Clinical Exercise Testing

62 Revill SM, Singh SJ, Morgan MD: Random- 76 Sueta CA, Gheorghiade M, Adams KF Jr, 89 Flaherty KR, Wald J, Weisman IM, Zeballos ized controlled trial of ambulatory oxygen and Bourge RC, Murali S, Uretsky BF, Pritzker RJ, Schork MA, Blaivas M, Rubenfire M, an ambulatory ventilator on endurance exer- MR, McGoon MD, Butman SM, Grossman Martinez FJ: Unexplained exertional limita- cise in COPD. Respir Med 2000;94:778–783. SH, et al: Safety and efficacy of epoprostenol in tion. Characterization of patients with a mito- patients with severe congestive heart failure. chondrial myopathy. Am J Respir Crit Care 63 National Heart, Lung and Blood Institute. Na- Epoprostenol Multicenter Research Group. Med 2001;164:425–432. tional Asthma Education Program. Expert Pa- Am J Cardiol 1995;75:34A–43A. nel Report: Guidelines for the Diagnosis and 90 Wasserman K, Zhang YY, Gitt A, Belardinel- Management of Asthma. National Heart, Lung 77 Kadikar A, Maurer J, Kesten S: The six-minute li R, Koike A, Lubarsky L, Agostoni PG: Lung and Blood Institute. National Asthma Educa- walk test: A guide to assessment for lung trans- function and exercise gas exchange in chronic tion Program. Expert Panel Report. J Allergy plantation. J Heart Lung Transplant 1997;16: heart failure. Circulation 1997;96:2221– Clin Immunol 1991;88:425–534. 313–319. 2227. 64 National Heart, Lung and Blood Institute. Na- 78 Sciurba FC, Rogers RM, Keenan RJ, Slivka 91 Robbins M, Francis G, Pashkow FJ, Snader tional Asthma Education and Prevention Pro- WA, Gorcsan J, 3rd, Ferson PF, Holbert JM, CE, Hoercher K, Young JB, Lauer MS: Venti- gram. Expert Panel Report 2: Guidelines for Brown ML, Landreneau RJ: Improvement in latory and heart rate responses to exercise: the Diagnosis and Management of Asthma. Be- pulmonary function and elastic recoil after Better predictors of heart failure mortality thesda, National Institutes of Health, 1997, lung-reduction surgery for diffuse emphysema. than peak oxygen consumption. Circulation Publ No 97-4051. N Engl J Med 1996;334:1095–1099. 1999;100:2411–2417. 65 American Thoracic Society: Standards for the 79 Criner GJ, Cordova FC, Furukawa S, Kuzma 92 Stelken AM, Younis LT, Jennison SH, Miller diagnosis and care of patients with chronic AM, Travaline JM, Leyenson V, O’Brien GM: DD, Miller LW, Shaw LJ, Kargl D, Chaitman obstructive pulmonary disease and asthma. Prospective randomized trial comparing bilat- BR: Prognostic value of cardiopulmonary ex- Am Rev Respir Dis 1987;136:225–244. eral lung volume reduction surgery to pulmo- ercise testing using percent achieved of pre- nary rehabilitation in severe chronic obstruc- dicted peak oxygen uptake for patients with 66 European Respiratory Society: Airway respon- tive pulmonary disease. Am J Respir Crit Care ischemic and dilated cardiomyopathy. J Am siveness. Standardized challenge testing with Med 1999;160:2018–2027. Coll Cardiol 1996;27:345–352. pharmacological, physical and sensitizing stim- uli in adults. Eur Respir J Suppl 1993;16:53– 80 Roomi J, Johnson MM, Waters K, Yohannes 93 Ziesche S, Cobb FR, Cohn JN, Johnson G, 83. A, Helm A, Connolly MJ: Respiratory rehabili- Tristani F: Hydralazine and isosorbide dini- tation, exercise capacity and quality of life in trate combination improves exercise toler- 67 Cypcar D, Lemanske RF Jr: Asthma and exer- chronic airways disease in old age. Age Ageing ance in heart failure. Results from V-HeFT I cise. Clin Chest Med 1994;15:351–368. 1996;25:12–16. and V-HeFT II. The V-HeFT VA Coopera- tive Studies Group. Circulation 1993;87: 68 American Thoracic Society: Guidelines for me- 81 Cahalin LP, Mathier MA, Semigran MJ, Dec VI56–VI64. thacholine and exercise challenge testing GW, DiSalvo TG: The six-minute walk test –1999. Am J Respir Crit Care Med 2000;161: predicts peak oxygen uptake and survival in 94 Richardson RS, Sheldon J, Poole DC, Hop- 309–329. patients with advanced heart failure. Chest kins SR, Ries AL, Wagner PD: Evidence of 1996;110:325–332. skeletal muscle metabolic reserve during 69 Zeballos RJ, Weisam IM, Connery SM, Brad- whole body exercise in patients with chronic ley JP: Standard treadmill vs. incremental cy- 82 Kessler R, Faller M, Fourgaut G, Mennecier B, obstructive pulmonary disease. Am J Respir cle ergometry in the evaluation of airway hy- Weitzenblum E: Predictive factors of hospitali- Crit Care Med 1999;159:881–885. perreactivity in unexplained dyspnea (ab- zation for acute exacerbation in a series of 64 stract). Am J Respir Crit Care Med 1999;159: patients with chronic obstructive pulmonary 95 O’Donnell DE, Lam M, Webb KA: Spiromet- A419. disease. Am J Respir Crit Care Med 1999;159: ric correlates of improvement in exercise per- 158–164. formance after anticholinergic therapy in 70 ACC/AHA/ACP-ASIM: Guidelines for the chronic obstructive pulmonary disease. Am J management of patients with chronic stable 83 Miyamoto S, Nagaya N, Satoh T, Kyotani S, Respir Crit Care Med 1999;160:542–549. angina: Executive summary and recommenda- Sakamaki F, Fujita M, Nakanishi N, Miyatake tions. A Report of the American College of Car- K: Clinical correlates and prognostic signifi- 96 Marciniuk DD, Gallagher CG: Clinical exer- diology/American Heart Association Task cance of six-minute walk test in patients with cise testing in interstitial lung disease. Clin Force on Practice Guidelines (Committee on primary pulmonary hypertension. Comparison Chest Med 1994;15:287–303. Management of Patients with Chronic Stable with cardiopulmonary exercise testing. Am J Angina). Circulation 1999;99:2829–2848. Respir Crit Care Med 2000;161:487–492. 97 Nixon PA, Orenstein DM, Kelsey SF, Doer- shuk CF: The prognostic value of exercise 71 American College of Sports Medicine: ACSM’s 84 O’Keeffe ST, Lye M, Donnellan C, Carmichael testing in patients with cystic fibrosis. N Engl Guidelines for Exercise Testing and Prescrip- DN: Reproducibility and responsiveness of J Med 1992;327:1785–1788. tion, ed 5. Baltimore, Williams & Wilkins, quality of life assessment and six-minute walk 1995. test in elderly heart failure patients. Heart 98 Harris-Eze AO, Sridhar G, Clemens RE, Gal- 1998;80:377–382. lagher CG, Marciniuk DD: Oxygen improves 72 American Heart Association: Exercise stan- maximal exercise performance in interstitial dards. A statement for healthcare professionals 85 Bernstein ML, Despars JA, Singh NP, Avalos lung disease. Am J Respir Crit Care Med from the American Heart Association. Writing K, Stansbury DW, Light RW: Reanalysis of the 1994;150:1616–1622. Group. Circulation 1995;91:580–615. 12-minute walk in patients with chronic ob- structive pulmonary disease. Chest 1994;105: 99 Belman MJ, Botnick WC, Shin JW: Inhaled 73 LoRusso TJ, Belman MJ, Elashoff JD, Koerner 163–167. bronchodilators reduce dynamic hyperinfla- SK: Prediction of maximal exercise capacity in tion during exercise in patients with chronic obstructive and restrictive pulmonary disease. 86 Chang JA, Curtis JR, Patrick DL, Raghu G: obstructive pulmonary disease. Am J Respir Chest 1993;104:1748–1754. Assessment of health-related quality of life in Crit Care Med 1996;153:967–975. patients with interstitial lung disease. Chest 74 Risk C, Epler GR, Gaensler EA: Exercise al- 1999;116:1175–1182. 100 Bolliger CT, Perruchoud AP: Functional eval- veolar-arterial oxygen pressure difference in in- uation of the lung resection candidate. Eur terstitial lung disease. Chest 1984;85:69–74. 87 Montgomery PS, Gardner AW: The clinical Respir J 1998;11:198–212. utility of a six-minute walk test in peripheral 75 Myers J, Froelicher VF: Hemodynamic deter- arterial occlusive disease patients. J Am Ger- 101 Weisman IM: Cardiopulmonary exercise test- minants of exercise capacity in chronic heart iatr Soc 1998;46:706–711. ing in the preoperative assessment for lung failure. Ann Intern Med 1991;115:377–386. resection surgery. Semin Thorac Cardiovasc 88 Jones NL, Killian KJ: Exercise limitation in Surg 2001;13:116–125. health and disease. N Engl J Med 2000;343: 632–641. Modalities of Clinical Exercise Testing 41

102 Sciurba FC: Early and long-term functional 106 American Thoracic Society: Pulmonary reha- 109 Ortega F, Montemayor T, Sanchez A, Cabello outcomes following lung volume reduction bilitation –1999. American Thoracic Society. F, Castillo J: Role of cardiopulmonary exer- surgery. Clin Chest Med 1997;18:259–276. Am J Respir Crit Care Med 1999;159:1666– cise testing and the criteria used to determine 1682. disability in patients with severe chronic ob- 103 Ramos-Barbon D, Fitchett D, Gibbons WJ, structive pulmonary disease. Am J Respir Latter DA, Levy RD: Maximal exercise test- 107 Maltais F, LeBlanc P, Jobin J, Berube C, Crit Care Med 1994;150:747–751. ing for the selection of heart transplantation Bruneau J, Carrier L, Breton MJ, Falardeau candidates: Limitation of peak oxygen con- G, Belleau R: Intensity of training and physi- R. Jorge Zeballos, MD sumption. Chest 1999;115:410–417. ologic adaptation in patients with chronic ob- Department of Internal Medicine structive pulmonary disease. Am J Respir Texas Tech University Health Sciences Center 104 Williams TJ, Patterson GA, McClean PA, Za- Crit Care Med 1997;155:555–561. 4800 Alberta Avenue mel N, Maurer JR: Maximal exercise testing El Paso, TX 79905 (USA) in single and double lung transplant recipi- 108 Strzelczyk TA, Quigg RJ, Pfeifer PB, Parker Tel. +1 915 833 3350, Fax +1 775 243 1377 ents. Am Rev Respir Dis 1992;145:101–105. MA, Greenland P: Accuracy of estimating E-Mail [email protected] exercise prescription intensity in patients 105 Howard DK, Iademarco EJ, Trulock EP: The with left ventricular systolic dysfunction. J role of cardiopulmonary exercise testing in Cardiopulm Rehabil 2001;21:158–163. lung and heart-lung transplantation. Clin Chest Med 1994;15:405–420. 42 Zeballos/Weisman

Weisman IM, Zeballos RJ (eds): Clinical Exercise Testing. Prog Respir Res. Basel, Karger, 2002, vol 32, pp 43–59 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Methods for Cardiopulmonary Exercise Testing Kenneth C. Becka Idelle M. Weismanb aPhysiological Imaging Laboratory, Department of Radiology, University of Iowa, Iowa City, Iowa; bHuman Performance Laboratory, Department of Clinical Investigation, Pulmonary-Critical Care Service, William Beaumont Army Medical Center, El Paso, Tex., and Department of Medicine, Pulmonary-Critical Care Division, University of Texas Health Science Center at San Antonio, San Antonio, Tex., USA Summary gen saturation of arterial blood using pulse oximetry (SpO2), electrocardiograph (ECG), blood pressure, and Over the last 20 years, CPET has expanded to include a possibly arterial blood gases or additional specialized tests wide spectrum of clinical applications. This has challenged clini- such as spirometry and exercise tidal flow volume loops. cal exercise testing laboratories to provide flexible, yet stan- dardized methodological approaches relevant to clinical deci- Measuring Exercise Intensity: Ergometry sion-making. Standardization of important methodological practices/processes is necessary to optimize clinical application Exercise uses the body’s internal energy stores to per- [1]. This chapter will review the practical aspects of setting up a form useful external work. Work is equal to the force clinical exercise-testing laboratory for the evaluation of both times the distance over which it acts; the rate of perfor- healthy subjets and patients and will utilize the most widely mance of the external work is defined as power. The unit accepted/applied criteria, as standardization is an evolving pro- of work is the joule (equal to 1 NewtonW meter) and power cess. The following topics will be included: equipment, method- is measured in joules Ws–1 or watts. Power output is also ology for the determination of metabolic responses to exercise commonly expressed in kilopond meters per minute, and for quantifying external work, protocols, monitoring, con- where 100 W = 612 kpW m Wmin–1 [2]. The external power duct of the test, patient safety, and emerging methodology for output should be quantifiable by measuring force (on the evaluation of ventilatory limitation. pedals, or on the tread), distance (crank length, tread length) and time. Exercise Testing Equipment The quantitative assessment of power output is made Layout of a typical clinical exercise testing laboratory using either a cycle ergometer or a treadmill. With cycle is shown in figure 1. During a cardiopulmonary exercise ergometers, the power output is measured directly by test (CPET), an external work load is imposed on the pa- measuring the resistance required to turn the pedals (tor- tient while physiological monitoring documents changes que, usually with an internal force transducer) and crank in external work intensity, metabolic gas exchange, oxy- revolutions per minute (RPM). Power output is torque times RPM. With the treadmill, determining the precise The opinions or assertions contained herein are the private views of power output is more difficult. External energy consumed the authors and are not to be construed as official or as reflecting the while walking or running on a flat motorized treadmill is views of the Department of the Army or the Department of De- essentially zero, although clearly the metabolic energy fense. requirement increases with walking speed. The metabolic requirement for walking is derived entirely from work

Fig. 1. Cardiopulmonary exercise testing (CPET) laboratory. Shown are a treadmill, electrically braked cycle ergometer (right) and a mechanically braked cycle ergometer (‘Monarch’, left). At rear is the collection of monitoring equipment including computer for acquisition of breath-by-breath metabol- ic data, and ECG machine. Photograph courtesy of the Human Performance Labora- tory, William Beaumont Army Medical Cen- ter, El Paso, Tex. Table1. Comparison of ergometers used in CPET cally braked cycle ergometry is preferable to treadmill testing for several reasons summarized in table 1: direct Cycle Treadmill quantitation of work rate, less noise (artifact) on ECG, ergometer easier to collect blood samples during exercise, less expen- sive, and safer across a wide spectrum of clinical patient V˙ O2max lower higher populations. Leg muscle fatigue often limits less often limits performance performance Metabolic and Ventilatory Responses Work rate quantification yes estimation only There are four basic measurements that are essential in Blood gas collection easier more difficult quantitating the response to exercise: Oxygen consump- tion (V˙ O2), carbon dioxide production (V˙ CO2), heart rate Instrumentation noise less more (HR), and expired minute ventilation (V˙ E). In turn, an and artifacts impressive number of derived variables can be measured safer less safe (?) during CPET; their impact on the interpretative process Safety less more and clinical decision-making has been noted previously Weight bearing in obese patients active normals [5–7] and reviewed in the chapter on Interpretation. His- More appropriate for torically V˙ O2 and V˙ CO2 measurements were made by timed collection of expired gas into large collection bags [8]. required to move the limbs, and from the up and down Today, the two most common methods for providing rap- motion against gravity. When the treadmill is inclined, id on-line analysis are the mixing chamber technique [6, power output increases due to the work against gravity 9] and breath-by-breath technique [10], both of which needed to prevent downward motion. Thus, power output require continuous measurement of expiratory flow (or is related to body weight in addition to treadmill speed occasionally volume) and continuous gas concentration. and elevation [3, 4]. The treadmill can impose noise on Expiratory Flow and Volume Measurement. Most instrumentation signals such as ECG, blood pressure, and equipment measures instantaneous flow using a pneumo- gas exchange measurements. V˙ O2max is often 5–10% low- tachograph and integrates that signal to obtain volume. er on the cycle as only the leg muscles are used during Pneumotachographs operate on a number of different exercise; consequently, leg fatigue is often the limiting fac- principles, the most common of which are listed in tor to exercise performance. For quantitative assessment table 2. Note that each type exploits different properties of exercise response in the clinical laboratory, electroni- 44 Beck/Weisman

Table 2. Methods for continuous measurement of flow and volume (types of pneumotachographs) Principle of operation Advantages Disadvantages Screen-type Differential pressure across screen Linear response Seriously affected by moisture or element sputum impaction Bernoulli or Proportional to gas density and Not disposable Pitot tube viscosity Hot wire Not seriously affected by moisture or sputum Highly nonlinear response requires anemometer Differential pressure within gas Turbine stream Lightweight computerized compensation Proportional to gas density Disposable Current maintains temperature of resistance wires Not seriously affected by mosture buildup May be affected by sputum impaction Exploits heat capacity of gas Linear response Not disposable Registers volume, not flow Least sensitive to changes in gas species (may May be affected by sputum impaction Counts revolutions of lightweight be affected slightly by gas density or viscosity ‘Over spin’ due to intertia requires spinning fan blade change) computerized compensation Not seriously affected by moisture Not disposable Constant calibration (one revolution = fixed volume) of the gas (density for screens and Pitot tubes, heat capaci- and Scholander techniques). Real time data analysis re- ty for the hot wire). If a study is being undertaken with an quires more rapid analyzers. Essential requirements for a inspired gas other than room air (e.g. high O2, or He/O2 gas analyzer include speed of response, linearity, stability breathing), the manufacturer of the equipment must be of calibration, and independence of output (i.e. presence consulted and validation should be performed. of CO2 does not affect O2 measurement). Validation of these properties can be accomplished with a set of 3–4 The calibration curve of any pneumotachograph can precision grade gas tanks. Testing at least one of these be affected by its mounting location in the system because gases both dry and humidified will identify how the ana- of entrance and exit effects of flowing gases. Thus, care lyzer is affected by water vapor. must be taken to ensure proper attachment to the exercise equipment prior to daily calibration. A common means Basic Concepts of Metabolic Measurements for daily calibration is with a 3L gas syringe. The calibra- Bag collection, mixing chamber, and breath-by-breath tion procedure should include pumping the syringe at var- methods all use the same set of basic equations to calcu- ious rates to confirm that output of the pneumotacho- late V˙ O2 and V˙ CO2. These equations express the mass bal- graph is independent of flow over the range expected dur- ance of O2 and CO2 by quantifying the amount of the gas ing exercise [11, 11A]. The internal resistance and dead taken in during inspiration less the amount given off dur- space volume of the pneumotachograph and associated ing expiration. The following basic equations are used [2, valves must be taken into consideration. Although good 6]: data regarding the effect of breathing resistance on exer- cise response are not available, a resistance of more than V˙ O2 = V˙ I W FIO2 – V˙ E W FEO2, 1–2 cm H2O Wl–1 W s–1 will likely be sensed by subjects at high levels of exercise. Added dead space of equipment where over-dotted V’s indicate timed collections of gas will manifest itself as an increase in ventilatory require- volumes and the subscript E indicates mixed expired gas, ment, particularly at rest and lower levels of exercise. as in a well mixed bag of expired gas. This equation has both inspired and expired gas volumes, but only one of Gas Concentration. Measurement of V˙ O2 and V˙ CO2 these volumes needs to be measured, usually expired vol- requires determination of concentrations of O2 and CO2 ume: in the expired and inspired air in addition to flow or vol- ume. If the classic bag collection technique is being used, V˙ I(STPD) = V˙ E(STPD) W FEN2/FIN2 = the time taken to analyze the gas is not a critical issue, and V˙ E(STPD) W (1.0 – FEO2 – FECO2)/FIN2, chemical gas absorption methods are often used (Haldane Exercise Testing Methodology 45

where FIN2 is inspired N2 concentration (0.79 for room dried before analysis because the analyzers are either air) and FEN2 is the mixed expired N2 (in a bag). damaged or their output affected by water vapor. It is common for manufacturers to use drying gas sample lines, The equation for V˙ CO2 is the reverse of the one for V˙ O2 in which water vapor is absorbed in transport from the (expired CO2 volume minus inspired CO2 volume), but mouth to the gas analyzers. Drying sample lines are prone the CO2 concentration is near zero in room air, so the to failure, and this failure to correct adequately for water inspired volume is often omitted: vapor in the expired gas can be a surprisingly large (20– 25%) source of error in either breath-by-breath or mixing V˙ CO2 = V˙ E W FECO2 – V˙ I W FICO2, but since FICO2 \" 0, chamber systems [12]. V˙ CO2 = V˙ E W FECO2 Breath-by-Breath and Effects of Gas Analyzer Time Delay. The breath-by-breath method samples gas flow The metabolic measurements V˙ O2 and V˙ CO2 are ex- and concentration over the breath to obtain inspired and pressed in standard temperature and pressure conditions expired volumes of CO2 and O2 per breath. These quanti- (STPD). The conversion from ATPS (ambient tempera- ties are then used in mass balance equations combined ture and pressure, saturated), to STPD is performed with breath timing information to determine V˙ O2 and using: V˙ CO2 extrapolated to 1 min. A technical hurdle is the fact that flow measurement occurs nearly instantaneously, but ATPS to STPD = Pbar – PH2O(t) W 273 , gas concentration signals are delayed by the transit time of 760 273 + t the gas along the sampling tubing into the instrument. When performing breath by breath integration, it is im- where Pbar is barometric pressure (mm Hg), t is room tem- portant to temporally realign the gas concentration and perature in°C and PH2O(t) is determined from lookup flow signals [13, 14]. Most automated commercial sys- tables of water vapor pressure against temperature. tems have computer software that accomplishes this re- alignment with a built-in calibration routine that deter- The minute ventilation of the lungs is usually reported mines the delay time that is used. in body temperature pressure saturated conditions. The conversion from SPTD to BTPS is straightforward, since Mixing Chamber. The mixing chamber technique di- it involves three numbers that do not vary: rects expiratory gas into a 5- to 10-liter chamber with internal baffles to facilitate mixing. Mixed gas concentra- STPD to BTPS = 760 310 863 tion and expiratory flow are measured continuously near W= the outlet of the box. In addition to the delay time for gas Pbar – PH2O(37 ° C) 273 Pbar – 47 concentration measurement mentioned above, the gas concentration signal of a mixing chamber system is de- Example: In a laboratory at about 700 feet altitude layed relative to the flow signal by the time to transport exhaled gas from the breathing valve to the mixing cham- above sea level (Pbar = 740 mm Hg) and typical room tem- ber. Because of changing V˙ E during exercise, the latter perature of 22°C, a one minute collection of expired gas, delay is not a fixed time period, but is determined from the ratio of (tubing volume + box volume)/V˙ E. Computer- or measurements obtained from a mixing chamber sys- ized systems can make real time adjustments for this tem, might be as follows: V˙ E = 50 liters/min, FEO2 = 16.7%, varying delay. FECO2 = 4.0%. E The V˙ E in STPD conditions is found from the conver- Comparing Metabolic Measurement Techniques Of the three methods for assessing V˙ O2, V˙ CO2 and V˙ E, sion of ATPS to STPD: (740 – 20)/760 ! 273/295 = the breath-by-breath technique has become the most com- 0.8767, so V˙ E(STPD) is 43.84 liters. mon, although there are still many mixing chamber sys- tems in use (table 3). Both are acceptable for clinical exer- E The inspired volume is obtained from 43.84 ! (1.0 – cise testing. Direct bag collection is usually reserved for validation studies. The breath-by-breath method requires 0.167 – 0 .04)/0.79 = 43.98 liters/min. a digital computer to carry out the calculations in real E V˙ CO2 = 43.84 ! 0.040 = 1.75 liters/min. time but has the advantages of flexibility and high time E V˙ O2 = 43.98 ! 0.2095 – 43.84 ! 0.167 = 1.89 liters/ resolution for changes in metabolic rate or ventilation. min. E Reported V˙ E(BTPS) = 43.84 ! 863/(740 – 47) = 54.6 liters/min. Effects of Water Vapor. Expired gas is warmed and humidified compared to inspired gas. Water vapor repre- sents about the same fractional concentration as CO2 in expired air, but is a much lower concentration in inspired air of most laboratory environments. Water vapor is nev- er measured directly, and rapid gas analyzers used in most commercial systems (with the exception of the few that rely on mass spectrometry) require the gas sample to be 46 Beck/Weisman

Table 3. Comparison of techniques for measuring V˙ O2, V˙ CO2 and V˙ E Feature Breath-by breath Mixing chamber Bag collections Accuracy variable, depends on calibration of flow variable, depends on calibration of flow high for steady state conditions meter, gas analyzer and handling of water meter, gas analyzer and handling of water vapor vapor high, needs means to collect gas volumes in low, needs computer or strip chart recorder, large bags, measurement of large gas Laboratory space low, needs computer, compact gas compact gas analyzers, calibration tanks volumes (Tissot spirometer) and high analyzers, calibration tanks precision gas analysis equipment requires expired gas tubing to mixing requires expired gas tubing to expired bag, Patient interface lightweight pneumotachograph and gas chamber, restricting movement restricting movement sampling lines allow freedom of movement equipment is usually portable, requires requires setup of bulky equipment; gas Technical ease daily calibration analysis methods (Haldane, Scholander) are of use equipment is usually portable, and requires usually laborious daily calibration variable, may be 30–60 s at low ventilation usually 60 s bag collection, can do 30 s rates, but nearly breath-by-breath at high collections at high intensities Time resolution may be breath-by breath ventilations computerized real time computations and data are usually calculated after exercise Immediacy computerized real time computations and displays testing of results displays However, there is also a high degree of breath-to-breath chamber systems have improved time resolution, while variability in the data. Some of the breath-to-breath vari- true breath-by-breath time resolution is generally not nec- ability is caused by variations in mismatch between essary in clinical testing situations. inspired and expired volume of each breath, leading to variation in gas stores in the lung [14, 15]. For practical Quality Control and Validation of Equipment purposes, most laboratories use averaging techniques to smooth the noise. There are two broad choices for averag- The equipment manufacturer should provide data ing methods: average over time or average over fixed sheets indicating results of validation testing. It is the number of breaths. For both, the longer the averaging responsibility of the laboratory to insure continuous accu- interval and the more breaths in the average, the more the racy over time as the equipment ages. Periodic quality data are smoothed [16–18]. However, there is a trade-off: control (QC) testing often makes the assumption that as data are smoothed, rapid changes may be obscured. As either the ergometer or the metabolic measuring equip- peak V˙ O2 usually occurs during a period of non-steady ment is operating properly. For instance, it is common to state metabolic response, it could be underestimated. validate metabolic measurement systems by having a nor- When averaging by fixed number of breaths, the time mal subject exercise at a given intensity while measuring interval of the average decreases as breathing rate in- their V˙ O2, V˙ CO2, and V˙ E and comparing the values ob- creases late in exercise. A 20- to 30-second moving aver- tained with those expected at the work intensity setting [3, age is probably a good choice for routine testing, though 6, 11A]. If the ergometer is not properly calibrated, an up to 60 s averaging may be appropriate to mimic tradi- ‘error’ would be detected in metabolic data. tional bag collections [2]. Additional studies are required to determine the clinical significance of these different Reproducibility of Measurements interval-averaging techniques. As with all laboratory measurements, there is some inherent ‘noise’ or uncertainty in the measurement of The V˙ O2 and V˙ CO2 data coming from mixing chamber responses to exercise. Consideration of the uncertainty of systems are smoother than unaveraged breath-by-breath CPET variables is important because of its impact on the data simply because the mixing chamber provides a phys- interpretation of CPET results. Numerous studies have ical averaging mechanism. In addition to the inability to shown the test-retest variability of both metabolic mea- measure PETO2 and PETO2, the only real disadvantage of surements (V˙ O2, V˙ CO2, and V˙ E) and external work intensi- mixing chamber systems over breath-by-breath systems is the lack of time resolution. However, modern mixing Exercise Testing Methodology 47

Table 4. Reproducibility of maximal Ref. Sample * VO2max * VEmax AT Maximal Disease exercise capacity in normals and selected power patient populations No. size 20 6 8.4% 4.4% 12.1% 5.5% normal 88 10 5.0% 7.0% 13.0% 7.0% normal 89 11 3.0% 5.0% 3.7% COPD 90 20 9.0% 8.1% – 9.7% COPD 91 13 6.6% 6.3% – 13.8% COPD 19 6 5.3% 5.5% – 5.6% ILD 92 11 4.1% 6.3% – 3.6% CHF – * Table adapted from Marciniuk et al. [19]. Table entries are coefficient of variation (SD/ mean) of the indicated variable. AT = The V˙ O2 at the anaerobic threshold; COPD = chronic obstructive pulmonary disease; ILD = interstitial lung disease; CHF = chronic heart failure. ties is quite low in both normal and patient populations torque. Such calibrators are commercially available when individuals are retested in the same laboratory [19] through independent vendors who will provide on-site (table 4). Because exercise testing is strongly dependent validation or manufacturers who will loan or rent these on patient motivation, which in turn can be affected by devices for validation studies [21, 22]. motivational skills of testing personnel, there may be some variability within a laboratory among testing per- Quality Control of Metabolic Measurements: sonnel. In addition, variability between laboratories is The Physiologic QC affected by adequacy of equipment calibration and main- Optimizing quality control of metabolic measurements tenance. The variability among testing personnel and is best accomplished by establishing a program of regular among laboratories has not been extensively studied. ‘physiologic calibration’ tests in one or more individuals Finally, time of day [20] and laboratory environment (usually healthy, willing, and available hospital em- (temperature, humidity) may affect test results. To ployees) at a fixed number of exercise intensities. Because achieve good comparable data for comparison purposes, the effect of small errors in the time delay adjustment laboratories should strive to standardize as many of these between the pneumotachograph and gas analyzer is am- variables as possible within the testing laboratory. plified at high breathing rates [13], it is relatively easy (and preferable) to test at both normal and high breathing Ergometer Validations rates for a more complete accuracy check. Gas exchange Treadmill validation is relatively easy, requiring simulators have become available that can be used for checking of the accuracy of the % elevation and speed of periodic validation [23]. However, use of these simulators the belt. This evaluation should be performed with a sub- is not a replacement for adequate physiologic validation. ject running on the treadmill, because the weight of the The gas from the simulator is usually at room tempera- subject can alter treadmill speed. Validation of belt speed ture, with normal humidity, so errors that can occur with is easily performed by counting the number of times a defective gas drying mechanism might not be picked up mark or piece of tape affixed to the tread cycles around with the devices. per minute. A reasonable algorithm for a physiologic QC program Cycle ergometers (arm or leg) are more difficult to vali- would proceed as follows. First, ensure that the ergometer date. Timed counting of pedal revolutions validates the is functioning properly and registering accurate power tachometer. Some manufacturers provide a ‘static’ cali- output. Select a subject who is likely to remain available bration device that allows the user to perform checks of for periodic testing and determine his/her V˙ E, V˙ O2, and the internal force transducer. Static calibration does not V˙ CO2 while exercising at either a fixed intensity or prefera- check the internal resistance of the cycle, nor does it check bly at several intensities (e.g. 50, 100, 150 W on a cycle) at the accuracy of the controller mechanism. Dynamic cali- least 4 times. The data are then entered into a QC data- bration is more difficult to perform and requires an exter- base to obtain an average for each intensity level includ- nal means of turning the cycle’s crank while measuring the ing range (95% confidence interval = 1.96 ! SD). At regu- 48 Beck/Weisman

lar intervals, the same person should perform the same (shorter in trained subjects, longer in disease [24]) 2- to exercise, and the results of V˙ E, V˙ O2, and V˙ CO2 are com- 3-min stages will result in nearly complete adaptation to pared with the established 95% confidence interval. If changes in intensity at the end of the test interval. data are outside this interval, repeat the test either in the same person or in another subject with established data. If It has become popular to impose ‘ramp’ increases in both measures are outside the limits, then an equipment external power output, especially when using cycle ergom- check is in order (pneumotachograph, gas analyzer, dry- eters. Ramp protocols increase power output nearly con- ing gas sampling line, etc.). If one or both measurements tinuously, so true steady state is never attained. It can be are within tolerance, add them to the QC database. When shown, however, that cardiac and metabolic responses six or more data points have thus been added to the data- during ramp protocols represent steady state values with a base, recalculate the 95% confidence interval [11A]. time lag that is determined by the time constant of the response to step change in power output [25]. Similar met- The most frequent causes for error in the metabolic abolic and cardiopulmonary values have been obtained measurements in our experience are failure of the gas ana- using the 1-min incremental or ramp protocols and there- lyzer or failure of the mechanism for drying the gas sam- fore, either is acceptable for clinical purposes [25–28]. ple. Other errors that can occur are in the alignment of Maximal exercise capacity determined by ramp protocols flow and gas concentration signals, in gas flow measure- has been shown to be reasonably accurate when compared ment, or in ergometer calibration (rare). with incremental protocols in both elderly subjects [29] and patients with COPD [30]. Exercise Protocols After attaining the maximal or peak exercise intensity, The choice of testing protocol is guided largely by the the patient should be allowed to ‘cool down’ by perform- goals of the exercise evaluation; either treadmill or cycle ing unloaded pedaling or slow walking on the level tread- ergometry can be used. Several protocols for clinical exer- mill while monitoring ECG, SpO2 and heart rate. Impor- cise testing are available and include: (1) a progressive tantly, the heart rate recovery may provide important incremental exercise test in which workload is increased independent prognostic information for mortality and (usually every 1–2 min) or continuously in a ramp fash- morbidity [31, 32]. Heart rate recovery has been defined ion; (2) a multi-stage (usually every 3 min or ‘pseudo as peak HR – HR at 1 min [31] or peak HR – HR at 2 min steady-state’ exercise protocol, and (3) a constant work [32]. Furthermore, current available data has come from rate protocol in which the work rate is maintained con- treadmill testing; although it is likely valid heart rate stant for a variable period of time (5–30 min). recovery still requires further testing for application in cycle ergometry. Incremental testing protocols change the exercise in- tensity in regular time intervals so that intensity increases Cycle Ergometry slowly until volitional exhaustion. Design of an incremen- Incremental or Ramp Protocols. Cycle ergometry is the tal protocol has two major elements: the size of the incre- preferable mode of exercise testing in patients with respi- ment in work intensity and the time period for each stage. ratory diseases (table 1). Exercise protocols designed to In cycle ergometry, the size of the increment can be determine V˙ O2max as one end point should be designed to changed by altering either the resistance on the pedals or, last approximately 10 min [33]. Patients will exercise in case of simple devices with frictional bands, the pedal until volitional exhaustion is achieved or the test is termi- cadence. Modern feedback controlled cycles control the nated based on the judgment of the monitor. A predaling power output in watts by adjusting resistance on the frequency of 60 rpm appears optimal although this can pedals to keep power output constant within a range of vary between 40 and 70 rpm. A popular protocol for pedal cadences. With treadmills, speed and elevation of obtaining cardiopulmonary measurement using cycle er- the treadmill determine the increment, which is usually gometry is outlined in table 5. The patient sits for 3 min expressed in METS, or metabolic equivalents compared on the cycle with the nose clip and mouthpiece in place for to rest. The time period for each increment is usually 1– baseline measurements. A warm-up period of usually 3 min. Because the cardiac and metabolic responses to a 3 min follows; some prefer unloaded cycling for this sudden change in work intensity have a nearly exponen- warm-up whereas others prefer low-intensity exercise (5– tial time course with a time constant (time to attain 63% 20 W). However, patients with advanced lung disease are of the total response) of about 30 s in normal subjects often unable to warm-up for 3 min even at unloaded cycling and therefore, this must be individualized. An Exercise Testing Methodology 49

Fig. 2. Comparison of V˙ O2 during ramp, incremental and constant Pre-test estimations of maximal exercise capacity can work exercise. Schematic graph which demonstrates that at any sub- be made by starting with predicted V˙ O2max then adjusting maximal work rate, V˙ O2 will be higher during constant work com- the prediction based on clinical assessment of the sub- pared to incremental or ramp exercise testing (vertical line). Though ject’s capabilities. For instance, a young fit subject should V˙ O2max may be comparable among the three protocols, the maximal be able to attain a V˙ O2max of about 40 ml Wmin–1 Wkg–1. A power output will be lowest with the constant load protocol (horizon- subject weighing 75 kg would have an estimated V˙ O2max of 3,000 mlW min–1. Using a rough rule of thumb that V˙ O2 tal line). increases about 9–11 ml Wmin–1 per watt of external power output, the estimated maximal power would be (3,000 – Table 5. Incremental exercise protocol (cycle ergometry) 500)/10 = 250 W. A V˙ O2 of 500 mlWmin–1 is subtracted from the estimated V˙ O2max to correct for V˙ O2 during Cardiopulmonary Familiarization warm-up cycling (10–20 W). Thus, a 20–25 W/min proto- ↓ col would bring the subject to near maximal within the measurements 10–12 min guideline. A very fit subject might require 30 (V˙ O2, V˙ CO2, V˙ E) Patient preparation to as high as 50 W/min protocols, whereas less fit, smaller, (ECG, pulse oximetry, or older subjects might require lesser increments. The blood pressure, ? arterial line) increase in power output can be a steady ramp or incre- ↓ menting intensity at the end of each 1- or 2-min period. Finally, recovery is monitored for at least 10 min during 1–3 min resting data which the patient performs unloaded cycling for at least ↓ 3 min to prevent blood pooling in the legs. 1–3 min unloaded pedaling Constant Work Protocols. Increasingly, constant work ↓ exercise at a standardized submaximal work load (based on an initial incremental exercise test) is being uilized to \"10 min incremental exercise evaluate the impact of therapeutic interventions includ- (5–30 W/min) ing exercise training, oxygen therapy, and lung volume ↓ reduction surgery. The relationship between V˙ O2 and work rate for a ramp protocol, a 1-min incremental exer- 10 min recovery cise protocol and a constant work rate exercise protocol (3 min unloaded cycling) are shown in figure 2. At any submaximal external power (ECG, blood pressure, oximetry) output, V˙ O2 will be higher during a constant work protocol compared to either ramp or 1-min incremental protocols. automated flywheel apparatus may be helpful to over- The difference between constant work and the ramp or come the inertia of the pedals under these circumstances. incremental protocols will likely be larger above com- Subsequently, the rate of increase in intensity depends on pared to below the anaerobic threshold [34]. Constant the pre-test estimation of the subject’s capacity. Most work protocols can be performed on either a cycle ergom- patients with respiratory disease will exercise at work rate eter or treadmill. Recent work has suggested that Borg increments of 5 W (very debilitated disease) to 25–30 W/ dyspnea ratings, measurements of inspiratory capacity (as min (mild disease and/or very fit). a reflection of dynamic hyperinflation) and endurance times during submaximal exercise were highly reproduc- ible and sensitive in patients with severe COPD [35, 36]; furthermore, constant work endurance times were more sensitive than 6-min walk times in determining thera- peutic effectiveness of pharmacological intervention [36]. A constant work protocol is also useful for corrobora- tion of pulmonary gas exchange and as a possible alterna- tive approach to obtain PaO2, P(A-a)O2, and VD/VT in lieu of arterial line placement during incremental exercise testing [37]. A work rate of 70% of the maximum work 50 Beck/Weisman

rate achieved during the incremental exercise test is used ventilation is necessary; patients in whom a non-invasive for the constant work test, which results in V˙ O2peak 1 90% exercise test suggests the presence of a pulmonary gas of the value obtained in the maximal exercise evaluation. exchange abnormality and the need for invasive measure- An arterial blood gas obtained at rest is used for the rest- ments; patients in whom pulse oximetry may not be reli- exercise comparison (see chapter on Interpretation – case able (blacks, smokers, patients with poor perfusion) and studies). Additional validation of this approach is war- in whom a more accurate SaO2 assessment is required for ranted. O2 prescription [43, 44] (see case studies in chapter on ‘Interpretation’). Treadmill Protocols Treadmill protocols are more complex because there A helpful blood sampling strategy includes samples at are two variables that affect work intensity: speed and rest, during unloaded cycling, and then every other min- elevation. In the extreme, speed or elevation can be left ute during incremental exercise and after 2 min of recov- constant while altering only the other variable (e.g. Balke ery. Arterial catheter placement into the radial artery is protocol). The most common treadmill protocol is that preferable as collateral circulation to the hand through the designed by Bruce [3, 38]. However, there are other alter- ulnar artery mitigates complications due to rare occur- natives, listed in the guidelines published by the Ameri- rences of arterial occlusion. The catheter is inserted after can College of Sports Medicine [3]. A comparison study of performing an Allen’s test, which evaluates for the pres- treadmill protocols found a slightly higher V˙ O2max using ence of such collateral circulation. the Taylor vs. Bruce and Balke protocols [39]. The Taylor protocol is a discontinuous protocol involving 3-min Valuable pulmonary gas exchange information may stages of exercise separated by 5-min rest periods [39]. also be provided by a single radial arterial stick, which is Consequently, it takes much longer to complete than con- usually attempted near peak exercise. However, assessing tinuous protocols and is, therefore, used less often in clini- exercise gas exchange from a sample obtained from a sin- cal exercise labs. gle arterial stick at peak exercise when the patient is strug- The Bruce and Balke protocols are the two most widely gling to finish the test or immediately post exercise when used treadmill protocols. The Bruce protocol is the stan- the gas exchange milieu is already different from peak dard for cardiac ischemia testing with the increment exercise conditions is problematic and should be discour- increase per stage approximately 50 W, too great for most aged. Misleading information may result as PaO2 changes respiratory patients [40]. In turn, protocols which approx- occur rapidly following the end of exercise and clinically imate a constant rate of increase in work rate are better significant abnormalities present at peak exercise can be suited for cardiopulmonary exercise testing. The Balke missed [45]. An alternative strategy utilizing a single stick protocol [41] maintains a constant speed at 3.3 mph as during minute 5 of constant work may be preferable as elevation is increased 1% every minute. This protocol can breathing pattern alterations are minimal compared to be modified to accommodate different levels of fitness, peak exercise using this approach (see constant work and selecting an appropriate speed level (walking or running) chapter on ‘Interpretation’, case 2). as elevation is increased 1% every minute [5, 42]. Monitoring Invasive vs. Noninvasive CPET Blood Pressure and ECG An important decision is whether a clinical exercise Heart rate is most accurately obtained from the R-R test requires the placement of an indwelling arterial cathe- interval on the 12-lead electrocardiogram. Movement ar- ter for blood gas analysis in order to answer relevant clini- tifacts must be minimized by adequate skin preparation cal question(s) [2, 7] (see chapter on ‘Interpretation’). and use of electrodes designed for exercise testing [11A]. Clinically, this is usually not necessary. However, there For optimal detection of myocardial ischemia and cardiac are some clinical situations in which the additional infor- arrhythmias, serial 12-lead ECG tracings should be ob- mation provided by arterial line placement may be help- tained during the exercise test. Care should be exercised ful and, therefore, should be considered. These include: when using averaged signals, and the raw tracings should patients with pulmonary disease and suspicion of pulmo- always be consulted for final ECG interpretation. nary gas exchange abnormalities like pulmonary vascular Arterial blood pressure can be measured using either disease, COPD with low DLCO, interstitial lung disease; cuff (noninvasive) or indwelling arterial cannula and pres- patients in whom documentation of psychogenic hyper- sure transducer. Modern transducers are pre-calibrated, Exercise Testing Methodology 51

but they always need to be positioned at the mid-heart In general, pulse oximeters are good for monitoring level for accurate readings. Care should also be exercised trending phenomena but not reliable for determining to exclude air bubbles from tubing and the transducer absolute magnitude of change. Quality assurance for pulse housing to optimize the frequency response of the signal. oximeters should include validation with arterial oxygen Because of peripheral amplification of the pulse pressure, saturation and linear regression plots for each oximeter. direct readings from indwelling catheters may register This is important because some pulse oximeters underes- higher (\"8–10 mm Hg) systolic and diastolic values com- timate while others overestimate arterial oxygen satura- pared to cuff [46, 47]. Cuff pressure measurement may tion especially at SaO2 !88%. Pulse oximetry during exer- also be problematic; motion associated with exercise can cise testing is considered standard of care: in the evalua- make auscultation difficult. Use of a miniature micro- tion of patients with exertional dyspnea, in patients with phone attached to the arm and systems that filter out resting hypoxemia, and patients with suspected exertional noise and amplify the pulses sometimes help, but these hypoxemia. and automated blood pressure systems should be vali- dated against standard cuff measurements before use. Arterial Blood Gases Exercise blood pressure criteria (see tables 7, 9) are based mostly on cuff measurements. As previously noted, it may be desirable to document Pulse Oximetry the changes in pulmonary gas exchange during exercise in The pulse oximeter provides valuable real-time read- out of changes in oxygen saturation (SpO2) during exer- patients with heart or lung disease. Whereas most labora- cise. As a result of the shape of the oxyhemoglobin disso- ciation curve, O2 saturation is a relatively insensitive indi- tories will use pulse oximetry (see above) to monitor cator of gas exchange abnormalities (see arterial blood gases). Oximeters operate by measuring light absorbance arterial saturation, it is well appreciated that SpO2 by of tissues at several wavelengths, allowing differentiation between oxygenated and unoxygenated blood [48]. Pulse itself provides a relatively crude index of changes in pul- oximeters perform rapid analysis of the signals to allow differentiation between arterial and mixed venous blood monary gas exchange. PaO2 and the alveolar to arterial during each pulse beat. Factors that will compromise their oxygen tension difference (P(A-a)O2 are much more sensi- performance therefore include deficiencies in perfusion to tive indicators of gas exchange abnormalities. Arterial the selected measuring site (usually finger or ear, but instruments are becoming available that attach to the blood sampling including measurement of PaO2, PaCO2, forehead), interference from bright ambient lights, some pH, HCO3, SaO2 and calculation of P(A-a)O2 and VD/VT types of finger nail polish, and skin pigmentation [11A, is optimal. Suggested guidelines and caveats for arterial 43, 48, 49]. The accuracy of pulse oximetry compared to direct sampling of arterial blood is generally good (B 4%) blood gas values during CPET appear in the chapter on as long as a good pulse signal is obtained, but the measure- ment is generally thought to be less accurate at saturations ‘Interpretation’ [table 3 in ref. 7; also 5, 50, 51]. To calcu- below about 88%; this is exacerbated in blacks [43]. Pulse oximeters measure both COHb and O2Hb, unlike co-oxi- late PAO2, use the alveolar air equation for O2: meters which can separate the two. Consequently, a patient with 5% COHb and 85% SaO2 will have SpO2 ͫ ͬ1 1 – R approximately 90%. Also, during exercise in some heart PAO2 = PIO2 – PaCO2 W R – FIO2 W R , failure patients, the instrument may register a desatura- tion because of diminishing peripheral perfusion [44]. where PIO2 indicates the inspired partial pressure of O2 Thus, care should be exercised in interpreting any changes (FIO2 ! (Pbar – 47), obtained by pulse oximetry alone. Significant desatura- tion, defined as ¢SpO2 65 mm Hg, should be confirmed R = V˙ CO2 with arterial blood gases [50]. V˙ O2 from metabolic measurements obtained at the same time as blood sampling, and FIO2 is the inspired fraction of O2 (usually 0.2095). A simpler form that is nearly as accurate especially when R \" 1.0 is PAO2 = PIO2 – PaCO2 . R The P(A-a)O2 is normally less than 10 mm Hg at rest, but may increase to about 20 even in normal subjects near maximal exercise. Widening of the P(A-a)O2 more than this suggests significant gas exchange defects such as worsening V˙ A/Q inequalities, diffusion limitation, right- to-left shunt, or compounding of these factors by the inev- itable fall in PvO2 that occurs progressively with exercise. 52 Beck/Weisman

If hypoxemia worsens without widening of the P(A-a)O2, Table 6. Conduct of a cardiopulmonary exercise test the hypoxemia is likely due to an inadequate ventilatory response and CO2 retention rather than a primary gas Pre-test patient check list exchange abnormality. Reason(s) for CPET Diagnosis, history, physical examination The VD/VT is also calculated using blood gases. To ECG, PFTs, salient laboratory test results complete the calculation, an estimate of the mixed ex- Health questionnaire and physical activity profile pired PCO2 obtained at the same time as arterial blood Compliant with pre-test instructions, especially medications sampling (PECO2) is necessary. Traditionally, this came Consent form signed from a bag collection of expired air. With mixing chamber systems, the number can be taken directly from the mix- Equipment and protocol selection ing chamber. Breath-by-breath systems will often provide Incremental vs. constant work or both (1 h between tests) that data, obtaining from the ratio of VCO2/VT, where VCO2 Invasive vs. noninvasive test is expired volume of CO2, averaged over a number of breaths, and both VCO2 and VT must be expressed in Pre-test laboratory procedures STPD conditions. The equation for VD/VT is Quality control Equipment calibration PaCO2 – PECO2 . PaCO2 Patient preparation Familiarization The VD/VT typically declines during exercise in healthy Monitoring devices – ECG, pulse oximetry, blood pressure individuals, but it is higher in older subjects compared to Arterial line placement (if warranted) younger subjects. VD/VT calculation should be based on PaCO2 because PETCO2 is unreliable [52]. Cardiopulmonary exercise testing Interpretation of integrative CPET results Conduct of the Exercise Test on the bike, or walking on the treadmill while wearing a noseclip, mouthpiece, pulse oximeter, etc. This is also a An overview for the conduct of a CPET appears in good opportunity to explain communication techniques table 6. Reason(s) for CPET are usually given by the refer- during exercise including use of hand signs and symptoms ring physician but is/are often obtained by consultation scoring [53, 54]. The patient should be encouraged to ask between the referring physician and the physician in questions and feel comfortable in the exercise-testing lab- charge of the exercise testing laboratory. A medical diag- oratory. nosis and a summary of the salient features of the physical examination/medical history should accompany the exer- Absolute and relative contraindications to exercise cise test request. The results of pulmonary function test- testing are provided as a reference source and appear in ing, EKG, chest X-ray and pertinent laboratory results table 7 [3, 6, 55–57]. Systemic hypertension is included, should also be noted. The patient with the assistance of although there is disagreement about the level (systolic exercise lab personnel should respond to a short medical 1200–250 mm Hg and diastolic 1115–120 mm Hg mea- questionnaire. This should include questions related to sured at rest). Care should be exercised in testing pregnant cardiopulmonary and major systemic disease and current women [3, 58] (see chapter on ‘Exercise in Pregnancy’). therapy, with special attention to medications that alter The clinical judgment of the physician in charge of the test heart rate and blood pressure, those used for the treat- should prevail. ment of cardiac insufficiency, and for treatment of pul- monary conditions such as asthma or COPD. Physical Preliminary Requirements for CPET activity and symptom(s) levels, as well as risk factors for When the test is pre-scheduled, the subject should be coronary artery disease should also be noted. A brief phys- issued some basic information about the conduct of the ical examination of the patient may be required to identi- test and a list of instructions for preparation. Importantly, fy contraindications to performing CPET such as knee or the patient is instructed on which medications should be other joint problems that can limit exercise performance. taken or withheld prior to exercise testing. For cardiac Patients may benefit from an exercise familiarization transplantation or impairment-disability evaluation, pa- screening ‘mini-session’ which includes practicing cycling tients should be tested on an optimized medication regi- men. For evaluation of exercise-induced asthma, respira- tory medications are usually withheld prior to exercise Exercise Testing Methodology 53

Table 7. Contraindications to exercise testing Table 8. Borg perceived exertion scales Absolute contraindications Category ratio scale Linear scale 1 Recent myocardial infarction 2 Changes in the resting ECG that suggest acute or recent numerical descriptor numerical descriptor myocardial event rating rating 3 Unstable angina pectoris 4 Uncontrolled cardiac rhythm disturbances, either 0 nothing at all 6 supraventricular or ventricular, particularly when 0.5 very, very slight 7 very, very light compromising cardiac output 5 Severe aortic stenosis and known or suspected dissecting aortic (almost none) 8 aneurysm 1 very slight 9 very light 6 Active or suspected acute pericarditis or myocarditis 2 slight 10 7 Acute congestive heart failure 3 moderate 11 fairly light 8 Recent systemic or pulmonary embolus 4 somewhat severe 12 9 Acute febrile illness 5 severe 13 somewhat hard 6 14 10 Third degree a-v block without pacemaker control 7 very severe 15 hard 11 Pulmonary edema or significant cor pulmonale 8 16 12 Physical disabilities or other inability to cooperate such as 9 very,very severe 17 very hard 10 maximal 18 severe emotional or psychological distress 19 very, very hard 20 Relative contraindications 1 Systemic hypertension Note: May substitute ‘weak’ for ‘slight’ and ‘strong’ for ‘severe.’ 2 Resting tachycardia (1 120 min–1) Table adapted from ref. 53, 54. 3 Frequent ventricular or atrial ectopy 4 Moderate aortic stenosis ECG electrodes proposed by Mason and Likar [59] is rec- 5 Moderate to severe pulmonary hypertension ommended. A supine resting 12-lead ECG should be 6 Moderate valvular heart disease obtained as the standard ECG for determination of rest- 7 Pregnancy ing abnormalities [60]. An upright baseline ECG is taken 8 Known electrolyte abnormalities and severe anemia as resting baseline. The patient should be familiarized (Hb ! 10 g/100 ml) with the protocol including the incrementing work rate, monitoring devices, and the equipment. Additional absolute or relative contraindications can exist and clinical judgment should always be used to determine the appro- If cycle ergometry is being used, the seat height should priateness of performing an exercise test. be carefully adjusted to optimum by having the patient slowly turn the pedals and look for slight angle of the knee Table compiled from ref. 3, 6, 55, 56, 57. when the pedal is at bottom of its swing, with little hip movement on the seat. Whether cycle ergometry or tread- testing (see chapter on ‘Exercise-Induced Asthma’). The mill is the modality, the patient needs to be instructed in patient should refrain from exercise the day of the test and proper exercise technique: regular, constant revolution of to abstain from smoking for at least 8 h. The subject pedals for cycle ergometry, and relaxed stride on a tread- should avoid eating a heavy meal within 3 h of testing and mill using the hands only for balance by touching the rails should wear comfortable loose fitting clothing and shoes lightly with no tension in the arms. appropriate for exercise. If clinically warranted, PFTS including MVV and spirometry may be performed. De- Symptom scoring and blood pressure are usually ob- pending on the clinical question(s), lung volumes and tained at the end of each stage, or every minute to 2 min in DLCO should also be obtained. A resting arterial blood the case of continuous ramp protocols. Other measure- gas may be appropriate if hypoxemia is suspected. ments, such as arterial blood gases or flow-volume loop measurements can be obtained less frequently, such as eve- Performing the Test ry other minute. Symptoms can be assessed using a variety A consent document is signed, if required. Patient of instruments; the most popular are the original Borg per- preparation includes the placement of ECG electrodes ceived exertion scale or the modified category ratio scale and if necessary, an arterial catheter. Skin preparation for the ECG lead placement is important to decrease the noise on the ECG tracing. The position of the 12-lead 54 Beck/Weisman

Table 9. Criteria for early termination of an exercise test terminating signs or symptoms, testing should be contin- ued until volitional exhaustion. Verbal encouragement is 1 Clinical observation and judgement often helpful in assuring a patient’s maximal effort. A 2 Moderate to severe angina maximal exercise test is usually reflected by a patient 3 1 2 mm horizontal or downsloping ST segment depression or achieving at least one of the following criteria: maximal predicted V˙ O2, heart rate, or V˙ E (relative to MVV), RER elevation 11.15 or lactate 18 mEq/l [63]. A test may be discontin- 4 Serious arrhythmias ued because of serious ECG changes and/or significant symptoms. Despite a patient’s best efforts, physiologic Second or third degree atrioventricular block limitation due to symptom limitation is often not Sustained ventricular tachycardia achieved; assessing patient effort under these circum- Frequent premature ventricular contractions stances is underscored. Finally, if medical complications Atrial fibrillation with a rapid ventricular response arise, the medical monitor should terminate the test. In 5 Severe hypertension (systolic pressure 1 260 mm Hg; diastolic those situations, the patient should be observed until sta- pressure 1 115–120 mm Hg) ble and physiologic variables have returned to baseline. 6 Drop in systolic blood pressure with increasing intensity Based on the judgment of the physician, admission to the accompanied by signs or symptoms, or drop below pre-exercise hospital maybe warranted. Resuscitation equipment values should always be available in the exercise laboratory. 7 Severe wheezing in the chest or upper airways 8 Unusual or severe shortness of breath Patient Safety 9 Ataxia, vertigo, visual or gait problems, confusion or other Extensive literature suggests that symptom-limited ex- signs of central nervous system disorder; and acute myocardial ercise testing in otherwise healthy individuals is a rela- infarction tively safe procedure. In patients, a death rate of approxi- 10 Physical or verbal manifestations of severe fatigue or shortness mately 0.5 per 10,000 has been reported in a survey of of breath 1,375 clinical exercise testing facilities [64]. In more than 11 Signs of poor peripheral perfusion (pallor, cyanosis, cold 70,000 maximal exercise tests performed in a preventive clammy skin) medicine clinic, no deaths were reported, with only 6 12 Leg cramps or extreme pain suggesting claudication major medical complications [3]. In a very large study of cycle ergometry exercise testing involving 11 million Modified from ref. 3, 68. sports persons and patients, 2 deaths per 100,000 tests were reported in patients with a chronic disease [65]. Sud- (table 8) [53, 54]. Increasingly, the modified category ratio den death during exercise testing was evaluated in eight scale is also being used for leg fatigue and chest pain. Alter- studies analyzed by the American Heart Association re- natively, these symptoms can be assessed with simple 0–4 vealing a rate of 0–5 per 100,000 exercise tests [66]. More scales (0 = none, 4 = severe). Testing personnel should recently, analysis of 75,828 exercise tests performed in the observe the patient for signs of fatigue or other symptoms, Veterans Affairs Health Care System revealed an event and monitoring ECG and gas exchange data. rate of 1.2 per 10,000 exercise tests performed (myocar- dial infarction, ventricular tachycardia) with no deaths When the patient stops exercise, it is important to [67]. It can be reasonably concluded therefore that exer- determine the main causes for termination (shortness of cise testing is safe, that the risk of medical complications breath, chest discomfort, leg fatigue or leg pain, etc.). The during exercise testing is related to the underlying disease, patient should be observed for signs of shortness of and that the morbidity for patients resulting from exercise breath, faintness, unusual weakness or other symptoms. testing is 2–5 per 100,000 clinical exercise tests. The subjective impression of the testing personnel of why exercise stopped should also be noted as this can differ Evaluation of Ventilatory Limitations from the patient’s assessment. In some cases, auscultation can be performed to detect breath sounds either over the Assessing the degree of ventilatory limitation has tradi- lung fields or the trachea and upper airway. As noted pre- tionally been based on the ventilatory reserve or on how viously (‘Exercise Protocols’), monitoring heart rate in the close the maximal V˙ E achieved during exercise ap- recovery period may provide valuable prognostic infor- mation [31, 32]. Criteria for terminating an exercise test appear in table 9 [2, 6, 61, 62]. If monitoring fails (e.g. ECG or blood pressure), consideration should be given to stopping, de- pending on the clinical circumstances. In the absence of Exercise Testing Methodology 55

MVV HEAVY EXERCISE Flow MFVL Flow MFVL 12 MVV FRC 12 8 RV 8 Pk Exercise 4 FRC 0 4 -4 TLC Fig. 3. Flow vs. volume loops obtained dur- 0 RV ing a maximal voluntary ventilation (MVV) -4 TLC EELV test and near maximal exercise in a healthy -8 young adult. Note the encroachment on both -8 EELV the expiratory and inspiratory MFVL enve- lope, the higher lung volume and attendant -12 0 24 6 -12 0 24 6 increased elastic load which increases the -2 -2 work of breathing during the MVV maneuv- VOLUME VOLUME er compared with exercise. Figure adapted from Johnson et al. [86]. proaches the maximal voluntary ventilation (MVV) or maximum voluntary ventilation (MVV) or a rise in some estimate of the MVV that is used as an index of PaCO2) and that ventilatory limitation is not an ‘all or ventilatory capacity. The standard maximal voluntary none’ phenomenon. One approach that has gained popu- ventilation (MVV) test is obtained at rest by having a larity is the measurement of the exercise tidal flow volume patient breathe in and out of a spirometer as hard as possi- loop (extFVL) and plotting it within the maximal flow ble for 12–15 s [5, 6, 11]. A rough estimate of the MVV volume loop (MFVL) [74–78]. This technique provides a can be obtained from the FEV1 obtained from spirometry good visual index of the degree of ventilatory constraint, multiplied by about 35–40 [68]. As there is considerable allows a more detailed approach to defining ventilatory variability in the relationship of MVV to FEV1 among limitation (relative to the V˙ E/MVV relationship), and has subjects, especially those with lung disease [68], the ‘cal- gained popularity due to the ease of measurement using culated’ MVV should be used only when directly mea- many of the commercially available automated exercise sured MVV or other direct methods (see below) are not systems. Figure 4 shows an example of the rest and peak available. exercise flow volume responses in a healthy, average fit adult plotted within the MFVL. Although the MVV is practical, simple, and widely applied, it most probably overestimates ‘true’ ventilatory Table 10 lists indices of ventilatory constraint using capacity for two reasons: (1) the maneuver is a short high extFVL plotted within the MFVL [73]. In addition, other intensity effort that cannot be sustained [69], and (2) the parameters can be devised that provide more continuous breathing pattern adopted by most subjects is different information, such as the area between the extFVL and the from exercise ventilation in that breathing occurs at high- MFVL. Further work must be done to determine which er lung volumes and at higher respiratory rates (fig. 3) [70, parameters in addition to visual inspection of the loops 71], and involves a different pattern of muscle activation are useful diagnostic tools. [72]. Another method for determining degree of flow limita- There has been a growing trend in both research and tion is the ‘negative expiratory pressure’ or NEP method clinical laboratories to find alternative ways to evaluate [79–81]. This technique requires a negative pressure ventilatory limitation during exercise [73]. This results source to be connected at the mouth so that at randomly from the appreciation that patients may discontinue exer- selected times during exercise, mouth pressure can be cise due to ventilatory constraints and dyspnea prior to forced to be negative, usually about –10 cm H2O, during achieving the classic indices associated with ventilatory expiration of one breath. An increase in expiratory flow limitation (i.e. minute ventilation (V˙ E) that reached the during a breath where the NEP is applied compared with 56 Beck/Weisman

a normal breath is taken as evidence that flow limitation Fig. 4. Calculation of flow limitation (FL) and inspiratory capacity is not present. (IC) from exercise test flow-volume loops (ext FVL) and maximal flow-volume loops (MFVL). Flow limitation is quantified by measur- Each of these techniques assesses ventilatory limita- ing the volume over which the ext FVL flows meet or exceed the tion in different ways. extFVL analysis is hampered by MFVL flows (Vol of FL) expressed as a % of the tidal volume. Anoth- the fact that the MFVL determined by routine spirometry er useful index of exercise adaptation is the change in inspiratory is affected by gas compression within the chest during the capacity (IC), here indicate for rest (rest IC) and exercise (ext IC). In maximal expiratory effort [82–84], and that the MFVL this normal individual, the IC gets larger during exercise as end-expi- curve itself can be affected by exercise [81, 84, 85]. In ratory lung volume (EELV) drops. In patients, EELV may rise as FL turn, the NEP technique is essentially an ‘all or none phe- occurs earlier in exercise, causing IC to fall. nomenon’ unable to detect the approach to flow limita- tion, but rather only when flow limitation is present; fur- thermore, NEP by itself does not document changes in breathing strategy in response to actual or impending flow limitation [73]. Both techniques have their technical diffi- culties: the NEP requires additional instrumentation and means for measuring and generating the negative mouth pressures during expiration only, whereas the MFVL curve analysis requires drift-free volume and flow signals, and each measurement must be accompanied by a full inspiratory effort from the patient to determine inspirato- ry capacity [86]. The IC measurement, as a reflection of end-expiratory lung volume and overall operational lung volumes is becoming increasingly important in clinical decision-making [35, 75, 87]. A combination of the NEP technique and the use of extFVL/MFVL may provide the greatest amount of information on ventilatory constraints imposed by the lung and chest wall. Acknowledgements Supported by a General Clinical Research Center grant from the National Institutes of Health (MO1-RR00585) and NHLBI grants HL-52230 and HL064368. Table10. Indices of ventilatory constraint using ext FVL plotted within the MFVL Index How it is assessed Role in ventilatory contraints Expiratory flow limitation (%FL) % of ext FVL that meets or FL may indicate airway collapse, increasing WOB, and could exceeds the MFVL trigger reflexes, increasing sensation of dyspnea Elastic load Dynamic rise in EELV EILV/TLC ratio high WOB, increased load on respiratory muscles Inspiratory flow reserve Fall in IC (EELV = TLC – IC) reduces inspiratory muscle length, and increases elastic load area between inspiratory limb of a measure of breathing reserve extFVL and MFVL EILV = End-inspiratory lung volume; TLC = total lung capacity; IC = inspiratory capacity; WOB = work of breathing. Other terms defined in text or figure 4. Table adapted from Johnson et al. [73]. Exercise Testing Methodology 57

References 1 Weisman IM, Zeballos RJ: Cardiopulmonary 18 Myers J, Walsh D, Sullivan M, Froelicher V: 33 Buchfuhrer MJ, Hansen JE, Robinson TE, Sue exercise testing – the need for standardization: Effect of sampling on variability and plateau in DY, Wasserman K, Whipp BJ: Optimizing the Pulmonary perspectives. Am Coll Chest Phys oxygen uptake. J Appl Physiol 1990;68:404– exercise protocol for cardiopulmonary assess- 1992;8:5–8. 410. ment. J Appl Physiol 1983;55:1558–1564. 2 Zeballos RJ, Weisman IM: Behind the scenes 19 Marciniuk DD, Watts R, Gallagher CG: Re- 34 Roston WL, Whipp BJ, Davis JA, Cunning- of cardiopulmonary exercise testing. Clin Chest producibility of incremental maximal cycle er- ham DA, Effors RM, Wasserman K: Oxygen Med 1994;15:193–213. gometer testing in patients with restrictive lung uptake kinetics and lactate concentration dur- disease. Thorax 1993;48:894–898. ing exercise in humans. Am Rev Respir Dis 3 American College of Sports Medicine: Guide- 1987;135:1080–1084. lines for Exercise Testing and Prescription, 20 Garrard CS, Emmons C: The reproducibility of ed 6. Philadelphia, Lippincott Williams & Wil- the respiratory responses to maximum exer- 35 O’Donnell DE, Lam M, Webb KA: Spirometric kins, 2000. cise. Respiration 1986;49:94–100. correlates of improvement in exercise perfor- mance after anticholinergic therapy in chronic 4 Astrand PO: Textbook of Work Physiology. 21 Russell JC, Dale JD: Dynamic torquemeter obstructive pulmonary disease. Am J Respir New York, McGraw-Hill, 1977. calibration of bicycle ergometers. J Appl Physi- Crit Care Med 1999;160:542–549. ol 1986;61:1217–1220. 5 Wasserman K, Hansen JE, Sue DY, Casaburi 36 Oga T, Nishimura K, Tsukino M, Hajiro T, R, Whipp BJ: Principles of Exercise Testing 22 Van Praagh E, Bedu M, Roddier P, Coudert J: Ikeda A, Izumi T: The effects of oxitropium and Interpretation, ed 3. Philadelphia, Lippin- A simple calibration method for mechanically bromide on exercise performance in patients cott Williams & Wilkins, 1999. breaked cycle ergometers. Int J Sports Med with stable chronic obstructive pulmonary dis- 1992;13:27–30. ease. A comparison of three different exercise 6 Jones NL: Clinical Exercise Testing, ed 4. Phil- tests. Am J Respir Crit Care Med 2000;161: adelphia, Saunders, 1997. 23 Huszczuk A, Whipp BJ, Wasserman K: A re- 1897–1901. spiratory gas exchange simulator for routine 7 Weisman IM, Zeballos RJ: An integrated ap- calibration in metabolic studies. Eur Respir J 37 Zeballos RJ, Weisman IM, Connery SM: Com- proach to the interpretation of cardiopulmo- 1990;3:465–468. parison of pulmonary gas exchange measure- nary exercise testing. Clin Chest Med 1994;15: ments between incremental and constant work 421–445. 24 Otsuka T, Kurihara N, Fujii T, Fujimoto S, exercise above the anaerobic threshold. Chest Yoshikawa J: Effect of exercise training and 1998;113:602–611. 8 Mitchell JH, Sproule BJ, Chapman CB: The detraining on gas exchange kinetics in patients physiological meaning of the maximal oxygen with chronic obstructive pulmonary disease. 38 Bruce RA, Kusumi F, Hosmer D: Maximal intake test. J Clin Invest 1958;37:538–547. Clin Physiol 1997;17:287–297. oxygen intake and nomographic assessment of functional aerobic impairment in cardiovascu- 9 Hughson RL, Kowalchuk JM, Prime WM, 25 Whipp BJ, Davis JA, Torres F, Wasserman K: lar disease. Am Heart J 1973;85:546–562. Green HJ: Open-circuit gas exchange analysis A test to determine parameters of aerobic func- in the non-steady-state. Can J Applied Sports tion during exercise. J Appl Physiol 1981;50: 39 Froelicher VE Jr, Brammell H, Davis G, No- Sci 1980;5:15–18. 217–221. guera I, Stewart A, Lancaster MC: A compari- son of the reproducibility and physiologic re- 10 Beaver WL, Wasserman K, Whipp BJ: On-line 26 Zhang YY, Johnson MC, Chow N, Wasserman sponse to three maximal treadmill exercise pro- computer analysis and breath-by-breath graph- K: Effect of exercise testing protocol on param- tocols. Chest 1974;65:512–517. ical display of exercise function tests. J Appl eters of aerobic function. Med Sci Sports Exerc Physiol 1973;34:128–132. 1991;23:625–630. 40 Bruce RA, McDonough JR: Stress testing in screening for cardiovascular disease. Bull NY 11 American Thoracic Society: Standardization of 27 Myers J, Buchanan N, Walsh D, Kraemer M, Acad Med 1969;45:1288–1305. spirometry, 1994 update. Am J Respir Crit McAuley P, Hamilton-Wessler M, Froelicher Care Med 1995;152:1107–1136. VF: Comparison of the ramp versus standard 41 Nagle FS, Balke B, Naughton JP: Gradational exercise protocols. J Am Coll Cardiol 1991;17: step tests for assessing work capacity. J Appl 11A Wanger J, Crapo RO, Irvin CG: Pulmonary 1334–1342. Physiol 1965;20:745–748. function laboratory management and proce- dure manual: A project of the American Tho- 28 Tanner CS, Heise CT, Barber G: Correlations 42 Pollock ML, Wilmore JH, Fox SM III: Exercise racic Society. American Thoracic Society, New of the physiologic parameters of a continuous in Health and Disease. Philadelphia, Saunders, York, NY, 1998/2001. ramp versus an incremental James exercise 1984. protocol in normal children. Am J Cardiol 12 Beaver WL: Water vapor corrections in oxygen 1991;67:309–312. 43 Zeballos RJ, Weisman IM: Reliability of non- consumption calculations. J Appl Physiol invasive oximetry in black subjects during ex- 1973;35:928–931. 29 Bader DS, Maguire TE, Balady GJ: Compari- ercise and hypoxia. Am Rev Resp Dis 1991; son of ramp versus step protocols for exercise 144:1240–1244. 13 Proctor DN, Beck KC: Time delays adjustment testing in patients 1 or = 60 years of age. Am J to minimize errors in breath-by-breath mea- Cardiol 1999;83:11–14. 44 Hansen JE, Cassaburi R: Validity of ear oxime- surement of V˙ O2 during exercise. J Appl Physiol try in clinical exercise testing. Chest 1987;91: 1996;81:2495–2499. 30 Miyahara N, Eda R, Takeyama H, Maeda T, 333–337. Aoe K, Kunichika N, Kohara H, Harad M: 14 Hughson RL, Northey DR, Xing HC, Dietrich Cardiorespiratory responses during cycle er- 45 Ries AL, Fedullo PF, Clausen JL: Rapid BH, Cochrane JE: Alignment of ventilation gometer exercise with different ramp slope in- changes in arterial blood gas levels after exer- and gas fraction for breath-by-breath respirato- crements in patients with chronic obstructive cise in pulmonary patients. Chest 1983;83: ry gas exchange calculations in exercise. Comp pulmonary disease. Int Med 2000;39:15–19. 454–456. Biomed Res 1991;24:118–128. 31 Nishime EO, Cole CR, Blackstone EH, Pash- 46 Rasmussen PH, Staats BA, Driscoll DJ, Beck 15 Beaver WL, Lamarra N, Wasserman K: kow FJ, Lauer MS: Heart rate recovery and KC, Bonekat HW, Wilcox WD: Direct and Breath-by-breath measurement of true alveolar treadmill exercise score as predictors of mortal- indirect blood pressure during exercise. Chest gas exchange. J Appl Physiol 1981;51:1662– ity in patients referred for exercise ECG. 1985;87:743–748. 1675. JAMA 2000;284:1392–1398. 47 Robinson TE, Sue DY, Huszczuk A, Weiler- 16 Sciurba FC, Owens GR, Ondriezek J: The 32 Cole CR, Foody JM, Blackstone EH, Lauer Ravell D, Hansen JE: Intra-arterial and cuff effect of sample interval on maximal values MS: Heart rate recovery after submaximal ex- blood pressure responses during incremental obtained during incremental exercise. Am Rev ercise testing as a predictor of mortality in a cycle ergometry. Med Sci Sports Exerc 1988; Respir Dis 1991;143:A176. cardiovascularly healthy cohort. Ann Int Med 20:142–149. 2000;132:552–555. 17 Slivka WA: Sciurba FC: Parameters and power 48 Sinex JE: Pulse oximetry: Principles and limi- to measure a therapeutic effect (lung reduction tations. Am J Emerg Med 1999;17:59–67. surgery – LVRS) in advanced emphysema. Am J Respir Crit Care Med 1998;157:A92. 58 Beck/Weisman

49 Ralston AC, Webb RK, Runciman WB: Poten- 65 Scherer D, Kaltenbach M: Frequency of life- 81 Koulouris NG, Dimopoulou I, Valta P, Finkel- tial errors in pulse oximetry. III. Effects of threatening complications associated with stein R, Cosio MG, Milic-Emili J: Detection of interference, dyes, dyshaemoglobins and other stress testing. Dtsch Med Wochenschr 1979; expiratory flow limitation during exercise in pigments. Anaesthesia 1991;46:291–295. 104:1161–1165. COPD patients. J Appl Physiol 1997;82:723– 731. 50 Am Assoc Resp Care (AARC): Clinical practice 66 American Heart Association: Exercise stan- guideline: Exercise testing for evaluation of dards. A statement for healthcare profession- 82 Coates AL, Desmond KJ, Demizio D, Allen P, hypoxemia and/or desaturation. Respir Care als. Circulation 1995;91:580–615. Beaudry PH: Sources of error in flow-volume 1992;37:907–912. curves: Effect of expired volume measured at 67 Myers J, Voodi L, Umann T, Froelicher VF: A the mouth vs. that measured in a body plethys- 51 Hansen JE, Sue DY, Wasserman K: Predicted survey of exercise testing: Methods, utilization, mograph. Chest 1988;94:976–982. values for clinical exercise testing. Am Rev interpretation, and safety in the VAHCS. J Respir Dis 1984;129(suppl):S49–S55. Cardiopulm Rehab 2000;20:251–258. 83 Hyatt RE: Effort independence and forced ex- piratory flow. Chest 1980;77:246–248. 52 Lewis DA, Sietsema KE, Casaburi R, Wasser- 68 Beck KC: Evaluating exercise capacity and air- man K: Inaccuracy of non-invasive estimates way function in the athlete; in Weiler JM (ed): 84 Rodarte JR: Detection of expiratory flow limi- of VD/VT in clinical exercise testing. Chest Allergic and Respiratory Disease in Sports tation during exercise in COPD patients (in- 1994;106:1476–1480. Medicine. New York, Marcel Dekker, 1997. vited editorial). J Appl Physiol 1997;82:721– 722. 53 Borg GAV: Psychophysical bases of perceived 69 Freedman S: Sustained maximum voluntary exertion. Med Sci Sports Exerc 1982;14:377– ventilation. Resp Physiol 1970;8:230–244. 85 Johnson BD, Scanlon PD, Beck KC: Regula- 381. tion of ventilatory capacity during exercise in 70 Hyatt RE: The interrelationships of pressure, asthmatics. J Appl Physiol 1995;79:892–901. 54 Borg GAV: Perceived exertion as an indicator flow and volume during various respiratory of somatic stress. Scan J Rehab Med 1970;2–3: maneuvers in normal and emphysematous sub- 86 Johnson BD, Weisman IM, Zeballos RJ, Beck 92–98. jects. Am Rev Respir Dis 1961;83:676–683. KC: Emerging concepts in the evaluation of ventilatory limitation during exercise: The ex- 55 Froelicher VF, Marcondes GD: Manual of Ex- 71 Olafsson S, Hyatt RE: Ventilatory mechanics ercise tidal flow-volume loop. Chest 1999;116: ercise Testing. St Louis, Mosby Year Book, and expiratory flow limitation during exercise 488–503. 1989. in normal subjects. J Clin Invest 1969;48:564– 573. 87 Babb TG: Mechanical ventilatory constraints 56 Zavala DC: Manual on Exercise Testing: A in aging, lung disease, and obesity: Perspective Training Handbook, ed 3. Iowa City, Universi- 72 Klas JV, Dempsey JA: Voluntary versus reflex and brief review. Med Sci Sports Exerc 1999; ty of Iowa, 1993. regulation of maximal exercise flow: Volume 31(suppl):S12–S22. loops. Am Rev Respir Dis 1989;139:150–156. 57 Pina IL, Balady GJ, Hanson P, Labovitz AJ, 88 Nordrehaug JE, Danielson R, Strangeland L, Madonna DW, Myers J: Guidelines for clinical 73 Johnson BD, Beck KC, Zeballos RJ, Weisman Rosland GA, Vik-Mo H: Respiratory gas ex- exercise testing laboratories: A statement for IM: Advances in pulmonary laboratory testing. change during treadmill exercise testing: Re- healthcare professionals from the Committee Chest 1999;116:1377–1387. producibility and comparison of different exer- on Exercise and Cardiac Rehabilitation, Amer- cise protocols: Technical notes. Scand J Clin ican Heart Association. Circulation 1995;91: 74 Johnson BD, Reddan WG, Seow KC, Dempsey Lab Invest 1991;51:655–658. 912–921. JA: Mechanical constraints on exercise hyper- pnea in a fit aging population. Am Rev Respir 89 Cox NJM, Hendriks JCM, Binkhorst RA, Fol- 58 American College of Obstetricians and Gyne- Dis 1991;143:968–977. gering HTM, van Herwaarden CLA: Repro- cologists Technical Bulletin: Exercise during ducibility of incremental maximal cycle ergom- pregnancy and the postpartum period. 1994; 75 Babb TG: Ventilatory response to exercise in eter tests in patients with mild to moderate Number 189. subjects breathing CO2 or HeO2. J Appl Physi- obstructive lung disease. Lung 1989;167:129– ol 1997;82:746–754. 133. 59 Mason RE, Likar I: Experimental and laborato- ry reports: A new system of multiple-lead exer- 76 Johnson BD, Saupe KW, Dempsey JA: Me- 90 Noseda A, Carpiaux JP, Prigogine T, Schmer- cise electrocardiography. Am Heart J 1966;71: chanical constraints on exercise hyperpnea in ber J: Lung function, maximum and submaxi- 196–205. endurance athletes. J Appl Physiol 1992;73: mum exercise testing in COPD patients: Re- 874–886. producibility over a long interval. Lung 1989; 60 Gamble P, McManus H, Jensen D, Froelicher 167:247–257. V: A comparison of the standard 12-lead elec- 77 Marciniuk DD, Watts R, Gallagher CG: Dead trocardiogram to exercise electrode place- space loading and exercise limitation in pa- 91 Owens MW, Kinasewitz GT, Strain DS: Evalu- ments. Chest 1984;85:616–622. tients with interstitial lung disease. Chest 1994; ating the effects of chronic therapy in patients 105:183–189. with irreversible air-flow obstruction. Am Rev 61 Lollgen H, Ulmer H-V, Crean P (eds): Eur Respir Dis 1986;134:935–937. Heart Assoc Report of the Task Force Confer- 78 Marciniuk DD, Sridhar G, Clemens RE, Zintel ence on Ergometry. Recommendations and TA, Gallagher CG: Lung volumes and expira- 92 Meyer K, Westbrook S, Schwaibold M, Hajric standard guidelines for exercise testing. Eur tory flow limitation during exercise in intersti- R, Peters K, Roskamm H: Short-term repro- Heart J 1988;9(suppl K):1–37. tial lung disease. J Appl Physiol 1994;77:963– ducibility of cardiopulmonary measurements 973. during exercise testing in patients with severe 62 Am College Card/Am Heart Assoc Guidelines chronic heart failure. Am Heart J 1977;134: for exercise testing: A report of the American 79 Eltayara L, Becklake MR, Volta CA, Milic- 20–26. College of Cardiology/American Heart Asso- Emili J: Relationship between chronic dyspnea ciation Task Force on Practice Guidelines and expiratory flow limitation in patients with Kenneth C. Beck, PhD (Committee on Exercise Testing). J Am Coll chronic obstructive pulmonary disease. Am J Physiological Imaging Laboratory Cardiol 1997;30:260–311. Respir Crit Care Med 1996;154:1726–1734. Department of Radiology University of Iowa Hospitals and Clinics 63 Shephard RJ: Test of maximum oxygen intake: 80 Mota S, Casn P, Drobnic F, Giner J, Ruiz O, 200 Hawkins Drive A critical review. Sports Med 1984;1:99–124. Sanchis J, Milic-Emili J: Expiratory flow limi- Iowa City, IA 52242-1077 (USA) tation during exercise in competition cyclists. J Tel. +1 319 356 1381, Fax +1 319 356 1503 64 Stuart RJ Jr, Ellestad MH: National Survey of Appl Physiol 1999;86:611–616. E-Mail [email protected] exercise stress testing facilities. Chest 1980;77: 94–97. Exercise Testing Methodology 59

Weisman IM, Zeballos RJ (eds): Clinical Exercise Testing. Prog Respir Res. Basel, Karger, 2002, vol 32, pp 60–71 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Deconditioning, and Principles of Training Thierry Troostersa,b Rik Gosselinka,b Marc Decramera,b aRespiratory Division and Respiratory Rehabilitation, University Hospital Gasthuisberg, Leuven, bFaculty of Physical Education and Physiotherapy, Department of Rehabilitation Sciences, Katholieke Universiteit Leuven, Belgium Summary vent deconditioning, rather than to improve exercise perfor- mance, significantly less exercise is warranted. Once-weekly Exercise capacity and muscle strength decline rapidly after endurance training, or just few resistance exercises, may be onset of immobilization, bed rest or significant reduction in dai- sufficient to prevent deterioration of muscle function during ly activities. The functional consequences of deconditioning are periods of inactivity, or when formal training is ended. This is increased mortality and comorbidity risk. In patients with an important feature in the prevention of deconditioning and underlying chronic disease (like myocardial infarction, chronic needs much more attention. obstructive pulmonary disease, congestive heart failure and renal failure), deconditioning has been associated with impaired It is generally accepted that regular exercise improves prognosis, independently of the underlying primary disease functional capacity, prevents morbidity and positively process. Deconditioning is highly prevalent in elderly, where it influences survival [1]. Deconditioning results in signifi- is often catalyzed by sarcopenia, observed from the 6th to 7th cant morbidity and has been associated with compro- decades on. The mechanisms of the rapid decline in skeletal mised survival, and reduced quality of life. In a longitudi- muscle function with inactivity are yet not fully understood, but nal study by Blair et al., who followed participants for 15 oxidative stress, suppression of insulin-like growth factor I and years [2], the least fit subjects had a threefold increased II and up-regulation of myostatin are currently suggested as key risk of dying compared to the most fit subjects. In a con- actors in the atrophy process. Because of the significant conse- secutive study the authors showed that subjects who quences of deconditioning, all efforts should be made to pre- improved their fitness over time reduced their relative vent and treat the onset, or further aggravation of the decondi- risk for dying by approximately 50% [3]. The present tioning process. Exercise training is a potent therapy for decon- chapter deals with both above-mentioned issues: decondi- ditioning. Training programs can consist of whole body exer- tioning on the one hand, and exercise training as an cise (endurance training) or resistance training, where isolated important tool to treat deconditioning, and improve exer- muscle groups are trained. It is widely accepted that exercise cise performance on the other hand. training is conducted at least 3 times weekly at high training intensity in order to improve exercise capacity and/or muscle Reduction in exercise performance or skeletal muscle strength. Special attention is needed for patients suffering from function may result from disease-specific processes, or severe deconditioning or underlying illness. In these cases, may be the result of disuse, prolonged bed rest or sedenta- more specific guidelines are proposed. The impact of both ry lifestyle. For the purpose of this chapter we will refer to endurance and resistance training on the cardiovascular system deconditioning as reduced exercise capacity and skeletal and the skeletal muscle is impressive, and deconditioning is muscle function due to disuse (immobilization, bed rest, restored within weeks of adequate training. In order to pre-

Table1. Summary of most important Impact Ref. impacts of deconditioning, inactivity or bed rest on different organ systems (a list of Cardiocirculatory Reduction in peak cardiac output, primarily 129, 130, review articles is added) system through reduction of stroke volume 5 Heart rate is increased at iso-work Pulmonary system Plasma volume is reduced, as well as Hb content Endocrine system Orthostatic intolerance is present, probably as a Skeletal consequence of cardiac atrophy and hypovolemia Central nervous system Little impact on resting pulmonary function, 131, 132, reduced ventilatory efficiency 16 Androgen levels decrease 133, 16 Insulin sensitivity decreases Bone loss, stiffness tendons 134 Mental concentration impaired, sleep quality reduced 135 or sedentary lifestyle) in the absence of any underlying due to significant reductions in ß-hydroxyacyl-CoA dehy- primary disease. The impact of specific diseases will be drogenase and citrate synthase, without changes in glycolyt- dealt with in other chapters. Skeletal muscle adaptations ic enzyme activity. Conflicting data are reported on capil- due to reduced activation (e.g. paralysis or motor neuron larization after relatively short-term (weeks to 3 months) blockade) are beyond the scope of the present review. The immobilization or detraining [14, 15], but since exercise focus of the present chapter will be on the skeletal mus- training increases capillarization, one may expect that long- cles. However, the impact on different organ systems, and term inactivity may reduce capillary density [16]. Although organ system responses, cannot be overlooked. Table 1 reduced peak exercise performance has been associated summarizes briefly the general impact of inactivity on the with reduced survival both in healthy subjects and patients otherwise healthy body, along with a concise list of review with heart disease [17], consequences of deconditioning at papers dealing with the impact of inactivity on these spe- submaximal exercise, and loss of skeletal muscle strength cific organ systems. may be more relevant to the daily functioning. Skeletal Muscle Abnormalities Associated with Models of deconditioning using bed rest show rapid Deconditioning (within days to weeks) and impressive loss of skeletal In clinical practice, deconditioning is characterized by muscle mass, a negative protein balance [18], accompa- impaired functional status and reduced peak oxygen con- nied by a disproportionately greater (\"30%) reduction in sumption [4]. In the classical study by Bengt Saltin et al. muscle strength [19, 20]. The latter has been attributed to [4], the authors reported a reduction of 28% in VO2peak the impaired maximal neural input (motor neuron re- after 20 days of bed rest. In a very recent follow-up study cruitment) [20] resulting in a decreased ratio ‘muscle the authors suggest that 30 years of aging may have less strength to muscle cross-sectional area’. Reduction in spe- impact on VO2peak than 20 days of bed rest [5]. These cific tension of the muscle is evidenced by a significantly data were later replicated by several other studies. The increased submaximal activation (EMG activity to elicit a reduction in VO2peak was reported to be related to the torque of 100 nm) after bed rest. Even after a period as initial VO2peak, and the duration of the bed rest [6]. short as 10 days, these findings were observed (albeit less Besides the reduction in peak exercise capacity, early impressive) [21]. Occasionally, necrotic changes, cellular onset of lactic acidosis [7–10] during incremental exercise edema, extracellular mitochondria and disorganized testing has been reported. Typically the onset of lactate myofibrils were reported after simulated microgravity, accumulation starts at a VO2 less than 40% of the pre- suggesting presence of myopathic changes besides atrophy dicted maximal VO2 [11]. After a period of decondition- [14]. ing, the lactate threshold was reported to shift to lower workloads [12]. Muscle bioenergetics are impaired [13], After 42 days of bed rest, Ferretti et al. [22] observed a 16% reduction in peak oxygen consumption. These au- thors studied comprehensively all factors contributing to Deconditioning, and Principles of Training 61

Fig. 1. Changes after 37 days of bed rest in healthy volunteers in sion [30] or bed rest [31] failed to show any effect on the (1) cross-sectional area of the thigh (CSA), (2) volume density of total atrophy. Protein balance, however, was better preserved mitochondria (VDM), (3) total capillary length (CL) in an identical by anabolic hormone administration, but there were no muscle slice from [22] and (4) extraction ratio at iso (submaximal) benefits in terms of muscle strength or function. These work rate (100 W) (ER(100 W)) from the same study, published in Fer- findings indicate that the impact of the mechanical stimu- retti et al. [138]. lation of the muscle cannot be replaced by increasing pro- tein synthesis alone. It has been suggested that increased the loss in peak VO2. Figure 1 summarizes some of the oxidative stress induced by the inactivity may play a role important consequences of bed rest on the skeletal mus- [32], since administration of antioxidant vitamin E signif- cle, which contribute significantly to the reduced peak icantly counteracts the atrophy observed after immobili- VO2. Besides the important decrease in peak cardiac out- zation of rat hindlimb [33]. Although appealing, oxidative put, the authors found a reduced oxidative capacity of the stress alone cannot explain the processes related to muscle limb muscles, and a reduced extraction of oxygen at iso- deconditioning and muscle wasting. Other mechanisms, work rate. It is noteworthy that in human muscle biopsies such as up-regulation of myostatin, a promoter of muscle taken after bed rest, atrophy is consistently observed in all atrophy, and down-regulation of insulin-like growth fac- muscle fibers of the m. quadriceps [20], whereas in animal tor II, have been suggested in recent literature to explain experiments atrophy is predominantly confined to type I the atrophy seen after 17 days of space flight [34]. Ele- (oxidative fibers). The changes seen in the muscle have vated growth differentiation factor (GDF)-8 or myostatin been associated to the unloading of the muscle. Indeed, levels have indeed been associated to sarcopenia in elder- bed rest itself did not result in changes in the deltoid mus- ly subjects and during hindlimb unloading [35]. Insulin- cle [23], a muscle which remains functional during these like growth factor administration onto previously immo- experiments. It is generally accepted that the abnormali- bilized muscle showed in older rat to increase the prolifer- ties observed after bed rest, are more most expressed in ation potential of satellite cells and reduced the immobili- weight-bearing (antigravity) muscles [24]. zation-induced atrophy [36]. Insulin-like growth factor I, on the other hand, may potentially suppress proteolysis Mechanisms of Deconditioning Induced Atrophy (and thus atrophy) via its interaction with the ubiquitin The mechanisms that underlie the muscle atrophy are pathways [37, 38]. A last family of factors which may play far from unraveled. Inactivity seems to be devastating to a role in the deconditioning observed after immobiliza- the muscle. Muscle wasting in humans starts almost tion or bed rest may be the myogenic regulatory factors immediately after application of bed rest [25]. It is impor- (MRFs). These regulatory factors bind to the regulatory tant, however, that the alterations in the signaling process, regions of certain muscle genes and activate their tran- detected at the mRNA levels, are often more impressive scription. Loughna and Brownson [39] showed that than the result at the protein level [26, 27]. Although MRF4 was dramatically down-regulated in the soleus decreased protein synthesis is probably the most impor- muscle (weight-bearing, oxidative muscle) after hindlimb tant contributor to the negative protein balance [18, 28, immobilization. This, however, was not observed in other 29], studies trying to stimulate protein synthesis by ad- muscles. Other muscle-specific regulatory factors such as ministrating anabolic hormone during hindlimb suspen- MyoD and myogenin, seem to play a less dominant role in the atrophy seen after hindlimb suspension in animal models [40]. In a recent report the serial analysis of gene expression (SAGE) technique was used to accurately mea- sure the expression of genes to evaluate the effect of immobilization on gene expression in rat hindlimb (gas- trocnemius) muscle [41]. Among other interesting find- ings, these authors report up-regulated genes involved in proteolysis after immobilization, whereas those involved in protein elongation were down-regulated. A threefold decrease in genes expression involved in energy metabo- lism was observed. Since most of the mechanisms remain far from elucidated, further research should focus on the mechanisms involved in the etiology of the muscle atro- 62 Troosters/Gosselink/Decramer

phy seen after short and longer periods of skeletal muscle (COPD) accumulate inactivity with aging, and disease- unloading. Further insight in the mechanisms of decondi- specific factors such as tissue hypoxia, inflammation, tioning could yield a first approach to prevent muscle excessive oxidative stress, associated with excessive cyto- wasting during periods of inactivity. This may significant- kine release, and deleterious effects of cytotoxic drugs ly reduce inactivity-related morbidity and/or mortality. such as corticosteroids [57]. This complex interaction of different factors makes it difficult to distinguish between Deconditioning in Clinical Practice the derangement in the skeletal muscle due to decondi- Transition from clean animal models and clinical ex- tioning alone, and these induced by disease-specific prob- periments to clinical practice is a complex step. Indeed, lems. The latter may result in myopathy rather than atro- deconditioning is seldom an isolated process. Often it is phy of the skeletal muscle [57]. The combination of sys- associated with aging, cytotoxic drugs [42, 43], chronic temic inflammation, increased oxidative stress, age-com- undernutrition (insufficient protein uptake), or chronic promised genetically affected muscle, may make the mus- disease inducing inactivity. Aging is in the context of the cle of patients with chronic diseases particularly vulnera- present chapter of particular interest, since subjects, as ble to even short bouts of inactivity. Disregarding its etiol- they get older, tend to decrease their accustomed level of ogy, it has been convincingly shown that skeletal muscle activity [44], especially if they are institutionalized [45]. A weakness is an important contributor to exercise intoler- recent analysis of the available literature revealed that in ance in COPD [58], congestive heart failure [59, 60] and elderly (165–70 years) only 30% of the total daily energy renal failure [61]. expenditure resulted from activities [46]. Aging itself has been associated with sarcopenia, defined as the age-spe- The consistent finding that relatively short (8 weeks) cific, unintentional, loss of skeletal muscle mass [47], exercise training in moderate COPD restores – to a large leading to reductions in muscle strength [48], functional extent – skeletal muscle bioenergetics [62], lactic thresh- exercise capacity [49] and peak oxygen consumption [50, old [63] and oxidative enzymes [64], favors the hypothesis 51]. Moreover, skeletal muscle of elderly showed to have that muscle abnormalities in chronic diseases such as less mitochondrial density, and decreased oxidative ca- COPD are the net result of deconditioning only. It is how- pacity per mitchondrial volume [52]. Unlike inactivity, ever important to indicate that skeletal muscle strength aging has been associated, in cross-sectional studies, with and CSA did improve after rehabilitation, but did not predominantly fiber type II atrophy. In a recent longitudi- return to normal values [65–67]. In healthy subjects, nal study, however, type I fiber proportion was signifi- training studies after previous deconditioning almost cantly reduced after 10 years of follow-up [48], and a unanimously report full recovery of skeletal muscle de- reduction of capillary to fiber ratio was observed. Aging is rangements after few weeks of exercise training [19, 68]. – as inactivity – associated with reduced satellite cell pro- This observation, in combination with the description of liferation in humans [53], which has been suggested as an myopathic changes [57], and the observation of patients important step in the age-associated sarcopenia [36]. In- in whom the relative muscle strength (strength/CSA) was terestingly, it was reported recently that aging is associat- clearly below normal, suggest that deconditioning alone ed with increased amount of genetic point mutations, a cannot explain the abnormalities observed in the skeletal phenomenon particularly present in the skeletal muscle muscle function observed in patients suffering from [54]. Moreover, aging showed to be associated with signif- chronic diseases [69]. Moreover, in animal models of icant denervation of muscle fibers [55, 56], which may be heart failure, inactivity itself has formally been excluded another potential pathway for the development of sarco- as a contributing factor [70]. In the clinical setting how- penia. The latter two observations may indicate that the ever, we feel that both inactivity and intrinsic disease- reduction in muscle function frequently observed in el- induced changes contribute to the skeletal abnormalities derly subjects is not only related to reduction in activity, and may strongly interact leading to clearly abnormal but may be promoted by the aging process itself. skeletal muscle function. Impact of Deconditioning in Chronic Disease Treatment for Deconditioning Patients suffering from a chronic disease have im- paired exercise performance, which can be at least partial- Exercise training has shown to be a potent therapy to ly attributed to deconditioning. Patients with chronic reverse the muscular changes induced by deconditioning. heart failure or chronic obstructive pulmonary disease Improved physical condition as a result of exercise train- Deconditioning, and Principles of Training 63

ing can be expected, and, as mentioned before, is related these effects, typically whole body endurance exercise to improved survival. Reversal of deconditioning in pa- training is applied. When the aim of the training program tients after myocardial infarction showed to be directly is to improve skeletal muscle force, or muscle cross-sec- related to reduced mortality [71] and subjects who im- tional area, resistance training can be applied, involving proved their activity status had significantly lower risk of small muscle groups. These two training types are dis- dying [3]. cussed consecutively in the section below. In the following section of this chapter, general princi- Aerobic Exercise ples of exercise training programs are discussed along The American College for Sports Medicine (ACSM) with their recognized effects in healthy (deconditioned) recently reviewed the available literature on aerobic exer- subjects. Aerobic training (endurance and interval) and cise training in healthy subjects [79], and subsequently strength (resistance) training will be discussed. It is impor- made guidelines for exercise training in elderly [80]. tant to consider that the reported effects of exercise train- Table 2 summarizes the generally accepted features of ing show a large variability. A highly standardized train- exercise training sessions in order to result in significant ing program in healthy subjects (age 17–65, n = 481) physiological training effects. reported a 17% increase in VO2 peak, with a range of –5 The above-mentioned guidelines aim at improving to +57% change in VO2peak after training. It is important physical fitness of unfit but otherwise healthy subjects. to mention that this study did not find any race, gender or Applying these guidelines will result in improvements in age effect when the improvements of exercise training peak oxygen consumption. The ACSM acknowledges that were expressed as relative gain, compared to baseline (% lower exercise intensity showed to be enough to reduce initial). From the same prestigious HERITAGE study, a risk for chronic degenerative diseases, and result in both genetic predisposition to VO2peak responses after train- quantity and quality of life benefits. Exercise training pro- ing was found [72], accounting for 47% of the variability grams aiming at treatment or prevention of chronic dis- in training response. The authors report 2.5 times more eases often apply somewhat different training strategies. variability between families than within families. Hence, Although the general principles remain the basis of these when effects of exercise training are appreciated in indi- training programs, the ACSM has introduced specific vidual patients, the intrinsic predisposition of a subject to guidelines for training in patients with coronary heart dis- present a training effect on the outcome used in the assess- ease [81], hypertension [82], osteoporosis [83] and obesity ment (e.g. VO2peak) should be considered. Therefore, a weight control. Training programs which can be used in large group of genetically unrelated subjects is needed to the treatment of COPD are reviewed in this book in the evaluate the general effects of a given exercise training chapter ‘The Importance of Exercise Training in Pulmo- program. Small studies possibly including genetically re- nary Rehabilitation’, and in a recent European Respirato- lated subjects can potentially be misleading. A second ry Society document [84]. Training programs in different comment refers to the impact of exercise training on daily diseases and frail elderly may successfully deviate from life activities. Exercise training has shown to improve the above-mentioned guidelines when the aim of the indices of physical function, and quality of life in virtually training program is to improve function, activities of dai- all patient groups in whom exercise training is applied as a ly living and general well-being, without improving therapeutic, or preventive approach. Transfer to actually VO2peak. In this case, lower exercise intensities or shorter increasing measured daily life activities is not always training time is allowed. From a meta-analysis summariz- guaranteed [73–76]. If the aim of a training program is to ing 34 training studies, Londeree [85] concluded that increase daily activities, special attention should be paid physiological benefits from exercise training are only ob- to the implementation of the physiologic gains in the tained when training intensity is at or above the ventilato- everyday life. This should be a point of attention for the ry (and probably lactic) threshold in healthy sedentary persons supervising the training program. subjects. A study investigating the effect of different exer- cise intensities on exercise capacity suggested that the General Principles of Exercise Training intensity threshold should be approximately 40% of the Training can aim at improving general cardiovascular VO2peak. In the same study, exercise training at 80% of fitness, including improvement in peak oxygen uptake peak VO2 resulted in more impressive benefits in exercise and muscle oxidative capacity, and may add to the pre- tolerance [86]. vention of chronic diseases such as cardiovascular disease [77], type II diabetes and colon cancer [78]. To achieve 64 Troosters/Gosselink/Decramer

Table 2. Summary of American College of Minimal requirement exercise training (ACSM recommendations) Sports Medicine Guidelines for developing and maintaining cardiorespiratory and Intensity 55–90% of HRmax; 40–85% of oxygen uptake reserve or heart rate muscular fitness [79] reserve (see footnote) Duration/session Frequency 20–60 min, continuous or intermittent Mode 3–5 days per week Whole body exercises (cycling, walking ...) The heart rate reserve is calculated as HRpeak – HRrest. The training intensity is calculated by adding a percentage of this value to the resting HR. Hence training pulse rate = resting pulse rate + [0.4 – 0.85 ! (maximum HR – resting HR)]. Suppose that the resting heart rate and peak heart rate at the end of an incremental exercise test in a 55 year-old man would be respectively 71 and 166 min–1. A reasonable training pulse would be 71 + [0.7 ! (166 – 71)] = 136 min–1. Training intensity according to VO2 reserve is done using essentially the same strategy. The guidelines on training intensity summarized in Table 3. Alternative tools that can be used to assure high training table 2 only apply for subjects in whom peak exercise per- intensity in the absence of maximal incremental exercise data formance is limited by reaching the boundaries of the car- diocirculatory system. In these subjects the peak heart Threshold for adequate training intensity rate reaches the predicted maximum heart rate, and car- diac output levels at peak exercise [87]. In subjects limited Blood lactate levels 4–5 mEq in their exercise performance for other reasons (heart fail- ure, ventilatory limitation, gas exchange in the lung, or in Gas exchange threshold At or around GET or ventilatory the skeletal muscle, or plain skeletal muscle weakness, (GET) threshold [136] including myopathies), these guidelines should be applied with caution. Heart rate may be unreliable as the only Symptoms Ability to talk while doing exercise variable in the appraisal of the training intensity. It is important to point out that high training intensity is war- Symptom scores 5–6 on 10-point scale [137] ranted, and feasible, in most chronic diseases in order to (e.g. Borg ratings) achieve physiologic benefits of endurance exercise train- ing. The intensity should be high relative to the perfor- rates. This, however, is not necessarily accompanied by mance capacity of the patients. Generally a training inten- increased systemic blood lactate levels, since the total sity at or around 70% of VO2peak is considered adequate, amount of working muscle mass may still be very small. both in health and chronic disease like COPD. Since peak Hence patients with chronic diseases, unlike decondi- work rate may considerably vary with different incremen- tioned, but otherwise healthy subjects may achieve signif- tal exercise testing approaches, exercise training at 70% of icant benefits from endurance training in the absence of the VO2 reserve (VO2rest + [0.7 ! (VO2rest – VO2peak)] whole body blood lactate accumulation during training, rather than at 70% of peak work rate should be achieved provided that the intensity is high, relative to the maximal to be methodologically correct [88]. It has been suggested workload patients can achieve. that lactate production is necessary to obtain physiologi- cal effects of exercise training [89]. However, more recent Measurements of VO2 during all exercise modalities studies consistently showed that high relative workloads used in a training program is in most exercise training set- were a prerequisite to achieve training effects, indepen- tings not feasible in the clinical routine. Since appropriate dently of the achieved blood lactate accumulation [90– training intensity is critical to achieve physiological bene- 93]. Patients with chronic disease, including COPD [62], fits, other means of estimating training intensity should and chronic heart failure [94], show reduced intramuscu- be used. Table 3 summarizes other tools that may be used lar pH, suggestive of intramuscular acidosis at low work to successfully target training intensity. Deconditioning, and Principles of Training 65

The training time should be minimally 20–30 min. available oxygen is the net result of increased number of Typically, endurance training consists of 30 min contin- mitochondria, increased oxidative activity of the existing uous exercise training, but recent studies support that in mitochondria, increased number (and CSA) of oxidative sedentary subjects, this duration of exercise can be spread (fiber I, IIa) muscle fibers. Peak muscle oxygen extraction over the day (e.g. three bouts of 10 min), with similar ratio only increases by approximately 10% in healthy sed- effectiveness [95]. entary subjects starting an endurance exercise program, while peak muscle VO2 increases by approximately 40%. Practical Implementation of Endurance Training. En- This illustrates that the increase in oxygen supply, and the durance training is preferably done on equipment which increase in oxygen demand are both important factors to involves large muscle mass (whole body ergometers). result in an appreciable effect of endurance training [99]. Amongst them treadmills, rowing machines, steppers or Hence in order to achieve large benefits in VO2peak as stair climbing, and bicycles are the most popular. Cycle seen in healthy young subjects, the peak oxygen supply ergometers have the advantage that workload can be con- should be increased. This is achieved by increasing peak trolled and adjusted easily. On a treadmill and stepper, cardiac output, and peak alveolar ventilation. In situa- the workload imposed is highly dependent on the body tions where the alveolar ventilation cannot be increased weight of the subject. Rowing devices engage a large (e.g. obstructive lung disease), or cardiac output is limited amount of skeletal muscle, which may result in difficulties by the disease (e.g. heart failure), improvement in in applying appropriate workload to each individual mus- VO2peak is limited. Submaximal exercise capacity and cle in elderly and severely impaired patients with chronic skeletal muscle performance may virtually normalize af- diseases. Training effects will be specific to the exercise ter exercise training [62, 100]. In many chronic diseases involved in the training, although some transfer to other VO2peak can thus not be used as a unique measure of activities may be achieved [96]. Hence, when training is skeletal muscle adaptations to exercise training. performed on a bicycle, training effects will be much larg- er when evaluated during incremental bicycle exercise Within the skeletal muscle fiber, endurance training [97]. In this context, training programs for patients and typically increases oxidative enzyme content. In younger frail elderly should preferably integrate activities relevant subjects however the latter seems to be the result of an to their daily life. Hence training programs consisting augmented mitochondrial density after training [101], solely of cycling in elderly and patients with chronic dis- whereas in elderly it may be the result of restoring mito- ease may result in important physiologic benefits, while chondrial function without apparent changes in mito- benefits towards increased activities of daily living may chondrial density [102, 103]. A recent report by Starritt et be limited. Walking and stepping exercises may be more al. [104] suggested that citrate synthase activity was appropriate in gaining ADL benefits. Whatever exercise increased after as few as 5 days of endurance training in modality (or combination of different modalities) is de- healthy, young subjects. Although research has focussed cided to be appropriate, supervising physiotherapists, or on the local effects of endurance training on the muscle, or physicians should reassure adequate (high) training inten- on whole body responses to exercise, systemic effects of sity for each session (see table 3). endurance training should not be excluded, especially in selected patient groups. Indeed, Linke et al. [105] showed Endurance Training: Impact on the Muscle. Aerobic correction of radial artery endothelial dysfunction after exercise training, as described above, has been studied lower limb exercise training. Other ‘unexpected’ second- extensively, both in humans and animal models. This ary beneficial effects of exercise training may be im- training modality showed to improve both cardiovascular proved immune system responses after training [106]. performance and general muscle oxidative capacity [98]. The clinical importance, of these systemic benefits of The latter is underscored by the larger peak arterial- endurance exercise training, merits further research. venous oxygen content difference, and by the shift of the point at which lactate accumulates in the blood to higher Although the changes in the skeletal muscles after exercise intensity [85]. This effect is the net result of intra- endurance training are described extensively, the precise muscle fiber and extra-muscle fiber adaptations to endur- mechanisms which form the basis of the observed pro- ance training. An increased oxygen supply to the muscle cesses are not yet fully understood. It remains unclear (through increased cardiac output, increased capillary vas- which of the acute responses to a single exercise bout are cular bed, and increased capillary to muscle fiber contacts) carried forward to result in the final training effect after matches the increased peak oxygen demand. The in- repeated bouts of endurance exercise over weeks. creased capacity of the skeletal muscle to deal with the 66 Troosters/Gosselink/Decramer

Resistance/Strength Training Table 4. Modalities for muscle strength measurements and training Resistance training targets isolated muscle groups. The modalities training load is adjusted to the maximal performance of the muscle rather than to the whole exercising body. Isometric Movement speed: zero; resistance variable* Three modalities of resistance training can be applied. Concentric Movement in the direction of the muscle These are classically subdivided based upon the speed of contraction: muscle shortening the movement, and the resistance applied over the range Free weights and Movement speed: free; movement of motion, as indicated in table 4. barbells resistance: variable* For concentric exercises the intensity of the training is Movement speed: constant; movement mostly determined by the number of contractions and the Isokinetic equipment resistance: variable* number of sets of contractions (e.g. three sets of 10 con- Movement speed: variable; movement tractions). The resistance is expressed relative to the max- CAM devices resistance: constant imal resistance (the maximal load that can be displaced Movement in the opposite direction of the once over the full range of motion, 1 repetition maximum, Eccentric muscle contraction 1RM). Training programs using few repetitions are typi- cal strength training programs, whereas lifting relatively * Frequently used in resistance training in the clinical setting. low percentages of 1RM with a high number of repetitions result in more endurance training. Impact on the Muscle. Resistance training is thought to Practical Implementation of Resistance Training. improve neural activation both through well-matched Weightlifting training was introduced by Captain Delorme pre- and postsynaptic adaptation in the neuromuscular in 1945 in an attempt to increase muscle strength in injured junction [110], enhanced neural facilitation and increased soldiers. Resistance training can be done using free weights. maximal motor unit discharge rates [87, 111]. These Alternatively, a device can be used where the movement is changes account for approximately 90% of the increase in limited to one specific muscle group, and the weight is skeletal muscle strength in the first days or weeks of a applied through a system of pulleys. In the latter setup both resistance training program, especially in elderly subjects movement and weights can be elegantly controlled and [111]. When training is continued, skeletal muscle adjusted. Training programs generally apply a load of 60– strength increases further through increases in muscle 80% of the weight that can be lifted once over the full range hypertrophy. It is well accepted that after a single bout of of motion (1RM). From a review of the literature, McDon- high-intensity resistance exercise both protein breakdown agh and Davies [107] concluded that to improve strength, a and synthesis are increased. The net protein balance, how- load of at least 66% of the 1RM was required, and had to be ever, is reported to be positive for 2 days following the lifted at least 10 times. Higher workloads however resulted acute exercise bout [112]. The effect of resistance training in significantly more training effect. Empirically most on the protein balance is unaffected by age. This suggests training programs use 20–30 repetitions. that both elderly and young subjects may benefit from this More sophisticated isokinetic (constant contraction form of exercise [113]. Protein synthesis could result from velocity) equipment has been described to do resistance increased mRNA transcriptional activity, or the acute training, but these apparatus are very costly. This limits exercise-induced increase in mixed muscle protein syn- the application of isokinetic resistance training in general thetic rate could be mediated through posttranscriptional clinical practice. In healthy but frail elderly subjects, resis- events [114]. This is probably due to an improved effi- tance training was successfully applied using 80% of 1RM ciency of mRNA translation after resistance exercise applied through a cheap cable pulley system. Exercise [115]. days were alternated by days of rest. This program increased strength and improved functional capacity in The functional effect of resistance training has histori- frail institutionalized elderly. A secondary analysis of the cally been characterized as increase in muscle strength, data of the women in this trial by Nelson et al. [108] without large crossover effects to peak oxygen consump- showed that the training program improved dynamic bal- tion, and tasks requiring predominantly oxidative (type I, ance and was an effective means of preserving bone min- IIa fiber) muscle work. This is probably due to the fact eral density in postmenopausal women. Hence resistance that most original observations were done in young sub- training may be a cheap and cost-effective training strate- jects undergoing periods of heavy resistance training [116, gy, especially in the elderly [109]. 117]. In young, healthy persons, the impact of resistance training on whole body oxygen uptake is negligible [118]. Deconditioning, and Principles of Training 67

Whether these observations hold in elderly subjects is cur- Prevention of Deconditioning and Maintaining rently debated. Literature supports clear oxidative capaci- Training Effects ty changes in skeletal muscles after resistance training in Abrupt ending of an exercise training period results in the elderly [102, 103]. In the last decade, resistance train- rapid loss of the benefits gained from it, even if the subject ing gained interest in the treatment of sarcopenia in frail returns to normal sedentary daily life (and thus remains elderly. Fiattarone et al. [45] convincingly showed that more active than during bed rest) [118]. It is accepted that resistance training improved strength, gait speed and mo- the volume of exercise necessary to maintain physical bility in frail subjects in their eight decade. An interesting condition is appreciably less than the amount of exercise recent observation by Greiwe et al. [119] may add to our necessary to improve exercise capacity. It has been gener- understanding of the efficiency that resistance training ally accepted that healthy subjects lose less muscle mass as showed in counteracting sarcopenia. The latter authors they remain more physically active [123, 124]. In the con- showed that TNF· expression in skeletal muscle of frail text of immobilization or bed rest, resistance training is elderly was dramatically reduced after a resistance train- an attractive tool since this form of training is adaptable ing program. Since the cytokine TNF· is a potent induc- to bedside situations. Among the many illustrations avail- tor of muscle wasting, the observation by Greiwe et al. able in current literature is the one by Akima et al. [125] may explain why resistance training showed to be success- who showed that 30 isometric leg press contractions were ful, especially in the frail elderly. In patients with chronic sufficient to prevent the atrophy seen after 20 days of bed diseases such as heart failure and COPD, resistance train- rest. Muscle protein synthesis was maintained during bed ing has been applied as a way to increase the training stim- rest when resistance training was associated every other ulus to skeletal muscles, when whole body exercise capaci- day during bed rest [28]. Alterations in fiber type compo- ty is limited by ventilatory and/or cardiocirculatory fac- sition and fiber atrophy were prevented in this setting. tors at relatively low workload. In patients with heart fail- Whether muscle weakness can be completely prevented ure, resistance training showed to increase peak oxygen using this approach is debated [28, 126]. On completion uptake [120]. In patients with COPD, strength training of exercise training programs, adequate strategies should improved whole body endurance [121]. The addition of be applied to prevent the loss of exercise training-induced strength training to a regular endurance training program benefits. Studies by Hickson et al. [127, 128] showed that showed to restore muscle strength and muscle cross-sec- reducing training frequency, without reducing training tional area more than endurance training alone in these intensity may be adequate to maintain training effect. patients [122]. Hence once a week training at high (see above) training intensity seems to be appropriate. References 5 McGuire DK, Levine BD, Williamson JW, 10 Convertino VA, Karst GM, Kirby CR, Gold- Snell PG, Blomqvist CG, Saltin B, Mitchell JH: water DJ: Effect of simulated weightlessness on 1 Blair SN: 1993 C.H. McCloy Research Lecture: A 30-year follow-up of the Dallas bedrest and exercise-induced anaerobic threshold. Aviat Physical activity, physical fitness, and health. training study. Effect of age on cardiovascular Space Environ Med 1986;57:325–331. Res Q Exerc Sport 1993;64:365–376. response to exercise. Circulation 2001;104: 1350–1357. 11 Wasserman K, Hansen JE, Sue DJ, Whipp BJ, 2 Blair SN, Kohl HW III, Paffenbarger RS Jr, Casaburi R: Principles of Exercise Testing and Clark DG, Cooper KH, Gibbons LW: Physical 6 Convertino VA: Cardiovascular consequences Interpretation, ed 2. Philadelphia, Williams & fitness and all-cause mortality. A prospective of bed rest: Effect on maximal oxygen uptake. Wilkins, 1994, p 479. study of healthy men and women. JAMA 1989; Med Sci Sports Exerc 1997;29:191–196. 262:2395–2401. 12 Karvonen J, Rauhala E, Chwalbinska-Moneta 7 Wasserman K, Whipp BJ: Exercise physiology J: The effects of three months slalom training 3 Blair SN, Kohl HW III, Barlow CE, Paffen- in health and disease. Am Rev Respir Dis on physical performance capacity. J Sports barger RS Jr, Gibbons LW, Macera CA: 1975;112:219–249. Med Phys Fit 1985;25:194–197. Changes in physical fitness and all-cause mor- tality. A prospective study of healthy and un- 8 Weisman IM, Zeballos RJ: An integrated ap- 13 Tartaglia MC, Chen JT, Caramanos Z, Taivas- healthy men. JAMA 1995;273:1093–1098. proach to the interpretation of cardiopulmo- salo T, Arnold DL, Argov Z: Muscle phospho- nary exercise testing. Clin Chest Med 1994;15: rus magnetic resonance spectroscopy oxidative 4 Saltin B, Blomqvist CG, Mitchell JH, Johnson 421–445. indices correlate with physical activity. Muscle RL, Wildenthal K, Chapman CB: Response to Nerve 2000;23:175–181. exercise after bed rest and after training. Circu- 9 Coyle EF, Martin WH III, Bloomfield SA, Lo- lation 1968;38:1–78. wry OH, Holloszy JO: Effects of detraining on responses to submaximal exercise. J Appl Phys- iol 1985;59:853–859. 68 Troosters/Gosselink/Decramer

14 Hikida RS, Gollnick PD, Dudley GA, Conver- 29 Goldspink DF: The influence of immobiliza- 44 Bortz WM, Bortz WM: How fast do we age? tino VA, Buchanan P: Structural and metabolic tion and stretch on protein turnover of rat skel- Exercise performance over time as a biomark- characteristics of human skeletal muscle fol- etal muscle. J Physiol 1977;264:267–282. er. J Gerontol A Biol Sci Med Sci 1996;51: lowing 30 days of simulated microgravity. Av- M223–M225. iat Space Environ Med 1989;60:664–670. 30 Bricout VA, Serrurier BD, Bigard AX, Guezen- nec CY: Effects of hindlimb suspension and 45 Fiatarone MA, O'Neill EF, Doyle Ryan N, Cle- 15 Coyle EF, Martin WH III, Sinacore DR, Joyner androgen treatment on testosterone receptors ments KM, Solares GR, Nelson ME, Roberts MJ, Hagberg JM, Holloszy JO: Time course of in rat skeletal muscles. Eur J Appl Physiol SB, Kehayias JJ, Lipsitz LA, Evans WJ: Exer- loss of adaptations after stopping prolonged Occup Physiol 1999;79:443–448. cise training and nutritional supplementation intense endurance training. J Appl Physiol for physical frailty in very elderly people. N 1984;57:1857–1864. 31 Zachwieja JJ, Smith SR, Lovejoy JC, Rood JC, Engl J Med 1994;330:1769–1775. Windhauser MM, Bray GA: Testosterone ad- 16 Mujika I, Padilla S: Cardiorespiratory and met- ministration preserves protein balance but not 46 Westerterp KR: Daily physical activity and abolic characteristics of detraining in humans. muscle strength during 28 days of bed rest. J ageing. Curr Opin Clin Nutr Metab Care 2000; Med Sci Sports Exerc 2001;33:413–421. Clin Endocrinol Metab 1999;84:207–212. 3:485–488. 17 Vanhees L, Schepers D, Fagard R: Comparison 32 Oishi K, Yokoi M, Maekawa S, Sodeyama C, 47 Moulias R, Meaume S, Raynaud-Simon A: Sar- of maximum versus submaximum exercise Shiraishi T, Kondo R, Kuriyama T, Machida copenia, hypermetabolism and aging. Z Geron- testing in providing prognostic information af- K: Oxidative stress and haematological tol Geriatr 1999;32:425–432. ter acute myocardial infarction and/or coro- changes in immobilized rats. Acta Physiol nary artery bypass grafting. Am J Cardiol 1997; Scand 1999;165:65–69. 48 Frontera WR, Hughes VA, Fielding RA, Fiata- 80:257–262. rone MA, Evans WJ, Roubenoff R: Aging of 33 Appell HJ, Duarte JA, Soares JM: Supplemen- skeletal muscle: A 12-year longitudinal study. J 18 Ferrando AA, Lane HW, Stuart CA, Davis- tation of vitamin E may attenuate skeletal mus- Appl Physiol 2000;88:1321–1326. Street J, Wolfe RR: Prolonged bed rest de- cle immobilization atrophy. Int J Sports Med creases skeletal muscle and whole body protein 1997;18:157–160. 49 Troosters T, Gosselink R, Decramer M: Six- synthesis. Am J Physiol 1996;270:E627–E633. minute walking distance in healthy elderly sub- 34 Lalani R, Bhasin S, Byhower F, Tarnuzzer R, jects. Eur Respir J 1999;14:270–274. 19 Vandenborne K, Elliott MA, Walter GA, Ab- Grant M, Shen R, Asa S, Ezzat S, Gonzalez- dus S, Okereke E, Shaffer M, Tahernia D, Cadavid NF: Myostatin and insulin-like 50 McClaran SR, Babcock MA, Pegelow DF, Red- Esterhai JL: Longitudinal study of skeletal growth factor-I and -II expression in the muscle dan WG, Dempsey JA: Longitudinal effects of muscle adaptations during immobilization and of rats exposed to the microgravity environ- aging on lung function at rest and exercise in rehabilitation. Muscle Nerve 1998;21:1006– ment of the NeuroLab space shuttle flight. J healthy active fit elderly adults. J Appl Physiol 1012. Endocrinol 2000;167:417–428. 1995;78:1957–1968. 20 Berg HE, Larsson L, Tesch PA: Lower limb 35 Baldwin KM: Effect of space flight on the func- 51 Fleg JL, Lakatta EG: Role of muscle loss in the skeletal muscle function after 6 weeks of bed tional, biochemical, and metabolic properties age-associated reduction in VO2 max. J Appl rest. J Appl Physiol 1997;82:182–188. of skeletal muscle. Med Sci Sports Exerc 1996; Physiol 1988;65:1147–1151. 28:983–987. 21 Berg HE, Tesch PA: Changes in muscle func- 52 Conley KE, Jubrias SA, Esselman PC: Oxida- tion in response to 10 days of lower limb 36 Chakravarthy MV, Davis BS, Booth FW: IGF- tive capacity and ageing in human muscle. J unloading in humans. Acta Physiol Scand I restores satellite cell proliferative potential in Physiol 2000;526:203–210. 1996;157:63–70. immobilized old skeletal muscle. J Appl Physi- ol 2000;89:1365–1379. 53 Decary S, Mouly V, Hamida CB, Sautet A, Bar- 22 Ferretti G, Antonutto G, Denis C, Hoppeler H, bet JP, Butler-Browne GS: Replicative poten- Minetti AE, Narici MV, Desplanches D: The 37 Wing SS, Bedard N: Insulin-like growth factor I tial and telomere length in human skeletal mus- interplay of central and peripheral factors in stimulates degradation of an mRNA transcript cle: implications for satellite cell-mediated gene limiting maximal O2 consumption in man after encoding the 14 kDa ubiquitin-conjugating en- therapy. Hum Gene Ther 1997;8:1429–1438. prolonged bed rest. J Physiol 1997;501:677– zyme. Biochem J 1996;319:455–461. 686. 54 Wang Y, Michikawa Y, Mallidis C, Bai Y, 38 Wing SS, Banville D: 14-kDa ubiquitin-conju- Woodhouse L, Yarasheski KE, Miller CA, As- 23 Desplanches D, Hoppeler H, Mayet MH, De- gating enzyme: Structure of the rat gene and kanas V, Engel WK, Bhasin S, Attardi G: Mus- nis C, Claassen H, Ferretti G: Effects of bedrest regulation upon fasting and by insulin. Am J cle-specific mutations accumulate with aging in on deltoideus muscle morphology and en- Physiol 1994;267:E39–E48. critical human mtDNA control sites for repli- zymes. Acta Physiol Scand 1998;162:135–140. cation. Proc Natl Acad Sci USA 2001;98:4022– 39 Loughna PT, Brownson C: Two myogenic reg- 4027. 24 Portero P, Vanhoutte C, Goubel F: Surface ulatory factor transcripts exhibit muscle-spe- electromyogram power spectrum changes in cific responses to disuse and passive stretch in 55 Brooks SV, Faulkner JA: Skeletal muscle weak- human leg muscles following 4 weeks of simu- adult rats. FEBS Lett 1996;390:304–306. ness in old age: Underlying mechanisms. Med lated microgravity. Eur J Appl Physiol Occup Sci Sports Exerc 1993;26:432–439. Physiol 1996;73:340–345. 40 Mozdziak PE, Greaser ML, Schultz E: Myoge- nin, MyoD, and myosin heavy chain isoform 56 Luff AR: Age-associate changes in the innerva- 25 Ferrando AA, Stuart CA, Brunder DG, Hill- expression following hindlimb suspension. tion of muscle fibers and changes in the me- man GR: Magnetic resonance imaging quanti- Aviat Space Environ Med 1999;70:511–516. chanical properties of motor units. Ann NY tation of changes in muscle volume during 7 Acad Sci 1998;854:92–101. days of strict bed rest. Aviat Space Environ 41 St Amand J, Okamura K, Matsumoto K, Shim- Med 1995;66:976–981. izu S, Sogawa Y: Characterization of control 57 Decramer M, de Bock V, Dom R: Functional and immobilized skeletal muscle: An overview and histologic picture of steroid-induced my- 26 Andersen JL, Schiaffino S: Mismatch between from genetic engineering. FASEB J 2001;15: opathy in chronic obstructive pulmonary dis- myosin heavy chain mRNA and protein distri- 684–692. ease. Am J Respir Crit Care Med 1996;153: bution in human skeletal muscle fibers. Am J 1958–1964. Physiol 1997;272:C1881–C1889. 42 Gayan-Ramirez G, Vanderhoydonc F, Ver- hoeven G, Decramer M: Acute treatment with 58 Gosselink R, Troosters T, Decramer M: Pe- 27 Andersen JL, Gruschy-Knudsen T, Sandri C, corticosteroids decreases IGF-1 and IGF-2 ex- ripheral muscle weakness contributes to exer- Larsson L, Schiaffino S: Bed rest increases the pression in the rat diaphragm and gastrocnemi- cise limitation in COPD. Am J Respir Crit amount of mismatched fibers in human skele- us. Am J Respir Crit Care Med 1999;159:283– Care Med 1996;153:976–980. tal muscle. J Appl Physiol 1999;86:455–460. 289. 59 Drexler H, Riede U, Munzel T: Alterations of 28 Ferrando AA, Tipton KD, Bamman MM, 43 Ferrando AA, Stuart CA, Sheffield-Moore M, skeletal muscle in chronic heart failure. Circu- Wolfe RR: Resistance exercise maintains skele- Wolfe RR: Inactivity amplifies the catabolic lation 1992;85:1751–1759. tal muscle protein synthesis during bed rest. J response of skeletal muscle to cortisol. J Clin Appl Physiol 1997;82:807–810. Endocrinol Metab 1999;84:3515–3521. Deconditioning, and Principles of Training 69

60 Magnusson G, Kaijser L, Rong H: Exercise 73 Gottlieb SS, Fisher ML, Freudenberger R, 89 Wasserman K, Sue DJ, Casaburi R, Moricca capacity in heart failure: Relative importance Robinson S, Zietowski G, Alves L, Krichten C, RB: Selection criteria for exercise training in of heart and skeletal muscle. Clin Physiol 1996; Vaitkevicus P, McCarter R: Effects of exercise pulmonary rehabilitation. Eur Respir J 1989; 16:183–195. training on peak performance and quality of 7(suppl):604S–610S. life in congestive heart failure patients. J Card 61 Diesel W, Emms M, Knight BK, Noakes TD, Fail 1999;5:188–194. 90 Punzal PA, Ries AL, Kaplan RW, Prewitt Swanepoel CR, van Zyl SR, Kaschula RO, Sin- LM: Maximum intensity exercise training in clair-Smith CC: Morphologic features of the 74 Meijer EP, Westerterp KR, Verstappen FT: patients with chronic obstructive pulmonary myopathy associated with chronic renal failure. Effect of exercise training on total daily physi- disease. Chest 1991;100:618–623. Am J Kidney Dis 1993;22:677–684. cal activity in elderly humans. Eur J Appl Phys- iol Occup Physiol 1999;80:16–21. 91 Maltais F, LeBlanc P, Jobin J: Intensity of 62 Sala E, Roca J, Marrades RM, Alonso J, Gon- training and physiological adaptation in pa- zalez De Suso JM, Moreno A, Barbera JA, 75 Morio B, Montaurier C, Pickering G, Ritz P, tients with chronic obstructive pulmonary Nadal J, de Jover L, Rodriguez-Roisin R, Fellmann N, Coudert J, Beaufrere B, Vermorel disease. Am J Respir Crit Care Med 1997; Wagner PD: Effects of endurance training on M: Effects of 14 weeks of progressive endur- 155:555–561. skeletal muscle bioenergetics in chronic ob- ance training on energy expenditure in elderly structive pulmonary disease. Am J Respir Crit people. Br J Nutr 1998;80:511–519. 92 Casaburi R, Porszasz J, Burns MR, Carithers Care Med 1999;159:1726–1734. ER, Chang RS, Cooper CB: Physiologic bene- 76 Van Etten LM, Westerterp KR, Verstappen fits of exercise training in rehabilitation of 63 Casaburi R, Patessio A, Ioli F, Zanaboni S, FT, Boon BJ, Saris WH: Effect of an 18-week patients with severe chronic obstructive pul- Donner CF, Wasserman K: Reductions in exer- weight-training program on energy expenditure monary disease. Am J Respir Crit Care Med cise lactic acidosis and ventilation as a result of and physical activity. J Appl Physiol 1997;82: 1997;155:1541–1551. exercise training in patients with obstructive 298–304. lung disease. Am Rev Respir Dis 1991;143:9– 93 Troosters T, Gosselink R, Decramer M: Exer- 18. 77 Surgeon General’s report on physical activity cise training in COPD: How to distinguish and health. From the Centers for Disease Con- responders from nonresponders. J Cardio- 64 Maltais F, LeBlanc P, Simard C, Jobin J, Be- trol and Prevention. JAMA 1996;276:522. pulm Rehabil 2001;21:10–17. rube C, Bruneau J, Carrier L, Belleau R: Skele- tal muscle adaptation to endurance training in 78 Booth FW, Gordon SE, Carlson CJ, Hamilton 94 Van der Ent M, Jeneson JA, Remme WJ, patients with chronic obstructive pulmonary MT: Waging war on modern chronic diseases: Berger R, Ciampricotti R, Visser F: A non- disease. Am J Respir Crit Care Med 1996;154: Primary prevention through exercise biology. J invasive selective assessment of type I fibre 442–447. Appl Physiol 2000;88:774–787. mitochondrial function using 31P-NMR spec- troscopy. Evidence for impaired oxidative 65 Troosters T, Gosselink R, Decramer M: Short- 79 Pollock ML, Gaesser GA, Butcher JD, Després phosphorylation rate in skeletal muscle in pa- and long-term effects of outpatient rehabilita- J, Dishman RK, Franklin BA, Garber CE: Po- tients with chronic heart failure. Eur Heart J tion in patients with chronic obstructive pul- sition stand: The recommended quantity, and 1998;19:124–131. monary disease: A randomized trial. Am J Med quality of exercise for developing and main- 2000;109:207–212. taining cardiorespiratory and muscular fitness, 95 DeBusk RF, Stenestrand U, Sheehan M, Has- and flexibility in healthy adults. Med Sci Sports kell WL: Training effects of long versus short 66 Bernard S, Whittom F, LeBlanc P, Jobin J, Bel- Exerc 1998;30. bouts of exercise in healthy subjects. Am J leau R, Berube C, Carrier G, Maltais F: Aero- Cardiol 1990;65:1010–1013. bic and strength training in patients with 80 Mazzeo RS, Cavanagh P, Evans WJ, Fiatarone chronic obstructive pulmonary disease. Am J MA, Hagberg J, McAuley E, Startzell J: Posi- 96 Nieuwland W, Berkhuysen MA, Landsman Respir Crit Care Med 1999;159:896–901. tion stand ACSM: Exercise and physical activi- ML, Lie KI, Rispens P: Training effects on ty for older adults. Med Sci Sports Exerc 1998; peak VO2, specific of the mode of movement, 67 O’Donnell DE, McGuire M, Samis L, Webb 30:992–1008. in rehabilitation of patients with coronary KA: General exercise training improves venti- artery disease. Int J Sports Med 1998;19:358– latory and peripheral muscle strength and en- 81 American College of Sports Medicine position 363. durance in chronic airflow limitation. Am J stand. Exercise for patients with coronary ar- Respir Crit Care Med 1998;157:1489–1497. tery disease. Med Sci Sports Exerc 1994;26: 97 Pierce EF, Weltman A, Seip RL, Snead D: i–v. Effects of training specificity on the lactate 68 Hakkinen K, Alen M, Kallinen M, Newton threshold and VO2 peak. Int J Sports Med RU, Kraemer WJ: Neuromuscular adaptation 82 American College of Sports Medicine. Position 1990;11:267–272. during prolonged strength training, detraining Stand. Physical activity, physical fitness, and and re-strength-training in middle-aged and el- hypertension. Med Sci Sports Exerc 1993;25: 98 Saltin B: Physiological effects of physical con- derly people. Eur J Appl Physiol 2000;83:51– i–x. ditioning. Med Sci Sports Exerc 1969;1:50– 62. 56. 83 American College of Sports Medicine Position 69 Bernard S, LeBlanc P, Whittom F, Carrier G, Stand. Exercise and physical activity for older 99 Poole DC, Musch TI: Pulmonary and periph- Jobin J, Belleau R, Maltais F: Peripheral mus- adults. Med Sci Sports Exerc 1998;30:992– eral gas exchange during exercise; in Roca J, cle weakness in patients with chronic obstruc- 1008. Rodriguez-Roisin R, Wagner PD (eds): Pul- tive pulmonary disease. Am J Respir Crit Care monary and Peripheral Gas Exchange in Med 1998;158:629–634. 84 Donner CF, Decramer M: Pulmonary Rehabil- Health and Disease, ed 1. New York, Dekker, itation. Eur Respir Monogr 2000;13:1–200. 2000, pp 469–517. 70 Simonini A, Long CS, Dudley GA: Heart fail- ure in rats causes changes in skeletal muscle 85 Londeree BR: Effect of training on lactate/ven- 100 Minotti JR, Johnson EC, Hudson TL, Zu- morphology and gene expression that are not tilatory thresholds: A meta-analysis. Med Sci roske G, Murata G, Fukushima E, Cagle TG, caused by inactivity. Circ Res 1996;79:128– Sports Exerc 1997;29:837–843. Chick TW, Massie BM, Icenogle MV: Skele- 136. tal muscle response to exercise training in 86 Sady S, Katch V, Freedson P, Weltman A: congestive heart failure. J Clin Invest 1990; 71 Vanhees L, Fagard R, Thijs L, Amery A: Prog- Changes in metabolic acidosis: Evidence for an 86:751–758. nostic value of training-induced change in peak intensity threshold. J Sports Med Phys Fitness exercise capacity in patients with myocardial 1980;20:41–46. 101 Hoppeler H, Howald H, Conley K, Lindstedt infarcts and patients with coronary bypass sur- SL, Claassen H, Vock P, Weibel ER: Endur- gery. Am J Cardiol 1995;76:1014–1019. 87 McArdle WD, Katch FI, Katch VL: Exercise ance training in humans: Aerobic capacity Physiology, Energy Nutrition and Human per- and structure of skeletal muscle. J Appl Physi- 72 Bouchard C, Leon AS, Rao DC, Skinner JS, formance, ed 4. Baltimore, Williams & Wil- ol 1985;59:320–327. Wilmore JH: Familial aggregation of VO2max kins, 1996, pp 1–849. response to exercise training: Results from the 102 Jubrias SA, Esselman PC, Price LB, Cress HERITAGE family study. J Appl Physiol 88 Debigare R, Maltais F, Mallet M, Casaburi R, ME, Conley KE: Large energetic adaptations 1999;87:1003–1008. LeBlanc P: Influence of work rate incremental of elderly muscle to resistance and endurance rate on the exercise responses in patients with training. J Appl Physiol 2001;90:1663–1670. COPD. Med Sci Sports Exerc 2000;32:1365– 1368. 70 Troosters/Gosselink/Decramer

103 Orlander J, Aniansson A: Effect of physical 115 Welle S, Bhatt K, Thompson C: Stimulation 127 Hickson RC, Foster C, Pollock ML, Galassi training on skeletal muscle metabolism and of myofibrillar synthesis by exercise is me- TM, Rich S: Reduced training intensities and ultrastructure in 70- to 75-year-old men. Acta diated by more efficient translation of loss of aerobic power, endurance and cardiac Physiol Scand 1980;109:149–154. mRNA. Am J Physiol Endocrinol Metab growth. J Appl Physiol 1985;58:492–499. 1999;274:E673–E683. 104 Starritt EC, Angus D, Hargreaves M: Effect of 128 Hickson RC, Rosenkoetter AM: Reduced short-term training on mitochondrial ATP 116 Luthi JM, Howald H, Claassen H, Rosler K, training frequencies and maintenance of in- production rate in human skeletal muscle. J Vock P, Hoppeler H: Structural changes in creased aerobic power. Med Sci Sports Exerc Appl Physiol 1999;86:450–454. skeletal muscle tissue with heavy-resistance 1981;13:13–16. exercise. Int J Sports Med 1986;7:123–127. 105 Linke A, Schoene N, Gielen S, Hofer J, Erbs 129 Fortney S, Scneider VS, Greenleaf JF: The S, Schuler G, Hambrecht R: Endothelial dys- 117 Chilibeck P, Syrotuik DG, Bell GJ: The effect physiology of bed rest; in Fregley MJ, Blatteis function in patients with chronic heart fail- of strength training on estimates of mitochon- CM (eds): Environmental Physiology. New ure: Systemic effects of lower-limb exercise drial density and distribution throughout York, Oxford University Press, 1996, pp training. J Am Coll Cardiol 2001;37:392– muscle fibers. Eur J Appl Physiol 1999;80: 889–939. 397. 604–609. 130 Pawelczyk JA, Zuckerman JH, Blomqvist 106 Bruunsgaard H, Pedersen BK: Special feature 118 LeMura LM, von Duvillard SP, Andreacci J, CG, Levine BD: Regulation of muscle sympa- for the Olympics: Effects of exercise on the Klebez JM, Chelland SA, Russo J: Lipid and thetic nerve activity after bed rest decondi- immune system. Effects of exercise on the lipoprotein profiles, cardiovascular fitness, tioning. Am J Physiol Heart Circ Physiol immune system in the elderly population. Im- body composition, and diet during and after 2001;280:H2230–H2239. munol Cell Biol 2000;78:523–531. resistance, aerobic and combination training in young women. Eur J Appl Physiol 2000;82: 131 Linnarsson D: Pulmonary function and car- 107 McDonagh MJN, Davies CTM: Adaptive re- 451–458. diopulmonary interactions at microgravity. sponses of mammalian skeletal muscle to ex- Med Sci Sports Exerc 1996;28:S14–S17. ercise with high loads. Eur J Appl Physiol 119 Greiwe JS, Cheng B, Rubin DC, Yarasheski 1984;52:139–155. KE, Semenkovich CF: Resistance exercise de- 132 Teasell R, Dittmer DK: Complications of im- creases skeletal muscle tumor necrosis factor mobilization and bed rest. 2. Other complica- 108 Nelson ME, Fiatarone MA, Morganti CM, · in frail elderly humans. FASEB J 2001;15: tions. Can Fam Physician 1993;39:1440– Trice I, Greenberg RA, Evans WJ: Effects of 475–482. 1446. high-intensity strength training on multiple risk factors for osteoporotic fractures. A ran- 120 Gordon A, Tyni-Lenne R, Jansson E, Jensen- 133 Ferrando AA: Effects of inactivity and hor- domized controlled trial. JAMA 1994;272: Urstad M, Kaijser L: Beneficial effects of monal mediators on skeletal muscle during 1909–1914. exercise training in heart failure patients with recovery from trauma. Curr Opin Clin Nutr low cardiac output response to exercise – a Metab Care 2000;3:171–175. 109 Robertson MC, Devlin N, Gardner MM, comparison of two training models. J Intern Campbell AJ: Effectiveness and economic Med 1999;246:175–182. 134 Bikle DD, Halloran BP: The response of bone evaluation of a nurse delivered home exercise to unloading. J Bone Miner Metab 1999;17: programme to prevent falls. 1. Randomised 121 Simpson K, Killian K, McCartney N, Stub- 233–244. controlled trial. BMJ 2001;322:697–699. bing DG, Jones NL: Randomised controlled trial of weightlifting exercise in patients with 135 DeRoshia CW, Greenleaf JE: Performance 110 Deschenes MR, Judelson DA, Kraemer WJ, chronic airflow limitation. Thorax 1992;47: and mood-state parameters during 30-day 6° Meskaitis VJ, Volek JS, Nindl BC, Harman 70–75. head-down bed rest with exercise training. FS, Deaver DR: Effects of resistance training Aviat Space Environ Med 1993;64:522–527. on neuromuscular junction morphology. 122 Bernard S, Whittom F, LeBlanc P, Jobin J, Muscle Nerve 2000;23:1576–1581. Belleau R, Berube C, Carrier G, Maltais F: 136 Vallet G, Ahmaidi S, Serres I, Fabre C, Bour- Aerobic and strength training in patients with gouin D, Desplan J, Varray A, Prefaut C: 111 Patten C, Kamen G, Rowland DM: Adapta- chronic obstructive pulmonary disease. Am J Comparison of two training programmes in tions in maximal motor unit discharge rate to Respir Crit Care Med 1999;159:896–901. chronic airway limitation patients: Standard- strength training in young and older adults. ized versus individualized protocols. Eur Res- Muscle Nerve 2001;24:542–550. 123 Rantanen T, Era P, Heikkinen E: Physical pir J 1997;10:114–122. activity and the changes in maximal isometric 112 Parise G, Yarasheski KE: The utility of resis- strength in men and women from the age of 137 Horowitz MB, Littenberg B, Mahler DA: tance exercise training and amino acid sup- 75 to 80 years. J Am Geriatr Soc 1997;45: Dyspnea ratings for prescribing exercise in- plementation for reversing age-associated de- 1439–1445. tensity in patients with COPD. Chest 1996; crements in muscle protein mass and func- 109:1169–1175. tion. Curr Opin Clin Nutr Metab Care 2000; 124 Hughes VA, Frontera WR, Wood M, Evans 3:489–495. WJ, Dallal GE, Roubenoff R, Fiatarone Singh 138 Ferretti G, Girardis M, Moia C, Antonutto G: MA: Longitudinal muscle strength changes in Effects of prolonged bed rest on cardiovascu- 113 Hasten DL, Pak-Loduca J, Obert KA, Yara- older adults: Influence of muscle mass, physi- lar oxygen transport during submaximal exer- sheski KE: Resistance exercise acutely in- cal activity and health. J Gerontol A Biol Sci cise in humans. Eur J Appl Physiol Occup creases MHC and mixed muscle protein syn- Med Sci 2001;56:B209–B217. Physiol 1998;78:398–402. thesis rates in 78–84 and 23–32 year olds. Am J Physiol Endocrinol Metab 2000;278:E620– 125 Akima H, Kubo K, Kanehisa H, Suzuki Y, Marc Decramer, MD, PhD E626. Gunji A, Kukunaga T: Leg press resistance Respiratory Division and training during 20 days of 6° head down tilt Respiratory Rehabilitation 114 Chesley A, McDougall JD, Tarnopolski MA: bed rest prevents muscle deconditioning. Eur University Hospital Gasthuisberg Changes in human muscle protein synthesis J Appl Physiol 2000;82:30–38. Herestraat 49, B–3000 Leuven (Belgium) after resistance exercise. J Appl Physiol 1992; Tel. +32 16 34 68 07, Fax +32 16 34 68 03 73:1383–1388. 126 Bamman MM, Clarke MSF, Feeback DL, E-Mail [email protected] Talmadge RJ, Stevens BR, Lieberman SA, Greenisen MC: Impact of resistance exercise during bed rest on skeletal muscle sarcopenia and myosin isoform distribution. J Appl Physiol 1998;84:157–163. Deconditioning, and Principles of Training 71

Weisman IM, Zeballos RJ (eds): Clinical Exercise Testing. Prog Respir Res. Basel, Karger, 2002, vol 32, pp 72–80 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Mechanisms and Measurement of Exertional Dyspnea Donald A. Mahlera Gustavo Fierro-Carriona John C. Bairda–c aDartmouth Medical School, Lebanon, N.H.; bPsychological and Brain Sciences, Dartmouth College, Hanover, N.H., and cPsychological Applications, Newport, N.H., USA Summary uncomfortable awareness of breathing’ [3]. Clinicians as well as investigators generally believe that dyspnea is a Dyspnea during physical activities is a common complaint of sensory experience that is likely perceived similar to the individuals with various respiratory diseases. One approach to perception of pain. In this article we review the possible examine and quantify this unpleasant sensation is to have the mechanisms contributing to breathlessness based on a patient perform cardiopulmonary exercise testing (CPX) and neurophysiological model. Then, we consider the princi- provide ratings of breathlessness during exercise in the labora- ples of sensory psychophysics (stimulus → response rela- tory. Currently, the patient is instructed to give ratings of tionship) that can be applied to the measurement of dys- breathlessness each minute or increment of the CPX. How- pnea during exercise testing. Information is provided on ever, computer technology has evolved to a point where a values obtained at peak exercise and as a continuum patient can provide continuous ratings throughout the exercise throughout the course of exercise. Finally, the clinical test. Investigations have demonstrated numerous clinical appli- applications of measuring breathlessness as part of cardio- cations resulting from the measurement of dyspnea during pulmonary exercise testing (CPX) are considered in eval- CPX. As examples, bronchodilator therapy, pulmonary reha- uating treatments such as bronchodilator therapy and pul- bilitation, oxygen, and bullectomy/volume reduction surgery monary rehabilitation in patients with respiratory dis- have been shown to measurably reduce the severity of exer- ease. tional dyspnea. Based on this extensive information, we recom- mend that the perception of breathlessness should be routinely Neurophysiological Model measured as part of CPX in all symptomatic individuals. Mechanisms of Exertional Dyspnea The mechanisms of dyspnea can be considered from a neurophysiological perspective [4]: What Is Exertional Dyspnea? Receptor Dyspnea is a subjective experience of breathing dis- ↓ comfort comprised of qualitatively distinct sensations that can vary in intensity [1, 2]. The various phrases used Afferent impulse to describe this sensation include ‘breathlessness’, ‘short- ↓ ness of breath’, ‘labored or difficult breathing’ and ‘an Integration/processing in the CNS ↓ Efferent impulse ↓ Dyspnea

Receptors Pulmonary Vascular Receptors. Receptors exist in the pulmonary arterial system that respond to increases in The major receptor sites considered in the sensation of pressure and/or flow that result in increased ventilation. dyspnea include chemoreceptors, mechanoreceptors and This process likely contributes to the sensation of dyspnea lung receptors [3, 4]. experienced by patients with pulmonary hypertension and acute pulmonary embolism. Chemoreceptors Hypoxemia stimulates respiration through its effects Lung Receptors on the peripheral chemoreceptors, and may thereby cause The lung contains various types of receptors that breathlessness in patients with lung disease [2–4]. For transmit information to the central nervous system example, the oxygen desaturation that may occur in many (CNS). Pulmonary stretch receptors in the airways re- patients with chronic obstructive pulmonary disease spond to lung inflation; irritant receptors in the airway (COPD), interstitial lung disease (ILD), and pulmonary epithelium respond to a variety of mechanical and chem- vascular disease can stimulate the peripheral chemorecep- ical stimuli and mediate bronchoconstriction, and C- tors and contribute to dyspnea. On the other hand, many fibers (unmyelinated nerve endings) located in the alveo- patients with respiratory disease may experience dyspnea lar wall and blood vessels respond to interstitial conges- during exercise, but not develop hypoxemia. tion. Numerous studies suggest that afferent impulses Breathing carbon dioxide (CO2) stimulates the central from these vagal-mediated receptors may contribute to chemoreceptors and increases ventilation [4, 5]. An in- dyspnea [9–11]. For example, stimulation of vagal irri- crease in dead space ventilation can develop during exer- tant receptors appears to intensify the sensation of tion in patients with severe respiratory impairment; al- breathlessness and may impart a sense of chest tightness veolar ventilation may be therefore inadequate, resulting or constriction. in hypercapnia. This enhances central respiratory drive and may cause breathlessness. Interestingly, patients with Afferent Impulse respiratory disease, especially COPD, may have hyper- Various nerve pathways can transmit the information capnia at rest, but describe little if any dyspnea despite the after stimulation of the receptor site(s) to the CNS. As increase in PaCO2. mentioned earlier, the vagal nerve transmits afferent Taken as a whole, the foregoing suggests that chemore- information from the lung receptors. ceptors probably have limited contribution to the experi- ence of exertional dyspnea. Integration/Processing in the CNS Although afferent signals associated with breathless- Mechanoreceptors ness are received, integrated and processed in the CNS, There are a number of receptors distributed through- little is actually known about these processes. However, it out the respiratory system that respond to mechanical is believed that the motor cortex or brainstem respiratory stimuli. neurons transmits a signal to the sensory cortex (i.e., cor- Upper Airway Receptors. Clinical observations are sup- ollary discharge) which may contribute to a ‘sense of ported by clinical research studies showing that upper air- effort’ to breathe [12]. This sensation increases whenever way and facial receptors modify the sensation of dyspnea the central motor command is increased (e.g., the load [6]. Patients sometimes report a decrease in the intensity placed on the respiratory muscles is increased), or when- of their breathlessness when sitting by a fan or open win- ever the respiratory muscles become weak or fatigued dow. (e.g., prolonged increase in ventilation or dynamic hyper- Chest Wall Receptors. The brain receives projections inflation which may occur with exercise). from a variety of receptors in the joints, tendons and mus- cles of the chest that influence ventilation. Mechanical Efferent Impulse stimuli, such as vibration, are known to activate these In response to afferent information, the CNS sends an receptors, and may affect the sensation of breathlessness. efferent impulse via the phrenic nerve to the diaphragm For example, investigators have shown that in-phase and other respiratory muscles to increase respiration. chest wall vibration may reduce dyspnea in patients with How this impulse affects the level of ventilation, pattern COPD [7, 8]. of breathing, lung volume, flow rates, etc., is not com- pletely understood. Clearly, the mechanisms of exertional Mechanisms and Measurement of 73 Exertional Dyspnea

dyspnea in patients with cardiorespiratory diseases are tored over time and the patient provides ratings that can multifactorial and complex. Stimulation of one or more of be analyzed from either perspectives. the aforementioned receptors could potentially contribute to dyspnea at rest as well as during exercise. In 1963, The purposes of measuring exertional dyspnea include Campbell and Howell [13] proposed the concept of differentiating between people who have less dyspnea and length-tension inappropriateness as the cause of breath- those who have more dyspnea, and determining how dys- lessness. According to their theory, dyspnea arises from a pnea changes as a function of stimulus parameters and disturbance in the relationship between the force or ten- medical intervention. sion generated by the respiratory muscles and the result- ing change in muscle length and lung volume. A current What Is the Stimulus for Dyspnea during Exercise? hypothesis proposes that a ‘mismatch’ occurs between As described above, the exact mechanisms and precise afferent information to the CNS and the outgoing motor stimuli for exertional breathlessness have not been com- command to the respiratory muscles. In brief, this theory pletely identified. Nevertheless, for measurement pur- of ‘neuroventilatory dissociation’ suggests that the brain poses it is reasonable to consider that an exercise test per- anticipates or expects a certain ventilatory response based formed on the cycle ergometer or treadmill (used to simu- on the associated afferent information [14]. Any dissocia- late a physical task) is a direct stimulus for both physiolog- tion or deviation from this system may cause and/or ical and perceptual responses [17, 18]. During CPX the intensify the perception of dyspnea. individual produces power, or generates work, which is a direct stimulus for both physiological and perceptual Measurement of Exertional Dyspnea responses. At the present time it is reasonable to use such variables (e.g., power production, work, or oxygen con- Psychophysical measurement pertains to the quantita- sumption) as the putative ‘stimulus’ for provoking dys- tive relationship between sensory experience and external pnea during exertion [18–20]. stimulus events (outer psychophysics) as well as the rela- tionship between sensory experience and physiological Instruments Used to Measure Dyspnea during Exercise variables (inner psychophysics) [15, 16]. For example, the Subjects generally provide ratings of dyspnea during relationship between a person’s breathlessness (sensory exercise using a visual analog scale (VAS) or a category experience) and work performed in watts during cycle scale. In addition to rating the intensity of breathlessness, ergometry falls within the realm of outer psychophysics, subjects can also be instructed to rate leg discomfort or whereas the relationship between breathlessness and oxy- chest pain if it is a predominant complaint. gen consumption is considered a topic of inner psycho- physics. VAS The VAS is a continuous scale represented by either a Another dichotomy in the field is the distinction be- vertical or horizontal line, usually 100 mm in length, with tween ‘global’ and ‘local’ processes. A global psychophysi- descriptors positioned as anchors [21]. For example, des- cal method is employed when a person estimates the mag- criptors may be ‘none’ or ‘no breathlessness’ and ‘very nitude of breathlessness associated with each of a number severe’ or ‘greatest breathlessness’. The subject places a of different magnitudes of a stimulus event (e.g., levels of mark on the VAS with a pen or can adjust a linear poten- work or oxygen consumption). A local method refers to a tiometer (or the cursor on a computer screen) to indicate situation where the ‘change’ in breathlessness is the de- his/her level of dyspnea on the VAS displayed on a moni- pendent variable associated with a change in stimulus tor. magnitude. With global methods the investigator is inter- ested in the quantitative relationship between breathless- 0 to 10 category-ratio scale (CR-10) ness and the precipitating stimulus magnitude. Local The most widely used scale for enabling individuals to methods focus on the amount of change in stimulus mag- rate dyspnea during CPX is the category-ratio scale (CR- nitude required to trigger a change in breathlessness for 10) developed by Borg [22]. This scale consists of a verti- each of a number of different stimulus magnitudes [16]. cal line labeled 0–10 with nonlinear spacing of verbal des- Typically, these two aspects of breathlessness are mea- criptors of severity corresponding to specific numbers sured using different procedures, though in the most that can be chosen by the subject to reflect presumed ratio recent applications breathlessness is continuously moni- properties of sensation or symptom intensity. 74 Mahler/Fierro-Carrion/Baird

Table1. Peak ratings of dyspnea on the Author Ref. Patients Age Diagnosis Peak dyspnea CR-10 scale during cycle ergometry (first author) rating (mean B SD) LeBlanc 45 18 NA CRD 7.7B2.0 Asthma 7.4B1.9 Mahler 27 15 46B16 COPD 6 (median) COPD 8.5B0.3 Killian 26 97 64B10 COPD 6.8 COPD 5.3B0.3 Dean 41 12 NA COPD 5.1B0.3 ILD 4.0B0.6 Mador 28 6 63B6 ILD 5.2B1.4 O’Donnell 20 30 66B1 30 69B1 Marciniuk 46 6 44B15 O’Donnell 33 12 64B3 NA = Not available; CRD = cardiorespiratory disease; COPD = chronic obstructive pul- monary disease. Investigators have shown that the VAS and the CR-10 of 5 and 9. Peak values for dyspnea were similar for scale provide similar scores during incremental CPX in patients with different severity of COPD. However, as healthy subjects [23] and in patients with COPD [24]. expected, the peak power output on the cycle ergometer However, the CR-10 scale has at least two advantages for was almost twice as high in the healthy subjects compared measuring dyspnea during CPX. First, the presence of with the patients with COPD. Mahler et al. [27] found descriptors on the CR-10 scale permits comparisons that 15 patients with asthma (46 B 4 years) gave a mean among individuals under the assumption that the verbal rating of 7.4 for dyspnea at peak exercise. Mador et al. descriptors on the scale correspond to the same subjective [28] noted that 6 patients with COPD had mean ratings of experience in different subjects. For example, two sub- 6.8 for ‘breathing discomfort’ on the CR-10 scale at maxi- jects may have different peak levels of cardiorespiratory mal effort on the ergometer. However, O’Donnell et al. fitness as measured by oxygen consumption, but nonethe- [20] observed somewhat lower values (5.1 B 0.3) in 30 less both may select the number ‘10’ on the CR-10 scale as patients with COPD who stopped exercise because of the proper indication of his or her subjective maximum breathlessness. breathlessness. Second, a numerical value or descriptor on the CR-10 scale may be easier to use as a dyspnea ‘tar- It is not clear why subjects terminate their exercise sub- get’ (as opposed to a measured length in mm on the VAS) stantially below the maximum point on the scale inas- for prescribing and monitoring exercise training [25]. much as they are instructed to exercise until they can no longer continue due to symptom limitation (i.e., reach Measurement of Dyspnea During Exercise their maximum breathlessness). Also, the existence of Peak Values large individual variation in stopping points found in Initial investigations measured peak values of dyspnea some studies brings into question the assumption that on the VAS or the CR-10 scale during CPX. Although a subjects all interpret the ‘maximum’ descriptor in the wide range of values have been reported, both healthy same manner. individuals and patients with cardiorespiratory disease usually stop exercise on the cycle ergometer at submaxi- An additional consideration during CPX is the deter- mal (!100 mm on the VAS and !10 on the CR-10 scale) mination of which symptom is more ‘limiting’ in the per- intensities of dyspnea and/or leg discomfort. formance of an exercise task. It is generally believed that Most older subjects (both healthy individuals and pa- healthy individuals are more often limited by leg discom- tients with cardiorespiratory disease) reach ‘symptom fort and/or general fatigue than by breathlessness. Of 40 limitation’ at ratings between 5 and 8 on the CR-10 scale patients with obstructive airway disease studied by Mah- (table 1). For example, Killian et al. [26] reported that 320 ler and Harver [29], 18 patients noted higher ratings for healthy subjects (63 B 4 years) had a median dyspnea rat- ‘leg fatigue’ than for dyspnea, 14 reported dyspnea as the ing of 6 at peak exertion, with 25–75th percentile values major symptom, and the remaining 8 patients indicated that leg fatigue and dyspnea were equal in intensity. Of 578 patients studied by Hamilton et al. [30], 265 (46%) Mechanisms and Measurement of 75 Exertional Dyspnea

Fig. 1. Intensities of leg effort and dyspnea at peak exercise in patients with cardiac disease (a) and with pulmonary disease (b). LD = Leg discomfort; BD = breathing discomfort [from 32, with permission]. rated leg effort greater than dyspnea, 222 (38%) rated the crease in power output on the cycle ergometer each 1– intensity of leg effort and dyspnea as equal, and 91 (16%) 2 min (incremental test) or by a continuous increase rated the intensity of dyspnea greater than leg effort. Fig- (ramp test). In this procedure, the subject is instructed to ure 1 shows the subjective ratings of leg effort and dys- provide ratings typically each minute ‘on cue’ during the pnea at peak exercise in patients with a pulmonary diag- CPX. Thus, a series of discrete dyspnea ratings is ob- nosis (n = 85) and with a cardiac diagnosis (n = 111) [30]. tained throughout exertion. Together, these data suggest that healthy individuals The most common approach to examine the breath- and patients with lung disease experience similar intensi- lessness continuum has been to determine the slope and ties for dyspnea and for leg discomfort at peak exercise. intercept of the stimulus → response relationship over a Furthermore, either symptom may actually be a percep- range of stimulus values [18, 19]. In general, the slope of tual limit to an individual’s exercise capacity. However, the regression between dyspnea ratings and power is high- in those patients with more severe airflow obstruction, er in patients with respiratory disease compared with dyspnea was more frequently reported as being limiting healthy individuals (fig. 2, 3) [17, 32, 33]. compared with leg discomfort [26]. Continuous Measurement Using a Computerized Although subjects are typically reliable in providing System dyspnea ratings at peak exercise when tested at different In 1993, Harty et al. [34] described the methodology time periods [17], Belman et al. [31] found that 9 patients and results of the continuous measurement of breathless- with COPD gave lower ratings of breathlessness on the ness during exercise. Six healthy subjects used a poten- CR-10 scale with successive tests over 10 days while per- tiometer to give their ratings on a VAS displayed on a forming treadmill walking. monitor. In 2001, Mahler et al. [35] reported on a continuous Continuum of Dyspnea Ratings method in which subjects throughout exercise moved a Although data on peak values of symptoms are clinical- computer mouse that controlled the length of a bar whose ly useful, there are clearly limits to such information, par- lower edge coincided with a value along the CR-10 scale ticularly when evaluating the effect of an intervention. to represent the current level of perceived dyspnea (fig. 4). Consequently, the next step in the development of meth- This approach allows the subject to provide ratings spon- ods for obtaining breathlessness ratings was to instruct taneously and continuously while performing the CPX patients to give ratings throughout the CPX. The most without waiting for a cue or request from the examiner. frequently applied exercise protocols incorporate an in- 76 Mahler/Fierro-Carrion/Baird

Fig. 2. Relationship between power output (% predicted) and dyspnea on the CR-10 scale in 85 patients with a pulmonary disease and 109 healthy normal individuals [from 32, with permission]. Fig. 3. Relationship between oxygen consumption (VO2) and breath- Fig. 4. Schematic diagram of subject pedaling on cycle ergometer lessness on the CR-10 scale in 12 patients with interstitial lung dis- with computerized system to enable the subject to provide contin- ease (ILD) and 12 age-matched normal individuals [from 33, with uous measurement of dyspnea during CPX. A mouse is positioned on permission]. a platform by the handlebars; the person can move the mouse in order to position the vertical bar to correspond to an intensity of This method creates a more thorough mapping of the dyspnea as measured on the CR-10 scale. The system is described in stimulus-response relationship because data can be re- greater detail in Mahler et al. [35]. corded continuously throughout the course of exercise rather than only at discrete points in time selected by the uals and in patients with COPD, ratings of breathlessness investigator. using the continuous method were both reliable and com- parable to those obtained by the discrete method during We compared the continuous method with the discrete incremental CPX. The continuous method also was more method (rating each minute on ‘cue’) using the CR-10 responsive in showing the expected increase in dyspnea scale presented on a computer screen. In healthy individ- Mechanisms and Measurement of 77 Exertional Dyspnea

Fig. 5. Dyspnea ratings provided each minute during submaximal exercise at 50–60% of peak exercise capacity pre- and post-ipratropium bromide (IB) (a) and pre- and post-placebo (P) solution (b) in 29 patients with COPD. * p ! 0.05 [from 36, with permission]. when an inspiratory load was added. Finally, patients Clinical Applications with COPD gave significantly more ratings (mean = 157) The following examples illustrate selected clinical ap- with the continuous method than with the discrete meth- plications in which patients provide ratings of exertional od (mean = 5) during 5 min of an incremental CPX. breathlessness during CPX. The exercise test simulates activities of daily living and provides the same standard There are at least three important advantages of the stimulus for evaluating the efficacy of different treat- continuous method for measuring dyspnea. First, the per- ments on the experience of dyspnea. ception of breathlessness almost certainly changes throughout the entire course of exercise rather than only Bronchodilator Therapy at arbitrary 1-min time intervals. Thus, the standard Mahler et al. [27] studied 15 patients with asthma dur- approach obtaining discrete ratings each minute may not ing incremental CPX and examined responsiveness of the accurately reflect the perceptual changes in dyspnea. Sec- work → dyspnea relationship at a baseline visit and then ond, the continuous method enables subjects to provide after patients received a bronchodilator (metaproterenol) substantially more dyspnea ratings compared with the and a bronchoconstrictor (methacholine) prior to exercise discrete method. This is important because many patients testing. No differences were observed for the slope of with cardiopulmonary diseases may only be able to exer- work → dyspnea at the three testing periods; however, the cise for 4 or 5 min; consequently, only 4 or 5 dyspnea rat- intercept of work → dyspnea was significantly higher after ings can be obtained with the discrete method. Further- bronchoconstriction. O’Donnell et al. [36] showed that more, statistical considerations become a concern when dyspnea ratings were reduced with 500 Ìg of ipratropium fitting a quantitative function to such a small number of bromide delivered via nebulizer as compared with place- data points. A third advantage of the continuous method bo therapy during endurance exercise at 50–60% of peak is the ability to calculate ‘just noticeable differences’ work rate in 29 patients with COPD (fig. 5). (JNDs) and the Weber fraction which may offer new sta- tistical indices for use in clinical measurement of dyspnea during CPX [35]. 78 Mahler/Fierro-Carrion/Baird

Pulmonary Rehabilitation tients were breathing room air during incremental Studies have demonstrated the benefits of exercise CPEX. training as part of a pulmonary rehabilitation program on reducing dyspnea during exercise [20, 37–39]. For exam- Bullectomy and Lung Volume Reduction Surgery ple, Ries et al. [37] showed that breathlessness ratings on At least three different studies have reported lower rat- the CR-10 scale were significantly lower during endur- ings of breathlessness during exercise following bullecto- ance treadmill exercise in those with COPD who did exer- my [42, 43] and after lung volume reduction surgery cise training compared with those who received only edu- (LVRS) [44]. For example, Teramoto et al. [42] showed cation. O’Donnell et al. [20] reported that the slope of that the slope of VO2-dyspnea was significantly decreased oxygen consumption → breathlessness fell significantly in and the absolute threshold load of dyspnea (defined as the patients who performed 2.5 h of exercise training three x-intercept of the regression line of VO2-dyspnea relation- times per week for 6 weeks compared with a control ship) was significantly increased after bullectomy in 8 group. Similarly, Ramirez-Venegas et al. [38] demon- patients with unilateral giant bulla. O’Donnell et al. [43] strated a reduction in the slope of power (W) → dyspnea reported that ratings of breathlessness fell by 45% (from relationship (pre: 0.09; post: 0.12) during incremental 6.1 B 0.8 to 3.2 B 0.7 on the CR-10 scale) at a standard- CPX after exercise training in 44 patients with COPD. ized work rate (39% of predicted maximal work rate) dur- ing incremental exercise after unilateral bullectomy (n = Oxygen 4) and after bullectomy plus ipsilateral lung reduction (n = Oxygen therapy has been shown to reduce dyspnea rat- 4). Finally, Martinez et al. [44] described reductions in ings on the VAS and on the CR-10 scale during exercise dyspnea during exercise (from 7.1 B 0.6 pre-LVRS to [40, 41]. For example, O’Donnell et al. [40] observed that 3.5 B 0.6 post-LVRS) in 12 patients with COPD selected ratings of breathlessness on the CR-10 scale were signifi- for volume reduction surgery. The changes in dyspnea cantly decreased in 11 patients with severe COPD when were significantly correlated with changes in end-expira- breathing 60% oxygen compared to when the same pa- tory lung volume (% predicted of total lung capacity). References 1 O’Connell JM, Campbell AH: Respiratory me- 7 Manning HL, Basner R, Ringler J, Rand C, 13 Campbell EJM, Howell JBL: The sensation of chanics in airways obstruction associated with Fencl V, Weinberger SE, Weiss JW, Schwartz- breathlessness. Br Med Bull 1963;19:36–40. inspiratory dyspnea. Thorax 1976;31:669– stein RM: Effect of chest wall vibration on 677. breathlessness in normal subjects. J Appl Phys- 14 O’Donnell DE: Exertional breathlessness in iol 1991;71:175–181. chronic respiratory disease; in Mahler DA (ed): 2 Lane R, Cockcroft A, Adams L, Guz A: Arterial Dyspnea. New York, Dekker, 1998, pp 97– oxygen saturation and breathlessness in pa- 8 Sibuya M, Yamada M, Kanamaru A, Tanaka 147. tients with chronic obstructive airways disease. K, Suzuki H, Noguchi E, Altose MD, Homma Clin Sci 1987;72:693–698. I: Effect of chest wall vibration on dyspnea in 15 Baird JC, Noma E: Fundamentals of Scaling patients with chronic respiratory disease. Am J and Psychophysics. New York, Wiley Inter- 3 Swinburn CR, Wakefield JM, Jones PW: Rela- Respir Crit Care Med 1994;149:1235–1240. science, 1978. tionship between ventilation and breathless- ness during exercise in chronic obstructive air- 9 Manning HL, Shea SA, Schwartzstein RM, 16 Baird, JC: Sensory psychophysics; in Kotses H, ways disease is not altered by prevention of Lansing RW, Brown R, Banzett RB: Reduced Harver A (eds): Self-Management of Asthma. hypoxaemia. Clin Sci 1984;67:515–519. tidal volume increases ‘air hunger’ at fixed New York, Dekker, 1998, pp 231–267. PCO2 in ventilated quadriplegics. Respir Phys- 4 Swinburn CR, Mould H, Stone TN, Corris PA, iol 1992;90:19–30. 17 Mahler DA, Jones PW, Guyatt GH: Clinical Gibson GJ: Symptomatic benefit of supple- measurement of dyspnea; in Mahler DA (ed): mental oxygen in hypoxemic patients with 10 Davies SF, McQuaid KR, Iber C, McArthur Dyspnea. New York, Dekker, 1998, pp 149– chronic lung disease. Am Rev Respir Dis 1991; CD, Path MJ, Beebe DS, Helseth HK: Extreme 198. 143:913–915. dyspnea from unilateral pulmonary venous ob- struction. Demonstration of a vagal mecha- 18 Killian KJ: Th objective measure of breathless- 5 Chonan T, Mulholland MB, Cherniack NS, Al- nism and relief by right vagotomy. Am Rev ness. Chest 1985;88(suppl):84S–90S. tose MD: Effects of voluntary constraining of Respir Dis 1987;136:184–188. thoracic displacement during hypercapnia. J 19 Mahler DA: The measurement of dyspnea dur- Appl Physiol 1987;63:1822–1828. 11 Taguchi O, Kikuchi Y, Hida W, Iwase N, Satoh ing exercise in patients with lung disease. Chest M, Chonan T, Takishima T: Effects of bron- 1992;101(suppl):242S–247S. 6 Schwartzstein RM, Lahive K, Pope A, Wein- choconstriction and external resistive loading berger SE, Weiss JW: Cold facial stimulation on the sensation of dyspnea. J Appl Physiol 20 O’Donnell DE, McGuire MA, Samis L, Webb reduces breathlessness induced in normal sub- 1991;71:2183–2190. KA: The impact of exercise reconditioning on jects. Am Rev Respir Dis 1987;136:58–61. breathlessness in severe chronic airflow limita- 12 Killian KJ, Gandevia SC, Summer E, Camp- tion. Am J Respir Crit Care Med 1995;152: bell EJM: Effect of increased lung volume on 2005–2013. perception of breathlessness. J Appl Physiol 1984;57:686–691. Mechanisms and Measurement of 79 Exertional Dyspnea

21 Gift AG: Visual analogue scales: Measurement 31 Belman MJ, Brooks LR, Ross DJ, Mohsenifar 40 O’Donnell DE, Douglas JB, Webb KA: Factors of subjective phenomena. Nurs Res 1989;38: Z: Variability of breathlessness measurement contributing to the relief of exertional breath- 286–288. in patients with chronic obstructive pulmonary lessness during hyperoxia in chronic airflow disease. Chest 1991;99:566–571. limitation. Am J Respir Crit Care Med 1997 22 Borg GAV: Psychological bases of perceived 155:530–535. exertion. Med Sci Sport Exerc 1982;14:377– 32 Hamilton AL, Killian KJ, Summers E, Jones 381. NL: Symptom intensity and subjective limita- 41 Dean NC, Brown JK, Himelman RB, Doherty tion to exercise in patients with cardiorespira- JJ, Gold WM, Stulbarg MS: Oxygen may im- 23 Wilson RC, Jones PW: A comparison of the tory disorders. Chest 1996;110:1255–1263. prove dyspnea and endurance in patients with visual analogue scale and modified Borg scale chronic obstructive pulmonary disease and for the measurement of dyspnea during exer- 33 O’Donnell DE, Chau LKL, Webb KA: Qualita- only mild hypoxemia. Am Rev Respir Dis cise. Clin Sci 1989;76:277–282. tive aspects of exertional dyspnea in patients 1992;146:941–945. with interstitial lung disease. J Appl Physiol 24 Muza SR, Silverman MT, Gilmore GC, Heller- 1998;84:2000–2009. 42 Teramoto S, Fukuchi Y, Nagase T, Matuse T, stein HK, Kelsen SG: Comparison of scales Shindo G, Orimo H: Quantitative assessment used to quantitate the sense of effort to breathe 34 Harty HR, Heywood P, Adams L: Comparison of dyspnea during exercise before and after bul- in patients with chronic obstructive pulmonary between continuous and discrete measure- lectomy for giant bulla. Chest 1992;102:1362– disease. Am Rev Respir Dis 1990;141:909– ments of breathlessness during exercise in nor- 1366. 913. mal subjects using a visual analogue scale. Clin Sci 1993;85:229–236. 43 O’Donnell DE, Webb KA, Bertley JC, Chau 25 Mejia R, Ward J, Lentine T, Mahler DA: Tar- LKL, Conlan AA: Mechanisms of relief of exer- get dyspnea ratings predict expected oxygen 35 Mahler DA, Mejia-Alfaro R, Ward J, Baird JC: tional breathlessness following unilateral bul- consumption as well as target heart rate values. Continuous measurement of breathlessness lectomy and lung volume reduction surgery in Am J Respir Crit Care Med 1999;159:1485– during exercise: Validity, reliability and re- emphysema. Chest 1996;110:18–27. 1489. sponsiveness. J Appl Physiol 2001;90:2188– 2196. 44 Martinez FJ, Montes de Oca M, Whyte RI, 26 Killian KJ, LeBlanc P, Martin DH, Summers Stetz J, Gay SE, Celli BR: Lung-volume reduc- E, Jones NL, Campbell EJM: Exercise capacity 36 O’Donnell DE, Lam M, Webb KA: Spirometric tion improves dyspnea, dynamic hyperinfla- and ventilatory, circulatory, and symptom lim- correlates of improvement in exercise perfor- tion and respiratory muscle function. Am J itation in patients with chronic airflow limita- mance after anticholinergic therapy in chronic Respir Crit Care Med 1997;155:1984–1990. tion. Am Rev Respir Dis 1992;146:935–940. obstructive pulmonary disease. Am J Respir Crit Care Med 1999;160:542–549. 45 LeBlanc P, Bowie DM, Summers E, Jones NL, 27 Mahler DA, Faryniarz K, Lentine T, Ward J, Killian KJ: Breathlessness and exercise in pa- Olmstead E, O’Connor GT: Measurement of 37 Ries AL, Kaplan RM, Limberg TM, Prewitt tients with cardiorespiratory disease. Am Rev breathlessness during exercise in asthmatics: LM: Effects of pulmonary rehabilitation on Respir Dis 1986;133:21–25. Predictor variables, reliability and resonsive- physiologic and psychosocial outcomes in pa- ness. Am Rev Respir Dis 1991;144:39–44. tients with chronic obstructive pulmonary dis- 46 Marciniuk DD, Watts RE, Gallagher CG: Re- ease. Ann Intern Med 1995;122:823–832. producibility of incremental maximal cycle er- 28 Mador MJ, Rodis A, Magalong VJ: Reproduc- gometer testing in patients with restrictive lung ibility of Borg scale measurements of dyspnea 38 Ramirez-Venegas A, Ward JL, Olmstead EM, disease. Thorax 1993;48:894–898. during exercise in patients with COPD. Chest Tosteson AN, Mahler DA: Effect of exercise 1995;107:1590–1597. training on dyspnea measures in patients with Donald A. Mahler, MD chronic obstructive pulmonary disease. J Car- Section of Pulmonary and Critical Care 29 Mahler DA, Harver A: Prediction of peak oxy- diopulm Rehabil 1997;17:103–109. Medicine gen consumption in obstructive airway disease. Dartmouth-Hitchcock Medical Center Med Sci Sports Exerc 1988;20:574–578. 39 Readon J, Awad E, Normandin E, Vale F, One Medical Center Drive Clark B, ZuWallack RL: The effect of compre- Lebanon, NH 03756-0001 (USA) 30 Hamilton AL, Killian KJ, Summers E, Jones hensive outpatient pulmonary rehabilitation Tel. +1 603 650 5533, Fax +1 603 650 0580 NL: Symptom intensity and subjective limita- on dyspnea. Chest 1994;105:1046–1052. E-Mail [email protected] tion to exercise in patients with cardiorespira- tory disorders. Chest 1996;110:1255–1263. 80 Mahler/Fierro-Carrion/Baird

Weisman IM, Zeballos RJ (eds): Clinical Exercise Testing. Prog Respir Res. Basel, Karger, 2002, vol 32, pp 81–88 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Cardiopulmonary Exercise Testing in Unexplained Dyspnea Steven E. Gaya Idelle M. Weismanb Kevin R. Flahertya Fernando J. Martineza aUniversity of Michigan Health System, Department of Internal Medicine, Division of Pulmonary and Critical Care Medicine, Ann Arbor, Mich., bHuman Performance Laboratory, Department of Clinical Investigation, Pulmonary-Critical Care Service, William Beaumont Army Medical Center, El Paso, Tex., and Department of Medicine, Pulmonary-Critical Care Division, University of Texas Health Science Center at San Antonio, San Antonio, Tex., USA Summary Breathlessness is an extremely common symptom; one large study identified dyspnea as the third most frequent The evaluation of unexplained dyspnea may be clarified and complaint (following fatigue and back pain) from outpa- enhanced by evaluation with cardiopulmonary exercise testing. tients evaluated in medical clinics [5]. The causes of acute An organized, systematic and stepwise approach to the evalua- dyspnea can often be distinguished by history, clinical tion of dyspnea, beginning with the medical history and physical examination, chest radiography or physiologic testing [6– examination, and including specific laboratory, radiographic 9]. However, the evaluation of dyspnea present for at least and physiologic testing, can help to clarify what can be a com- 1 month (chronic dyspnea) can be very challenging. plex and potentially multifactorial problem. The utility of car- diopulmonary exercise testing lies in its sensitivity in differen- The etiologies of chronic dyspnea identified in three tiating between multiple etiologies of disease and its ability to studies are listed in table 1. Airway disease (asthma or study in greater detail the potential causes of disease and exer- chronic obstructive pulmonary disease (COPD)) repre- cise limitation. sents the majority of cases, followed by cardiovascular disease, interstitial lung disease (ILD), deconditioning, Dyspnea describes the sensation of breathlessness, psychogenic disorders, gastroesophageal reflux, neuro- feelings of air hunger, uncomfortable sensations of breath- muscular disease and pulmonary vascular disease. It ing, or awareness of respiratory distress [1–3]. A recent should be appreciated that a referral bias is likely present consensus statement from the American Thoracic Society in these studies which originated from pulmonary referral defined dyspnea as ‘... a term used to characterize a sub- clinics. As such, physicians specializing in cardiovascular jective experience of breathing discomfort that consists of disease will likely identify a higher proportion of cardio- qualitatively distinct sensations that vary in intensity. vascular disease in their population of patients [10]. Simi- The experience drives from interactions among multiple lar data are lacking from a primary care setting. physiological, psychological, social, and environmental factors and may induce secondary physiological and be- The opinions or assertions contained herein are the private views of havioral responses’ [4]. the authors and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of De- fense.

Table1. Primary etiology of chronic dyspnea in three published stud- Table 2. Potential sequence of diagnostic testing in the evaluation of ies and the number of patients evaluated with comprehensive cardio- chronic dyspnea pulmonary exercise testing Step Suggested diagnostic studies Study Patients Number n% evaluated with CPET Initial evaluation History and physical examination Chest radiograph Pratter et al. [15] 25 29 15 Initial diagnostic Electrocardiogram Asthma 12 14 15 testing Laboratory studies COPD 12 14 50 Spirometry/diffusing capacity Interstitial lung disease Subsequent Arterial blood gas Cardiomyopathy 9 11 diagnostic testing Upper airway 78 Pulmonary function tests: Psychogenic 45 Lung volumes Deconditioning 45 MVV Gastroesophageal reflux 34 Mouth respiratory pressures Extrapulmonary 34 Bronchial provocation testing Miscellaneous 56 Cardiopulmonary exercise testing Echocardiography DePaso et al. [31] Functional cardiac testing: Hyperventilation syndrome 14 19 Exercise thallium Exercise radionuclide ventriculography Unexplained 14 19 Exercise echocardiography Cardiac catheterization Asthma 12 17 Pulmonary vascular evaluation: Ventilation/perfusion lung scanning Cardiac disease 10 14 Pulmonary angiography Pulmonary parenchymal evaluation: Pulmonary vascular disease 4 6 High resolution computed tomography of chest Lung biopsy Interstitial lung disease 23 Gastroesophageal reflux evaluation Chronic obstructive disease 3 4 Neuromuscular disease 34 Gastroesophageal 34 Thyroid disease 23 Deconditioning 23 Upper airway 23 Miscellaneous 11 Martinez et al. [33] 14 28 Initial Evaluation Deconditioning 12 24 Medical History and Physical Examination Asthma Different descriptors of dyspnea exist in patients with Psychogenic 9 18 various types of cardiopulmonary disease [12, 13]. There- Cardiac disease 7 14 fore an evaluation should always begin with a detailed his- Unexplained 7 14 tory and physical examination. Interestingly, a prelimina- Interstitial lung disease 48 ry report of 11 patients suggested that a 15-item question- Gastroesophageal reflux 12 naire, which assessed dyspnea descriptors, may provide Miscellaneous 12 additional complementary information to the medical history [14]. The value of this form of evaluation in Evaluation of Chronic Dyspnea patients presenting with breathlessness requires further prospective validation. Several groups have suggested a stepwise approach to The timing of symptoms should be sought from every the evaluation of patients with unexplained exertional patient. An acute onset of symptoms may suggest bron- dyspnea [6, 11]. These approaches take the most common choconstriction, pulmonary embolism, cardiac ischemia, causes of dyspnea into consideration during the evalua- or airway obstruction due to a foreign body or secretion. tion process. Table 2 enumerates such an approach; fig- In contrast, chronic symptoms are more likely to reflect ure 1 illustrates an adaptation of these published recom- slowly progressive disorders such as COPD, congestive mendations [9]. heart failure (CHF), or ILD. Precipitating factors such as body position may provide additional diagnostic clues. 82 Gay/Weisman/Flaherty/Martinez

Fig. 1. Algorithm for the evaluation of chronic dyspnea [from 9, with permission]. Orthopnea is common in patients with CHF, severe history of wheezing and the diagnosis of asthma as 42 and COPD, ascites, obesity, anterior mediastinal masses and 83% respectively. Past medical history, concurrent condi- respiratory muscle weakness. Similarly, trepopnea (dys- tions, previous surgeries, social information (including pnea in one lateral position but not in the other) can be cigarette smoking, previous and current occupation, fami- seen with patients with unilateral lung disease, unilateral ly/living status), and medication history should also be pleural effusion, and unilateral obstruction of the airway. reviewed. In the study of Pratter et al. [15] a smoking his- Platypnea (dyspnea in the upright position, which may be tory had an excellent negative predictive value (100%) for relieved by recumbency) can be seen in patients with an excluding the diagnosis of COPD, however, only 20% of intracardiac shunt, parenchymal lung shunts or hepato- patients with a smoking history had COPD. pulmonary syndrome [9]. A detailed physical examination with special attention Associated symptoms, such as cough, wheezing, or to the heart and lungs, is essential. Other organ systems sputum production, can provide additional information pertinent to the history should also be investigated. After in the evaluation process. Cough supports a diagnosis of completing the history and physical examination, physi- airway disease, ILD, or gastroesophageal reflux disease. cians may be able to generate a list of probable causes for Similarly, wheezing suggests asthma, COPD, or CHF. In a chronic dyspnea, which can be used to direct diagnostic prospective study of 100 patients evaluated for chronic testing. For example, a strong history of cardiac risk fac- breathlessness in a subspecialty clinic, Pratter et al. [15] tors or appropriate symptoms may lead to further, spe- reported the positive and negative predictive values for a cific cardiac testing. In contrast, a strong suspicion of Cardiopulmonary Exercise Testing in 83 Unexplained Dyspnea

respiratory disease may lead to further pulmonary spe- respiratory musculature [22]. The evaluation of a de- cific testing. This approach may allow for a rapid and creased MVV should also include assessment for upper cost-effective narrowing of the evaluation of chronic dys- airway abnormalities [23]. Important information can be pnea. obtained by examining the flow-volume curve to look for evidence of poor effort, vocal cord dysfunction, and extra- Initial Diagnostic Testing thoracic airway obstruction [23]. Although a decreased Laboratory/Radiographic Testing MVV can be a useful clue to the etiology of chronic dys- Routine laboratory testing such as a complete blood pnea, the effort dependence limits the diagnostic accuracy count, thyroid function and renal testing, may be helpful of this study. in screening for basic systemic disorders (anemia, hypo-/ hyperthyroidism, and metabolic acidosis) which can Measurement of the DLCO assesses the ability of the cause dyspnea [9]. Furthermore, the presence of a pneu- alveolar-arterial interface to transfer gas [24]. There is mothorax, hyperinflation, interstitial fibrosis, pulmonary potential utility to the identification of an isolated reduc- edema, pulmonary artery enlargement, cardiomegaly, or tion in DLCO as this finding may be suggestive of several diaphragm elevation can provide additional clues to the possible etiologies of dyspnea (see below) [24]. Interest- cause of chronic dyspnea. Although chest radiographs are ingly, in an asymptomatic patient, an isolated reduction helpful in determining the presence of certain abnormali- in DLCO has not been shown to be clinically significant ties, the absence of changes should not rule out potential and does not demand further evaluation [25]. In addition, disorders. Mild ILD, for example, may not be seen on a decreased DLCO may identify a group of patients more chest x-ray but may be suggested by pulmonary function likely to demonstrate abnormal gas exchange during testing [16]. CPET (see below) [26]. Pulmonary Function Testing Subsequent Diagnostic Testing The high prevalence of respiratory disease in patients The results from the history, physical exam, and initial presenting with dyspnea makes spirometry and flow-vol- testing can be used to direct further testing if the etiology ume loop analysis essential during the initial work-up of of the dyspnea remains elusive. For example, if the initial chronic dyspnea [9]. Spirometry can be used to diagnose evaluation suggests a cardiac etiology, then further evalu- obstructive lung disease. However, the suggestion of re- ation for CHF or cardiac ischemia is warranted. If, on the strictive lung disease requires further evaluation of lung other hand, initial pulmonary function testing demon- volumes and respiratory muscle strength to differentiate strates an isolated decrease in DLCO, further evaluation ILD from respiratory muscle weakness. Evaluation of for pulmonary hypertension (echocardiography, right lung volumes by body plethysmography or by gas dilution heart catheterization) [27], ILD with emphysema (high- techniques [17, 18] can give an accurate assessment of the resolution computed tomography) [28, 29] or imaging for presence/severity of restrictive lung disease. Similarly, the recurrent pulmonary emboli can be pursued [30]. measurement of maximal inspiratory and maximal expi- If the initial evaluation does not suggest a likely disor- ratory pressures is a useful tool to screen for respiratory der, the age of the patient can be used to guide further muscle dysfunction. In fact, in patients with neuromuscu- testing. For a young patient, particularly if dyspnea is lar disease, the earliest physiologic abnormality is a de- intermittent, there is a greater likelihood of asthma [31]. crease in respiratory pressure measured at the mouth [19, Methacholine challenge testing (MCT) allows for a sensi- 20]. As such, syndromes of respiratory muscle weakness tive, but not specific tool to evaluate for the presence of may be uncovered with this simple test. Unfortunately, in asthma [32]. MCT can also be used in older patients with the setting of patients presenting with dyspnea, the sensi- a recent series identifying airway hyperreactivity in pa- tivity and specificity of this test is unknown. The mea- tients with a median age of 54 years [33]. surement of maximum ventilatory ventilation can serve In older individuals or those with potential cardiovas- as an additional surrogate measurement of impaired re- cular risk factors, specific cardiac testing should be con- spiratory muscle function [21]. Data from our institution sidered. In a recent evaluation of ambulatory and inpa- have confirmed that an isolated decrement in MVV (com- tient university-based family practice patients, the fre- pared to that expected for the measured FEV1 ! 40) is quency of CHF increased with age [34]; 74% of the seen in patients with mitochondrial myopathy presenting patients were older than 65 years. Importantly, 40% of the with unexplained exertional intolerance and involving the patients had preserved systolic function, a condition that was more commonly seen in women. This condition, dia- 84 Gay/Weisman/Flaherty/Martinez

stolic heart failure, is more likely to be seen in older unexplained dyspnea was the indication for testing in 32 patients particularly in the presence of hypertension, dia- patients. The most frequent exercise pattern was ‘inap- betes mellitus, obesity or valvular heart disease [35]. A propriate hyperventilation’ (n = 14); 9 additional patients recent report has documented the presence of exertional demonstrated a normal exercise response while 3 demon- pulmonary hypertension in a small series of patients eval- strated ventilatory limitation, 2 demonstrated a cardiac uated for dyspnea [36]. A separate report described 103 limitation and 4 were felt to have deconditioning. The consecutive patients evaluated in an emergency room set- authors concluded that CPET provided information in ting with new onset dyspnea; 14 patients had pericardial many patients that negated the need for further testing. effusions with 4 demonstrating clinically significant effu- sions (mean age 54 years) [37]. As such, some patients Given the distribution of etiologies of dyspnea (ta- may be best evaluated with specific cardiac studies early ble 1), CPET should be optimally suited to serve an in the evaluation process. important diagnostic role in the evaluation of this symp- tom. In fact, figure 2 demonstrates possible diagnostic Cardiopulmonary Exercise Testing categories derived from CPET in patients evaluated for Cardiopulmonary exercise testing (CPET) is a diagnos- dyspnea [11, 39]. Numerous chapters in this monograph tic modality that is ideally suited for the evaluation of provide specific data regarding the pattern of response in unexplained dyspnea. Appropriate analysis of the pat- cardiac, respiratory, and myopathic disorders. terns of response, as described elsewhere in this volume (see chapter 25), can suggest various disorders that can A review of the available literature suggests that a com- contribute to the sensation of breathlessness. Interesting- pletely normal study does not exclude early disease but ly, in many series of unexplained dyspnea the utilization should serve to reassure the patient that a major disorder of CPET is infrequent. In 100 consecutive patients evalu- is not likely present [33, 38]. Most patients with psycho- ated at a tertiary referral center for chronic dyspnea CPET genic dyspnea will have a normal response to exercise was used in the evaluation of only 15 patients [15]. CPET although an abnormal pattern of ventilation may be was felt to be particularly useful in diagnosing psychogen- instructive [11]. Figure 3 demonstrates an erratic pattern ic dyspnea (n = 4) and deconditioning (n = 4). In a subse- of respiratory flow generation during the course of a maxi- quent study of 77 patients referred to a tertiary referral mal CPET in a patient with psychogenic dyspnea. Simi- pulmonary clinic with unexplained dyspnea, CPET was larly, a hyperventilation syndrome may be suggested dur- utilized in 15 patients [31]. An abnormal study was noted ing CPET [40], although this diagnosis is generally one of in 4 patients but was only felt to be diagnostic or direct exclusion. Abnormal electrocardiographic tracings can further evaluation in 2 of these subjects. [31] In a prospec- suggest the presence of ischemic heart disease although tive study of 50 patients with unexplained dyspnea, Mar- other findings on CPET are nonspecific. One prospective tinez et al. [33] examined the diagnostic value of CPET. study has demonstrated the similar response of patients Seven patients had a cardiac cause for their symptoms, 5 with deconditioning and those with non-ischemic heart of whom demonstrated electrocardiographic changes of disease [33], therefore additional testing is likely indi- ischemia during CPET. In 17 patients, a pulmonary pro- cated. cess was felt to be the cause of dyspnea; 9 of these patients demonstrated diagnostic CPETs (4 with exertional bron- The diagnostic difficulty between diagnosing cardio- chospasm and 5 with gas exchange defects). The largest vascular disease, deconditioning, and metabolic myopa- group of subjects (n = 24) demonstrated a pattern consis- thies using CPET has been highlighted by recent data that tent with poor conditioning or occult cardiovascular dis- confirm a similar hyperdynamic and hyperventilatory ease. In 14 of these patients obesity and/or deconditioning response in patients with abnormal peripheral muscle were felt to be causal while 3 had cardiovascular disease oxygen utilization [41, 42]. Patients with histologically or and 7 had hyperactive airways disease confirmed with enzymatically confirmed mitochondrial disease can additional testing. Importantly, these investigators noted present with unexplained dyspnea and decreased maxi- that CPET was useful in identifying a cardiac or pulmo- mal VO2, a low anaerobic threshold and an abnormally nary process, although it was insensitive in distinguishing steep heart rate response [22, 41–43]. Data from our insti- deconditioning from a cardiac limitation. This latter tution confirm that metabolic myopathies represented group formed the largest group of patients. Sridhar et al. 8.5% (28/331) of cases of unexplained exertional limita- [38] described results of 100 randomly chosen CPETs; tion evaluated in a tertiary specialty clinic over a 4-year period. Interestingly, these patients can demonstrate sig- nificant improvement in exercise capacity after pulmo- nary rehabilitation [44]. This finding has led some to sug- Cardiopulmonary Exercise Testing in 85 Unexplained Dyspnea

Fig. 2. Results from CPET which may help guide subsequent evaluation of unexplained dyspnea. CPET = Cardiopul- monary exercise test; ILD = interstitial lung disease; HRCTS = high-resolution computed tomography scan; PVD = pulmonary vascular disease; V/Q = ventilation/perfusion, +/– or incomplete resolution on the arm under treatment [from 39, with permission]. gest that an aggressive pulmonary rehabilitation program patients with a decreased DLCO may be best assessed with be considered in patients with suspected mitochondrial collection of arterial blood samples during CPET. disease before muscle biopsy is performed [39]. Further data are required to address these evolving concepts. Additional data collected during CPET can have im- portant diagnostic significance. Measurement of pleural CPET can be used to identify potential pulmonary dis- and diaphragmatic pressures can identify unexpected re- orders in patients presenting with dyspnea. Evaluation of spiratory muscle dysfunction [48]. Serial spirometry after potential pulmonary etiologies for unexplained dyspnea exercise may identify patients with exercise-induced should include consideration of exercise-induced reactive bronchospasm, although the sensitivity of CPET in this airway disease, which may become apparent from mea- setting appears to be less than other forms of bronchopro- surement of flow volume loops during or after exercise, vocation testing [49, 50]. The addition of tidal flow vol- evaluation of gas exchange abnormalities from parenchy- ume loop analysis may improve diagnostic accuracy, mal lung disease and consideration of pulmonary hyper- although specific data are lacking [51]. Interestingly, ex- tension manifest only during exercise. Measurement of amination of vocal cord function during exercise or flow- arterial blood gases can provide useful information in volume loops during and after exercise may identify identifying parenchymal lung disease [45, 46] or pulmo- patients with vocal cord dysfunction [52, 53]. A recent nary vascular disease (pulmonary dead space (VD/VT) report identified vocal cord dysfunction in 5/33 young abnormality) [47]. Importantly, a decreased DLCO ap- military personnel evaluated for exertional dyspnea [52]. pears to identify a group of patients more likely to demon- strate abnormal gas exchange during CPET [26]; those As highlighted by the above studies, the utility of CPET in the evaluation of chronic dyspnea is decreased 86 Gay/Weisman/Flaherty/Martinez

Fig. 3. Tidal inspiratory and expiratory flows during the final minute of a maximal cardiopulmonary exercise test demonstrate an erratic pattern of respiratory efforts. Further testing revealed no specific cardiopulmonary pathology and the patient’s symptoms abated with psychotherapy. The most likely diagnosis was psychogenic dyspnea [from 9, with permission]. by its lack of specificity. Nonetheless, it can serve as a some series [33] may have more than one diagnosis, both decision node in the overall evaluation of chronic dyspnea bronchoprovocation testing and CPET may be necessary as outlined in figures 1 and 2. It is notable that figures 1 in such scenarios (see case 1 in the chapter An Integrative and 2 support a scheme in which bronchoprovocation Approach to the Interpretation of Cardiopulmonary Exer- testing precedes CPET in the evaluation of unexplained cise Testing, pp 300–322). It is evident that further pro- dyspnea, as it appears that bronchoprovocation testing is spective study is required to better define the role of phys- more sensitive than CPET in the diagnosis of hyperactive iologic testing in the evaluation of unexplained dyspnea. airways disease [50]. Importantly, as 10% of patients in References 1 Wasserman K, Casaburi R: Dyspnea: Physio- 8 McNamara RM, Cionni DJ: Utility of the peak 16 Orens JB, Kazerooni EA, Martinez FJ, Curtis logical and pathophysiological mechanisms. expiratory flow rate in the differentiation of JL, Gross BH, Flint A III Lynch JP: The sensi- Annu Rev Med 1988;39:503–515. acute dyspnea. Cardiac vs. pulmonary origin. tivity of high-resolution CT in detecting idio- Chest 1992;101:129–132. pathic pulmonary fibrosis proved by open lung 2 Mahler DA, Harver A, Lentine T, Scott JA, biopsy: A prospective study. Chest 1995;108: Beck K, Schwartzstein RM: Descriptors of 9 Alhamad EH, Gay SE, Flaherty KR, Martinez 190–115. breathlessness in cardiorespiratory diseases. FJ: Evaluating chronic dyspnea: A stepwise ap- Am J Respir Crit Care Med 1996;154:1357– proach. J Respir Dis 2001;22:79–88. 17 Aaron SD, Dales RE, Cardinal P: How accu- 1363. rate is spirometry at predicting restrictive pul- 10 Cook DG, Shaper AG: Breathlessness, angina monary impairment? Chest 1999;115:869– 3 Simon PM, Schwartzstein RM, Weiss JW, La- pectoris and coronary artery disease. Am J Car- 873. hive K, Fencl V, Teghtsoonian M, Weinberger diol 1989;63:921–924. SE: Distinguishable sensations of breathless- 18 Irvin CG: Lung volumes. Semin Respir Med ness induced in normal volunteers. Am Rev 11 Weisman IM, Zeballos RJ: Clinical evaluation 1998;19:325–334. Respir Dis 1989;140:1021–1027. of unexplained dyspnea. Cardiologia 1996;41: 621–634. 19 Demedts M, Beckers J, Rochette F, Bulcke J: 4 Society American Thoracic. Standardization of Pulmonary function in moderate neuromuscu- Spirometry, 1994 Update. American Thoracic 12 Elliott MW, Adams L, Cockcroft A, Macrae lar disease without respiratory complaints. Eur Society. Am J Respir Crit Care Med 1995;152: KD, Murphy K, Guz A: The language of J Respir Dis 1982;63:62–67. 1107–1136. breathlessness. Use of verbal descriptions by patients with cardiopulmonary disease. Am 20 Martinez FJ: Neuromuscular diseases of the 5 Kroenke K, Arrington ME, Mangelsdorff AD: Rev Respir Dis 1991;144:826–832. chest; in Goldstein R, O’Connell J, Karlinsky J The prevalence of symptoms in medical outpa- (eds): A Practical Approach to Pulmonary tients and the adequacy of therapy. Arch Intern 13 Simon PM, Schwartzstein RM, Weiss JW, Medicine. Philadelphia, Lippincott-Raven, Med 1990;150:1685–1689. Fencl V, Teghtsoonian M, Weinberger SE: Dis- 1997, pp 323–344. tinguishable types of dyspnea in patients with 6 Mahler DA: Diagnosis of dyspnea; in Mahler shortness of breath. Am Rev Respir Dis 1990; 21 Celli BR: Clinical and physiologic evaluation DA (ed): Dyspnea. New York, Dekker, 1998, 142:1009–1014. of respiratory muscle function. Clin Chest Med pp 221–259. 1989;10:199–214. 14 Scott JA, Mahler DA: Prospective evaluation 7 Ailani RK, Ravakhah K, DiGiovine B, Jacob- of a descriptor model to diagnose the etiology 22 Flaherty KR, Wald J, Weisman KM, Zeballos sen G, Tun T, Epstein D, West BC: Dyspnea of dyspnea. Chest 1995;188S. RJ, Schork A, Blaivas M, Rubenfire M, Marti- differentiation index: A new method for the nez FJ: Unexplained exertional limitation: rapid separation of cardiac vs. pulmonary dys- 15 Pratter MR, Curley FJ, Dubois J, Irwin RS: Characterization of patients with a mitochon- pnea. Chest 1999;116:1100–1104. Cause and evaluation of chronic dyspnea in a drial myopathy. Am J Respir Crit Care Med pulmonary disease clinic. Arch Intern Med 2001;164:425–432. 1989;149:2277–2282. Cardiopulmonary Exercise Testing in 87 Unexplained Dyspnea

23 Martinez FJ: Pulmonary function testing in the 35 Vasan RS, Benjamin EJ, Levy RD: Prevalence, 46 Keogh BA, Lakatos E, Price D, Crystal RG: evaluation of upper airway obstruction; in Nor- clinical features and prognosis of diastolic Importance of the lower respiratory tract in ton M (ed): Atlas of the Difficult Airway. St heart failure: An epidemiologic perspective. J oxygen transfer. Exercise testing in patients Louis, Mosby, 1996, pp 125–133. Am Coll Cardiol 1995;26:1565–1576. with interstitial and destructive lung diseases. Am Rev Respir Dis 1984;129(suppl):S76–S80. 24 Crapo RO: Carbon monoxide diffusing capaci- 36 James KB, Maurer J, Wolski K, Lutton SR, ty (transfer factor). Semin Respir Crit Care Haas G, Schilz R, Rubin D, Young JB: Exercise 47 Wasserman K, Hansen JE, Sue DY, Casaburi Med 1998;19:335–347. hemodynamic findings in patients with exer- R, Whipp BJ: Principles of Exercise Testing tional dyspnea. Tex Heart Inst J 2000;27:100– and Interpretation. Including Pathophysiology 25 Ansari A, Collier J, Mohsenifar Z: Isolated 105. and Clinical Applications. Philadelphia, Lip- reduction in single-breath diffusing capacity in pincott Williams & Wilkins, 1999. young, healthy, asymptomatic women. Am J 37 Blaivas M: Incidence of pericardial effusion in Med Sci 1995;310:226–228. patients presenting to the emergency depart- 48 Knobil K, Becker FS, Harper P, Graf LB, Wolf ment with unexplained dyspnea. Acad Emerg GT, Martinez FJ: Dyspnea in a patient years 26 Mohsenifar Z, Collier J, Belman MJ, Koerner Med 2001;8:1143–1146. after severe poliomyelitis. The role of cardio- SK: Isolated reduction in single-breath diffus- pulmonary exercise testing. Chest 1994;105: ing capacity in the evaluation of exertional dys- 38 Sridhar MK, Carter R, Banham SW, Moran F: 777–781. pnea. Chest 1992;101:965–969. An evaluation of integrated cardiopulmonary exercise testing in a pulmonary function labo- 49 Eliasson AH, Phillips YY, Rajagopal KR: Sen- 27 Rubin LJ: Primary pulmonary hypertension. N ratory. Scott Med J 1995;40:113–116. sitivity and specificity of bronchial provoca- Engl J Med 1997;336:111–117. tion testing. An evaluation of four techniques 39 Weisman IM, Zeballos RJ: A step approach to in exercise-induced bronchospasm. Chest 28 Hansell DM: High-resolution computed to- the evaluation of unexplained dyspnea: The 1992;102:347–355. mography in the evaluation of fibrosing alveo- role of cardiopulmonary exercise testing. Pulm litis. Clin Chest Med 1999;20:739–760. Perspect 1998;15:8–11. 50 Zeballos RJ, Weisman IM, Connery SM, Brad- ley JP: Standard treadmill (STE) vs. incremen- 29 Klein JS, Gamsu G, Webb WR, Golden JA, 40 Gardner WN: The pathophysiology of hyper- tal cycle ergometry in the evaluation of airway Müller NL: High-resolution CT diagnosis of ventilation disorders. Chest 1996;109:516– hyperreactivity in unexplained dyspnea. Am emphysema in symptomatic patients with nor- 534. Rev Respir Crit Care Med 1999;159:A419. mal chest radiographs and isolated low diffus- ing capacity. Radiology 1992;182:817–821. 41 Hooper RG, Thomas AR, Kearl RA: Mito- 51 Johnson B, Weisman IM, Zeballos RJ, Beck chondrial enzyme deficiency causing exercise KC: Emerging concepts in the evaluation of 30 Fedullo PF, Auger WR, Channick RN, Kerr limitation in normal-appearing adults. Chest ventilatory limitation during exercise: The ex- KM, Rubin LJ: Chronic thromboembolic pul- 1995;107:317–322. ercise tidal flow-volume loop. Chest 1999;116: monary hypertension. Clin Chest Med 2001; 488–503. 22:561–581. 42 Dandurand RJ, Matthews PM, Arnold DL, Ei- delman DH: Mitochondrial disease. Pulmo- 52 Morris MJ, Deal LE, Bean DR, Grbach VX, 31 DePaso WJ, Winterbauer RH, Lusk JA, Dreis nary function, exercise performance, and blood Morgan JA: Vocal cord dysfunction in patients DF, Springmeyer SC: Chronic dyspnea unex- lactate levels. Chest 1995;108:182–189. with exertional dyspnea. Chest 1999;116: plained by history, physical examination, chest 1676–1682. roentgenogram, and spirometry. Analysis of a 43 Elliott DL, Buist NRM, Goldberg L, Kenna- seven-year experience. Chest 1991;100:1293– way NG, Powell BR, Kuehl KS: Metabolic 53 Bacharier LB, Strunk RC: Vocal cord dysfunc- 1299. myopathies: Evaluation by graded exercise tion: A practical approach to diagnosis. J Res- testing. Medicine 1989;68:163–172. pir Dis 2001;22:93–103. 32 Sterk PJ: Bronchoprovocation testing. Semin Respir Med 1998;19:317–324. 44 Taivassalo T, DeStefano N, Argov Z, et al: Fernando J. Martinez, MD Effects of aerobic training in patients with mi- 1500 E. Medical Center Drive 33 Martinez FJ, Stanopoulos I, Acero R, Becker tochondrial myopathies. Neurology 1998;50: 3916 Taubman Center FS, Pickering R, Beamis JF: Graded compre- 1055–1060. Ann Arbor, MI 48109 (USA) hensive cardiopulmonary exercise testing in Tel. +1 734 763 2540, Fax +1 734 936 5048 the evaluation of dyspnea unexplained by rou- 45 Risk C, Epler GR, Gaensler EA: Exercise al- E-Mail [email protected] tine evaluation. Chest 1994;105:168–174. veolar-arterial oxygen pressure difference in in- terstitial lung disease. Chest 1984;85:69–74. 34 Diller PM, Smucker DR, David B, Graham RJ: Congestive heart failure due to diastolic or sys- tolic dysfunction: Frequency and patient char- acteristics in an ambulatory setting. Arch Fam Med 1999;8:414–420. 88 Gay/Weisman/Flaherty/Martinez

Weisman IM, Zeballos RJ (eds): Clinical Exercise Testing. Prog Respir Res. Basel, Karger, 2002, vol 32, pp 89–98 OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Respiratory System Responses to Exercise in Aging Bruce D. Johnson Division of Cardiovascular Diseases, Department of Internal Medicine, Mayo Clinic and Foundation, Rochester, Minn., USA Summary mal review of baseline changes in the pulmonary system with aging, the reader is referred to several previous Aging results in a progressive fall in vital capacity and maxi- papers [5–9]. mal expiratory flows, at a rate which may accelerate in the more advanced years of life. In addition, there is an apparent Table 1 summarizes the changes in lung and chest wall reduction in pulmonary capillary blood volume, a gradual rise in function, gas exchange and ventilatory control with aging pulmonary vascular pressures, and an altered ventilation distri- and the impact of these age-related changes on the bution as aging progresses. These changes result in a reduction response to exercise in the healthy, active, older adult. in ventilatory reserve and in theory would make the older adult more susceptible to gas exchange abnormalities during exer- Pulmonary Mechanics cise. Nevertheless, the healthy older adult is generally able to maintain an appropriate alveolar ventilation to maintain arterial With the decline in pulmonary function (due primarily oxygenation and reduce arterial carbon dioxide levels, even to a loss of lung elastic recoil), older subjects have a during heavy exercise. The primary limitation resulting from reduced ability to increase tidal volume (VT) and to the age-related changes in the pulmonary system may be relat- increase flow rates during exercise. In the average 70-year- ed to the large work and cost of breathing which could limit old subject, the reduction amounts to a 30% loss in vital exercise performance by competing for blood flow with the capacity (VC) and forced expiratory volume in 1 s (FEV1) locomotory muscles. relative to the 20-year-old adult [4, 9, 10]. The relative reduction in flow rates is particularly great over the lower The large capacities of the respiratory system allow for lung volumes (functional residual capacity, FRC, and significant erosion in function between maturity and below) and thus limits the ability to increase flow in the senescence with minimal impact on normal breathing. older adult over this lung volume range [1]. Interestingly, Only during moderate to heavy exercise does it appear ventilatory (VE) demand tends to fall with aging commen- that the aged-induced changes significantly impact nor- surate with the fall in metabolic demand so that it may be mal breathing, and then primarily through an effect on theorized that the relative demand/capacity balance may breathing strategy, work and cost of breathing rather than be similar to the average fit young adult. It is this relation- on alveolar to arterial gas exchange [1–4]. The focus of ship of demand to capacity that in part determines the this review will be the impact of age-related changes in the adequacy of the ventilatory compensation to exercise in respiratory system on the response to exercise. For a for- the older adult. Interestingly, recent studies have suggest- ed that the decline in pulmonary function with aging may accelerate beyond age 50–60 years and that this decline in

Table1. Influence of age-related changes in the respiratory system on the response to heavy exercise Baseline change (or proposed change) Result Response to exercise in the older adult Pulmonary Mechanics ↓ Lung volumes (TLC, VC) ↑ Expiratory flow limitation ↓ Flow rates Altered regulation of end-expiratory lung volume ↓ Elastic recoil ↓ Pressure generation capacity ↑ Chest wall stiffness ↓ Expiratory and inspiratory pressure reserve ↓ Respiratory muscle strength ↑ VT/VC ↓ Intervertebral space ↑ Work and oxygen cost of breathing ↑ Competition for blood flow between locomotor and Gas Exchange/Hemodynamics respiratory muscles ↓ Elastic recoil (nonuniform) ↑ Alveolar duct diameter ↓ Pulmonary capillary blood volume ↑ Dead space ventilation ↓ Alveolar septa ↓ Surface area ↑ VE to maintain alveolar PO2 ↑ Stiffness of pulmonary arteries and ↑ VA/Qc inhomogeneity Arterial PO2 maintained within 5 mm Hg resting values ↑ Pulmonary pressures Alveolar to arterial PO2 difference ↑ threefold capillaries Diastolic dysfunction ↓ Response to chemical and VE response generally adequate to maintain PaO2 near mechanical stimuli resting values and to ↓ PaCO2 below resting values Ventilatory Control ↓ Integration of sensory inputs in CNS ↓ Perceptual sensitivity to inspiratory and expiratory loads ↓ Inspiratory neuromuscular output function is not modified by habitual physical activity nor 30-year-old may have a peak exercise VO2 of 45 ml/kg/ high aerobic capacity [4, 7]. Thus in advanced age, the min and the average 70-year-old, 25 ml/kg/min. This loss in capacity may begin to play a role in limiting human results in a comparable decline in VCO2 and thus it can be performance, especially in the elderly that maintain rela- predicted (from the above equation) that the peak ventila- tively high levels of activity [2–4]. tion necessary for maintaining normal alveolar oxygen levels during peak exercise (assumes a similar dead space Ventilatory Demand to tidal volume ratio, VD/VT) will fall by approximately Ventilatory demand is dependent foremost on meta- 30–40% or a decrease from 120 l/min for the normally bolic demand (quantified by measurements of oxygen active 30-year-old to 70 l/min for the average fit 70-year- consumption, VO2, or carbon dioxide production, VCO2), old adult. However, given the increase in dead space ven- but also on the dead space ventilation (VD) and regula- tilation with aging (↑ 0.1–0.6%/year beyond age 25–30), it tion of arterial CO2 levels (PaCO2). This relationship is is expected the decline in ventilatory demand for the aver- summarized in the following equation: VE = (K WVCO2)/ age 70-year-old will only be 25–30% [9, 15]. With in- [PaCO2W(1 – VD/VT)] (K = the constant 0.863 and repre- creased fitness it can subsequently be predicted that for sents the factor needed to transform fractional gas con- every 500 ml increase in metabolic demand (F50 W on a centration to partial pressure and to express gas volumes cycle ergometer), the ventilatory requirements will in- at body temperature and pressure saturated with water crease by F15 l/min. The ventilatory demands would fur- vapor). Many studies have evaluated changes in maximal ther increase as pH falls with heavy exercise and arterial oxygen consumption with aging and most demonstrate a CO2 is reduced in an attempt to compensate for the acido- decline of approximately 0.4–0.6%/year beyond the age of sis. 30–35 and attribute the decline primarily to a reduced cardiac output as a result of a decline in heart rate, Breathing Pattern although a loss of muscle mass and altered mitochondrial The typical response to exercise is to increase both the function may play a role [11–14]. Thus the average 25- to frequency of breathing and the tidal volume. Most studies 90 Johnson