP. Circulatory adjustments to dynamic exercise and effect ofphysical training in normal subjects and in patients with coronary arterydisease. Prog Cardiovasc Dis 18: 459495, 1976.

8. di Prampero PE. Factors limiting maximal performance in humans. EurJ Appl Physiol 90: 420429, 2003.

9. di Prampero PE. Metabolic and circulatory limitations to V O2 max at thewhole animal level. J Exp Biol 115: 319331, 1985.

9a.Ekblom B, Ekblom O . Stroke volume and the endurance athlete. Scand JMed Sci Sports 16: 7071, 2006.

10. Ekblom B, Goldbarg AN. The influence of physical training and otherfactors on the subjective rating of perceived exertion. Acta Physiol Scand83: 399406, 1971.

11. Ekblom B, Huot R, Stein EM, Thorstensson A. Effect of changes inarterial oxygen content on circulation and physical performance. J ApplPhysiol 39: 7175, 1975.

12. Ekblom B, Wilson G, strand PO. Central circulation during exerciseafter venesection and reinfusion of red blood cells. J Appl Physiol 40:379383, 1976.

13. Ekelund LG. Circulatory and respiratory adaptation during prolongedexercise. Acta Physiol Scand Suppl 292: 138, 1967.

14. Grubbstrom J, Berglund B, Kaijser L. Myocardial oxygen supply andlactate metabolism during marked arterial hypoxaemia. Acta PhysiolScand 149: 303310, 1993.

15. Hammond HK, White FC, Bhargava V, Shabetai R. Heart size andmaximal cardiac output are limited by the pericardium. Am J PhysiolHeart Circ Physiol 263: H1675H1681, 1992.

16. Hawkins M, Raven P, Snell P, Stray-Gundersen J, Levine B.Maximaloxygen uptake as a parametric measure of cardiorespiratory capacity. MedSci Sports Exerc. In press.

17. Hill A, Lupton H. Muscular exercise, lactid acid and the supply andutilization of oxygen. QJM 16: 135171, 1923.

18. Jensen K, Jorgensen S, Johansen L. A metabolic cart for measurementof oxygen uptake during human exercise using inspiratory flow rate. EurJ Appl Physiol 87: 202206, 2002.

19. Kaijser L, Grubbstrom J, Berglund B. Myocardial lactate release duringprolonged exercise under hypoxaemia. Acta Physiol Scand 149: 427433,1993.

20. Kitamura K, Jorgensen CR, Gobel FL, Taylor HL, Wang Y. Hemo-dynamic correlates of myocardial oxygen consumption during uprightexercise. J Appl Physiol 32: 516522, 1972.

21. Nelson RR, Gobel FL, Jorgensen CR, Wang K, Wang Y, Taylor HL.Hemodynamic predictors of myocardial oxygen consumption during staticand dynamic exercise. Circulation 50: 11791189, 1974.

22. Noakes JB, Wolffe TD. 1996 Memorial Lecture. Challenging beliefs: exAfrica semper aliquid novi. Med Sci Sports Exerc 29: 571590, 1997.

23. Noakes TD. Maximal oxygen uptake: classical versus contemporaryviewpoints: a rebuttal. Med Sci Sports Exerc 30: 13811398, 1998.

24. Noakes TD. Physiological models to understand exercise fatigue and theadaptations that predict or enhance athletic performance. Scand J Med SciSports 10: 123145, 2000.

25. Noakes TD, Peltonen JE, Rusko HK. Evidence that a central governorregulates exercise performance during acute hypoxia and hyperoxia. J ExpBiol 204: 32253234, 2001.

26. Stray-Gundersen J, Musch TI, Haidet GC, Swain DP, Ordway GA,Mitchell JH. The effect of pericardiectomy on maximal oxygen consump-tion and maximal cardiac output in untrained dogs. Circ Res 58: 523530,1986.

27. Taylor CR, Karas RH, Weibel ER, Hoppeler H. Adaptive variation inthe mammalian respiratory system in relation to energetic demand. RespirPhysiol 69: 1127, 1987.

28. Taylor CR, Weibel ER. Design of the mammalian respiratory system. I.Problem and strategy. Respir Physiol 44: 110, 1981.

29. Taylor HL, Buskirk E, Henschel A. Maximal oxygen intake as anobjective measure of cardio-respiratory performance. J Appl Physiol 8:7380, 1955.

30. Wagner PD. Determinants of maximal oxygen transport and utilization.Annu Rev Physiol 58: 2150, 1996.

31. Wagner PD. New ideas on limitations to V O2 max. Exerc Sport Sci Rev 28:1014, 2000.

32. Weibel ER, Taylor CR, Hoppeler H. The concept of symmorphosis: atestable hypothesis of structure-function relationship. Proc Natl Acad SciUSA 88: 1035710361, 1991.

33. Vella CA, Robergs RA. A review of the stroke volume response toupright exercise in healthy subjects. Br J Sports Med 39: 190195, 2005.

786 CARDIAC PERFORMANCE DURING MAXIMAL EXERCISE

J Appl Physiol VOL 102 FEBRUARY 2007 www.jap.org

at Southampton Solent Univ on January 8, 2013

http://jap.physiology.org/D

ownloaded from