0031-3998/95/3806-0926$03.00/0 PEDIATRIC RESEARCH Vol. 38, No. 6, 1995 Copyright O 1995 International Pediatric Research Foundation, Inc Printed in U.S.A.

Effect of Theophylline on Metabolic and Ventilatory Response to Hypoxemia during Quiet Sleep in Piglets

Division of Respiratory Medicine and The Jeremy Rill Centre for SIDS and Respiratory Control Disorders, Department of Pediatrics, Montreal Children's Hospital-Research Institute, McGill University, Montrial, Canada

Theophylline, a drug frequently used in newborns, stimulates significant increase in respiratory exchange ratio (R)in response respiration and increases the metabolic rate in a sustained fash- to hypoxemia (from 0.87 -C 0.05 to 0.97 2 0.04, p < 0.001). ion; hypoxemia, on the other hand, decreases metabolic rate and However, after the administration of theophylline, additional increases ventilation slightly and, at times, only transiently. This exposure to hypoxemia did not result in a change in R. In study looked at the effect of theophylline on the metabolic and summary, our results show that, in sleeping newborn piglets, ventilatory response to hypoxemia in piglets. We studied two theophylline does not abolish the decrease in oxygen consump- groups of piglets during normoxia and hypoxemia: first during a tion observed in response to hypoxemia; nor does it enhance the baseline period; and second, after the infusion of either theophyl- ventilatory response to a moderate degree of hypoxemia. line or a placebo. All studies were done in quiet sleep, 2 d after (Pediatr Res 38: 926-931, 1995) instrumentation was performed to place vascular catheters and electroencephalographic electrodes. 0, consumption (Vo,) and Abbreviations CO, production (Vco,) were measured in a metabolic chamber, Fio,, fractional concentration of inspired oxygen and alveolar ventilation (VA)was then derived from Vco, and Fico,, fractional concentration of inspired CO, Paco,. We found that theophylline did not abolish the small Paco,, partial pressure of arterial CO, decrease in oxygen consumption brought about by hypoxemia. PAco,, partial pressure of alveolar CO, Nor did theophylline augment the ventilatory response to hypox- VO,, oxygen consumption emia. In fact, the percent change in alveolar ventilation decreased ~co,,CO, production slightly: going from 17 2 8% during the baseline period to 9 i- VA, alveolar ventilation 6% (p < 0.005) after theophylline administration. We found a EEG, electroencephalograph

Compared with adults, newborns have high demands for newborn piglets in which Paco, was measured along with oxygen. They also, however, are prone to diseases in which minute ventilation (4), the use of theophylline, although stim- hypoxemia occurs. These include pulmonary involvement sec- ulating the ventilatory response to hypoxemia, did not yield the ondary mostly to prematurity but also infections and apnea of expected diminution of Paco,. Therefore, even though meta- infancy. These conditions are often treated with theophylline, a bolic rate was not measured, the results suggest that the respiratory stimulant (1, 2). metabolic rate (C02 production) had increased, thereby in- The effect of theophylline on the response to hypoxemia in creasing ventilation after theophylline administration. newborns has been studied largely from the point of view of Clearly, it is important to know the metabolic rate when the ventilation. The studies, however, have shown conflicting re- effect of theophylline during hypoxemia is being assessed in sults: both a stimulation of the ventilatory response to hypox- newborns, for at that age theophylline and hypoxemia have emia (3, 4) and no effect at all (5). In fact, in one study of different effects on ventilation and metabolic rate. Indeed, although hypoxemia increases ventilation slightly in new- Received September 21, 1994; accepted June 20, 1995. born-sometimes only transiently-(reviewed in Refs. 6 and Correspondence and reprints requests: Aurore CBtt, Division of Respiratory Medicine 7) and also depresses metabolic rate (8, 9), theophylline in- (D-380). Montreal Children's Hosp~tal,2300 rue Tupper, Montrtal, Canada H3H 1P3. Supported.- by the Association Pulmonaire du Qutbec and, in part,. by. the Medical Creases both ventilation and metabolic rate in a sustained Research Council of Canada (Grant MA-10258) and the Heart and Stroke Foundation of fashion (10, 11). The decrease in metabolic rate in response to Canada. hypoxemia is viewed as an adaptation of the newborn to low ' Visiting physician from the University of Cali, Colombia. Research Scholar of the Fonds de la recherche en santt du Qutbec. Oxygen Content (9, 12, 13). Nonetheless, if theophylline abol- METABOLIC EFFECT OF THE(3PHYLLINE IN HYPOXEMIA 927 ishes the decrease in metabolic rate that occurs in response to We measured total body 0, consumption and CO, produc- hypoxemia, the consequences could be deleterious for those tion by continuously recording the drop in Fio, and the in- hypoxemic newborns so often given theophylline. So far, no crease in Fico, in the animals' metabolic chamber. An electro- study has explored the interaction of theophylline and hypox- chemical cell oxygen analyzer (model S-3M1) and sensor emia with metabolic rate and ventilation in newborns. (model N-22 M; AMETEK, Applied Electrochemistry, Pitts- The purpose of the present study was to determine whether, burgh, PA) was used to measure the Fio, (dry). An infrared in newborn piglets, theophylline modifies the metabolic re- CO, analyzer (model CD-3A) and sensor (model P-61B; sponse to hypoxemia. We elected to measure metabolic and AMETEK, Applied Electrochemistry) was used to measure the ventilatory parameters during quiet sleep to avoid any stimu- Fico,. This equipment is sensitive to a +0.01% change in Fio, lation of respiration caused by the general stimulation of or Fico, and has an accuracy of +0.02% and a 90% response activity, secondary to theophylline, during the awake state. We time of 100 ms. We calibrated these instruments with three have shown in the past that, after administration of theophyl- different gas mixtures of 0, and CO,, the range of which line, quiet sleep is present in sufficient amounts to make such corresponded to the range of values used during the studies. To study feasible, but that active sleep is greatly diminished (10). ensure that all measurements of Vo, and Vco, were accurate, we regularly tested our system's ability to measure a small, METHODS precisely known change in 0, and CO, levels, both in room air and with a hypoxic mixture similar to that used for the study. Animal preparation. A total of 16 piglets were studied at a For that purpose, we added 0, (or CO,) to the chamber, mean age of 7 f 1 d (mean f SD) and a mean weight of 2.7 through a precisely calibrated flowmeter, at a rate of 20 and 30 + 0.5 kg. The animals were obtained a few days before the mL1min for a known period of time (usually between 5 and 10 experiments and kept in our animal quarters. We provided min). This flow rate represents a value for oxygen consumption them with sow milk replacer (Wetnurse; JEFF0 Co. St- (or CO, production) of 10-15 mL/kg/min in our piglets. We Hyacinthe, QuCbec, Canada) ad libitum throughout the day and were able to measure the expected changes exactly within night and monitored their weight gains closely. Only those f0.01% of Fio, or Fico,. In the past, we obtained very animals that had a normal weight gain (=I00 glday) were used reproducible Vo, and vco, results using that method (10). for the experiments. The environmental temperature in the Measurements were made over a period of approximately 15 animal quarters was kept between 26 and 30°C; the sleeping min (sufficient time to measure 0, consumption and CO, area was provided with sufficient additional heat to maintain a production), during which time neither Fio, nor Fico, changed thermoneutral environment (32-33°C) (14). by more than 0.3 (gas circulation system closed). The mea- We performed surgery for catheter insertion 2-3 d before the surements were made when the piglets were in quiet sleep, experiments, using halothane/N,O/O, anesthesia. One catheter starting at least 30 s after that state had been clearly identified was placed in the axillary vein for theophylline or placebo by EEG and behavioral criteria. The time delay of 30 s was infusion; a second cathether was placed in the descending aorta chosen because, with our system, it represents three times the from a femoral artery for blood gas sampling and monitoring of delay to see a change in the Fio, or Fico,. We calculated 0, blood pressure and heart rate (polyvinyl, TYGON microbore consumption as follows: tubing; Norton Co., Wayne, NJ). We also instrumented the animals with s.c. electrodes (platinum subdermal type E2; Vo, (mLO, . minp' . kgp') Grass Instruments, Quincy, MA) for EEG recording. Once the anesthesia was stopped, the animals woke rather rapidly (with- A Fio, (%/min) X [vol of chamber (mL) - vol of animal] - in 15-20 min); a few hours after surgery, they seemed to weight of animal (kg) behave, walk, and feed normally. Measurement of total body 0, consumption and CO, pro- We calculated CO, production (Vco,) in much the same way, duction, and alveolar ventilation. We used the closed-system replacing AFio, in the equation by AFico,. We derived alveo- method for the measurement of these parameters. The animals larventilation from the value of CO, production and Paco, as were studied in a totally sealed plexiglass chamber (44.6 X 45 follows: X 60 cm; volume, 120.4 L). We kept the chamber's tempera- ture in the thermoneutral range for piglets; that is, between 32 CO, production Alveolar ventilation = X K and 33°C with the ambient temperature adjusted to fO.l°C PAco, (Telethermometer; Yellow Spring Instruments, Yellow where K is a constant used as a correcting factor to convert Springs, OH). We used a gas circulation system, so the CO, Vco, (and Vo,) to standard temperature and pressure dry. We and 0, levels in the chamber could be adjusted between the used Paco, to estimate PAco,. Blood gases and pH were periods of metabolic rate measurement. By circulating the analyzed with an ABL30 blood gas analyzer (Radiometer ambient air of the chamber through a container of CaSO, Copenhagen, Copenhagen, Denmark), and values were cor- (Drierite), we maintained the H,O vapor at approximately 20 rected for the body temperature of the animals. f 5% of relative humidity. During the recording periods, care was taken to maintain a constant level of ambient temperature Body temperature. Throughout the studies, we recorded and humidity in the chamber, thereby restricting the variables rectal temperature in each animal. For that purpose, a flexible that might cause Fio, changes. rectal probe (electronic thermometer; Physitemp, Clifton, NJ) 928 PORRAS ET AL. was inserted 3-4 cm deep. It was fixed carefully to the tail, line, diluted to a volume of 15 rnL with a normal saline before being extended with catheters and electrodes outside the solution (0.9%) and given over 30 min, would give us a stable chamber. The animals adapted to the probe immediately and blood level of approximately 40 pM/L (7.3 mg/L) for the next appeared unaffected by it throughout the study. 3 h (10). We did not aim at higher blood levels of theophylline Hypoxemia induction. We induced hypoxemia by adding a because we had previously noted that, at higher levels, our mixture of nitrogen and oxygen to the chamber, thereby pro- piglets scarcely slept. gressively decreasing the Fio, so as to obtain a Pao, between Data and statistical analysis. We considered as valid, each 45 and 50 mm Hg (11-12% Fio,). This induction was done quiet sleep episode that lasted longer than five minutes (ex- over a period of 15 min. In preliminary experiments, we had cluding the first 30 s). For calculation of VO, and Vco,, a mean noted that a more rapid induction of hypoxemia (less than 5 value was obtained for each episode of quiet sleep (range of min) usually agitated the animals, preventing appropriate sleep two to four episodes) achieved during the baseline period and for several minutes thereafter. We established the level of Pao, after the drug administration (postinfusion period), during both at no lower than 45 mm Hg, inasmuch as the piglets' length of normoxia and hypoxemia. For heart rate, respiratory rate, and time asleep diminished significantly below that level and agi- blood pressure, we systematically analyzed the 2nd and 3rd tation emerged in the theophylline group. min of each episode of quiet sleep. Determination of behavioral states. We used both EEG and Before analyzing the effect of theophylline on the response behavioral criteria to determine each piglet's state of con- to hypoxemia, we did two statistical tests. First, we examined sciousness (10, 15). We studied quiet sleep only. Quiet sleep whether the animals had returned to baseline values at the end was determined according to the following criteria: eyes of their 2-3-h rest period, that is, before the start of the infusion closed, no movement, regular respiratory and heart rates, and of drug or placebo (paired t test). No difference was found. high amplitude slow waves on the EEG. Second, to evaluate whether time affects the various parameters Experimental protocol. We separated the animals into two during hypoxemia, we performed a single factor analysis of groups: placebo (n = 6) and theophylline (n = 10). The variance on the results of each hypoxemia exposure. Because placebo group was used to establish whether a repeat exposure no significant effect of time was found, we then averaged all to hypoxemia in the same piglets would result in similar the values obtained during hypoxemia for each variable-in metabolic and cardiorespiratory responses. During all experi- each group and for each exposure. At that point, we felt ments, the animals were permitted to move freely and to stay justified in comparing the different experimental conditions. awake or fall asleep spontaneously. Next, we performed a two-way analysis of variance across The baseline study started at about 1000 h, usually lasting conditions to determine whether the second exposure to hy- about 3 h. After an initial 45-min recording, during which three poxemia in the placebo group differed from the first. We measurements of metabolic and cardiorespiratory parameters performed the same type of analysis for the theophylline group, were made, we began the induction of hypoxemia. After a and we used the Newman-Keuls test to evaluate differences stabilization period of approximately 15 min, we started the between specific means. To evaluate whether theophylline had recordings again, repeating the measurement of metabolic and an effect on the ventilatory response to hypoxemia, we did a cardiorespiratory parameters. The total duration of the hypoxic paired t test analysis on the percent difference between nor- exposure (including the induction time) lasted 90 min. The moxia and hypoxemia from the two exposures (baseline and animals were then returned to their quarters and allowed to eat postinfusion) for each variable. Differences between mean and recuperate for at least 2.5 h. values were considered significant when p < 0.05. The second part of the study started with another 45 min of baseline measurements. We then administered the theophylline preparation, or the equivalent amount of normal saline as a RESULTS placebo, infusing continuously over 30 min. We did three measurements in the following 45 min and then induced Placebo group. For the six piglets that received the placebo, hypoxemia again and repeated the measurements. no significant difference emerged from a comparison of the During all studies, we recorded respiratory rate with respi- responses to the first and second exposure to hypoxemia (Fig. ratory inductive plethysmography (RESPITRACE) and blood 1). As well, the placebo group's response to hypoxemia did not pressure with a pressure transducer (model P23ID; Gould, differ from that of the theophylline group's baseline period. We Oxnard, CA). All data were recorded on paper (Grass model therefore felt justified in comparing the theophylline group's 7E Polygraph; Grass Instrument, Quincy, MA) for further response to hypoxemia during the baseline period with that analysis. Heart rate was recorded continuously by a cardiota- recorded after the administration of theophylline. chometer (model 7P4H, Grass Instruments) triggered from the Theophylline group. The mean serum levels of theophylline arterial pressure pulse. This protocol was approved by the were as follows: 42 -t 4 pMIL (7.6 pg1mL) after the end of the Animal Care Committee at our institution. infusion; and 38 + 5 pM/L (6.9 pg/mL) at the end of the Determination of blood levels of theophylline. We deter- study, approximately 2.5 h later. mined theophylline blood levels by the Tdx Theophylline I1 Effect of theophylline. Both VA and Vo, increased signifi- assay (an immunoassay). Blood samples were obtained twice: cantly after the administration of theophylline, whereas Paco, immediately after the drug infusion and at the end of the study. decreased significantly. The other variables did not change In the past, we determined that a 9 mglkg dose of aminophyl- significantly, and there was no change in body temperature. METABOLIC EFFECT OF THEOPHYLLINE IN HYPOXEMIA 929

-a Alveolar Vent~lat~on Respiratory exchange ratio Alveolar Ventilat~on Respiratory exchange ratio 30 3 30 B !!

5 3 1 g 10 m ;..I.c I; .I 0 0 s. 0 s. 0 1st exposure 2nd exposure O 1st exposure 2nd exposure Baselme Post-lnfuslon O Baselme Post-~nfus~on Figure 2. Percent change with hypoxemia of both alveolar ventilation and m PaCO, Oxygen consumption respiratory exchange ratio (results expressed as mean i SD). Note that in the baseline period, for both alveolar ventilation and respiratory exchange ratio, the percent change with hypoxemia was significantly higher than that after theophylline administration. *p < 0.005, paired t test between baseline and postinfusion.

DISCUSSION G -30-30u 1st exposure 2nd exposure 1st exposure 2nd exposure The major finding of this study is that, in our sleeping Figure 1. Comparison of the percent change with hypoxemia for the two piglets, the use of theophylline modified neither the metabolic i exposures in the placebo group (results expressed as mean SD). Note that nor ventilatory response to hypoxemia. These results differ there was no significant difference for any of the variables. somewhat from those of other investigators. We will therefore consider the methodologic differences between the studies Effect of hypoxemia. Hypoxemia, in the theophylline group, before discussing possible explanations of our results. increased the VA, respiratory exchange ratio (R), respiratory Methodologic differences. First, unlike other studies of new- rate, and heart rate (see Table 1 for cardiorespiratory and born animals exposed to theophylline and hypoxemia, we used metabolic variables), whereas both Paco, and Vo, decreased. nonanesthetized animals. Because most anesthetic agents have Neither VCO, nor the systolic and diastolic blood pressures been shown to depress both ventilation and the ventilatory were affected. There was also no effect of hypoxemia on body response to hypoxemia (reviewed in Ref. 16), we thought that, temperature. in the anesthetized animal studies, theophylline might have For VA and R, however, the response to hypoxemia differed reversed the depression of ventilation brought about by anes- before and after the administration of theophylline (Fig. 2). In thesia, thereby modifying the ventilatory response to hypox- all 10 animals studied, the percent change in VA with hypox- emia. Second, our piglets had intact upper airways and we did emia was lower after the administration of theophylline than not use any respiratory apparatus. Note that, although such during the baseline period (8.9 +- 6.1% versus 17.0 +- 8.1%, p apparatuses might not modify the magnitude of the ventilatory < 0.005). Also, whereas R increased significantly with hypox- response, they likely modify the pattern of breathing and emia in the baseline part of the study, no significant change in perhaps the dead space ventilation as well. Finally, we exposed R occurred with hypoxemia after the administration of theo- our animals to a moderate degree of hypoxemia for approxi- phylline. The changes in VO, and VCO, during hypoxemia were mately 2 h. In addition, we induced alveolar hypoxia over a not significantly different before and after the administration of period of 15 min and studied the animals after a stabilization theophylline. The percent drop in VO, with hypoxemia was 9.6 period of at least 15 min. Thus our results do not show the 2 4.0% and 11.3 ? 3.7%, respectively, during baseline and classical biphasic ventilatory pattern obtained by other inves- after theophylline administration. Finally, there were no sig- tigators in the first 10 min of hypoxemia in piglets (4, 5) and in nificant changes in the heart and respiratory rates, blood pres- newborns of several mammal species (17-20). We used such a sure, and body temperature throughout the study. progressive induction of hypoxemia because, once we had seen

Table 1. Metabolic and cardiorespiraton, response to hypoxemia in the theophylline group Baseline Postinfusion Response Normoxia Hypoxemia Normoxia Hypoxemia Vo, (mL . min-' . kg) Vco, (mL . min-' . kg) Paco, (kPa) VA (mL . min-' . kg) R (Vco,No,) Respiratory rate (breathslmin) Heart rate (beatslmin) Blood pressure (mmHg) Systolic Diastolic *p< 0.01. ** p < 0.005 (hypoxemia vs normoxia) 930 PORRAS from our preliminary studies that a quick exposure to alveolar adult humans (27-29). Although it is not known whether this is hypoxia invariably awakens the animals who then become the case for young piglets, our results do suggest such a agitated, we realized that our planned study in quiet sleep phenomenon. Nonetheless, before any conclusive statement would be impossible, especially after theophylline administra- can be made about substrate utilization during hypoxemia with tion. and without theophylline, direct measurements of substrates Metabolic rate response. The metabolic rate response to would be required. hypoxemia was not significantly modified after the adminis- Ventilatory response to hypoxemia. It was expected that, by tration of theophylline to our piglets. Indeed, the small de- stimulating both ventilation and metabolic rate, theophylline crease in oxygen consumption with hypoxemia during the would increase the ventilatory response to hypoxemia. How- baseline period was still present after theophylline administra- ever, although an increase in ventilation and oxygen consump- tion, despite the slight overall increase in metabolic rate. A tion after theophylline administration did occur in normoxia, decrease in oxygen consumption upon exposure to hypoxemia we saw no increase in the ventilatory response to hypoxemia. has been described in various newborn mammals (8, 12, 21, Below we discuss several explanations for that phenomenon. 22) and in piglets studied in similar conditions and at an age First, it is possible that the ventilatory drive in response to similar to ours (9). It has been shown that this decrease in hypoxemia was increased by theophylline but counterbalanced oxygen consumption is not accompanied by either lactate in the end by hypocapnia's depressive effect on respiration. In accumulation (21) or accumulation of an 0, deficit (13, 22), our piglets, the level of Paco, after theophylline administration both signs of tissue hypoxia. This led to the suggestion that, was significantly lower than that of the baseline period. The during periods of 0, limitation, newborns actually decrease presence of hypocapnia can then explain the smaller increase in some of their oxygen consumption-but without signs of 0, the ventilatory response to hypoxemia after theophylline ad- deficit- by decreasing nonessential metabolism (processes re- ministration. We elected not to correct the hypocapnia, for our quiring 0, that are part of normal metabolism but not essential preliminary experiments had shown that sleep was almost for survival) (13, 22). If this phenomenon does represent an impossible during isocapnic hypoxemia. In the only study of adaptation of the newborn to low levels of oxygen, the strategy newborn animals in which Paco, levels were measured and no may well be preserved at the theophylline blood level achieved attempts made to correct hypocapnia (4), no conclusion con- in our study-a level also used in treatment of human new- cerning the effect of theophylline on the hypoxic ventilatory borns. drive can be drawn, because the large increase obtained in the It could be argued, however, that the decrease in metabolic ventilatory response to hypoxemia after theophylline adminis- rate observed in our piglets was really rather slight. Several tration was not accompanied by a proportional decrease in species (mice, rats, and kittens, for instance) have a 30-50% Paco,; this strongly suggests an increase in metabolically decrease in oxygen consumption with hypoxemia (8). The produced CO, as the cause of the ventilatory increase. major characteristics of these species are their small size and Second, perhaps it is the degree of hypoxemia, and hence the poor overall development at birth. We believe, however, that degree of hypoxic ventilatory depression, that acts as one of the our piglet model of hypoxemia is closer to what occurs in human newborns, who decrease their oxygen consumption by major determinants of the effects of theophylline. The level of only 10-15% (23) upon exposure to mild hypoxemia. ventilation achieved during hypoxemia is thought to depend on We found that hypoxemia exposure alone was associated the interaction between various factors (30), including the with an increased R, suggesting an increased utilization of stimulatory effect of peripheral chemoreceptors; the depressant carbohydrates as a source of energy. Although an increased R effect of hypoxemia, if severe, at the CNS level; and the change is only indirect evidence of increased carbohydrate use, it was in metabolic rate, if any. Depending on the level of hypoxemia, a consistent and stable finding throughout the 2 h of hypoxic respiratory depression could then become the predominant exposure. The increased R was present in all animals-in the factor in the ventilatory response, an effect possibly reversible baseline study of the theophylline group as well as the placebo by theophylline. This could explain why Darnall (4) showed an group, even with a repeat exposure to hypoxemia. In adults, an increase in the ventilatory response to hypoxemia after theo- increased R (24) and a greater dependence on carbohydrates to phylline administration, for his piglets were exposed to severe derive energy (25) has already been reported upon exposure to hypoxemia. In contrast, Long and Lawson's (5) study, with high altitude hypoxemia. In newborns as well, an increase in R 18% Fio,, which certainly represents a mild degree of hypox- with hypoxemia has been reported in several instances (12,21, emia, showed no increase in the ventilatory response to hy- 23). It seems that using carbohydrates as fuel might be advan- poxemia with theophylline. tageous in hypoxemic situations because slightly more energy Finally, one might argue that we used a dose of theophylline (per liter of oxygen used) is liberated by the burning of that was too low for the determination of any increase in the carbohydrates as opposed to fat (26). ventilatory response to hypoxemia, despite increased ventila- Interestingly, after pretreatment with theophylline, we found tion and metabolic rates. However, we did not attempt to no increase in R with hypoxemia. This finding might be evaluate the effect of higher doses because sleep would have explained by the opposing effects of hypoxemia (which in- been difficult to obtain at higher blood levels of theophylline creases R) and theophylline (which decreases R). The other (10). Nevertheless, Long and Lawson (5) did use high doses of well known methylxanthine, caffeine, has already been shown theophylline in the newborn piglets but without an increased to promote fat utilization, thereby decreasing R (toward 0.7) in ventilatory response to hypoxemia. METABOLIC EFFECT OF THEOPHYLLINE IN HYPOXEMIA 93 1

It appears, therefore, that theophylline may be able to slim- (eds) The Pharmacological Basis of Therapeutics. Pergnmon Press, New York, pp 618-637 ulate the ventilatory response to hypoxemia only in the pres- 12. Mortola JP, Renon~coR 1988 Metabol~cand ventilatory rates In newbom kittens ence of a preexisting depression of respiration secondary to during acute hypaxia. Respir lrnysiol 7355-68 severe hypoxernia or anesthesia, and in the absence of hy- 13. Frappell P, Saiki C, Mortola JP 1991 Metabolism during norrnoxia, hypoxia and recovery in the newborn kitten. Respir Physiol 86:115-124 pocapnia. We have insufficient data to draw conclusions con- 14. Stanton HC, Mueller RL 1977 Performance of swine chilled during artificial rearing. cerning a dose-dependent effect of theophylline. Am J Vet RCS 38:100L1006 15. Car6 A, Haddad GG 199(1 Effect of sleep on regional blood flow distribution in In summary, our results show that, during quiet deep in p~gletsPedlatr Res 28:218-222 newborn piglets, theophylline does not abolish the smalI de- 16. Pavlin EG, Hmbeln TF 1986 Anesthesia and the oonttol of ventilation. In: Fishman crease in 0, consumption brought about by hypoxernia; nor AP (ed) Handbook of Physiology. The Respiratory Syslem. Sect. 3. Vol. 11. American Physiolo~icSoctcly. Rcthesda, MD, pp 793-813 does it enhance the ventilatory response to a moderate degree 17. Bureau MA, Zinman R, Foulon P, Bcgin R 1984 Diphas~cventilntory response to of hypoxemia. hypoxin in newbom lambs. J Appl Physiol 5684-90 18. Grunsteln MM, Hazinsk~TA. Schlueter MA 1981 Respimtory control during hypoxia in newborn rabbits: implied aclion of endorphins. I Appl Physiol 51 : 122-130 Acknowledgmemts. The authors are grateful to Drs. Imman- 19. bframboise WA, Guthrie RD, Siandaen TA, Wmtlrum DE 1983 Pulmonaty me- uela Moss and Robert T. Brouillette for their helpful criticism chanics during thc ventllatoly response to hypoxcmia in thc ncwhorn monkcy. 1 Appl Physiol 55:1008-1014 of this manuscript and to Richard Nadler for his skillful 20. Rigatto H. Brady JP. dn la Torre Verduzco R 1975 Chemoreceptor reflexes In the technical assistance. pretem infants. I. The e&ct of gestational age on the vent~latoryresponse 10 inhalation of lOOI oxygen md 15% oxygen. Pediahics 55:604-613 21. Moss M. Moreau G, Lister G 1987 Oxygen transpod nnd metabolism in Ihe wnscio~~s lamb: the effects or hypoxemra. hdralr Res 221177-183 22. Sidl D, Kuipers JRG, Htymann MA, Rudolf AM 1983 Developmental change!! in I. Marclial F, Bairam A, Vert P 1987 Neon~talapnea and apnerc syndrome. In Stern L oxygenatron and circulatory response lo hypoxemia in lambs. Am I Physiol (ed) Cllnlcs in Perinatology. The Respiratory System in the Newborn. WB Saunders. 245:H674-H682 Phlladeiphia, pp 509-529 23. Cross KW. JPM. 2. Rlanchard PW,Aranda SV 1992 Pharmacolherapy of respiratory control disorders. In: Titad Trythall DAH 1958 The gawous metabolism or the newborn Beckerman RC. Brouillette RT, Hunt CE (eds) Respimwry Control Disorders in infant brcathlng 15% oxygen. Acta Pacdiatr Scand 47.271-273 Infants and Children. Williams & Wilkins, Baltimore, pp 352-370 24. Picon-Reategu~E 18fA Insulln. epinephrine and glktcagon In the metabolism of 3. Bruce RD. D'mall RA, AlthauS JS 1986 Aminophyll~nereduces hypoxic ventilatory carbohydmtes at high altitude. Fed Proc 25:1233-1238 depress~onw~lhout increasing catecholamines. Padiatr hlmonol 2:218-224 25. Brooks GA. Butterfield GE, Wolfe RR. Groves DM. Marzeo US,Sutton 3R. Wolfel 4. Damall RA 1985 Aininophyllrne rcduccs liypoxic vcnrtlatory depression: pssrble EE. Reeve! JT1991 Fncreased dependance on blood gluco~cafter acclrmatization lo role of adenos~ne.Pedlatr Res 19306-710 4,300. J Appl Physiol 70.919-927 5. Lang WR, tawson EE I984 Nturotranrmilten and biphnsic respuatory response to 26. Bursztein S, Blwyn DH. Askanazi J, Kinney 1M 1989 Fucl utilization m normal. hypoxia. J Appl Phys~ol57213-222 smrving and parhologic states. In. Bursztein S, Elwyn DH, Askanazi I. Kinney IM 6. Haddad GG. Melllns RB 1984 Hypoxia and respiratory contml in eatly hie. Annu (eds) Energy Metabokisrn, Indirect Cntorimetry and Nutrillon. Williams & Wilkins, Rev Physiol46:629-643 Baltimore, pp 119-171 7. Rigatto H 1992 Mnturatton ol breathing contml in the fetus and newbom infnnt. In: 27. Acheson KJ. Zahorsh-Mark~rklewiczEi. Pinel P. Anantharaman K. l€qu~erE 1980 Beckermnn RC, Broulllette RT, Hunt CE (eds) Respiratory Control D~sordersIn Caffcinc and coffee: their influence on mctaboltc rare and substrate utit~zation in Infants and Children. W~lttams& Wilkins. Bnl~imorc.pp 61-75 normal weight and obcsc individuals. Am J Cliri Nutr 33.989-W7 8. MonoLa IF. Rcz7onim R, Lanthier C 1989 Ventilation and oxygen consumption during acute hypoxla in newborn mammals: a comparative analysis. Raspir Physiol 28. Beavo JA, Rogers NL. Grofford OB 1970 Effects of xanth~nedenvht~ves on l~polys~s 78:3 1-43 and on adenosine 3'5'-monophosphate phosphodiesterase activ~ty.Mol Phxrnacol 9. Rosen CL. Schecter WS, Mellins RB, Haddad GG 1993 ERwt of acute hypoxta on 6597-604 mctntmlism and ventilatton In nwake piglets. Respir Physiol 91:307-319 29. teBlanc J. Jobin M, C8t6 1. Samson P. Lnbne A 1985 Enhanced metabolrc raponsc 10. C6tt A. Blanchard PW, Meehan B 1992 Cardioresp~ratoryand metabolic effects of to cnffctnc in exercirc-tratnad human subjectq. I Appl Physiol 59:832-837 doxapram and theophylline in slecping ncwhorn p~glctrI Appl Physiol 72.410-415 30. Bryan AC, Bowes C,Maloney JE 1886 Control of brealhing in the fetus and rhe 11, Rall TW 1990Drugsused sn the umtment of asthma. The rnerhylxnnrh~ne?,cmmolyn newborn. In: Flshman AP (ed) Handbook of Phys~olagy,The ResplrAtory System. sodlurn and orher agents. In: Goodmw A. G~lmnnP, RaII W,N~es AS. Taylor P Sect. 3. American Physiologic Society. Belhesdtr, MD. pp 621-647