Assessment of Exercise Intensity Formulas By Use of Ventilatory Threshold

Assessment of Exercise Intensity Formulas By Use of Ventilatory Threshold* Linn Goldberg, M.D.; Diane L. Elliot, M.D.; and Kerry S. Kuehl, M.S.

Guidelines for training heart rate (HR) during aerobic exercise are often determined by predictive formulas. Measurement of the heart rate at ventilatory threshold (VT) by expired gas analysis provides a direct index of the upper limits of conditioning mtensity We evaluated 115 nonsmoking, healthy adults with measurement of peak oxygen uptake to classify groups as low- (0 = 45), average(n = 45), and high (n = 25) -fitness, Heart rate at VT was compared with the approximate midpoint (77 percent) of recommended training intensity as estimated by the Karvonen equation, predicted maximal (220- age), and meas-

ured maximal HR formulas. No significant difference among the various HR formulas at 77 percent and HR at VT were found for high-Stness individuals. Among the lowand average-6tness groups, the Karvonen formula at 77 percent was signi6cantly higher (p <0.001) than HR at ventilatory threshold. Predicted and measured maximal HR at 77 percent were not above the vr among the lowand average-6tness individuals and are appropriate for training intensitj, However, the Karvonen formula appears to overestimate heart rate intensity among those of low and average 6tness and may be excessive for these groups.

Recommendations for aerobic conditioning are based on the intensity frequency and duration of exercise.':" The intensity should be sufficient to elicit a training response, yet if exertion is too intense, the duration of training will be compromised.v" The heart rate (HR) is often a reliable indicator of both myocardial and total oxygen requirements during exercise. Heart rate changes are commonly used to quantify and monitor endurance training. 1-3,6 Several formulas (Table 1) employ either predicted or measured maximal heart rates to prescribe exercise intensity The Karvonen equation7 uses predicted maximal HR (220- age) along with measured resting HR and introduces the concept of a heart rate reserve: [(220- age) - (resting heart rate)] X (60 to 90%) + (resting HR) Also, a percentage of either the predicted maximal (220-age) X (70 to 85 percent) or the maximal HR measured during exercise testing (HR X 70 to 85 percent) has been suggested as an appropriate training level. 1-3 Although these heart rate formulas predict the training intensity respiratory parameters during physical exertion provide a direct measurement of exertion levels," While exercise intensity below 50 percent of peak oxygen consumption is not likely to result in significant aerobic conditioning.v'? determination of the ventilatory threshold (VT) is a preferred, noninvasive method to assess the upper limits of training. 11

This threshold, found by locating the point of a nonlinear rise in minute ventilation (VE) with respect to oxygen consumption, can be identified during progressive exercise.v" Since vr correlates with an acute rise in levels of venous lactate, metabolic acidosis, and an accelerated rate of muscle glycogen depletion, exertion above this point results in an inability to sustain exercise.v'" In addition, a potential arrhythmogenic increase in catecholamine concentration occurs during exercise above the ventilatory threshold. 14,15 To assess the validity of the predicted methods for training intensity we compared the suggested midpoint (77 percent) of these formulas with the exercise HR at the ventilatory threshold. If the predicted HR intensity at this midintensity level is significantly higher than the HR at VI: the duration of exertion could be limited, with a resultant decrease in endurance benefits.v" Excessive training intensity may also be hazardous to those with coronary ischemia or exercise-induced arrhythmias.

*From the Human Performance Laboratory Division of General Medicine, Oregon Health Sciences Unfversity Portland. Supported in part by DHHS grant 1 D28 PE10057-Ol. Manuscript received August 7; revision accepted January 14. Reprint request: Dr. Goldberg, 3181 SW Sam jackson Park Road, Portland, Oregon 97201

MATERIAL AND METHODS

One hundred fifteen healthy subjects (64 men and 51 women) consecutively referred for a fitness evaluation were studied over 14 months. Inclusion criteria included no medication use, no symptoms or history of cardiopulmonary disease, and no contraindication or

Table I-Intensity Prescription Formulas Predicted maximal (220-age) x 77% Karvonen formula [(220-age) - (resting heart rate) x 77%] + resting heart rate Measured maximal Measured maximal heart rate x 77%

CHEST I 94 I 1 I JUL'(, 1988

95

Table 2-E:rercise Testing Results in Three Groups Varying Fitness ( ± SEM)

musculoskeletal limitation to maximal exertion. Exercise Testing

Subjects fasted for at least four hours before testing and performed an exercise test to volitional exhaustion by either motorized treadmill (Bruce protocol) or Corival 300 electrically braked cycle ergometer using a standard protocol (two-minute stages with 3O-W workload increases/stage) suggested by the American College of Sports Medicine .' Resting heart rate was determined after five minutes of supine rest. A 12-lead ECG recording (Marquette Case 11) was continuously displayed during exercise . Subjects breathed through a low-resistance two-way valve (Hans-Rudolph), with expired air passing through a 10-L mixing chamber. Spirometric parameters and expired gas analysis values were measured with a Gould 9000 IV metabolic measurement cart and calibrated with standardized gas values before each test. Expired oxygen and carbon dioxide were continuously measured by paramagnetic and infrared analyzers, respectively. Ventilatory measurements were averaged over 20second intervals. Exercise Intensity Determination

Ventilatory threshold was defined as the point where a nonlinear increase in ventilation occurred as compared with oxygen consumption.·· 11 This was visually assessed by independent observers, using the point of departure from the best line drawn through the initial VE points ." Observers were blinded to the study;' hypotheses and mutually agreed on any contested threshold determinations. Data Analysis

The approximate midpoints (77 percent) of standard recommended HR training formulas were compared with the HR at ventilatory threshold by analysis of variance with post hoc, twotailed t test comparisons. Data are presented as mean:t SEM .

Fitness Measurement N

\'02, mVkglmin

Age, yr Male/Female ratio Cycle/Ireadmill Max RER* Max HR, bpm

Low-fitness Group Forty-five subjects with a mean (± SEM) peak oxygen uptake (V0 2) (mllkg/min) of 21.7 ± 1.8 were identified. Their measured maximal HR of 166±3 bpm was significantly lower (p<0.05) than both the average- and high-fitness groups (Table 2). Training heart rate from the exercise prescription formulas at 77 percent intensity was compared with the HR at ventilatory threshold (Fig 1). The heart rate (138 ± 3 bpm) at vr was similar to the predicted formula of (220 - age) x 77 percent (137 ± 2 bprn), However, the Karvonen formula at 77 percent was significantly higher (150± 1 bpm) (p
Low

Average

High

45 21.7:t 1.8 44:t2 1.1:1 1:1 1.32:t .02 166:t3t

45 31.8:t 1.7 46:t3 1.1:1 1:0.8 1.26:t .03 176:t3

25 51.2:t2.3 42:t2 2.1:1 1:0.9 1.23:t .02 178:t 2

*Respiratory exchange ratio. tp < .05.

Average Fitness Group Forty-five subjects with a mean peak V0 2 of31.8 ± 2 mllkg/min and mean HR of 176 ± 3 bpm, respectively, were evaluated. Comparisons among predicted heart rate formulas and ventilatory measurements are shown in Figure 2. Heart rate at the ventilatory threshold was 145 ± 3 bpm and not significantly different from the predicted training heart rate ([220- age] x 77 percent) formula (143±2 bprn), However, the Karvonen formula at 77 percent was significantly higher (155 ± 2 bpm) (p
INTENSITY PRESCRIPTION LOW FITNESS

RESULTS

Patients were stratified to either low, average, or high fitness by standard tables of maximal oxygen consumption (ml/kg/min)." Fitness level, age, male/ female ratio in each group, and use of cycle or treadmill testing, as well as respiratory exchange ratio and HR at peak exertion, are shown in Table 2. No differences in age, testing modalities, or maximal respiratory exchange ratio were present among the three fitness groups.

of

160 155 •• p
150

e 145

e ... a":: ...... :z:"

Q,

CD

140 135

CD

130 125 120 Predicted Karvone n nOlo

220-age

x 77%

Measured Max HR

HR at VT

x 77%

FIGURE 1. Predicted training heart rate (HR) formulas at 77 percent intensity compared with HR at the ventilatory threshold (VI} for the low-fitness group. Assessment of Exercise Intensity Formulas (Goldberg,

EJllo~

Kuehl)

160

INTENSITY PRESCRIPTION AVERAGE FITNESS

155 •• p

150

e e

Q.

.. ....

CD III

It

< 0.00 1

145 140 135

III CD

:l: 130

125 120 Predicted Predicted Measured HR at VT Karvonen 22O-age Max HR x 77% x 77% nOlo FIGURE 2. Predicted training heart rate (HR) formulas at 77 percent intensity compared with HR at the ventilatory threshold (VT) for the average-fitness group.

the ventilatory threshold. Paabo and Karpmans study" of women also found that use of the Karvonen method resulted in consistently higher exercise levels than the maximal observed HR formula. Fitness stratification among our subjects does not appear to reflect differences in testing procedures or end points. Maximal or peak heart rates were adequately identified for each fitness group as shown by equivalence of respiratory exchange ratios (RER) and all subjects exercising beyond their ventilatory threshold. 19 The finding of lower HR among less conditional individuals is consistent with other cross-sectional studies that also reported a lower maximal HR among the less fit. 20.21 In addition, all groups had an equivalent percentage of peak exercise HR at vr, similar to previous studies in which the percentage of peak HR at the ventilatory threshold was independent of fitness level. 11 The ventilatory threshold corresponds to the upper limit of aerobic conditioning intensity.4 .s.13 At intensities beyond this threshold, anaerobic metabolism could limit training duration owing to accumulation of lactic acid, with resultant metabolic acidosis.v!" In addition, exertion above vr could be arrhythmogenic secondary to increased catecholamine concentrations. 14.IS Green et al22 presented evidence that lactate accumulation in exercising muscle actually precedes both the vr and the blood lactate thresholds, a finding

High Fitness Group This group of 25 subjects had a peak V0 2 and heart rate of51.2±2.3 ml/kg/rnin and 178±2 bpm, respectively. Predicted HR formulas at mid-intensity and HR at the ventilatory threshold are shown in Figure 3. Heart rate at VT (150±4 bpm) was not different than Karvonen (155 ± 1) and predicted maximal (145 ± 2) HR formulas ([220 - age] x 77 percent). However, 77 percent of measured peak HR was significantly lower (139±2 bpm) (p
Heart rate at the ventilatory threshold was compared with standard formulas predicting the midpoint (77 percent) of exercise intensity for various fitness groups. We found that, among low- and average-fitness individuals, the predicted training heart rate at the approximate midpoint of the Karvonen formula was significantly higher than the HR at the ventilatory threshold. Other formulas, using 77 percent of either predicted (220 - age) or measured maximal heart rates, resulted in training levels either near or below the ventilatory threshold. Coplan et al," studying mostly men and analyzing only treadmill testing, similarly reported that an exercise prescription based on 80 percent of peak measured HR was likely to be below

160

INTENSITY PRESCRIPTION HIGH FITNESS

155

•• P < 0.001

150

e e

.. ....

145

Q. CD III

It

III CD

:l:

140 135 130 125 120 Predicted Karvonen nOlo

Predicted 22O-age

x 77%

Measured MaxHR

HR at VT

xn%

3. Predicted training heart rate (HR) formulas at 77 percent intensity compared with HR at the ventilatory threshold (VI) for the high-fitness group. FIGURE

CHEST / 94 / 1 / JULY. 1988

97

that strengthens the contention that prescribed training should not exceed the ventilatory threshold. Suggested exercise protocols for determining ventilatory threshold have differed among investigators. Although some evidence suggests that the vr may be better detected during long-term rather than progressive exercise.P the ventilatory threshold determinations during either test are highly correlated and coincide with the threshold for plasma lactate production. 24 In addition, visual interpretation of the ventilatory threshold has been reliable and correlates highly with computer-determined regression models. 16 This threshold determined as the point at which nonlinear rapid increase in ventilation occurs also correlates with other noninvasive ventilatory variables to determine the VI: 23 Although maximal oxygen uptake is reduced during cycle ergometry when compared to treadmill testing, 25 the ventilatory threshold occurs at the same percent of maximal V02 with similar maximum HR, VE, and lactate production.Pr" Davis et al l 2 found that there was no significant difference between vr obtained by either cycle ergometry or treadmill exercise. Visual assessment of the ventilatory threshold is subjective, which is a limitation of this methodology owing to potential variability in the point of slope ehange.P However, we compared the midpoint rather than the upper limits of the HR formulas to establish whether the predicted HR equations were inappropriately high. If midexercise intensity prescription heart rates are above the ventilatory threshold, the maximal suggested training HR would likely be above VI: despite some slurring of this level. Our results are derived from individuals who were without cardiopulmonary disease and medication use. Application of these data for medically impaired populations or for those who use medication that could alter exercise parameters may not be appropriate. For most individuals, training intensity is determined from the predicted or measured maximal heart rate. Our findings validate the use of either maximal predicted (220 - age) or measured HR at the midpoint of recommended intensity However, we found that a 77 percent training intensity as calculated by the Karvonen formula, resulted in a level incompatible with prolonged aerobic conditioning for those of low or average fitness. Since the midpoint of recommended exercise intensity of the Karvonen equation was significantly higher than the ventilatory threshold among these groups, prescribing exercise above this level (78 to 90 percent) should not be recommended. REFERENCES 1 American College of Sports Medicine. Guidelines for exercise testing and prescription. 3rd ed. Philadelphia: Lea ~ Febiger,

98

1986:18-19 2 Goldberg L, Elliot DL. Prescribing exercise. West J Med 1984; 141:383-86 3 Fischer AA, Goldberg L, Lyles BD Jr, McKeag DB. Mobilizing the sedentary patient. Patient Care 1987; 21:14-31 4 Wasserman K, Whipp BJ, Koyal SN, Beaver WL. Anaerobic threshold and respiratory gas exchange during exercise. J Appl Physiol 1973; 35:236-43 5 Davis JA. Anaerobic threshold: a review of the concept and directions for future research. Med Sci Sports Exerc 1986; 17:618 6 Astrand ~ Rodahl K. Textbook of work physiology New York: McGraw-Hill, 1977 7 Karvonen M, Kentala K, MustaIa o. The effects of training heart rate: a longitudinal study Ann Med Exp BioI Fenn 1957; 35:30715 8 Dwyer J, Bybee R. Heart rate indices of the anaerobic threshold. Med Sci Sports Exerc 1983; 15:72-78 9 Davies CfM, Knibbs AV. The training stimulus: the effects of Intensity duration and frequency of effort of maximum aerobic power output. Int Z Agnew Physiol1971; 29:299-305 10 The recommended quantity and quality of exercise for developing and maintaining fitness in health adults. Med Sci Sports 1978; 10:7-10 11 Coplan NL, Gleim G~ Nicholas jA. Using exercise respiratory measurement to compare methods of exercise prescription. Am J Cardioll986; 58:832-36 12 Davis jA, Vodak ~ Wilmore JH, Vodak J, Kurtz E Anaerobic threshold and maximal aerobic power for three modes of exercise. J Appl Physiol1976; 41:544-50 13 Wasserman K. The anaerobic threshold measurement to evaluate exercise performance. Am Rev Respir Dis 1984; 129(suppl):3540 14 Lehman M, Keul J, Huber J, DaPrada M. Plasma catecholamines in trained and untrained volunteers during graduated exercise. Int J Sports Med 1981; 2:143-47 15 Lehman M, Schmid ~ Kuel J. Plasma catecholamines and blood lactate cumulation during incremental exhaustive exercise. Int J Sports Med 1985; 6:78-81 16 Orr G~ Green HJ, Hughson RL, Bennet SW A computer linear regression model to determine ventilatory anaerobic threshold. J Appl Physioll982; 52:1349-52 17 Exercise testing on apparently healthy individuals: a handbook for physicians. Dallas: American Heart Association, 1975 18 Paabo S, Karpman MB. The relationship between exercise intensity levels of two predictive heart rate equations and percent maximal oxygen consumption. J Sports Med 1981; 21:226-30 19 Wasserman K, Hansen JE, Sue DY, Whipp 8J. Principles of Exercise Testing and Interpretation. Philadelphia: Lea & Febiger, 1987 20 Cooper KH, Purdy J, White S, Pollock M, Linnerud AC. Agefitness adjusted maximal heart rates. Med Sci Sports 1977; 10:78-

86

21 Hansen JE, Sue DY, Wasserman K. Predicted values for clinical exercise testing. Am Rev Respir Dis 1984; 129(suppl):S49-S55 22 Green HJ, Hughson RL, Orr G~ Ranney DA. Anaerobic threshold, blood lactate and muscle metabolites in progressive exercise. J Appl Physioll983; 54:1032-38 23 Hughson RL. Methodologies for measurement of the anaerobic threshold. Physiologist 1984; 27:304-11 24 Buchfuhrer MJ, Hansen JE, Robinson TE, Sue K, Wasserman K, Whipp BJ. Optimizing the exercise protocol for cardiopulmonary assessment. J Appl Physioll983; 55:1558-64 25 Hermansen L, Saltin B. Oxygen uptake during maximal treadmill and bicycle exercise. J Appl Physioll969; 26:31-37

Assessment of Exercise Intensity Formulas (Goldberg, Bllot, Kuehl)