ISSN: 1524-4539 Copyright © 1994 American Heart Association. All rights reserved. Print ISSN: 0009-7322. Online

72514Circulation is published by the American Heart Association. 7272 Greenville Avenue, Dallas, TX

1994;90;2166-2179 CirculationGidding, J Isabel-Jones, RE Kavey and GR Marx

RL Washington, JT Bricker, BS Alpert, SR Daniels, RJ Deckelbaum, EA Fisher, SS Disease in the Young, the American Heart Association

on Atherosclerosis and Hypertension in Children, Council on Cardiovascular Guidelines for exercise testing in the pediatric age group. From the Committee

http://circ.ahajournals.orglocated on the World Wide Web at:

The online version of this article, along with updated information and services, is

http://www.lww.com/reprintsReprints: Information about reprints can be found online at  

journalpermissions@lww.com410-528-8550. E-mail:

Fax:Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters 

http://circ.ahajournals.org/subscriptions/Subscriptions: Information about subscribing to Circulation is online at

by on January 15, 2008 circ.ahajournals.orgDownloaded from

2166

AHA Medical/Scientific StatementSpecial Report

Guidelines for Exercise Testing in thePediatric Age Group

From the Committee on Atherosclerosis and Hypertensionin Children, Council on Cardiovascular Disease in the Young,

the American Heart Association

Reginald L. Washington, MD; J. Timothy Bricker, MD; Bruce S. Alpert, MD;Stephen R. Daniels, MD, PhD; Richard J. Deckelbaum, MD; Edward A. Fisher, MD;

Samuel S. Gidding, MD; Josephine Isabel-Jones, MD; Rae-Ellen W. Kavey, MD;Gerald R. Marx, MD; William B. Strong, MD; Douglas W. Teske, MD;

Jack H. Wilmore, PhD; Mary Winston, EdD

T he purpose of this report is to update guidelinesfor exercise testing in both normal children andchildren with cardiovascular disease. This re-

port is intended for physicians, nurses, exercise special-ists, physical educators, technologists, and other healthprofessionals involved in the exercise testing and train-ing of children. These standards and guidelines areintended to complement the American Heart Associa-tion's adult exercise testing standards1 as well as tosupplement and update an earlier publication of theAHA on exercise testing in children.2

Because exercise is only one of many stresses to whichhumans can be exposed, we prefer the term "exercisetest" to "stress test." Exercise can be useful as aphysiological stress to elicit findings and abnormalitiesthat are not evident at rest. This report emphasizesexercise laboratory testing and does not attempt toaddress the uses of exercise in physical education classesor other nonlaboratory settings. But because laboratoryexercise testing may not reproduce the stress of aspecific sport or activity, evaluation may need to betailored for individuals who report the occurrence ofsymptoms during specific sports.

Basic Exercise PhysiologyAn understanding of the physiological changes that

occur during exercise is of practical importance to theclinician who performs exercise testing of children. Inparticular, the physiological adaptations of normal chil-dren to exercise provide insight into the abnormalitiesfound at rest in children with congestive heart failure.Because obtaining useful and accurate laboratory mea-surements and interpreting test results depend on an

"Guidelines for Exercise Testing in the Pediatric Age Group"was approved by the American Heart Association SAC/SteeringCommittee on February 16, 1994.

Requests for reprints should be sent to the Office of ScientificAffairs, American Heart Association, 7272 Greenville Avenue,Dallas, TX 75231-4596.X American Heart Association, 1994.

appreciation of normal physiological responses, thissection provides an overview of the terminology andmeasurements used in the exercise testing of childrenthat have come from the field of exercise physiology.

WorkExercise is defined as any activity involving the gen-

eration of force by activated muscle(s) that results in analteration of the homeostatic state. In dynamic exercise,the muscle may perform shortening (concentric contrac-tions) or be overcome by external resistance and per-form lengthening (eccentric contractions). When mus-cle force results in no movement, the contraction istermed static or isometric. Static exercise (eg, handgrip) imposes a relatively greater pressure-load thanvolume-load on the heart; dynamic exercise (eg, run-ning) results primarily in a volume-load. The cardiovas-cular response is proportional to the intensity of dy-namic exercise up to maximal levels of exercise.Dynamic exercise is preferred for most testing becauseit can elicit maximal or near-maximal cardiovascularand metabolic responses. With isometric exercise, incontrast, the subject is more likely to be limited by acutemuscular fatigue before cardiac limitations are ap-proached. Most activities combine varying amounts ofboth isometric and dynamic exercise. Estimates of theamount of exercise performed are made by using thefollowing definitions.

Work. Force expressed through a distance but with nolimitation on time. The units for work are joules orkilopond*meter.

Power. The rate of performing work, ie, the derivativeof work with respect to time or the product of force andvelocity. The units for power are unit of work per unit oftime. One joule per second equals one watt.Kilopond*meter per minute is also used to expresspower.Maximal power output. The highest rate of work

achieved during exercise testing.

by on January 15, 2008 circ.ahajournals.orgDownloaded from

AHA Council on CVDY Guidelines for Pediatric Exercise Testing

Endurance time. The total test time to exhaustion foran individual performing a continuous graded test. Theduration of testing is compared with the endurance timeobtained by normal subjects matched for age and gen-der who were tested by using an identical protocol.

Physical working capacity-i 70 (PWC-1 70). The highestrate of submaximal steady-state work corresponding toa heart rate of 170 beats per minute identified duringcontinuous graded cycle ergometry.

Total work. Accumulated work to exhaustion or to apredetermined end point during exercise testing.The units of work have traditionally varied with the

type of ergometer used during the test. With the use ofa cycle ergometer the units of work are expressed aswatts or joules, whereas units of kilopond*meter/minuteare typically used to express work with treadmill testing.The expected amount of work performed during a

particular test varies with the age, gender, and fitness ofthe subject and the type of protocol used.

Perceived ExertionThe subjective rating of intensity of perceived exer-

tion by the person exercising is a good indicator ofrelative fatigue and is used in conjunction with the heartrate. The concept of perceived exertion has been inter-preted to quantify effort during exercise.3 A 6- to20-point Borg scale may be used to assess perceivedexertion.4

Special verbal and written instructions about therating of perceived exertion are provided to the testsubject. Although perceived exertion varies among chil-dren at given rates of work, results are reproduciblefrom test to test. The Borg scale may be used in judgingthe degree of fatigue reached from test to test per-formed by the same child or to compare the level offatigue during testing with that experienced during dailyactivities or training.

Heart RateThe immediate response of the cardiovascular system

to exercise is an increase in heart rate due to a decreasein vagal tone. This response is followed by an increase insympathetic outflow to the heart and systemic bloodvessels. During dynamic exercise, the heart rate in-creases linearly with the rate of work. During low tomoderate levels of exercise at a constant work rate, theheart rate reaches a steady state within 1 minute andincreases proportional to the rate of work in subjectswith a normal sinus node. Variables that affect heartrate during exercise include the type of exercise, bodyposition during testing, gender, state of health andfitness of the subject, and environmental conditions(such as heat, cold, humidity, or altitude).Young children compensate for a smaller heart and

lower stroke volume by an increased heart rate at agiven rate of work; thus, they attain higher maximalheart rates than adults. After puberty, maximal heartrate decreases with age at a rate of 0.7 or 0.8 beats perminute per year of age. Females have a higher heart ratethan males at any given rate of work after puberty.Obese children have higher submaximal heart ratesthan lean children at the same rate of work. Exercise ina hot or humid climate results in higher heart rates fora given rate of work than exercise in a neutral thermal

A decrease in resting heart rate is characteristic ofconditioning. Peak rate that can be achieved by thesinus node is largely independent of the state of condi-tioning. Increased mechanical efficiency with retestingmay provide an increased intensity of exercise thatraises the observed peak heart rate at maximal volun-tary effort. The peak heart rate observed may decreaseon subsequent tests with endurance training in thosewith initial high maximal heart rates. Heart rate alone isan inadequate determinant of fitness. Additional phys-iological determinants to assess the level of conditioningmay include peak oxygen consumption, the anaerobicthreshold, endurance time, and total work.

Motivational factors, subject cooperation, mechanicalefficiency, and the type of protocol are also importantdeterminants of heart rate with exercise testing. Tread-mill exercise results in a slightly higher maximal heartrate than does cycle ergometry. Table 1 gives examplesof peak heart rates in children of various ages usingeither treadmill or cycle.

Electrocardiographic ChangesElectrocardiographic changes during ventricular re-

polarization have been used to identify mismatch ofmyocardial oxygen demand and supply.

Investigators differ in their analyses and interpretationof ST segment changes on exercise electrocardiography.20Fig 1 illustrates two methods of measuring the depressionof the ST segment and the J point. In the upper panel, thebaseline (P-R isoelectric line) is superimposed on the P-Rsegment of the QRS-T complex for identifying the J-pointdepression. In the lower panel, the baseline is drawnconnecting several P-Q points of at least three consecutiveQRS-T segments for identifying the J-point and ST de-pression. The criteria for significant ST-segment J-pointdepression include a J-point depression 22mm and an STdepression > 1 mm with a flat or downsloping ST segmentat 60 milliseconds.Normal physiological J-point depression was found in

9% of boys and 18% of girls with the PQ-PQ isoelectricline method and 2.3% of boys or girls in the same studyby using the PR isoelectric line method.21 The STsegment in normal subjects with "physiological" STdepression rapidly returns to baseline (within 80 milli-seconds) and is not flat or downsloping (Fig 2).

Interpretations of abnormal repolarization are impos-sible if depolarization is also abnormal. Bundle branchblock, which is often present in children who have heartdisease, renders the ST segments uninterpretable withexercise testing. This is also true with a ventricularpacemaker or with Wolff-Parkinson-White syndrome.

Cardiac Output and Stroke VolumeCardiac output (CO) is the product of stroke volume

(SV) and heart rate (HR). The relation is as follows:

CO=SVxHR

Stroke volume depends on preload, afterload, andcontractility.

Cardiac output increases in an almost linear mannerduring exercise, concomitant with increasing oxygenconsumption, increasing at each higher exercise levelduring a graded test. A steady-state cardiac output is

environment.

2167

achieved in normal children within 2 to 3 minutes of by on January 15, 2008 circ.ahajournals.orgDownloaded from

2168 Circulation Vol 90, No 4 October 1994

TABLE 1. Maximal Heart Rate With Treadmill, Cycle Ergometer, and Supine Cycle

Author, Year Age, y Population Sex Peak HR, bpm ProtocolTreadmill

Shephard, 19695

Skinner, 19716

Hermansen, 19717

Wilmore, 19748

Wilmore, 19829

Cumming, 197810

Riopel, 197911

11-13

6-15

10-12

8-12

13-15

4-18

4-21

Canada Both

USA Male

Female

Male

Female

Scandinavia Male

USA Male

Male

Canada Both

USA, black Male

Female

USA, white Male

Female

193

197.7-200.9

203.0-203.5

199.1

202.7

205.9-206.6

196.8±7.7 (177-213)

202.1 ±8.5 (173-222)

193-206

179-186

181-194

183-192

187-191

Continuous, progressiveContinuous, progressive

Intermittent, multistage

..

Bruce

Balke, modifications of Austin

Cycle Ergometer

Astrand, 195212

Goldberg, 196613

Wilmore, 196714

Ericksson, 197115

James, 198016

Alpert, 198217

7-9

7-9

10-11

10-11

12-13

12-13

6-16

7-9

10-11

12-13

6-15

5-33(mean, 14.3)

6-15

Scandinavia Male

Female

Male

Female

Male

Female

USA Male

Female

USA Female

Scandinavia

USA, 95%white, 5%

black

Male

Both

USA, Black Male

Female

USA, White Male

Female

208

211

211

209

205

207

194+9

193±8

195

196

194

200.5±2.9

187-199

188-194

185-195

191-194

191 -195

Continuous

Intermittent, multistage

Continuous

Continuous, progressive,Constant workload,electrically braked

Continuous, gradedMechanically braked

Supine Cycle

Cumming, 197718 5-16 Canada Male 170+17 Intermittent, graded

Female 174±1 1

Lock, 197819 5-16 USA Both 142+4.9 Fixed rate of work,submaximal

HR indicates heart rate; bpm, beats per minute.

each rate of work increase.622 Usually cardiac outputshows a threefold through fivefold increase from rest tomaximal exercise, as illustrated in Table 2.

Children typically have a markedly lower stroke volumebecause of smaller body size, but information is limited.

Table 3 provides values for stroke volume derived fromfive studies of healthy children performing either supine orsitting exercise using a cycle ergometer.The cardiac output in a given individual depends on

the type of exercise being performed. Upright exercise

by on January 15, 2008 circ.ahajournals.orgDownloaded from

AIA4 Council on CVDY Guidelines for Pediatric Exercise Testing 26

4 4~~~~~~~~~~~~~~~~~.

A~~~~~~~~~~~~~~~~~;J....

two~.....

measurement.. .. teP

iselctic to.tp,. from. set.o.the.

be~~~~~~~~~~~~~~~~~~~~~~~~~........

isoe.ectric.lin

one.. QRS..to.

A~~~~~~~~~~~~~........

for~~.....

0~~~~~~~~~~~~~~~~~~~~~~~~~~..

ulltils

oFI Atom ari giventw rateho fs wo rk,t r in n er is upinec

isex erc tise willmehave a( venousnetudawrnm an de higthe rPca o us tp th htplex erci et oie. illve rgto -eletry infepediatic getinearallycresuTse higheercmaimaln

cardi acin output nethan hi terg ometry v n io a m t od

BloodPrsliesisureed dfracpa l Tsg etm aue

R pr ssu red themprck ro ea0a

resistancrando okanaiacardiacoutput.htcrultrycanefroms rstpito exercise arecomplex.f mexerncise tefi

ofiesisoftane, wherich cause thelisystolacicebloo prgessrertoat ofi va ulaw rkbd, however,

therdiasotolcpressanuretoincreaseronly slightly. Inrsomenorma pdivtiduas,veeasdlationultwithaexrisher mayxeven

thes ueuo

unchsangedadcricotu.Th.icltr hne

FIG 2. Comparison of physiological ST segment depression

(top) with ischemic changes (bottom). lschemic changes are

characterized by flat or downsloping ST depression. Repro-duced from Bricker.20

A lack of increase or a decrease in systolic blood

pressure below the normal resting level has been widelyheld as an ominous indication of severe cardiac dysfunc-tion.23 A decrease in blood pressure during exercise

occurs when the peripheral vascular resistance dropsnormally but a cardiac abnormality limits the subject's

ability to increase cardiac output with exercise. This

may occur with severe aortic stenosis or cardiomyopa-thy. An exertional drop in blood pressure is, however,not as specific as once thought and can occur in normal

individuals.24

Systolic blood pressure increases with progressiveworkloads. In adults, a maximal exercise systolic pres-

sure above 220 mm Hg has been considered an excessive

rise. Maximal systolic blood pressure in children rarelyexceeds 200 mm Hg, and there is no evidence of dangerwhen the systolic blood pressure reaches the 250-mm

Hg range during exercise in an asymptomatic child or

adolescent. For optimal safety of the exercise testing

procedure, however, the systolic blood pressure gener-

ally should not be permitted to exceed a range that can

be measured during the test.

The normal diastolic blood pressure response to

exercise in children is not known with confidence be-

cause ambient noise and movement of the child's arm

during an exercise study make the response difficult to

(AN

0.

v

2169

by on January 15, 2008 circ.ahajournals.orgDownloaded from

2170 Circulation Vol 90, No 4 October 1994

TABLE 2. Exercise Cardiac Output

Author, Year Age, y Rest Value Maximum Value Instrument ProtocolEricksson, 197115 13-14 5.25±0.50 L/min per m2 17.41 ±0.88 L/min per m2 Cycle ContinuousEricksson, 197323 11-13 3.9 Llmin per m2 12.5 Llmin per m2 Seated cycle in Continuous

catheterization laboratory

determine accurately. This is generally a more difficultproblem with treadmill testing than with cycle ergome-try. However, it is unusual for children exercising on atreadmill to have a significant increase in diastolicpressure.The magnitude of increase of peak systolic and dia-

stolic blood pressure values during exercise increaseswith age and body size. Because males have a highermaximal stroke volume than females, they routinelyhave a higher systolic blood pressure response.

Oxygen ConsumptionOxygen uptake (Vo2) increases rapidly when dynamic

exercise begins. After the second minute at each level ofintensity of exercise, oxygen uptake usually reaches aplateau. In a steady state at a given workload, heartrate, cardiac output, blood pressure, and pulmonaryventilation are maintained at reasonably constantlevels.25The maximum oxygen uptake (Vo2max) is the highest

amount of oxygen that a given individual can consumewhile performing dynamic exercise; in contrast, peakoxygen consumption is the maximal amount of oxygenuptake observed during a specific exercise study. Peakoxygen consumption may or may not equal maximumoxygen uptake.

In adults, Vo2 reaches a plateau at Vo2max and nofurther increase is observed with further increases inthe rate of work. This plateau is infrequently foundwhen testing children, and Vo2max cannot be preciselydetermined in many pediatric studies; thus, the peakobserved oxygen uptake is used instead.

Children increase their oxygen consumption approx-imately 10-fold during exercise; adults typically obtain a10- to 15-fold increase. A trained athlete may achieve asmuch as a 20-fold increase in oxygen consumption.Oxygen consumption is strongly related to fat-free body

mass, and when Vo2 is indexed by body weight, thedifference in oxygen consumption between gendersbecomes minimal.

Table 4 lists ranges for peak oxygen consumptionmeasured in normal children using a variety of exercisetest protocols; these are reviewed in detail by Freedsonand Goodman.26

Ventilatory Anaerobic ThresholdThe ventilatory anaerobic threshold (VAT) is a lab-

oratory measurement that approximates an identifica-tion of the point at which the subject's oxygen supplybegins to be outstripped by demand during an exercisetest.27 In normal individuals, a maximal aerobic exerciseeffort is reached when the cardiovascular system hasattained its maximal capacity to deliver blood to exer-cising muscles, and cardiac output can increase nofurther. Increasing peripheral oxygen extraction occursand raises the difference between systemic arterial andvenous oxygen content. As the limit of tissue oxygendelivery is reached, additional energy is provided anaer-obically by glycolysis, which results in a rise in muscleand plasma lactic acid.The anaerobic threshold is a theoretical point that

occurs during dynamic exercise when the muscle usesanaerobic metabolism as an additional source of energy.A rise in tissue lactate level slightly precedes the rise inblood lactate level. All tissues do not shift simultane-ously to anaerobic metabolism, so that this "point" intime is actually a brief interval during the test in whichexercising tissues change from predominantly aerobicmetabolism to anaerobic metabolism.

Lactate concentration in the blood results from adynamic relation between lactate production and lac-tate clearance. If blood samples are obtained through-out exercise testing and lactate concentration is mea-sured, at an observable point the blood lactate level

TABLE 3. Exercise Stroke Volume in Children With Young Adult Values Included for Comparison

Rest, mL/beat Exercise,Author, Year Age, y Sex per m2 mL/beat per m2 Instrument ProtocolEricksson, 197115 13-14 Male 61.8±6.2 86.8±3.9 Cycle ContinuousEricksson, 197323 11-13 Male 49.4 66.9 Seated cycle in Continuous

catheterizationlaboratory

Cumming, 197718 5-16 Male 51.9±5.9 55.7±12.7 Supine cycle in Intermittent,catheterization graded

laboratoryFemale 45.3±6.3 46.3±3.1

Lock, 197819 5-16 Both 52.7±2.0 57.3±2.3 Supine cycle in Fixed rate ofcatheterization work,

laboratory submaximalAstrand, 196424 Mean, 24 Male Mean, 68 ... ...

Mean, 21 Female Mean, 57

by on January 15, 2008 circ.ahajournals.orgDownloaded from

AJI4 Council on CVDY Guidelines for Pediatric Exercise Testing

TABLE 4. Values of Peak Oxygen Uptake: Vo2max WithExercise in Children

Vo2max,Protocol mL/min per kg

Stepwise loading, cycle ergometer 35.6-60.6

Supramaximal, cycle ergometer 49-55.4

Discontinuous, cycle ergometer 41.8-56.6

Other, cycle ergometer 32.6-61.4

Increasing-grade stepwise loading,treadmill 47.7-61.0

Increasing-speed stepwise loading,treadmill 45.7-58.2

Increasing-grade and increasing-speedstepwise loading, treadmill 45.9-61.3

Walking, treadmill 43.1-53.5

See Reference 26 for details. Values are given as range.

abruptly increases above resting levels. This point hasbeen termed the onset of blood lactate accumulation.

Routine blood lactate measurement is impractical inexercise testing of children. As lactate is formed, it isbuffered in the serum by the bicarbonate system, andCo2 increases. This accumulation of Co2 results in reflexhyperventilation. The onset of reflex hyperventilation,or VAT, may be described as a point in time during theexercise study or can be expressed in terms of otherexercise test measurements, such as percent of Vo2max,heart rate at VAT, or percent of total exercise time.The VAT has been proposed as a more sensitive

marker of fitness than oxygen consumption, heart rate,or total work in pediatric training.27 (See Table 5.) Theheart rate at VAT has been used to determine the targetheart rate for rehabilitation training.

In summary, the true anaerobic threshold at themuscle cell level, the onset of the blood lactate accumu-lation, and the VAT are separate but related events thatoccur during exercise and should not be equated withone another.

Age and Lactate Level TolerancePeak attainable lactate levels with exercise are higher

for adults than for children. One explanation for this inthe past has been the suggestion that the ability totolerate increased serum lactate levels and to persistwith exercise in the anaerobic state depends on thedegree of the subject's sexual maturity. This differencewas attributed to developmental limitations of the ac-

tivity of key enzymes, such as phosphofructokinase, thatare involved in anaerobic metabolism. However, a majorsubjective factor confounding the interpretation ofblood lactate developmental data seems to be howwilling the exercising subject is to tolerate the uncom-

fortable symptoms associated with fatigue. Metabolicchanges associated with anaerobic metabolism, al-though not the blood lactate itself, will contribute to thesensation of fatigue. Subjects tolerate the discomfortassociated with fatigue to different degrees. Trainedadults achieve higher peak exercise lactate levels thanthose who are untrained, and elite adult athletes oftenhave considerably higher peak exercise lactate levelsthan the typical trained adult. An alternative explana-tion is that adults at peak exertion can usually becoerced to tolerate higher levels of fatigue symptomsthan a young child, and that symptoms of fatigue aremore easily tolerated by athletes after training than bythose unaccustomed to the sensation of fatigue.

Inferences regarding the developmental aspects ofmetabolic responses to intense exercise are, for the mostpart, based on cross-sectional rather than longitudinalstudies.

Testing ProceduresErgometers

Exercise laboratories find varying applications fordifferent types of ergometers. Three types of ergome-ters commonly used are cycles, treadmills, and arm

ergometers. Both the cycle ergometer and the treadmillproduce adequate, reliable, and reproducible maximalloads on the oxygen transport system for collection ofdiagnostic and functional information. Arm ergometerscan be used by subjects for whom lower extremityexercise is difficult or impossible; from a cardiopul-monary standpoint, the intensity of stress is lower thanwith leg exercise. Since each ergometer has differentcharacteristics, the selection depends on the needs ofthe user. There are four general considerations forchoosing a particular ergometer: (1) the size of thesubject; (2) the space available for the exercise labora-tory; (3) the specific reason(s) for performing a givenexercise test; and (4) the number of variables to bemonitored and recorded during the test.

Cycle ergometers are convenient for laboratories withlimited space. Mechanically braked ergometers are por-table and relatively silent. The standard cycle ergometermay require minor modification for testing childrenunder 6 years of age. The modifications include varying

TABLE 5. Ventilatory Anaerobic Threshold in Children

Vo2 at VATStudy VAT, % Vo2max HR at VAT mL/kg per min No. of Children, yWashington, 198828 Males, 75+13 169+15 34+7 151 Children, 7.5-12.5

Females, 71+9 167±16 30±5

Cooper, 198429 60±9 ... Males, 27±6 109 Children, 6-17

Females, 19+6

Reybrouck, 198630 Males, 58-74 ... 29-35 95 Children, 5-14

Females, 61-70 24-29

VAT indicates ventilatory anaerobic threshold; Vo2max, maximum oxygen uptake; and HR, heart rate. Children'sages are given as range.

2171

by on January 15, 2008 circ.ahajournals.orgDownloaded from

2172 Circulation Vol 90, No 4 October 1994

the height of the seat, the angle of the handlebar, thelength of the pedal arm, and the size of the seat.

Because of the stability of the trunk and arms and theeasy access to the patient, most sequential measure-ments (eg, radionuclide imaging and echocardiography)during exercise testing can be made more convenientlywhen a cycle ergometer is used.The mechanically braked cycle ergometer increases

metabolic requirements by friction. A belt surroundingthe flywheel of the ergometer is adjusted to vary theresistance against which the individual pedals. Mechan-ically braked cycle ergometers are less expensive thaneither electronically braked cycle ergometers or tread-mills, but they require frequent calibration. The pedal-ing speed must be constant to assure an unchanged rateof work. A metronome or other regular auditory soundsource is useful in maintaining pedaling frequency.The electronically braked ergometer delivers a con-

stant rate of work throughout a range of pedalingspeeds (generally 30 through 70 rpm). This is accom-plished through a feedback system that increases theresistance during slow pedaling speeds and decreasesthe resistance during faster pedaling speeds. This typeof ergometer is more expensive and requires specialequipment for calibration. The efficiency of exercise(and thus oxygen consumption) varies with pedalingrate. At very low pedaling speeds with electronicallybraked cycle ergometers, the braking force increasesand the work becomes more isometric and less aerobic.

Treadmills are usually noisier than cycle ergometersand require a larger space and a ceiling height that isadequate to accommodate tall subjects. The mechanicalefficiency of running is affected by body size and weight,gait, and length of stride. Modifications of the treadmillmay include an adjustment of speed for small childrenwho have shorter strides and should include side railsand or safety straps to prevent falling. A padded matshould be placed at the end of the treadmill for thesubject's protection. Because of the higher Vo2 ob-tained, some physicians prefer treadmill testing to cycleergometry for assessing the patient undergoing a car-diac rehabilitation training program. The type of er-gometer selected should be appropriate for the type oftraining; eg, a treadmill for a jogger and a cycle ergom-eter for a cyclist. A treadmill is easy to calibrate withoutspecial equipment.The requirement for coordination between the child

and the test personnel and the potential problem ofmovement artifacts during treadmill exercise make ittechnically more difficult to perform repeated measure-ments and specialized techniques than with a cycle. It ismore difficult to obtain consistent and accurate bloodpressure determinations with treadmill testing than withcycle ergometry.

ElectrocardiogramThe heart rate can be measured manually by averag-

ing several R-R intervals. Alternatively, the electrocar-diographic signal can be processed through a tachome-ter, and a direct recording of the heart rate based on

each R-R interval can be obtained. Many commerciallyavailable electrocardiographic units are equipped withsuch a tachometer.

Appropriate application of the electrocardiographicelectrodes and shielding of the cables is critical for

obtaining artifact-free electrocardiographic recordings.Movement artifacts occur when the electrode-electro-lyte interface is mechanically disturbed, resulting in avoltage change. Electrodes for exercise testing avoidmetal-to-skin contact by "floating" the electrode on anelectrolyte-gel mixture to minimize artifact from musclemovement. A large contracting muscle mass betweentwo electrodes will result in an artifact that cannot befiltered. It is occasionally necessary to move electrodesto avoid large muscle groups even though the leads areplaced on the torso. Although this may require somerepositioning of electrodes for optimal recordings whentesting a heavily-muscled adolescent, it is usually not amajor problem with testing smaller children. Silver-silver chloride electrodes have very low impedance foroptimal recording of low frequency electrical phenom-ena and perform better than the less expensive metalsthat have been used to fabricate electrodes. Electrodesthat are too small will have an excessively high imped-ance; the large adult electrodes will not provide goodcontact on a small child and will not stay on well. Thequality of electrode adhesion, which is essential forquality recordings in children, varies among the brandsavailable.

Skin preparation is the single most important factorin obtaining precise exercise electrocardiographic trac-ings from children. The epidermis possesses a layer ofdead skin cells (stratum corneum) with a very highelectrical resistance. The dermis itself has very lowelectrical resistance. Skin preparation should be done asgently as possible to avoid frightening the child. Clean-ing with alcohol or acetone should be followed by gentleabrasion and removal of the stratum corneum beforeelectrode placement. This can be done by brisk rubbingwith rough cloth, fine sandpaper, emery board, skincleaning gel containing a mild abrasive, dental drill, orcommercially available electrode placement guns. Ob-taining well-defined and reliable electrocardiographicrecordings in exercising children is impossible withoutmeticulous attention to skin preparation technique.The right and left arm electrodes are placed in the

right and left infraclavicular fossae. The left leg elec-trode is placed on the lower abdomen above the leftanterior superior iliac crest. The right leg lead may ormay not be used on the lower abdomen. Some labora-tories place the lower limb leads higher on the torso;this higher placement is of particular value in the testingof obese children. Precordial leads Vi through V6 areplaced in the same location as for a routine restingelectrocardiogram. The individual electrodes and theelectrocardiographic cable should be secured to thesubject's body to minimize artifact from movement ofthe cables during exercise. This may be accomplished byusing a commercially available knit shirt that is cus-tom-fit for the patient or by using tape.At least 3 leads of the standard 12-lead electrocar-

diogram should be displayed or recorded continuouslyat baseline, during testing, and for the first 10 minutesafter the completion of an exercise test. Ideally, theexaminer should have the option of viewing variouscombinations of leads, including at least one inferiorlead (standard leads II, III, or aVF), one anterior rightlead (V1 or V2), and one lateral left lead (Vs or V6). Acomplete 12-lead electrocardiogram should be recorded

by on January 15, 2008 circ.ahajournals.orgDownloaded from

AA4 Council on CVDY Guidelines for Pediatric Exercise Testing

before the test, at maximal exercise, and during therecovery period.

Cardiac Output MeasurementSeveral techniques have been used for noninvasively

measuring cardiac output during exercise.30 Cautionshould be used when comparing cardiac outputs mea-sured by different techniques and in different laborato-ries. These measurements are sensitive to small changesin methodology and exercise protocols.The techniques used most frequently to measure

cardiac output noninvasively and without the need forradioactive materials are Co2 rebreathing31 and acety-lene-helium rebreathing.32 When these measurementsare obtained in the gas phase, the Fick technique istermed "indirect." When using Co2, the cardiac outputis measured by analysis of expired gases. Arterial PCo2can be estimated by using the end-tidal PCo2. Mixedvenous content can be estimated by using a Co2-re-breathing procedure in which Co2 concentration in therebreathing bag equilibrates with the alveolar air andthe mixed venous blood.An alternate method for measuring cardiac output

uses inert gases that do not interact with the blood andare thereby transported in purely physical solutionrather than bound to carrier molecules such as hemo-globin. Transport of these gases is said to be "bloodflow-limited." When an inert gas is introduced into thelungs at a given concentration, that concentration fallsat a measurable and exponential rate as a result of theuptake of that gas by the blood flow. Commonly usedgases for this technique are acetylene and nitrous oxide.Doppler echocardiography has been attempted by

some exercise physiologists to assess cardiac output andprovide beat-to-beat measurements of stroke volumeand cardiac output.None of these noninvasive techniques offers compel-

ling advantages or disadvantages, and the 10% variationobservable from one test to another is comparable tothe variability found with invasive cardiac outputdeterminations.31

OximetryNoninvasive (ear or finger oximetry) measurement of

blood oxygen saturation can be useful during exercisetesting of children who have congenital heart disease toeither determine the presence or absence of hypoxemiaor to quantify the degree of hypoxemia during exercise.Although concerns have been raised about the reliabil-ity of this technology with exercise testing,34 there isgenerally good reliability of oxygen saturations mea-sured by pulse oximetry if the oximeter heart ratetracking is acceptable during exercise.35 Normal chil-dren maintain oxygen saturations greater than 90%during maximal exercise when monitored by pulse oxi-metry. Desaturation (less than 90%) during exercise isconsidered an abnormal response and may reflect pul-monary, cardiac, or circulatory compromise. However,some elite adult endurance athletes can desaturate toless than 90% with intense exercise, presumably due topulmonary ventilatory-perfusion mismatch with intenseexercise.Good surface contact of the sensor electrode is vital

for accurate measurements when using oximetry duringexercise testing. If finger oximetry is used, the subject

must be instructed not to make a "tight fist," as this willrender the measurements inaccurate.

Blood Pressure MeasurementBlood pressure can be measured indirectly by using a

cuff, a sphygmomanometer, and a stethoscope to detectthe Korotkoff sounds. During exercise the Korotkoffsounds often are difficult to hear. For this reason, theonset of the fifth Korotkoff phase is used to indicatediastolic blood pressure for children during exercisetesting. At peak effort, however, the onset of the fifthKorotkoff phase may not be heard in some children,with sounds still audible at 0 mm Hg, which makesaccurate measurement of diastolic pressure impossible.Numerous commercial electronic units have been avail-able for measuring blood pressure during exercise test-ing.24 Although the accuracy of some of these deviceshas been documented, many of those available areinadequate for accurate blood pressure determination.

Several different cuff sizes should be available. Thebladder of the cuff should completely encircle the armand the width of the bladder should be at least twothirds the length of the upper arm. The most commonsource of error in measuring blood pressure of childrenis the use of a cuff that is too small.

In most normal subjects systolic blood pressure de-clines rapidly after maximal exercise, usually reachingresting levels within 6 minutes or less. In some individ-uals blood pressure remains lower than pre-exerciselevels for several hours. Some exercise physiologistshave their subjects continue to perform low-level exer-cise for several minutes in the postexercise period tominimize venous pooling and resultant symptomatichypotension. Others measure the postexercise physio-logical data with their patients in the supine position tominimize orthostatic symptoms caused by low venousreturn from blood pooling in dilated vascular beds.

Respiratory Gas MeasurementVentilation and pulmonary gas exchange may be

measured during exercise by collecting the expired gasand determining the volume and fractional concentra-tions of oxygen and carbon dioxide in the inspired andexpired gas phases. This method yields minute ventila-tion, oxygen uptake, and carbon dioxide production.The respiratory gas exchange ratio (VCoJVo2) and theoxygen ventilatory equivalent (VE/VO2) can be calcu-lated. With the addition of flow measurement with aturbine or a pneumotachograph, the average respira-tory rate and tidal volumes can be measured. Thesemeasurements are relatively simple to perform andaffordable for most clinical laboratories.25'35Rapid gas analyzers for oxygen and carbon dioxide

are now commercially available. Commercially manu-factured metabolic carts with on-line computers are alsoavailable, but these must be adapted for pediatricsubjects. The most common systems use the average offractional concentration of collected gas over time oranalyze the gas of each breath-the so-called "breath-by-breath technique." Care must be used to assure thatthe mask or mouthpiece fits properly and that there areno air leaks. Closely fitting masks may be more suitablefor smaller children who do not easily tolerate a mouth-piece and nose clip.

2173

by on January 15, 2008 circ.ahajournals.orgDownloaded from

2174 Circulation Vol 90, No 4 October 1994

Determining the VentilatoryAnaerobic Threshold

Several techniques have been used to determine theventilatory anaerobic threshold (VAT), which requiresthe measurement of oxygen consumption and carbondioxide production. The ventilatory equivalent for oxy-gen and the ventilatory equivalent for carbon dioxideare plotted against time; initially, the two ventilatoryequivalents increase linearly during graded exercisetesting. The VAT is that point at which the slope for theventilatory equivalent for carbon dioxide separates fromthe slope for the ventilatory equivalent for oxygen. Atthis same point, minute ventilation also increases. Ap-proximately 20% of individuals will not demonstrate aclear-cut VAT because of erratic breathing.27

General Principles of Exercise TestingReasons for Exercise Testing

Exercise testing evaluates work performance andidentifies mechanisms that limit the work performanceof children with cardiac and other problems. Exercisetesting is performed for a variety of reasons: (1) toevaluate specific symptoms or signs that may be inducedor aggravated by exercise; (2) to identify abnormaladaptive responses occurring in children with cardiac orother disorders; (3) to assess the effectiveness of specificmedical and surgical treatments; (4) to estimate levelsof functional capacity for and to improve the safety ofvocational, recreational, and athletic recommendations;(5) to estimate prognosis; (6) to evaluate fitness levels;and (7) to establish baseline data and follow up effec-tiveness of cardiac rehabilitation.A properly supervised and monitored exercise test in

children can be conducted with very low risk.4,36,37 Theprecautions listed below will help to assure the safety ofthe tests.

Physical EnvironmentAn exercise laboratory is ideally at least 500 square

feet (46.4 m2) and must have adequate ventilation. Anambient temperature of 22°C through 24°C with approx-imately 50% humidity provides a satisfactory environ-ment for testing. The number of people in the room, theheat generated by the equipment, and the size of theroom are important considerations for optimal temper-ature control in the exercise laboratory. The noise levelmust be kept to a minimum. Smoking should be prohib-ited in the vicinity of the laboratory. There should be awide exit door from the laboratory to expedite patienttransport if needed.

Laboratory StaffingAt least one physician must assume the responsibility

of directing the laboratory. This physician should havetraining both in exercise testing in general and perform-ing exercise tests in young patients with disorders ofvarying severity. The director assures (1) that labora-tory personnel are adequately trained in the mechanicsof testing and in emergency procedures; (2) that equip-ment works properly; (3) that acceptable testing proce-dures are used; and (4) that reliable results are con-veyed to the referring physician, the patient, and thefamily.

In most instances, after proper screening of the child,the physician may delegate the actual conduct of theexercise test to personnel who have both an understand-ing of the pathophysiological responses to exercisetesting and expertise in emergency procedures. In otherinstances, the physician may elect to supervise theexercise test. Two or more competent personnel withtraining in pediatric cardiopulmonary resuscitation arerequired to perform an exercise test properly. A physi-cian and other medical personnel and the standardpediatric resuscitative equipment must be immediatelyavailable to the laboratory in the event of an emergency.These recommendations are intended for diagnosticexercise testing. Maximal exercise testing of normalpediatric age volunteers for research purposes cansafely be done by individuals with American College ofSports Medicine certification (or with equivalent train-ing in pediatric exercise testing) without direct physiciansupervision of the test procedure.

ConsentThe laboratory's personnel should respect the right of

the subject, parent, or guardian to be informed aboutthe test and the associated risks. A meaningful writtenconsent for the procedure is obtained by most labora-tories. The subject should always be allowed to termi-nate the test upon request.

Patient PopulationEach patient should be given a general explanation of

the laboratory equipment and the test procedure. Thepatient should be instructed to avoid food intake for atleast 2 hours before the study and to wear comfortableclothing and shoes, preferably flexible sports shoes. Ahistory that includes a review of medications, a physicalexamination, and a resting supine 12-lead electrocar-diogram provide information that will identify individ-uals for whom the test is either contraindicated orshould be performed only with special considerations. Acardiologist may be consulted before testing if thepatient's symptoms suggest a cardiovascular disorder.

Relative Contraindications andSpecial Considerations

Previous publications concerning pediatric exercisetesting have included contraindications, relative con-traindications, and special considerations that wereextrapolated from standards and guidelines for adultexercise testing.24 As with any stress-inducing proce-dure, the need for the information to be obtained byexercise testing must be carefully weighed against itsrisks. However, even for children whose diagnoses placethem in a high-risk group there may be compellingreasons to perform exercise testing. For example, youngpatients, even those for whom limitations have beenrecommended on competitive activity, may be morephysically active when unsupervised than their cardiol-ogist or parents are aware. A carefully observed labo-ratory exercise test performed under controlled circum-stances can be safer than the unsupervised activity ofhigh physical intensity on the playground. Althoughserious cardiac decompensation with exercise testingcan occur in adults, it is extremely rare in children.

Appropriate monitoring, equipment for the manage-ment of complications of testing, training of laboratory

by on January 15, 2008 circ.ahajournals.orgDownloaded from

AHA Council on CVDY Guidelines for Pediatric Exercise Testing

personnel in pediatric basic and advanced life support,staff experience in test performance and interpretation,and the availability of medical staff backup are allrequired for the safe diagnostic testing of childrenreferred to a hospital laboratory. A physician's presenceduring all pediatric exercise tests, however, is not rou-tine. For these guidelines, we have provided examples ofsituations in which the physician's availability but notactual presence is required. Such situations may include

(1) the assessment of physical working capacity inhealthy children for research purposes

(2) evaluation of chest pain that is not thought to beanginal in origin

(3) follow-up of postoperative patients assessed tohave good hemodynamics after repair with testing toevaluate physical working capacity, to screen for ar-rhythmias, or as a baseline for cardiac rehabilitationprograms

(4) evaluation of premature atrial or ventricular con-tractions in an otherwise healthy child (only if the QTinterval is normal)

(5) routine follow-up of children with known arrhyth-mias or with permanently implanted artificialpacemakers

(6) follow-up of patients with previous Kawasakidisease or with other known coronary abnormalities asa screening for ischemic electrocardiographic changeswith exercise (excluding those with giant coronary an-uerysms or known ischemia)

(7) testing of asymptomatic patients who have beenassessed as having mild to moderate aortic stenosis byclinical or echocardiographic criteria

(8) evaluation of physical working capacity in asymp-tomatic or minimally symptomatic cases with othercongenital or acquired cardiac malformations.

Other situations may occur in which testing by theexercise laboratory personnel may be appropriate afterreview by the physician. Criteria for such situationsshould be established for each laboratory based on thecapabilities of the technical personnel and the charac-teristics of the referral population. The procedure toorder testing in the laboratory should permit the refer-ring physician to request physician supervision if thereis reason for special concern.However, pediatric exercise testing is by no means

risk free. In some situations the risk of maximal exercisetesting is of such great magnitude that the significanceof the information to be obtained from the test cannotoutweigh the threat to the child's care. Examples ofdiagnoses for which the risk of exercise testing is likelyto result in a decision to defer or cancel testing include

(1) severe pulmonary vascular disease(2) poorly compensated congestive heart failure(3) recent myocardial infarction(4) active rheumatic fever with carditis(5) acute myocarditis or pericarditis(6) severe aortic stenosis(7) severe mitral stenosis(8) unstable arrhythmia, especially with compromised

hemodynamics(9) Marfan syndrome with suspected aortic dissection(10) uncontrolled, severe hypertension(11) evidence of hypertrophic cardiomyopathy with a

history of syncope

Indications to Terminate an Exercise TestThere are three general indications to terminate an

exercise test: when diagnostic findings have been estab-lished or a predetermined end point has been reached,when monitoring equipment fails, or when signs orsymptoms indicate a potential hazard to the patient thatmay result in injury. Warning symptoms include pain,headache, dizziness, syncope, excessive dyspnea, orfatigue. Signs indicating test termination include STsegment depression or elevation greater than 3 mm;significant arrhythmia precipitated or aggravated by theexercise testing, such as premature ventricular contrac-tions with increasing frequency, supraventricular tachy-cardia, ventricular tachycardia, or atrioventricular con-duction block; or a progressive decrease in bloodpressure.

Exercise Studies in Specific Disease StatesExercise studies provide useful clinical information

for monitoring response to therapy and about fitness,the potential for inducing arrhythmias, cardiac isch-emia, and blood pressure response. Diseases for whichstudies may be useful include congenital heart disease,acquired valvular heart disease, cardiomyopathy,chronic lung disease, Kawasaki disease, systemic orpulmonary hypertension, and sickle cell disease. Exer-cise studies can be individualized by varying protocols toanswer specific diagnostic questions. Table 6 outlinessome findings from exercise studies in several specificcardiac diseases and provides information on expectedresponses, pathological responses, and clinical use.

ProtocolsBecause performance on exercise tests largely de-

pends on the type of exercise and the testing protocol,exercise testing should be performed in a standardizedand systematic manner. Standardized protocols makecollaborative research efforts possible as well as permitthe reliable comparison of results of repeated testing ofthe same individual, of individual test findings withestablished norms, and of tests performed in differentlaboratories.Most protocols in wide use involve a progressive

increase in rates of work without a resting periodbetween changes of work rate increments. Intermittentor discontinuous graded protocols involve a rest periodbetween each change in rate of work. Although cum-bersome for routine diagnostic testing, these discontin-uous protocols have value in research projects in whichcomparatively long durations at steady state are re-quired for data accumulation. Several protocols usingthe cycle ergometer or treadmill have been used inchildren.38'39 Pediatric exercise testing laboratories mustaccommodate subjects of various sizes and ages. Thebest exercise testing protocol for children depends onthe information required and the age, health, andfitness level of the subject. The familiarity of thelaboratory staff with specific treadmill and cycle proto-cols must be considered. The protocols described hereare widely used in pediatric cardiac centers. Because ofsubtle variations in testing procedures, it is optimal touse normal data generated in one's own laboratoryrather than to rely solely on published normative data.

2175

by on January 15, 2008 circ.ahajournals.orgDownloaded from

2176 Circulation Vol 90, No 4 October 1994

TABLE 6. Exercise Test Findings in Patients With Cardiac Abnormalities

Lesion Common FindingsAortic stenosis None

Postoperative aortic stenosis

Aortic insufficiencyPulmonary stenosisPostoperative pulmonary stenosisAortic coarctation, preoperative orpostoperative

Atrial septal defectVentricular septal defect

Mitral regurgitation

Mitral prolapse

Hypertension

Cyanotic heart disease, palliated

Postoperative tetralogy of Fallot

Postoperative arterial switchprocedurePostoperative Mustard procedure

Postoperative Fontan

Kawasaki disease

Cardiomyopathy

Postoperative cardiac transplant

Cardiac pacemakers

ST changes may be seen without predictingresidual gradient

Same as postoperative ASNone expectedNone expectedHypertension

Arm-to-leg BP gradient

NormalNone expected

Low working capacity if large shuntNone expected

False-positive ST changes common; STchanges with hyperventilationNormal rise in BP with exertion;BP levels above normal for

rate of work

Low chronotropic responseDesaturation with exerciseLow working capacity

Decreased chronotropic responseLow physical capacity

Right bundle branch blockUncertain significance of suggestions of

ischemiaDecreased chronotropic response

Low physical capacityST changes in right precordial leads

Low physical working capacityDesaturation

Drop in BP during exerciseNone expected

Low physical working capacityLow peak HR; low Vo2Slow rise in heart rate

Normal working capacitySlightly low peak HR; normal Vo2

High postexercise HRWenkebach at upper rate limit

Findings That WarrantMore Concern

Ischemic ST changesDecreased BP responseExertional symptoms

ArrhythmiaLow V02

Decreased BP responseSame as postoperative ASRare with very severe PS

None expectedIschemia

Diminished working capacitySevere hypertension

None expectedArrhythmiaDesaturation

Low working capacityIschemia

Arrhythmia

Ischemia

ArrhythmiaLow working capacityProfound desaturation

Ischemia

Neurologic changesArrhythmiasDrop in BPDesaturation

Limited long-term data

ArrhythmiaDesaturation

Profound cyanosisArrhythmia

AnginaAbnormal wall motionIschemic ST changes

Perfusion abnormality (nuclear)Syncope, light-headedness

lschemic ST changeslschemic ST changesLow working capacity

Abnormal (over or under) sensingVentricular arrhythmias

BP indicates blood pressure; AS, aortic stenosis; PS, pulmonary stenosis; and HR, heart rate.

-

by on January 15, 2008 circ.ahajournals.orgDownloaded from

AHA Council on CVDY Guidelines for Pediatric Exercise Testing

Treadmill ProtocolsOn the treadmill, rate of work is advanced by in-

creases in either slope, usually expressed as a percent,belt speed, or both. Percent slope indicates the numberof feet in elevation for each 100 feet of distancetraveled.

Bruce ProtocolCumming et al10 have provided physiological data and

endurance time norms for the Bruce protocol whenused for children between 4 and 14 years of age. Thiscontinuous graded protocol uses stages at which thespeed and grade are progressively increased at 3-minuteintervals from 1.7 through 6 mph and from 10% through22%, respectively. The goal is to reach the level ofmaximal voluntary effort at which the subject is unableto continue.The advantages of this protocol are that it may be

used for all ages. Physiological responses to submaximalworkloads can be measured, and longitudinal data canbe obtained in a given subject by using the sameprotocol as he or she grows.The potential disadvantages include that the test is

longer than some other protocols, and younger subjectsmay become bored. This protocol may be inconvenientfor testing highly fit individuals because subjects mustwait until 12 minutes into the test before beginning aslow run (and then at an 18% slope).

Balke ProtocolThis type of treadmill protocol incorporates a con-

stant treadmill speed with increasing slope. Riopel etall' report normal heart rate and blood pressure data inhealthy children tested with a modified Balke protocol,which progressively increases the grade from 2% tomore than 10% at 2% increments until exhaustion whilemaintaining a constant speed of 3.5 mph. This continu-ous protocol is well suited for the unfit, the obese, thevery young child, or the chronically ill individual.A potential disadvantage is that active, fit, young

subjects may find the test duration too long and theslope too shallow. The test may be modified and short-ened by beginning the protocol at a high initial grade(between 6% and 10%). The speed can be adjusted tofit the subject's age and fitness level as another means ofcontrolling the test time. When a laboratory changes thestandard protocol, it is imperative that normative databe obtained on healthy matched controls.

Cycle ProtocolsMost of the various cycle protocols call for a cadence

rate of between 50 and 60 rpm, different stage lengths(between 1 and 3 minutes), and methods for adjustingthe initial and incremental work loads.

James ProtocolSpecific exercise protocols are based on body surface

area. When three progressive stages (each stage is 3

minutes long) have been successfully completed, workload is increased by either 100 or 200 kg-M/min until amaximal level of voluntary effort is reached, and then thetest is terminated. The goals of this protocol are to reacha state of exhaustion or predetermined end point, toestimate the highest rate ofwork (maximal power output),and to measure other physiological variables. Normativedata have been provided by James et al16 and Washingtonet al.28

Godfrey ProtocolThis continuous protocol consists of a specific exer-

cise program for each of three height groups. Workloads are increased at intervals of 1 minute until a levelof exhaustion is reached. The maximal power output orhighest rate of work achieved is measured and com-pared with normal values for size and gender. Godfreyhas published normative pediatric data obtained byusing this protocol.22

Strong ProtocolExercise programs are designed for each of four

groups according to body weight. The work load withineach program is estimated to produce a heart rate at ornear the expected value after successful completion oflevels 1 through 3. Subsequent work loads are increasedat increments of 75 to 300 kg-M/min at 1-minuteintervals according to the size and current performanceof the subject. The goal of this protocol is to determinephysical working capacity at a heart rate of 170 beatsper minute and to determine the highest rate of workproducing exhaustion or predetermined end points.Normative data have been provided by Alpert et al.'7,25

SummaryExercise testing of children differs from adult exercise

testing in many ways beyond the technical issues relatedto test performance that are addressed in this report.Disease processes that produce myocardial ischemia arerelatively rare in children compared with adults. Exer-cise testing may be useful in these cases, but the use oftesting to assess functional capacity or cardiac rhythmswill be encountered more often. Although the preciserole of exercise testing in patient evaluation or long-term management of the cardiac patient will varysomewhat from center to center, exercise testing is oftenessential to diagnose and to direct treatment in a widevariety of clinical problems. An understanding of therole of exercise testing for children with known orsuspected heart abnormalities is an essential part of thetraining of pediatric cardiologists. The staff of thepediatric exercise laboratory should be available todiscuss with the clinician when a test might be of valuein a specific case in addition to providing advice aboutthe specifics of the performance of the test and offeringage- and size-appropriate normal data from the labora-tory with test interpretation.

2177

by on January 15, 2008 circ.ahajournals.orgDownloaded from

2178 Circulation Vol 90, No 4 October 1994

SamplePediatric Exercise LaboratoryConsent for Exercise Testing

In order to evaluate the functional performance and capacity of the heart, lungs, and blood vessels, I hereby consentvoluntarily to the performance of an exercise test on my child, . I understand that the patient must walkor run on a treadmill, or pedal a bicycle until the limits of fatigue, breathlessness, chest pain, and/or other symptoms are ofsuch severity that the test should be stopped. The electrocardiogram will be continuously monitored and blood pressure will betaken intermittently during exercise and for a prescribed time following exercise. Occasionally other indicated measurementswill be performed. A physician will be present or in the immediate area.

I understand that the potential risks of testing include changes in the rhythm of the heart and the possibility of excessivechanges in blood pressure. Thus, there is a remote chance of fainting and an unlikely chance of a heart attack. I understandthat professional supervision by an Exercise Physiologist or physician will protect my child against injury, by providingappropriate precautionary measures. In the event that the precautionary measures are insufficient, hospital treatment will beavailable for my child.

I understand that the benefits of testing include quantitative assessment of working capacity and critical appraisal of thedisorders or diseases that impair capacity, and that this knowledge facilitates better treatment and more accurate prognosis forfurther cardiac events.

I have been assured that I have the right to withdraw my child from the test at any time, that confidential information aboutthe test result will not be given to non-medical persons without my consent, and that the welfare of my child will be protected.The procedures and risks of the graded exercise test have been explained to me and I have read this form and have had an

opportunity to ask any questions pertaining to the test. I understand the test procedures that will be performed and give myconsent voluntarily.Date:

Parent or Guardian

Witness

References1. Fletcher GF, Froelicher VF, Hartley LH, Haskell WL, Pollock

ML. Exercise standards: a statement for health professionals fromthe American Heart Association. Circulation. 1990;82:2286-2322.

2. James FW, Blomqvist CG, Freed MD, Miller WW, Moller JH,Nugent EW, Riopel DA, Strong WB, Wessel HU. Standards forexercise testing in the pediatric age group: American Heart Asso-ciation Council on Cardiovascular Disease in the Young: ad hoccommittee on exercise testing. Circulation. 1982;66:1377A-1397A.

3. Bar-Or 0, Ward D. Rating of perceived exertion in children. In:Bar-Or 0, ed. Advances in Pediatric Sport Sciences: BiologicalIssues. Champaign, Ill: Human Kinetics Publishers; 1989:151-168.

4. Tomassoni TL. Conducting the pediatric exercise test. In: RowlandTW, ed. Pediatric Laboratory Exercise Testing: Clinical Guidelines.Champaign, Ill: Human Kinetics Publishers; 1993:1-17.

5. Shephard RJ, Allen C, Bar-Or 0, Davies CTM, Degre S, HedmanR, Ishii K, Kaneko M, LaCour JR, Di Prampero PE, Seliger V.The working capacity of Toronto schoolchildren. Can MedAssoc J.1969;100:560-566, 705-714.

6. Skinner JS, Bar-Or 0, Bergsteinova V, Bell CW, Royer D, BuskirkER. Comparison of continuous and intermittent tests for deter-mining maximal oxygen intake in children. Acta Paediatr ScandSuppl. 1971;217:24-28.

7. Hermansen L, Oseid S. Direct and indirect estimation of maximaloxygen uptake in pre-pubertal boys. Acta Paediatr Scand Suppl.1971; 217:18-23.

8. Wilmore JH, McNamara JJ. Prevalence of coronary heart diseaserisk factors in boys, 8 to 12 years of age. JPediatr. 1974;84:527-533.

9. Wilmore JH, Constable SH, Stanforth PR, Tsao WY, Rotkis TC,Paicius RM, Mattern CM, Ewy GA. Prevalence of coronary heartdisease risk factors in 13- to 15-year-old boys. J Cardiac Rehab.1982:2:223-233.

10. Cumming GR, Everatt D, Hastman L. Bruce treadmill test inchildren: normal values in a clinic population. Am J Cardiol.1978;41:69-75.

11. Riopel DA, Taylor AB, Hohn AR. Blood pressure, heart rate,pressure-rate product and electrocardiographic changes inhealthy children during treadmill exercise. Am J Cardiol. 1979;44:697-704.

12. Astrand P-O. Experimental Studies of Physical Working Capacity inRelation to Sex and Age. Copenhagen, Denmark: Munksgaard; 1952.

13. Goldberg SJ, Weiss R, Adams FH. A comparison of the maximalendurance of normal children and patients with congenital cardiacdisease. J Pediatr. 1966;69:46-55.

14. Wilmore JH, Sigerseth PO. Physical work capacity of young girls,7-13 years of age. J Appl PhysioL 1967;22:923-928.

15. Eriksson BO, Grimby G, Saltin B. Cardiac output and arterialblood gases during exercise in pubertal boys. J Appl Physiol.1971;31:348-352.

16. James FW, Kaplan S, Glueck CJ, Tsay JY, Knight MJS, Sarwar CJ.Responses of normal children and young adults to controlledbicycle exercise. Circulation. 1980;61:902-912.

17. Alpert BS, Flood NL, Strong WB, Dover EV, DuRant RH, MartinAM, Booker DL. Responses to ergometer exercise in a healthybiracial population of children. J Pediatr. 1982;101:538-545.

18. Cumming GR. Hemodynamics of supine bicycle exercise in normalchildren. Am Heart J. 1977;93:617-622.

19. Lock JE, Einzig S, Moller JH. Hemodynamic responses to exercisein normal children. Am J Cardiol. 1978;41:1278-1284.

20. Bricker JT. Pediatric exercise electrocardiography. In: RowlandTW, ed. Pediatric Laboratory Exercise Testing: Clinical Guidelines.Champaign, Ill: Human Kinetics Publishers; 1993:43-65.

21. Thapar MK, Strong WB, Miller MD, Leatherbury L, Salehbhai M.Exercise electrocardiography of healthy black children. Am J DisChild. 1978;132:592-595.

22. Godfrey S. Exercise Testing in Children: Applications in Health andDisease. Philadelphia, Pa: WB Saunders Co; 1974.

23. Eriksson BO, Koch G. Cardiac output and intra-arterial bloodpressure at rest and during submaximal and maximal exercise in 11-13year old boys before and after physical training. In: Bar-Or 0, ed.Pediatric Work Physiology. Natanya, Israel: Wingate Institute;1973:139-150.

24. Astrand PO, Cuddy TE, Saltin B, Stenberg J. Cardiac outputduring submaximal and maximal work. J Appl Physiol.1964;19:268-274.

25. Alpert BS, Fox ME. Blood pressure response to dynamic exercise.In: Rowland TW, ed. Pediatric Laboratory Exercise Testing: ClinicalGuidelines. Champaign, Ill: Human Kinetics Publishers;1993:67-90.

26. Freedson PS, Goodman TL. Measurement of oxygen consumption.In: Rowland TW, ed. Pediatric Laboratory Exercise Testing: ClinicalGuidelines. Champaign, Ill: Human Kinetics Publishers;1993:91-129.

27. Washington RL. Anaerobic threshold. In: Rowland TW, ed.Pediatric Laboratory Exercise Testing: Clinical Guidelines.Champaign, Ill: Human Kinetics Publishers; 1993:115-130.

28. Washington RL, Van Gundy JC, Cohen C, Sondheimer HM,Wolfe RR. Normal aerobic and anaerobic exercise data for NorthAmerican school-age children. J Pediatr. 1988;112:223-233.

L%

by on January 15, 2008 circ.ahajournals.orgDownloaded from

AIIA Council on CVDY Guidelines for Pediatric Exercise Testing

29. Cooper DM, Weiler-Ravell D, Whipp BJ, Wasserman K. Aerobicparameters of exercise as a function of body size during growth inchildren. JAppl Physiol. 1984;56:628-634.

30. Reybrouck T, Ghesquiere J, Weymans M, Amery A. Ventilatorythreshold measurement to evaluate maximal endurance per-formance. Int J Sports Med. 1986;7:26-29.

31. Washington RL. Measurement of cardiac output. In: RowlandTW, ed. Pediatric Laboratory Exercise Testing: Clinical Guidelines.Champaign, Ill: Human Kinetics Publishers; 1993:131-140.

32. Zeidifard E, Silverman M, Godfrey S. Reproducibility of indirect(CO 2) Fick method for calculation of cardiac output. J AppiPhysiol. 1972;33:141-143.

33. Driscoll DJ, Staats BA, Beck KC. Measurement of cardiac outputduring exercise: a review. Pediatr Exerc Sci. 1989;1:102-115.

34. Curtis S, Orenstein D, Nixon P, Ross E. Accuracy of SaO2monitors during exercise testing in pediatric patients with cysticfibrosis. Am Rev Respir Dis. 1988;137:305. Abstract.

35. Orenstein DM. Assessment of exercise pulmonary function. In:Rowland TW, ed. Pediatric Laboratory Exercise Testing: ClinicalGuidelines. Champaign, Ill: Human Kinetics Publishers; 1993:141-163.

36. Alpert BS, Verrill DE, Flood NL, Boineau JP, Strong WB. Com-plications of ergometer exercise in children. Pediatr Cardiol.1983;4:91-96.

37. Freed M. Exercise testing in children: a survey of techniques andsafety. Circulation. 1981;64(suppl IV):IV-278. Abstract.

38. Cumming GR, Langford S. Comparison of nine exercise tests usedin pediatric cardiology. In: Binkhorst RA, Kemper HCG, SarisWHM, eds. Children and Exercise XI. Champaign, Ill: HumanKinetics Publishers; 1985:58-68.

39. Rowland TW. Aerobic exercise testing protocols. In: Rowland TW,ed. Pediatric Laboratory Exercise Testing: Clinical Guidelines.Champaign, Ill: Human Kinetics Publishers; 1993:19-41.

2179

by on January 15, 2008 circ.ahajournals.orgDownloaded from