Cardiac Output Determined by the CO2 Rebreathing Method with Correction of Lung-bag Volume Shrinkage

Transcription

1 Japanese Journal of Physiology, 34, ,1984 Cardiac Output Determined by the CO2 Rebreathing Method with Correction of Lung-bag Volume Shrinkage Tokuo YANG, Peter NORSK,* and Flemming BONDE-PETERSEN* Institute of Health and Sports Science, University of Tsukuba, Niihari, Ibaraki, 305 Japan *August Krogh Institute, University of Copenhagen, Universtetsporken 13, DK 2100 Copenhagen, Denmark Abstract The purpose of this study was to elucidate the effect of lungbag volume shrinkage on P'vco2 estimated by the C02 rebreathing method and to correct its effect in calculation of cardiac output. The P'vco2 was extrapolated with the linear relationship between dpaco2 Jdt and PAc o 2 during rebreathing. The P'vc o 2 was higher than real mixed P002 by 0.82 mmhg at rest, by 1.35 mmhg at 50W, 1.57 mmhg at 100W, and 3.06 mmhg at 150W considering the effect of lung-bag volume shrinkage detected by argon concentration. The cardiac output obtained by the C02 rebreathing method with correction of above differences was in good agreement with that determined by the acetylene method (r=0.930, p<0.001). When the cardiac output was plotted against 02 uptake, our results and the regression line were fairly close to the reported regression lines. Key Words: cardiac output, C02 rebreathing method, lung-bag volume shrinkage, mixed venous C02 pressure. DEFARES (1956) not only introduced the rebreathing technique to estimate oxygenated mixed venous C02 pressure (Pvco2), but also was the first to discuss the effect of lung-bag volume shrinkage on the estimation of the Pvco2. The effect of lung-bag volume shrinkage which derived from 02 uptake being higher than C02 elimination during rebreathing was then judged to be negligible. The recent development of a fast responding gas analyzer makes it possible to use the C02 rebreathing method even at high exercise intensities. At the same time, since exercise requires high 02 uptake, the problem of lung-bag volume shrinkage would be revealed. In fact, LUNDIN and THOMSON (1965) proved such effect on the Pvcp2 estimation in detecting the change of nitrogen concentration. Later, FARHI et al. (1976) proposed the equation to estimate the theoretical difference between the real Pvco2 and the Pvco2 determined by the C02 rebreathing Received for publication May 30,

2 884 T. YANG, P. NORSK, and F. BONDE-PETERSEN method (P'vco2). Nevertheless many investigators (KLAUSEN, 1965; FERGUSON et al., 1968; KNOWLTON and ADAMS, 1974; HEIGENHAUSER and FAULKNER, 1978; AUCHINCLOSS et al., 1980) have not paid attention to this effect. The purpose of the present study was, therefore, to confirm the study of LUNDIN and THOMSON (1965). Further, Farhi's equation was modified to an equation to estimate cardiac output by the C02 rebreathing method and the validity of the cardiac output was tested. METHOD The subjects were 6 healthy young men, and their maximal 02 uptake was !/min. At rest and during exercise, heart rate was monitored to ascertain the steady state in cardiac function. The steady state of heart rate was usually attained within 5 min. The exercise was performed using an electrically braked bicycle ergometer (BoNDE-PETERSEN, 1974) at each work load (50, 100, and 150 W) for 10 min. At least 10 min rest was taken between successive work loads. After 7 min of exercise, 02 uptake (Vo2) and C02 output (Vco2) were determined by the Douglas bag method. The gas analyses were performed by previously calibrated gas analyzers, type OA-184, Servomex for 02 and Beckman LB-2 for C02. Cardiac output measurement was started immediately after the expired gas collections. The gas composition of the rebreathing bag was around 1.5 % acetylene, 9 % argon, and 46% oxygen in nitrogen. The initial rebreathing bag volume was measured by displacement of gas mixture in a 9-liter Warren Collins wet spirometer. A thin catheter was fitted to the mouth piece connected the rebreathing bag and the gas was sampled for continuous analysis with a quadrupole 8-channel mass spectrometer (Centronic MGA 200). Farhi's equation to assess the difference between Pvco2 and P'vco2 is as follows. Q (C vco2-cvco2) PB-47?p'V 'P vco2 (1) ( co t) where Q =cardiac output, C'vc o 2 = equivalent C02 content to P'0 2, Cvc o 2 = mixed venous C02 content, and PB=barometric pressure. The Fick equation for C02 is valid immediately before rebreathing. Q (Cvco2-Caco,)-Pco2 (2) where Caco2=arterial C02 content. When combining Eqs. (1) and (2) in order to eliminate Cvco2, we have the equation to estimate cardiac output: Q (C'vco2-Caco2) Vco2 Vo2 P vco2 + (C`vco2-Caco 2) ( PB-47-P'v cot) (3) Japanese Journal of Physiology

3 CARDIAC OUTPUT BY THE CO2 REBREATHING METHOD 885 The second term is the correction factor for lung-bag volume shrinkage. P'vco2 was determined by a graphical analysis in which alveolar C02 pressure change per unit time (dpaco2/dt) plotted against PAco2 during rebreathing. The P'vco2 was the value when dpaco2/dt=0 (FARHI et al., 1976; AUCHINCLOSS et al., 1980). Since dpaco2/dt linearly related to PAco2, P'vco2 was estrapolated with the linear line. If the linear line is expressed as dpaco2/dt=--a PAco2+b, P'vco2 will become b/a. Arterial C02 pressure (Paco2) was estimated from end-tidal P002 (PET0 o 2) and tidal volume (VT) before rebreathing : Paco2= PETco2-2.1 VT (JONES et al.,1979). Cac o 2 and C'VC o 2 were estimated from Pac o 2 and P vc o 2 respectively using a standard C02 dissociation curve for oxygenated blood (ROOT, 1958). These values were substituted to solve Eq. (3). Cardiac output was also determined by the acetylene method. Since inadequate mixing of the inspired gas from the rebreathing bag and lung-bag volume Fig. 1. Relationship between PA002 and CO2 elimination given by the product of the CO2 concentration in the lungs and lung-bag volume (Vsys) during rebreathing. The figure shows that the gradual rise of PAco2 corresponds to a decrease of CO2 elimination. The CO2 elimination that was calculated by measured Vsys at each breath was used to estimate Pvco2 (, r= ) and that was calculated by a constant Vsys was also used to estimate P'vcc2 (0, r=-0.980). Estimated PVC02 or P'Vco2 were obtained as the intercept of PA002 (an example from one subject with 150 W). Vol. 34, No. 5, 1984

4 886 T. YANO, P. NORSK, and F. BONDE-PETERSEN shrinkage tend to distort the acetylene concentration, these effects were corrected in detecting argon concentration. The subjects were asked to wait for at least 10 min until the next cardiac output measurement to avoid insufficient washout of acetylene previously administered (BoNDE-PETERSEN et al., 1980). The difference between PW o2 and Pvco2 was graphically estimated (Fig. 1). CO2 elimination is the CO2 volume change per unit time (dfaco2 Vsys/dt) during rebreathing. The CO2 volume change is the product of the CO2 concentration (FAco2) and lung-bag volume (Vsys). Vsys was calculated from the initial bag volume and dilution rate of argon from the initial concentration to the mixed one. The measured Vsys at each rebreathing was used for the PvC02 estimation (see DISCUSSION). When multiplying constant Vsys/(PB-47) by the linear line to estimate P'vco2, we obtain, dpa0 d o 2 Vsys _ d(fac o 2 Vsys) PA b Vsys t PB-47 dt -(-a cot ) PB-47 This equation can be expressed in the figure illustrating CO2 elimination vs. PACO2 When the calculated CO2 elimination ceases (CO2 elimination=0), PAco2 becomes b/a. Accordingly, the PAco2 at that time is the PW 02 as determined from the relationship between dpaco2/dt and PAco2. Then constant Vsys was calculated by the dilution rate of argon from the initial concentration to the second or the third respiration during rebreathing. (4) RESULTS The cardiac output determined by the acetylene method (QC2H2) ranged from 5.7 to 21.41/mm. Two subjects at 150W failed to rebreathe and repeated the rebreathing after a few minutes. Since the intervals were too short to wash out the previously inspired acetylene, two QC,H2 at 150W were excluded. There were no significant differences between QC 2 H 2 and cardiac output determined by the CO2 rebreathing method (Qco2) at rest and during exercise (Table 1). There was significant correlation between QC02 and Qc2H2 (r=0.930, p<0.001, Fig. 2). The Qco2, except for 2 resting data points, lies within 20% of the line of identity. There was significant correlation between Vo2 and Qco2. The regression line was as follows : Qco2=5.15 V (r =0.936, p<0.001). The effects of lung-bag volume shrinkage shown by the pressure differences between P'vco2 and Pvco2 are presented in Table 1. The effects were gradually increased with increased exercise intensities. The mean pressure differences ranged from 0.82 mmhg at rest to 3.06 mmhg at 150 W. Japanese Journal of Physiology

5 CARDIAC OUTPUT BY THE CO2 REBREATHING METHOD 887 Table 1 Mean values of metabolic rate and circulatory functions at rest and during exercise. Fig. 2. Comparison of cardiac output determined by the acetylene method and determined by the CO2 rebreathing method at rest and during exercise (r=0.930). The solid line represents the line of identity. The differences between P'vco2 and Paco2 intensities. The mean differences ranged from at 150W. increased with increased exercise 13.7 mmhg at rest to 40.0 mmhg DISCUSSION The measured data, cardiac output using a 'O2, V002, PETco 2, and P'vco 2 were used to estimate the standard C02 dissociation curve (ROOT, 1958). The Vol. 34, No. 5, 1984

6 888 T. YANG, P. NORSK, and F. BONDS-PETERSEN Fig. 3. Cardiac output determined by the CO2 rebreathing method plotted against 02 uptake at rest and during exercise. The regression lines were: Q=5.15 V (--.-, present study) ; Q=5.9 V (------, BEVEGARD et al., 1963); Q =5.1 V0, +4.8 (---, HERMANSEN et al., 1970). deviation of P'vco2 from Pvco2 due to the lung-bag volume shrinkage was corrected in a theoretical equation in the calculation of cardiac output. Some of the assumptions existing in the measuring procedures and in estimating values, which may include sources of error, are considered in the following discussion. Pco2 at mouth level during rebreathing represents mean PAco2. The fluctuation of PAco 2 during a respiratory cycle has been theoretically considered one reason for the difference between the mean PAco, and PETco2. The PAco, increases toward Pvco2 during expiration, and decreases with dilution by new air during inspiration (DUBOIs et al., 1952). Due to the fluctuation during the respiratory cycle, PETco2 is close to the peak PAco2. Thus, it is necessary to correct the PETco2 to obtain a mean PAco2. The mean PAco, during rebreathing is gradually increased and approaches P'vco2. At the same time, inspired C02 increases and the difference between P'Aco2 and Pvco2 becomes small. Therefore, the fluctuation during rebreathing would be minimized and the PETco2 during rebreathing would be close to the mean PAco,. From this reason, the PETco2 during rebreating was used to estimate the P'vco2. Difference between alveolar and arterial C02 pressure (A-a Pcp2 diff,). A-a Pco2 dill. during rebreathing has been tested using the Collier method, which produces a plateau in PAco, after inspiring a high C02 gas mixture (10-20 %) from the rebreathing bag. The A-a Pco2 dill. has been used as the basis of two hypotheses: the "Delayed Equilibration Hypothesis" (HILL et al., 1973; FoRSTER and CRANDALL, 1975) and the "Charged Membrane Hypothesis" (GURTNER et al., 1969). Japanese Journal of Physiology

7 CARDIAC OUTPUT BY THE CO2 REBREATHING METHOD 889 The PETCO2 before rebreathing was corrected by the regression line proposed by JONES et al. (1979). Since the regression line was determined by the direct measurements of Paco2 and PETco2, the regression line can include the correction of an A-a Pco2 duff. as well as of fluctuation. However, the PETCO2 during rebreathing was not corrected in this study, since it is not known if an A-a Pco2 duff. exists during rebreathing when there is continuous gas exchange across the lungs (Defares method), and there is no available correction for this method. If an A-a PCO2 duff. exists during rebreathing, the Qco, estimated by the Defares method will be systematically low compared to other methods. Shift of C02 dissociation curve. The venous blood through the lungs exhausts C02 into the alveoli and flows away at the same C02 pressure as in the alveoli. When rebreathing is started, C02 content in the arterial blood is increased corresponding to the rise of Pco2 in the alveoli. At the same time, the blood through the lungs is also saturated with oxygen, so that the C02 dissociation curve can be shifted (Haldane effect). This shift allows us to use the C02 dissociation curve for oxygenated blood to estimate C'VCO2 from P'VCO2. Hemoblobin concentration, ph, and other possible changes, which take place during exercise, may affect the slope of the C02 dissociation curve. MIYAMURA and HONDA (1978) have demonstrated that the dissociation curve at rest shifted downward with exercise. However, the difference between C'VCO2 and Caco2, which is needed for the calculation of cardiac output, was constant since the curve shifted without changing its slope. Therefore, even if one C02 dissociation curve for oxygenated blood is used for the calculation of cardiac output during exercise, the error would be negligible. Effect of lung-bag volume shrinkage. C02 elimination during rebreathing is calculated as the product of C02 concentration and lung-bag volume (dfaco2 Vsys/dt). When C02 elimination from blood to lung ceases during rebreathing (dfaco2 Vsys/dt=0), arterial C02 content should be increased to the identical C02 content in mixed venous blood. The PA00, at that time should be Pv0O2. However, the C02 tracing during rebreathing is still increased due to the volume shrinkage related to the 02 uptake. If PAco2 during rebreathing exceeds the Pv-co2, C02 in the lung must begin to be absorbed into the blood and PAco2 decreases. When the decrease of PAco2 with the CO2 absorption during rebreathing is temporarily counteracted by the increase of PA00, with lung-bag volume shrinkage, the PA00, during rebreathing could appear steady in C02 tracing (dpaco2/ dt=0). The PAco2 at that time should exceed the PVco2, and we defined the PAco2 as P'Vco2. Recirculation. We assumed that CVco2 was constant over the period of rebreathing. This assumption will be affected by recirculation. An average recirculation time can be calculated by dividing total blood volume by cardiac output. Our criterion of sampling time was less than half the time of an average recirculation time, e.g sec at rest and sec during exercise. The linear Vol. 34, No. 5, 1984

8 890 T. YANG, P. NORSK, and F. BONDE-PETERSEN relationship between dpa0o2/dt and PACO2 can be affected by the recirculation. The points deviating from the linear line were excluded. Evaluation of the cardiac output determined by the CO2 rebreathing method. Using the CO2 rebreathing method, it is not necessary to wait long until the next measurement, as it is with other inert gas methods. Defares method does not require any technique and is safe even at high exercise intensities. These advantages have stimulated many investigators to test the validity of the cardiac output determined by the Defares method (JERNERUS et al., 1963; KLAUSEN, 1965; LUNDIN and THOMSON, 1965; FERGUSON et al., 1968; KNOWLTON and ADAMS, 1974; HEIGENHAUSER and FAULKNER, 1978). LUNDIN and THOMSON (1965) proved the effect of the lung-bag volume shrinkage on the P'Vc02 estimation by detecting nitrogen concentration. We confirmed the effect in this study. For the calculation of cardiac output, the effect of the lung-bag volume shrinkage could roughly be assessed by the ratio of (P vco2- PvCO2)/(P'vCO2--PaCO2). Our results showed that not only the difference between P'v0O2 and PvCO2 increased but also the difference between P'vc02 and Pac02, with the ratio about the same. Therefore, if this effect is neglected, the Qco2 will be estimated 6 % lower than real value regardless of exercise intensity. LUNDIN and THOMSON (1965) have also suggested that the effect of the lungbag volume shrinkage could be corrected by applying a fast rebreathing rate (50 breaths per min). However, a recent study has suggested that the fast rebreathing rate makes cardiac output increase (BONDE-PETERSEN et al., 1980). Therefore, we used a theoretical equation to correct the effect. The QCO2 obtained in this study was in good agreement with QC,H2 When the QCO2 was plotted against V02 (Fig. 3), our results and regression line were fairly close to the regression lines determined by the dye dilution method (HERMANSEN et al., 1970) and the direct Fick method (BEVEGARD et al., 1963). It is, therefore, likely that the CO2 rebreathing method with the theoretical correction of the lung-bag volume shrinkage is an appropriate procedure to estimate cardiac output. Furthermore, since the theoretical equations used in this study do not require measurements of any insoluble gas, it is not always necessary to use expensive rapid gas analyzers such as mass spectrometers. This advantage makes possible the wide use of the Defares CO2 rebreathing method. REFERENCES AUCHINCLOSS, J. H., GILBERT, R., KUPPINGER, M., and PEPPI, D. (1980) Mixed venous CO2 tension during exercise. J. App!. Physiol.: Respir. Environ. Exercise Physiol, 48: BEVEGARD, S., HOLMGREN, A., and JONSSON, B. (1963) Circulatory studies in well trained athletes at rest and during heavy exercise, with special reference to stroke volume and the influence of body position. Acta Physiol. Scand., 57: BONDE-PETERSEN, F. (1974) A new electrically braked bicycle ergometer with low voltage power supply. Eur. J. App!. Physiol., 33: Japanese Journal of Physiology

9 CARDIAC OUTPUT BY THE CO2 REBREATHING METHOD 891 BONDE-PETERSEN, F., NoRSK, P., and SUZUKI, Y. (1980) A comparison between freon and acetylene rebreathing for measuring cardiac output. Aviat. Space Environ. Med., 51: DEFARES, J. G. (1956) A Study of the Carbon Dioxide Time Course during Rebreathing, Kemink en Zoon N.U., Utrecht. DUBOIS, A. B., BRITT, A. G., and FENN, W. 0. (1952) Alveolar CO2 during the respiratory cycle. J. App!. Physiol., 4: FARHI, L. E., NESARAJAH, M. S., OLSZOWKA, A. J., METILDI, L. A., and ELLIS, A. K. (1976) Cardiac output determination by simple one-step rebreathing technique. Respir. Physiol., 28: FERGUSON, R. J., FAULKNER, J. A., JULIUS, S., and CONWAY, J. (1968) Comparison of cardiac output determined by CO2 rebreathing and dye-dilution methods. J. App!. Physiol., 25: FoRSTER, R. E. and CRANDALL, E. D. (1975) Time course of exchanges between red cells and extracellular fluid during CO2 uptake. J. App!. Physiol., 38: GURTNER, G. H., SONG, S. H., and FARHI, L. E. (1969) Alveolar to mixed venous PC02 difference under conditions of no gas exchange. Respir. Physiol., 7: HEIGENHAUSER, G. J. F. and FAULKNER, J. A. (1978) Estimation of cardiac output by the CO2 rebreathing method during tethered swimming. J. App!. Physiol.: Respir. Environ. Exercise Physiol., 44: HERMANSEN, L., EKBLOM, B., and SALTIN, B. (1970) Cardiac output during submaximal and maximal treadmill and bicycle exercise. J. App!. Physiol., 29: HILL, E. P., POWER, G. G., and LoNGO, L. D. (1973) Mathematical simulation of pulmonary 02 and CO2 exchange. Am. J. Physiol., 224: JERNERUS, R., LUNDIN, G., and THOMSON, D. (1963) Cardiac output in healthy subjects determined with a CO2 rebreathing method. Acta Physio!. Scand., 59: JoNEs, N. L., ROBERTSON, D. G., and KANE, J. W. (1979) Difference between end-tidal and arterial PC02 in exercise. J. App!. Physiol.: Respir. Environ. Exercise Physiol., 47: KLAUSEN, K. (1965) Comparison of CO2 rebreathing and acetylene methods for cardiac output. J. App!. Physiol., 20: KNOWLTON, R. G. and ADAMS, G. E. (1974) The consistency of carbon dioxide rebreathing as a non-invasive method to determine exercise cardiac output. Ergonomics, 17: LUNDIN, G. and THOMSON, D. (1965) Determination of the mixed venous CO2 pressure with a rebreathing method. Correction for volume changes of the lung-bag system. Acta Physiol. Scand., 64: MIYAMURA, M. and HONDA, Y. (1978) CO2 dissociation curves of oxygenated whole blood obtained at rest and in exercise. Eur. J. App!. Physiol., 39: RooT, R. W. (1958) Blood CO2 dissociation as function of CO2 pressure. In: Handbook of Respiration, ed. by DITTER, D. S. and GREBE, W. B. Saunders Co., Philadelphia and London, pp. 65. Vol. 34, No. 5, 1984