003 1-399818712106-0568$02.00/0 PEDIATRIC RESEARCH Copyright 0 1987 International Pediatric Research Foundation, Inc. Vol. 21, No. 6, 1987 Printed in U.S.A. Coupling of Ventilation and COz Production during Exercise in Children DAN M. COOPER, MARTIN R. KAPLAN, LEORA BAUMGARTEN, DANIEL WEILER-RAVELL, BRIAN J. WHIPP, AND KARLMAN WASSERMAN Division of Respiratory and Critical Care, Department of Pediatrics, Harbor-UCLA Medical Center, UCLA School of Medicine, Torrance, California 90509 ABSTRACT. T h e purpose of this study was t o determine how ventilation (VE) and C 0 2 production ( V C 0 3 in response to exercise change during the growth process in children and teenagers. Dynamic gas exchange responses were measured in two types of studies: 1) 128 healthy children ranging in age from 6 to 18 yr performed progressive exercise tests ("ramp" type protocol) for measurement of,the slope of the relationship between VE and VCOZAVE/AVCOZ;and 2) the response characteristics of VEand VC02 in the transition between rest and exercise were measured in 11 teenagers and 11 younger children. Gas exchange was measured breath by breath. We found a small but significant decrease in AVE/AVCOz with increasing body weight ( r = -0.46, p <0.05), height, or age (mean slope of 27 in the youngest in 21 in the .oldest subjects). The response characteristics of VE and VCOZ (measured as the time constant of the best-fit exponential response) were longer than for VOz in both younger children and teenagers; but the time constants for VE and VC02 were each approximately 30% faster in younger children compared to teenagers. In addition, end-tidal PCO, during exercise was significantly lower in the younger subjects (mean value of 39.6 torr) compared to the teenagers (mean value of 43.5 torr). The results suggest that the process of respiratory control in exercise matures to a small degree during childhood in that PCO, may be regulated at lower levels in younger children and there may be growth-related differences in the relative amounts of COz that can be stored in tissues. (Pediatr Res 21: 568-572, 1987) Abbreviations vE,ventilation VC02, COZ output V O ~O2 , uptake AT, anaerobic threshold RCP, respiratory compensation point Babies have different respiratory control than do adults. For example, the arterial PC02 is regulated at lower levels, the pattern of breathing is erratic and marked by periodicity, and the ventilatory response to hypoxia and hyperoxia differs from the adult (1, 2). Thus, while it is apparent that the respiratory control apparatus undergoes maturation in the normal human being, Received November 3, 1986; accepted January 6, 1987. Correspondence Dan Michael Cooper, M.D., A- 15 Annex, Harbor-UCLA Medical Center, 1000 W. Carson Street, Torrance, CA 90509. Supported by a Research Grant of the California Lung Association. D.M.C. is a Clinician-Scientist of the American Heart Association, Greater Los Angeles Affiliate. there is a dearth of information on the ontogeny of these mechanisms during childhood. We hypothesized, therefore, that maturation of respiratory control could be detected and characterized during growth in children. In adults, it is known that the degree of V, is linked to the metabolic production of C 0 2 (3-8). This linkage is marked by homeostasis for C 0 2 concentration in the blood, such that PC02 is kept within a very narrow range despite large fluctuations in the VC02 as occur during exercise. But the child is faced with problems not encountered by the mature individual-while the ventilatory apparatus in both the adult and child must quickly respond to the increased C 0 2 load induced by physical activity, the child must, in addition, deal with greater C02 loads imposed by the increasing body size of the growth process itself. This research is focused on the precise linkage between VC02 and VE during exercise in growing children. This was done by first examining the dynamic vEand V C O ~ response kinetics during progressive cycle ergometer exercise in a large group of children and teenagers. Following this, in a separate experiment, detailed analysis of the response kinetics in the transition between rest and constant work rate exercise was done in a small group of children and teenagers. These types of studies utilizing breath-by-breath analysis of gas exchange and dynamic exercise protocols are particularly useful in childrena population in which more invasive methods are unfeasible. METHODS Population. All subjects were volunteers obtained through local schools and community organizations. Obese children, children with a history of chronic disease, and children not allowed to participate in normal physical education programs at school were excluded from the study. No attempt was made to select subjects who were particularly active; i.e. there was no recruitment through physical education or sports programs. All children were within the normal range for height and weight by reference tables of the National Center for Health Statistics. The subjects were predominantly of the middle socioeconomic class. Eighty-six percent of the subjects were Caucasian; the remainder consisted of Oriental, Hispanic, and black children. This project was approved by the Human Subjects Committee of Harbor-UCLA Medical Center. Informed consent was obtained from each subject and, when appropriate, from a guardian before participation. Two types of exercise protocols were used: a) a progressive exercise test for which 128 children (68 boys and 60 girls, range 6-1 8 yr old) comprised the sample, b) a constant work rate test for which 11 younger children (five boys and six girls, range 710 yr old) and 1 1 teenagers (five boys and six girls, range 15- 18 yr old) comprised the sample. These latter subjects were randomly chosen from the larger group. In addition, we analyzed the relationship between VC02 and work rate from 15 ramp COUPLING OF VE AN11 V C O ~IN CHILDREN protocols randomly chosen from children ranging in age from 6 to 8 yr and from 15 ramp protocols randomly chosen from normal adults ranging in age from 18 to 20 yr previously tested in our laboratory. Measurement of gas exchange. The subjects breathed through a low-impedance turbine volume transducer for measurement of inspiratory and expiratory volumes. Deadspace of the mouthpiece and turbine device was 90 ml. Respired PO2 and PC02 were determined by mass spectrometry from a sample drawn continuously from the mouthpiece at 1 ml/s. The electrical signals from these devices underwent analog to digital conversion for the on-line breath to breath computation of V02 (STPD), VC02 (STPD), and VE (BTPS) as previously described (3). The external deadspace ventilation was subtracted in the calculation of VE. The data from each test were displayed on line and stored on digital tape for subsequent analysis. Progressive exercise protocol. The protocol consisted of a ramp pattern of increasing work rate (9, 10) utilizing an electromagnetically braked cycle ergometer. Subjects began with a minimum of a 3 min warm-up cycling at 0 W (unloaded) work rate. The work rate was then continuously increased at a constant rate. The increase in work rate per min was selected so that the total exercise duration would be greater than 6 min and less than 14 min. The mean time for the test (not counting the warm-up) was 9 min. The children were instructed to raise a hand when they could not continue, and on this signal, the work rate was reduced to 0 W. The children were actively encouraged throughout the test. The children were instructed to maintain as constant a pedaling rate as possible between 50 and 70 rpm. A pedaling rate meter was in full view of each subject, and a servomechanism in the electronic braking system of the ergometer maintained the work rate performed to an accuracy of 1% within a range of pedaling rates of 50-90 rpm. The progressive exercise tests were used to measure the AT. The AT indicates the point during exercise at which lactate concentrations begin to increase in the blood. As C 0 2is liberated by the bicarbonate buffering of the lactic acid, VE and VCO, increase out of proportion to the increase in V02, thereby allowing the noninvasive determination of the the AT (1 I). In each subject, the AT was measured by finding the V 0 2 above which VE/VO~ and increased without an increase in VE/VC02 or a decrease in PETC02, as previously described (9). Constant work rate exercise protocol. The work rate for the constant work rate protocols was taken as 75% of the work rate corresponding to the subject's AT. The mean work rate was 64 W for the teenagers and 22 W for the younger children. When normalized to body weight, the rest to steady-state exercise 0 Boys Girls I I I 20 I I I 40 I I I 60 I I I 80 I WEIGHT ( k g ) Fig. 2. Slope of the \jE-\jCOZrelationship as a function of body weight in 128 normal subjects ranging in age from 6 to 18 yr. Boys are represented by open circles;girls by closed circles. There was a small but significant negative correlation between the slope and body weight (see text). increase in \jo2 did not differ significantly between the younger children (mean, 16 ml O2 min-I kg-') and the teenagers (mean, 13 ml O2 min-I kg-'). In eight of the teenagers, constant work rate tests were also performed to study gas exchange dynamics at 20 W to match the work rate performed by the younger children. Each subject performed a minimum of six rest-to-constant work rate fransifions. Exercise periods were 6 min each, and heart rate, V02, VE,and VC02 returned to the preexercise resting values before a repetition was performed. The subjects began exercise with the activation (at the end of an exhalation) of a green light signal; there was no voice command. To avoid the expenditure of energy required to overcome the inertia of the fly-wheel at the start of exercise, the ergometer fly-wheel was motorized and maintained at a rate of 60 rpm until the onset of pedaling. When the subject starts to pedal, the motor maintaining the fly-wheel is turned off. Data analysis.. 1) Progressive exercise protocol-a graphical presentation of VE (liter, BTPS min-I) as a function of VC02 (liter, STPD min-') (Fig. 1) was used to determine the point at which VE increased out of proportion to VC02-the RCP. This occurs at higher work rates than the AT at which point VEis still coupled to VC02. Linear regression techniques were then used to find the best-fit line from the data starting at 60 s after the onset of exercise to the RCP, and the slope (AVE/AVC02) of the best-fit line was calculated for each subject. The slopes were then plotted as a function of body weight (Fig. 2), height, and age, and the linear regression and correlation coefficients were determined by standard techniques. The data were also analyzed using polynomial, logarithmic, and exponential fitting models. In 15 younge! children and in 15 young adults, the relationship between VC02and work rate during ramp protocols was quantified. This was done by using linear regression techniques to find the 0 slope (AVC02/AWR) of the best-fit line below the subjects' AT. 0.25 0.50 0.75 1 .O 1.25 2) Constant work rate protocol-VC02 and VE kinetics [folV C O ~(L. STPD. m ~ n - ' ) lowing the cardiodynamic phase (4, 12)] were studied at a work Fig. 1. Breath by breath measurement of vEas a function of VCOZ rate which was 75% of the subject's AT. The time constant (7) during a progressive exercise protocol (ramp test) in an 8-yr-old child. of the response (Fig. 3) was determined from the following Solid line indicates the best-fit line from I-min after the onset of the equation (4, 12): 5 5 ramp to the RCP (see text) as indicated. The slope of the line is given as AVE~AVCOZ. A V ( ~=) AVSS x (1 -e-'IT) 5 70 COOPER ET AL. / ' -60 Z 0 YOUNG CHILDREN 60 120 180 240 -60 0 TIME (sec) 60 120 180 240 Fig. 3. Kinetics of VOZ,vCOZ,and V, in children and teenagers. The response of v02(solid line), VC02 (dashed line), and VE (dotted line) are shown. Each line represents the averaged, superimposed data of either the 11 younger children or the 11 teenagers. Time 0 on the x-axis indicates the onset of constant work rate exercise. In order to compare the gas exchange responses of the different sized groups, the y-axis represents the gas exchange response as a percent of the rest to steadystate exercise difference. The responses of V C O ~and VE were slower than for vozin both younger children and the teenagers; however, the responses of VE and vcozwere somewhat faster in the younger children compared to the teenagers. above the prior where AV(~)is the increase in either V, or V C O ~ control values at any exercise time (t); AVss is the difference between rest and steady-state exercise V; and 7 is the tjme constant or the time to reach 63% [(l - l/e) x loo%] of AVss. The fitting window was between 20 to 120 s after the onset of exercise, and a best-fit exponential (characterized by 7)was found using iterative techniques (4). The model calculated a delay representing the difference in time between the onset of exercise and the extrapolated onset of the best-fit exponential. The mean end-tidal PC02 during the steady-state exercise was measured in each subject. 3) Statistical analysis-independent t tests were used to compare the time constants for VE and VC02 in the teenagers with those of the younger children. In the teenagers, dependent t tests were used to compare the time constants for V, and VC02 at the higher work rate with those obtained at the lower work rate. Statistical significance was taken at the p <0.05 level. RESULTS Progressive exercise protocol. Linear regression analysis demonstrated that the slope of the V,-VC02 relationship decreased to a small but significant degree with increasing body weight (Fig. 2). The regression line showed that the average values were 27 for the smallest subjects and 21 for the largest subjects. The data were also analyzed using polynomial and exponential regressions with no improvement in correlation. .Similar linear regression correlations were observed for AVE/AVC02 as a function of age and body weight. There were no significant differences in these values between the boys and girls of the study. The linear regression equations (Y is the slope, AV,/AVC02), SE estimates (Sy.x),and r were as follows: 1) for weight (kg) Y = -0.099 x weight + 28.7, SY.X= 3.26, r = -0.46, p <0.05 2) for height (cm) Y = -0.096 x height + 38.6, SY.X= 3.17, r = -0.50, p <0.05 and 3) for age (yr) There was no statistical difference between AVCO~/AWRin the 15 young adults [mean value 1 1.7 + 2.5 (SD) ml C 0 2 min-' W-'1 compared to the younger children (12.5 f 2.3 ml COz min-' W-I). Constant work rate exercise protocol (Table 1). The time constants for VC02 were significantly shorter in the 11 younger children compared to the 11 teenagers. Similarly, time constants for V, were significantly shorter in the 11 younger children compared to the 11 teenagers. The mean end-tidal PC02 during exercise was significantly lower in the I I younger children compared to the 1 1 teenagers. There were no significant differences in the steady-state respiratory exchange ratios during rest or exercise between the younger children and teenagers (mean resting r in younger children was 0.92 and in the teenagers, 0.94; mean exercise r in younger children was 0.92 and in the teenagers, 0.94). The difference in time constants was not work rate dependent since the time constants for VC02 in the teenagers did not differ between the higher work rate [mean 52 + 6 s (SEM)] and lower work rate [mean 54 ? 7 (s) (SEM)]. Similarly, the time constants for VE in the teenagers at the higher work rate [mean 56 2 s (SEM)] did not differ significantly from the lower work rate [mean 64 + 8 s (SEM)]. + DISCUSSION The slope of the V ~ - V C O relationship ~ during exercise was to a small but significant degree greater in younger children compared to teenagers (Fig. 2), and the time constants for VE and VC02 were faster in younger children compared to teenagers (Table 1). These were somewhat surprising results for two reasons. First, we have previously found other dynamic relationships of progressive exercise (e.g. work efficiency, derived from the relationship between AVO2 and Awork rate) to be independent of age or body size (19). Second, we had also observed that the time constant for oxygen uptake (7V02) was independent of age or body size in children (13).. The relationship between VE and V C O ~is given by a modification of the alveolar gas equation: where vEis ventilation, VCO, is C 0 2 production, PaC02 is arterial C 0 2 tension, and VD/VT is deadspace to tidal volume ratio. As. can be. seen, the factors affecting the relationship between VE and VC02 will be the PaC02 and the V,/VT. The equation implies that as the regulated level of PC02 decreases, more ventilation will be required for a given increase in VC02. We estimated the PaC02 with measurements of PETCO2 during steady-state exercise (direct measurements of arterial PC02 during exercise are not available in children), and lower values of end-tidal PC02 were found in the younger children (Table I). Steeper slopes in the younger children may be a manifestation of a lower COz set-point. Since the C 0 2 produced per increase in work rate is independent of age, our data suggest that younger children breathe more for a given increase in metabolic demand than do teenagers. Relatively larger deadspace ventilation (both the external and physiologic deadspace) could also account for increased slopes in the younger subjects, but previous studies suggest that this is not a major factor. First, as noted in the "Methods" section, the external deadspace ventilation is subtracted from the total ventilation in our calculation of V,. Moreover, it has been demonstrated that when the external deadspace ventilation is subtracted, the relationship between V, and VC02 is virtually unaffected even when the size of the external deadspace is varied experimentally (14). Finally, previous work in children using a variety of indirect measurements indicate that VD/VTduring exercise is constant throughout childhood (15- 17). Recently, Brischetto et al. (18) found that AV,/AVCO~was larger in older adults (aged 67-79 yr) than in younger adults (2237 yr). The mean value of the slope in the older individuals was similar to that obtained in the youngest children in our study. The difference in the ventilatory response was attributed to COUPLING OF Table 1. Age, wt, time constants for VE a, vCO,, and vE,preexercise and mean exercise VO~,and mean exercise end-tidal C02 in I I younger children and I I teenagers Preexercise Gender F F Age (yr) 10.1 9.2 8.4 7.7 8.2 9.5 8.3 8.2 7.8 9.8 7.7 F F F F M M M M M Younger children Mean SD - 8.6* 0.9 57 1 AND vCO~IN CHILDREN Wt (kg) 7 ~ 0 2 7~coz TVE ~ 0 2 Mean exercise ~ 0 2 Mean exercise pETc02 (mm Hg) (s) (s) (literlmin) (literlmin) 30.5 26.0 27.2 23.3 26.4 33.4 31.8 28.6 26.1 31.7 27.7 29.2 25.1 25.1 21.8 31.9 24.7 23.2 30.4 23.7 26.9 28.5 44.2 41.7 48.8 32.3 47.9 39.4 23.1 47.6 37.7 31.7 44.0 45.8 46.8 52.8 31.1 55.0 25.9 18.3 56.0 43.9 32.0 46.6 0.25 0.21 0.25 0.24 0.23 0.2 1 0.19 0.26 0.11 0.25 0.20 0.85 0.62 0.85 0.60 0.5 1 0.5 1 0.63 0.64 0.77 0.85 0.63 41 37 39 40 43 40 41 37 39 39 40 28.4* 3.1 26.4 3.2 39.9* 8.1 41.3* 12.5 0.22* 0.04 0.68* 0.13 39.6* 1.7 28.3 5.6 50.4 7.2 52.4 9.2 0.32 0.06 1.20 0.21 43.5 2.5 Teenaeers 17.5 65.9 Mean 1.O 10.8 SD * Significantly differed from teenagers. (s) increased VD/VTwhich is known to occur in elderly subjects as the steady-state values for R during rest and exercise were the (19). The results of our study, combined with those of Brischetto same in children and teenagers suggesting the same patterns of et al. (18), suggest that the AVE/AVC02 is at its minimum substrate utilization. A more likely explanation is that greater amounts of C 0 2 are beginning late in adolescence and extending through middle age. This may represent a period of ventilatory control and function stored in the teenagers compared to the younger children. A major site for rapid C 0 2 storage is the blood (21), and teenagers when C 0 2 excretion in man is most efficient. When the work rate input is below the subject's AT, the gas have generally higher levels of hemoglobin than do younger exchange dynamics following the onset of exercise are normally children (22). In addition, it is not known whether CO2 storage well described by a first-order exponential (after the early cardi- capacity of tissues such as muscle and fat are different in younger odynamic phase) (4, 20). The response characteristics become children compared to teenagers. In the presence of relatively more complex for work rates above the AT (I I), and the present smaller C01 stores, C02 originating in the exercising muscle cells study was designed so that that the work rate input was below may saturate the tissue stores more quickly to the new venous the AT for both the teenagers and young children. In adults, PC02 value and reach the central and pulmonary circulation kinetics for V02 are known to be faster than for VC02 (4, 1 I), faster as well, thereby accounting for the more rapid VE kinetics reflecting the much larger storage capacity for C 0 2 relative to O2 in the younger children. Previous studies from this laboratory (21). Similarly, we found that VCOz and VEkinetics were slower have demonstrated that ventilatory responses can be modified by changing the amount of C02 stored (e.g. by hyperventilation) than V02 kinetics in both the teenagers and younger children. In a previous study (13), we found that for the same rest-to- (23). The results of this study suggest that the complex process of exercise increase in metabolic demand (AVOJkg), younger children had the same time. constant of V02 as did teenagers. In respiratory control does undergo gradual change during childcontrast, the analysis of VCOz and VE kinetics show them to be hood. Differences in body stores for C 0 2 may explain the obmore rapid in younger children than in teenagers (Table I). The served differences in VE and VC02 kinetics, and, for reasons as differences in time coptants are unlikely to be attributable to yet unclear, younger children appear to regulate C02 at lower nonlinearities of the VE and VC02 response since these values levels. However, the magnitude of the observed differences bewere the same at two different work rates in the teenage subjects. tween the youngest and oldest subjects was small (AVE/AVCOZ The slower responses in the teenagers means that less C 0 2 (per decreased by 19%; 7VC02 increased by 3 1%) relatively to the 4kg body weight) was exhaled into the atmosphere in the transition fold increase in body mass suggesting that the respiratory control from rest to exercise. What is the fate of this "unaccounted for" mechanisms are close to maturity early in life. It seems, then, C02? One possibility is that in exercise transitions, younger that the coupling of cellular C 0 2 production to ventilation is children actually produce less COz per O2 consumed at the well regulated throughout childhood, and may be viewed as one cellular level than do teenagers. However, this is highly unlikely of the homeostatic mechanisms of the growth process itself. 572 COOPER ET AL. REFERENCES I . Southall DP, Richards JM, Shinebourne JA, Franks CI, Wilson JA, Alexander JR 1983 Prospective population-based studies into heart rate and breathing patterns in newborn infants: prediction of infants at risk of SIDS. In: Tildon JT, Roeder LM, Steinschneider A (eds) Sudden Infant Death Syndrome. Academic Press, New York, pp 621-652 2. Rigatto H, Brady JP, Verduzco R 1975 Chemoreceptor reflexes in preterm infants: I. The effect of gestational and postnatal age on the ventilatory response to inhalation of 100% and 15% oxygen. Pediatrics 55:604-613. 3. Beaver WL, Lamarra N, Wasserman K 1981 Breath-by-breath measurement of true alveolar gas exchange. J Appl Physiol 51:1662-1675 4. Lamarra N 1982 Ventilatory Control, Cardiac Outut, and Gas Exchange Dynamics During Exercise Transients in Man (PhD Thesis). UCLA, Los Angeles 5. Phillipson EA, Bowes G, Townsend ER, Duffin J, Cooper J D 1981 Role of metabolic C 0 2 production in ventilatory response to steady-state exercise. J Clin Invest 68:768-774 6. Whipp BJ 1981 The control of exercise hyperpnea. In: Hornbein T (ed) The Regulation of Breathing. Marcel Dekker, New York, pp 1069-1 139 7. Whipp BJ 1983 Ventilatory control during exercise in humans. Ann Rev Physiol 45:393-413 8. Wasserman DH, Whipp BJ 1983 Coupling of ventilation to pulmonary gas exchange during nonsteady-state work in men. J Appl Physiol 54587-593 9. Cooper DM, Weiler-Ravell D, Whipp BJ, Wasserman K. Aerobic parameters of exercise as a function of body size during growth in children. J Appl Physiol 56:628-634 10. Whipp BJ, Davis JA, Torres F, Wasserman K 1981 A test to determine parameters of aerobic function during exercise. J Appl Physiol 50:2 17-221 11. Wasserman K, Whipp BJ, Davis JA 198 1 Respiratory physiology of exercise: 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. metabolism, gas exchange, and ventilatory control. Int Rev Physiol 23:15021 1 Casaburi R, Weissman ML, Huntsman DJ, Whipp BJ, Wasserman K 1979 Determinants of gas exchange kinetics during exercise in the dog. J Appl Physiol46: 1054-1060 Cooper DM, Beny C, Lamarra N, Wasserman K 1985 Kinetics of oxygen uptake and heart rate at onset of exercise in children. J Appl Physiol59:21 I217 Ward SA, Whipp BJ 1980 Ventilatory control during exercise with increased external dead space. J Appl Physiol48:225-23 1 Anderson SD, Godfey S 1971 Cardio-respiratory response to treadmill exercise in normal children. Clin Sci 40:433-442 Godfrey S 1974 The growth and development of the cardio-pulmonary responses to exercise. In: Davis JA, Dobbing J (eds) Scientific Foundations of Paediatrics. WB Saunders, Philadelphia, pp 271-280 Godfrey S, Davies CTM, Wozniak E, Barnes CA 1971 Cardio-respiratory response to exercise in normal children. Clin Sci 40:4 19-43 1 Brishetto MJ, Millman RP, Peterson DD, Silage DA, Pack A1 1984 Effect of aging on ventilatory response to exercise and C02. J Appl Physiol 56: 11431150 Hams EA, Thompson JG 1958 The pulmonary ventilation and heart rate during exercise in healthy old age. Clin Sci 17:349-359 Wasserman K, Whipp BJ, Castagna J 1974 Cardiodynamic hyperpnea: hyperpnea secondary to cardiac output increase. J Appl Physiol 36:457-464 Chemick NS, Longobardo GS 1970 Oxygen and carbon dioxide gas stores of the body. Physiol Rev 50:196-243 Pearson HA 1983 Diseases of the blood. In: Behrman RE, Vaughn EC (eds) Textbook of Pediatrics. WB Saunders, Philadelphia, pp 1204-1257 Ward SA, Whipp BJ, Koyal S, Wasserman K 1983 Influence of body C 0 2 stores on ventilatory dynamics during exercise. J Appl Physiol 55:742-749 Erratum An error has been found in the recently published article by Lee Frank titled "Oxygen Toxicity in Neonatal Rats: the Effect of Endotoxin Treatment on Survival during and post-02 Exposure (Pediatr Res 2 1:109- 115, 1987). On page 1 10, top of column 2, the endotoxin dosage used for the electron micrographic study should read: neonatal rats treated with saline or endotoxin in progressive doses of 10 ~ g / k gat 0 time, 50 pglkg at 24 h, 250 wg/kg at 72 h, and 1.25 mg/kg at 72 h of O2 exposure instead of "neonatal rats treated in the same manner as described above with saline or endotoxin. . . ." The results obtained in the survival and light morphometric studies were similar with either endotoxin regimen (20 pg/kg at 0 time and 40 pg/kg at 24 h of O2 exposure, or the 4 progressive dose regimen).