J. Human Ergol., 4: 35-41,1975 SERIAL DETERMINATION OF CARDIAC OUTPUT DURING PROLONGED WORK BY A CO2-REBREATHING TECHNIQUE Brian A. WILSON and R. T. HERMISTON* Department of Human Kinetics, University of Guelph, Guelph, Ontario, Canada *Faculty of Health, Physical Education and Recreation, University of Windsor, Windsor, Ontario, Canada Cardiac outputs were calculated on eight male college students while walking on a treadmill at controlled heart rate levels. The heart rate levels were maintained through the use of a heart rate controller (Quinton Model AI-607), standard deviations for heart rate in beats/min ranged from +2.1 at the heart rate level 110 to +3.2 at 170. Test-retest correlations of the cardiac output values at the various heart rate levels were : at heart rate 110, 0.66 after 7 min and 0.85 at 30 min; at heart rate 150, 0.92 after 7 min and 0.68 at 30 min; at heart rate 130, 0.97 after 7 min, and at heart rate 170, 0.85 after 7 min. The calculated cardiac output values when plotted against oxygen uptake compare favorably with published reports. The data collected demonstrate that the CO2-rebreathing method used in the present investigation provides a reliable bloodless technique for determining multiple cardiac output during prolonged work of varied intensities. Calculation of many cardiac outputs on one subject during exercise presents a variety of problems for both subjects and investigators if direct cardiac output methods are employed. An indirect bloodless technique allows these determinations to be made repeatedly with relative ease and little discomfort to the subject. The lack of reported data examining the reproducibility of the CO2-rebreathing technique for the calculation of repeat cardiac outputs during prolonged work led to the present investigation. DEFARES (1958) first employed the rebreathing Pvco2 value in the Fick equation to determine cardiac output. Other investigators (JERNERUS et al.,1963; LUNDING and THOMPSON,1966) used the same technique, but the cardiac output values calculated were usually lower than those computed by direct Fick and dye-dilution methods (HAMILTON, 1962). Length of the rebreathing period and the initial Received for publication May 12, 1975 35
36 B. A. WILSON and R. T. HERMISTON CO2 percentage in the rebreathing system were found to be the important factors in improving the accuracy. Recent literature (FERGUSON et al., 1968; LUNDING, 1965; MUIESAN etal.,1968; BAR-OR et al.,1971) demonstrates that with rebreathing periods between 10-15 sec and initial gas percentage of 3-5% CO2 in oxygen, the cardiac output values computed by rebreathing are highly correlated (r>0.90) with those calculated by both direct Fick and dye-dilution techniques. Although the method has become an exceptable measure of cardiac output during short term work, its reliability for determining multiple outputs over extended exercise periods has yet to be examined. METHODS The basic steps employed in computing the cardiac output with the CO2- rebreathing technique were as follows : The partial pressure of CO2 in the arterial blood was determined using the Bohr formula and the dead space estimate from ASMUSSEN and NIELSEN's (1956) data: Paco2=Vt (Feco2) (Pb-47)/Vt-(VD+D), where Vt is tidal volume in liters at BTPS, Feco2 fraction of CO2 in the expired air, VD dead space in liters BTPS, D dead space of the valve apparatus in liters, and Paco2 partial pressure of CO2 in the arterial blood in mmhg. The venous CO2 tension was estimated using the method of DEFARES (1958) and a continuous sampling procedure of JERNERUS (1963). Due to the high oxygen concentration of the rebreathing mixture all CO2 tensions were converted to concentrations using a standard CO2 dissociation curve for oxygenated blood (COMROE et al., 1962). The cardiac output was calculated using the Fick equation for CO2: Q=Vco2/ (V-A)co2, where Q is cardiac output in liters/min, Vcp2 CO2-production in ml/ min, and (V-A)CO2 venous-arterial CO2 concentration difference in ml/liter. Eight male physical and health education majors, between the ages of 20 and 27 years volunteered to participate in this study. These subjects were all physically active, but none were involved in endurance type training. ASTRAND et al. (1964) and EKELOM and HERMASEN (1968) have reported the unique circulatory responses of the highly trained endurance athletes. The subjects practiced rebreathing and were put through a trial test which consisted of one seven minute walk at a controlled heart rate of 130 beats/min. The trial walk was performed in order to determine the electrocardiogram and treadmill acceleration settings on the heart rate controller, and to estimate the percentage grade to be used for the test walks. The rebreathing apparatus used is shown in Fig. 1. The five liter rebreathing bag contained 4.0 percent CO2 in oxygen. The gas was sampled from the bag during rebreathing, the water vapor was removed, then the sample was analyzed by a Godart infra-red Capnograph and returned to the system at a rate of approximately 2.0 liters per minute. The volume of gas in the bag was approximately 2.5 liters at heart rate level 110, and was increased for heart rate levels of 130 and
CARDIAC OUTPUT DURING PROLONGED WORK 37 Fig. 1. Schematic representation of the CO2-rebreathing apparatus. 150 approaching 5 liters at level 170. At the end of a deep expiration, the subject was switched from room air into the rebreathing system, and breathed in and out at a constant rate dependent upon the work level. The rebreathing period lasted from 8 to 15 seconds (LUNDING and THOMPSON,1965). Subjects were instructed to empty the bag with each inspiration, and to maintain a constant expired volume. This allowed for rapid mixing of lung and bag gases, plus an even time space interval for the breath by breath C 2 increase. Slight variations in the depth and frequency of breathing did not affect the calculation. During the prolonged walks of 30 min, an identical procedure was followed at each five minute interval. RESULTS The test, re-test cardiac output values calculated are plotted in Fig. 2; 83 of the Fig. 2. Test, re-test cardiac output values determined by the CO2 method. 20% from the line of identity. Dashed lines are
38 B. A. WILSON and R. T. HERMISTON 112 plots fell within a 20% error of the line of identity. For the seven minute tests, all plots above the 110 heart rate level were within the 20 % error range. Correlated t-tests revealed no significant differences (P<0.05) for any of the test re-test values calculated. The reproducibility of the method is illustrated in Table 1. Figure 3 shows the relationship between mean cardiac outputs and oxygen uptake for four independent studies : 1) The regression line developed by EKELUND and HOLMGREN (1967) using direct Fick. 2) The mean plots of the data by FERGUSON et al. (1968) for sitting work calculated by CO2 rebreathing. 3) Data of REEVES et al. (1961) for treadmill walking determined by direct Fick. 4) The results of the present investigation. Table 1. Test, re-test values for cardiac outputs. Fig. 3. Comparison of mean Q plotted against mean Vo2 for the present study and four earlier papers. \. direct Fick (REEVES), ~- ~ CO2 method (present study), direct Fick (EKELUND & HOLMGREN), - - - CO2 method (FERGUSON).
CARDIAC OUTPUT DURING PROLONGED WORK 39 DISCUSSION There were high correlations (0.65<r<0.97) and no significant differences between the means of duplicate cardiac outputs calculated at heart rate levels 110, 130,150, and 170, for seven minutes of exercise. At heart rate level 110, however, the accuracy of the method was affected by lower than criterion tidal volumes. For tidal volumes between 1.5 and 3.3 liters, ASMUSSEN and NIELSEN (1956) calculated the range of maximal error in estimating Paco2 to be between {1.7 and 1.6 mmhg. Some of the subjects at heart rate 110 were working with tidal volumes below one liter. This factor undoubtedly contributed to the experimental error for the lower work level. At the lower work loads an end tidal air sampling method should be employed to calculate PaCa2 (FERGUSON et al.,1968; JONES et al., 1966). Of the 29 test, re-test plots that fell outside the 20 % error range, 19 were at the 110 heart rate level. A high degree of variability had been reported in the cardiac response during prolonged work at a constant work load (GRIMBY et al., 1966; LEVY et al., 1961). Data examining the reliability of cadiovascular values during prolonged work were reported by LEVY et al. (1961). His comparison of cardiac outputs determined during the 10th minute of a 20 minute walk gave a correlation value of 0.64. The variations that occur in the cardiovascular components during prolonged work at a controlled heart rate level have not been fully examined. It appears from the present investigation that the alterations in work load required to maintain a constant heart rate contribute to the degree of variation. The only correlation values below 0.65 were recorded during the 10 and 15 minute intervals for heart rate 110 and the 15 minute interval for heart rate 150. It was during these time intervals that the highest degree of variation in treadmill speeds occurred in order to maintain the prescribed heart rate level. The cardiac outputs and mean oxygen uptake values for the 7th minute of the exercise periods are plotted in Fig. 3. In all cases the observed values are lower than those predicted by the regression line by EKELUND and HOLMGREN (1967). These results were expected since in the present study, subjects exercised in the erect posture causing a certain reduction in cardiac output values due to lower limb blood pooling (BEVERGARD et al., 1963; NILSSON et al.,1963). The plotted cardiac output values of the present investigation fall between those of the regression line, and the values from data of REEVES et al. (1961) for treadmill walking calculated by the direct Fick technique. No tendency to plateau at the 170 heart rate level was observed. This is contrary to the trend reported by FERGUSON et al. (1968) for sitting work using the CO2 method, but consistent with the relationship found from cardiac outputs determined using direct Fick (EKELUND and HOLMGREN, 1967; REEVES et al., 1961). The levelling off in cardiac output often found using indirect methods (FERGUSON et al.,1968; KLAUSEN, 1965) has been used to support the hypothesis that the heart
40 B. A. WILSON and R. T. HERMISTON cannot maintain stroke volume at high heart rate levels. Subjects in the present study were able to maintain stroke volume up to 90 % of the maximum heart rate values. This finding is in agreement with results of studies employing direct Fick (ASTRAND et al., 1964; EKELUND and HOLMGREN, 1967; REEVES et al.,1961). However, it must be remembered that the workload was not constant over the 30 minute period but showed a mean decrease of 8 % at heart rate 150. This percent decrease compares favorably with the usual heart rate increases during prolonged work under steady load conditions (GRIMBY et al., 1966). This usual increase in heart rate over time at a given work load was prevented by an automatic decrease in the treadmill speed. During constant load work bouts of this duration both Q and Vo2 have been reported to increase (GRIMBY et al., 1966). Neither of these parameters changed significantly during the thirty minute walks in the present study even with a decreasing work rate. This response may be partly explained by some shift in substrate utilization toward fat metabolism but must also be due to a real decrease in mechanical efficiency. More data are needed before any definite conclusions can be made concerning the relative contributions of heart rate and stroke volume to cardiac output during work at controlled heart rate levels. The CO2-rebreathing method did prove to be a reliable determinant of cardiac output during treadmill exercise. For work loads requiring tidal volumes greater than 1.5 liters, the method produced consistent outputs over the 30 minute exercise period. The data suggest that the CO2 method utilized in the present study is a reliable bloodless technique for determining multiple cardiac outputs during prolonged work. The authors are indebted to Dr. H. J. Montoye for his review of the article, to T. Cada for his excellent technical assistance and to L. Wilson and B. Gladdin for their help in the preparation of the manuscript. REFERENCES ASMUSSEN, E. and NIELSEN, H. (1956) Physiological dead space and alveolar gas pressure at rest and during muscular exercise. Acta Physiol. Scand., 38:1-21. ASTRAND, P, O., CUDDY, T. E., SALTIN, B., and STENBERG., J. (1964) Cardiac output during submaximal and maximal work. J. Appl. Physiol., 19:268-273. BAR-OR, O., SHEPARD, R. J., and ALLEN, C. L. (1971) Cardiac output of 10 to 13 year old boys and girls during submaximal work. J. Appl. Physiol., 30:219-223. BEVERGARD, S., HOLMGREN, A., and JONSSON, B. (1963) Circulatory studies in well trained athletes at rest and during heavy exercise with special reference to the influence of stroke volume and body position. Acta. Physiol. Scand., 57:26-50. COMROE, J. H., FORSTER, R. E., DUBOIS, A. B., BRISCOE, W. A., and CARLSEN, E. (1962) The Lung, 2nd Ed. Year Book, Medical Publishers, Inc., Chicago, p. 154. DEEARES, J. G. (1958) Determination of PvCO2 from the exponential CO2 rise during rebreathing J. Appl. Physiol., 13:159-164. EKBLOM, B. and HERMASEN, L. (1968) Cardiac output in athletes. J. Appl. Physiol., 25:619-625. EKELUND, L. G. and HOLMGREN, A. (1967) Central hemodynamics during exercise. Circulation
CARDIAC OUTPUT DURING PROLONGED WORK 41 Res., 20, Suppl. 1: 21-23. FERGUSON, R. S., FAULKNER, J. A., JULIUS, S., and CONWAY, J. (1968) Comparison of cardiac output determined by CO2 rebreathing and dye-dilution methods. J. Appl. Physiol., 25: 450-454. GRIMBY, G., NILSSON, N. J., and SANNE, H. (1966) Repeated serial determination of cardiac output during 30 minute exercise. J. Appl. Physiol., 21: 1750-1756. HAMILTON, W. F. (1962) Measurement of the cardiac output. In Handbook of Physiology. The Williams and Wilkins Co., Baltimore, Sec. II, Vol. 1, pp. 551-573. JERNERUS, R., LUNDING, G., and THOMPSON, D. (1963) Cardiac output in healthy subjects determined with a CO2 rebreathing method. Acta Physiol. Scand., 59: 390-399. JONES, N. L., MCHARDY, G. J. R., NAIMARK, A., and CAMPBELL, E. J. M. (1966) Physiological dead space and alveolar-arterial gas pressure differences during exercise. Clin. Sci., 31: 19-29. KLAUSEN, K. (1965) Comparison of CO2 rebreathing and acetylene methods for determination of cardiac output. J. Appl. Physiol., 20: 763-766. LEVY, A. M., TABKIN, B. S., and HANSON, S. S. (1961) Cardiac output in normal men during steady state exercise utilizing dye-dilution technique. Brit. Heart J., 23: 425-432. LUNDING, G. (1965) Mixed venous CO2 tension determined by a CO2 rebreathing method, influence of rebreathing time. Acta Physiol. Scand., 63: 55-57. LUNDING, G. and THOMPSON, D. (1965) Determination of the mixed venous CO2 pressure with a rebreathing method. Correction for volume changes for the lung-bag system. Acta Physiol. Scand., 64: 448-452. LUNDING, G, and THOMPSON, D. (1966) Cardiac output in the supine and sitting position determined by a CO2 method. Acta Physiol. Scand., 66:129-132. MUIESAN, G., SORBINI, C. A., SOLINAS, E., GRASSI, V., CASUCCI, G., and PETZ, E. (1968) Comparison of CO2 rebreathing and direct Fick method for determining cardiac outputs. J. Appl. Physiol., 24: 424-429. NILSSON, N. J., ARENANDER, E., CARLSTEN, A., GRIMBY, G., and WESTLING, H. (1963) Effect of body posture on the cardiac output in patients with varicose veins. Scand. J. Gun. Lab. Invest., Suppl.15: 69-193. REEVES, J. T., DROVER, R. F., BLOUNT, S. G., Jr., and FILLEY, G. F. (1961) Cardiac output response to standing and treadmill walking. J. Appl. Physiol., 16: 283-288.