Central chemical in term newborn
regulation
of respiration
M. A. BUREAU, R. BEGIN, AND Y. BERTHIAUME Unite’ de Recherche Pulmonaire, Faculte’ de Midecine, Sherbrooke, Quebec JlH 5N4, Canada
BUREAU, M. A., R. B&GIN, AND Y. BERTHIAUME. Central chemical regulation of respiration in term newborn. J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. 47(6): 1212-1217, 1979.-The role of the medullary H+-sensitive chemoreceptors on the drive of breathing was studied in 10 unanesthetized newborn animals (8 lambs and 2 kids). The experiment consisted of sequential measurements of ventilation (VE) during a progressive change in the arterial pH (pH,) and in the pH of the cisternal cerebrospinal fluid (pH&, induced by intravenous infusion of hydrochloric acid (HCl) followed after an 8-h steady state of acidosis by rapid bicarbonate [HCO:] infusion. It is shown that a rapid change in [HCO&F occurs during the infusion of HCl or NaHC03. As a consequence both CSF and arterial pH change in the same direction and large changes in pHcsF (from 7.331 to 7.227) were observed. Such CSF acidosis did not contribute to further increase VE beyond the level by hyperventilation induced by the initial fall of pH,. The ventilatory response to the decrease in pH, was found to fall off with moderate to severe acidosis (pH, < 7.20). In conclusion, this study demonstrates an instability of the pHcs* during neonatal metabolic acidosis and it suggests an immaturity of both the H’-sensitive medullary and peripheral chemoreceptors in the &day-old newborns. metabolic acidosis; medulky chemoreceptor; peripheral chemoreceptor; cerebrospinal fluid acid base; newborn lamb
OF THE PH of the cerebrospinal fluid (pHcsF) in the chemical control of breathing of newborn mammals is largely unknown. In adults, the blood-brain barrier maintains the pH csF within a narrow range despite induction of severe systemic metabolic acidosis (10, 14). This stability of the pH csF during metabolic acidosis is caused by a relatively low permeability of the bloodbrain barrier that tends to maintain a stable CSF bicarbonate concentration [HCO&F despite decrease in arterial bicarbonate [HCO&. Because of such a barrier, paradoxical changes (i.e., in opposite direction) occurring in pH, and in pHcs~ are observed during transient changes in pH, induced by H+ or HCOi infusion (1, 3,
THE ROLE
12)
Indeed, following a decrease in pH, occurring with the induction of systemic metabolic acidosis, the pHcs~ becomes alkaline for a few hours (3, 12). This is explained by the fall of pH, that increases the drive for breathing at the peripheral chemoreceptor level and causes a rapid decrease in the intravascular carbon dioxide tension (Pco2); carbon dioxide diffuses rapidly out of the CSF 1212
Universite’
de Sherbrooke,
space increasing the pHcs~, since the [HCO&sF does not decrease rapidly enough to compensate the initial central respiratory alkalosis. After a few hours of systemic acidosis, the [HCO&F falls, pHcs~ returns to normal and remains at the lower normal level during a steady state of metabolic acidosis (3, 12). The same phenomenon occurs with infusion of base in the correction of metabolic acidosis: the pH, returns towards normal whereas the pH csF drops rapidly because of infused HCO: does not penetrate the CSF as quickly as Pco2 accumulates in CSF due to a return of ventilation to lower level. Knowledge of the time course of these pHcs~ and pH, changes in metabolic acidosis have led to better comprehension of the dynamics of the control of breathing and have shown that ventilation changes result from the opposite or synergistic interactions of pH, and pHcs~ (3, 12), stimulating (or inhibiting), respectively, the peripheral chemoreceptors and the medullary chemoreceptors. Thus the biphasic increase of ventilation that was described years ago (11) in subjects undergoing a systemic metabolic acidosis (or during its recovery) has been explained by such interaction in both groups of chemoreceptors (11). Because of their locations in the blood compartment and beyond the blood-brain barrier, at any moment of metabolic acidosis these chemoreceptors receive different signals during systemic acidosis; these signals may be different qualitatively (i.e., acidosis versus alkalosis) or quantitatively (i.e., the degree of pH changes) (3, 12). Besides the difference in signals, the response in ventilation to a given pH change seems to differ: the ventilatory response induced by pH change at the level of the medullary chemoreceptors appears much more sensitive to pH variation than that of the peripheral chemoreceptors (1). In newborns, however, the efficiency of the regulation of the pHcs~ in the course of acidosis is not known and consequently the role of the medullary chemoreceptors on the newborn respiration is largely unknown. So far, very few studies have been carried out on the regulation of the pHcs~ in the neonatal age and the maturation of the H’sensitive medullary receptors has not been studied. Although an impermeability of the blood-brain barrier to H+ or HCO; has been suggested in studies of newborn lambs (5,6) and human newborns (8)) additional work is clearly needed to better understand the role of the blood-brain barrier in the regulation of the pHcs~
OlSl-7567/79/0000-0000$01.25
Copyright
0 1979 the American
Physiological
Society
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 12, 2019.
REGULATION
OF
RESPIRATION
IN
TERM
1213
NEWBORN
and consequently the effects of the pHcs~ on the drive of respiration during the neonatal period. It is the purpose of this work to analyze, in unanesthetized newborn lambs and goats, the relationships between the arterial and CSF acid-base status in the induction, the maintenance, the therapy, and the recovery of metabolic acidosis. An additional goal was to evaluate the contribution of the medullary receptors and the peripheral chemoreceptors to the control of breathing in this age group. MATERIALS
AND
METHODS
csf Y 0’ 0 1 Pa
- -
0 lambs and goats were chosen because their size permitted repeated CSF samplings 4c from the cisterna magna and because some knowledge of neonatal CSF acid-base physiology has already been established in these species (7, 8). The eight newborn f lambs and the two newborn goats of the study were born x 3e healthy by spontaneous delivery. At the time of the study 2 their mean age was 8.1 t 1.6 (SE) days, and their mean 3c weight 4.4 t 0.5 kg. They were left with their mothers beforethe experiments and during the recovery period of the experiments. Procedures. In each unanesthetized animal breathing 2E room air for the whole experiment, a hydrochloric acid (HCl) metabolic acidosis was induced with a 5 mmol/kg 7.40 -+ iv infusion over 1 h and the metabolic acidosis was maintained with a 2-mmol/kge h infusion, a total of 8 h of acidosis. The objective was to decrease the [HCO& . a -6to a level below 12 mmol/l. The HCl was infused in a 5% /i0 n 7.30 dextrose, 0.45% NaCl solution and the fluid was infused I Q at a constant rate of 4 ml/kg h. No oral fluids were given throughout the period of acidosis. After 8 h of acidosis, 7.20 a l5-min iv infusion of bicarbonate was given with a mean amount of 22.5 t 2.2 mm01 of bicarbonate. The bicarbonate given to return the arterial pH towards 7.10 _ normal has been calculated based on preliminary experiments as body weight ( kg) x base deficit x 0.5 factor of 0 2 4 6 8 10 24 HOURS distribution. At the end of the infusion the animals were ONTROll ACIDOSIS # RECOVERY 1 returned to their mothers to recover. FIG. 1. Comparison between acid-base status of arterial blood and Sampling and measurements. During the acidosis and cerebrospinal fluid in newborn lambs and kids in control conditions, during the recovery, 2 ml of arterial blood and 1 ml of during acidosis, during bicarbonate therapy, and during recovery. BroCSF were sampled un .der stric t anaerobic con iditions from ken Line refers to mean * SE for pH, Pcoz, and HCO: of cerebrospinal fluid and unbroken line to arterial blood. the femoral artery and from the cisterna magna by percutaneous punction after local subcutaneous instillation of 1% lidocaine (Xylocaine, Astra) according to techniques previously detailed (3, 4). Briefly, for CSF samdiffered by 0.005 pH unit or less. Samples were obtained pling, a BD-22G (Becton-Dickinson, NJ), disposable immediately before the induction of acidosis (time 0.40) spinal tap needle was introduced in the cisterna magna and at hours l:OO, 3:00, 6.90, and 8:OOof acidosis mainand, while CSF was allowed to flow under its natural tenance, at hour 8:15 which was at the end of the 15-min pressure, CSF was withdrawn into a l-ml Hamilton gas- bicarbonate infusion and during the spontaneous recovtight syringe (model 1001 LT, Hamilton, Beno, NV) ery from acidosis, hour 11:OO and hour 24:O0. equipped with a 22.gauge needle introduced at the outlet Ten minutes before the period of sampling, ventilation was recorded during 5 min to obtain respiratory freof the spinal needle shaft. The dead space of the syringe had been filled with mercury and immediately after quency (f), tidal volume (VT), and expired minute ventilation (VE) as well. Each animal was wearing a soft sampling the syringe was sealed with mercury. Immediately after sampling, blood and CSF were an- plastic cone mask attached to a pneumotachograph volalyzed for pH, Pco~, and POZ with a Corning 165 bloodume integrator system (Hewlett-Packard). The dead space of the mask was approximately 10 ml. gas apparatus. The pH csF was measured repeatedly without rinsing the electrode until two consecutive readings StatisticaL anaZysis. In the presentation of the data, AnimaZs. The newborn
l
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 12, 2019.
1214
BUREAU,
mean values are followed by the standard error of the mean (SEM) as an index of dispersion. The data was analyzed statistically by the paired Student t test for acid-base values of arterial blood, for acid-base value of CSF samples, and for minute ventilation values obtained at different times throughout the experiment. Differences with P < 0.05 were considered significant (20).
BaGIN,
AND
BERTHIAUME
-1c
,E w
RESULTS
Acid-base status. Figure 1 presents the time course of the changes occurring in the blood and CSF acid-base values during induction, steady-state, rapid correction, and recovery of metabolic acidosis. pH. In base-line control condition, the pH, was 7.397 and the pHosF 7.331. During HCl infusion, both arterial and pHos~ moved in the same direction but pHcs* was more stable. After 1 h of acidosis, pH, was 7.259 and pHcsF 7.318. At hour &OO, pH, had dropped to 7.145 and the pHcsF to 7.227. With infusion of bicarbonate, the pH, returned to 7.268 and the pHcsF to 7.258. At the end of the recovery period, hour 24:00, pH, was 7.326 and pHcsF
A a H’a
nMol
7.306. Pcoz. Carbon dioxide tension in both compartments changes in the same direction. Control values of Pace, AI H:sf nMol and Pcs~coz were 34.6 and 42.7 Torr. After 1 h of acidosis, 20 Pace, was 30.8 Torr and PcsFC02was 40.4 Torr, and after 3 h of acidosis, they were 27.6 and 38.5 Torr. Then, Pco2 did not change in blood or in CSF being, at hour 8:00, 10 27.1 Torr in arterial blood and 37.2 Torr in CSF. The infusion of bicarbonate caused an abrupt increase of both Pace, and Pcsfc02 to reach 30.1 Torr in blood and 41.0 0 tn B nmdm n n nn Torr in CSF. Thereafter both PCO~ slowly returned to normal. Bicarbonate. Control [HCO;] were 20.4 mmol/l for blood and 22.1 mmol for CSF. Within 1 h of acidosis, [HCOT], was 13.7 mmol but [HCO&SF was 20.3 mmol. After 8 h, they were, respectively, 9.3 and 15.5 mmol. The infusion of bicarbonate increased the [HCOi], to 13.6 FIG. 3. Relationship between change in arterial PCO~ in response to mmol and [HCO-]3 CSFto 18 mmol. After twenty-four h, changes in arterial and cerebrospinal hydrogen concentration. they were 16.9 and 20.1 mmol. Po2. Arterial POT was 66.0 t 1.5 Torr in control conditions and increased to 77.4 -t 2.3 Torr after 1 h of acidosis, then it remained high during acidosis (82.2 t 2.3 Torr at at hour 8:OO)but it decreased after bicarbonate infusion hour 3:OO; 80.4 t 2.7 Torr at hour 6:00, 80.9 t 1.9 Torr to 74.7 t 2.2 Torr at hour 9:00, 71.6 t 1.6 Torr at hour ll:OO, and 69.9 t 3.2 Torr at hour 24:OO. VentiZation. The time course for development of resVE 1- liter piratory compensation of acidosis and its correction is Kgamin presented in Fig. 2. Resting control ventilation was 0.487 0.6 1 ( BTPS) .kg-’ min. With acidosis VE increased in 1 h to 121% and reached its maximum at hour 3:O0. Ventilation increased in a monophasic manner and from hour 3100 0.5 no further increase of VE was observed despite continu+ ous rise in HL and pH+csF (Fig. 3). Following bicarbonate Z 0.4 therapy VE returned towards control levels (Fig. 2). Respiratory frequency and tidal volume. The mean (*SE) tidal volume (ml kg-’ emin-‘) in control period I I I I J was 8.7 t 0.7; it was 8.6 to 0.9 at hour l:OO, 8.9 t 0.8 at 0 2 4 6 8 10 HOURS 24 hour 3:00, 9.25 t 0.9 at hour 6:00, 9.2 t 0.9 at hour 8:00, lCONTROLl ACIDOSI S # RECOVERY 1 8.8 t 0.6 at hour 8:15, 8.5 t 0.7 at hour 1l:OO; 8.0 t 0.8 at hour 24:OO. The respiratory frequency (min-‘) was 58 FIG. 2. Minute ventilation (mean sfr SE) in course of metabolic + acidosis and its therapy with bicarbonate infusion in 10 newborn lambs - 6.1 in control period; it was 68 t 7.5 at hour l:OO, 68 and kids. t 7.3 at hour 3:00, 60 t 5.5 at hour 6:00, 62 t 6 at hour
t
1 1
NaHC03
l
l
NaHC03
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 12, 2019.
REGULATION
OF
RESPIRATION
IN
TERM
1215
NEWBORN
8:00,61 t 4.1 at hour 8:15,58 t 4.0 at hour ll:OO, and 55 t 6 at hour 2490.
of the CSF, presented in this study, was clearly different in newborns than that reported in adults but the difference is more quantitative than qualitative. First, the CSF paradoxical alkalosis of the early hours (induction) of metabolic acidosis as reported in studies in adult subjects (3, 12) does not exist in our newborn lambs and goats (Fig. 1). On the contrary, the infusion of HCl in blood is followed by an immediate fall of 0.013 pH unit in the pHcsF (hour 1~00).Then, with the steady state of acidosis (from hour 190 to 8.90) the pHcsF did not remain near normal but continued to decrease to a value (7.227) that is unusually low for CSF (11, 13, 14). Finally, the paradoxical CSF acidosis usually following the rapid infusion of bicarbonate (3, 11) is not found in our newborns, but base infusion caused similar pH changes toward normal in both blood and CSF. The absence of the “adult type” of paradoxical changes in pH, and pH csF in newborn is explained by the limitation (or immaturity) in mechanisms responsible for the stability of pH csF despite changes in pH,. Consequently, either acidosis in the arterial compartment or correction of arterial acidosis resulted in bicarbonate changes A[HC03] that are in CSF as high as slightly more than 50% of the A[HCOy] occurring in the arterial blood. As shown in Fig. 4, the ratios of A[HCO&F to A[HCO& reach 50% during H+ infusion and 57% following bicarbonate infusion. Also, as observed by others (3, 6), it appears that the rate of bicarbonate change from blood to CSF (occurring with intravenous infusion of bicarbonate) is faster than the change from CSF to blood (occurring during induction of acidosis). Indeed, after 6 h of acidosis A[HCO&F/A[HCO& reached a stable level whereas, after bicarbonate infusion, the stable level is achieved almost immediately (Fig. 4). Comparing the A[HCO&F/A[HCO& obtained in this study of newborns to those obtained in a previous study of metabolic acidosis of adult dogs using an identical method (3), the change in such ratio is greater in
DISCUSSION
The present data establish three particularities of the neonatal physiol .%Y related to the regulation of the CSF acid-base status and also related to the chemical control of breathing in this age group. First, it is shown that the mechanisms responsible for the stability of the pHcsF during metabolic acidosis are at this age inadequate to cope with metabolic acidosis. Second, it is found that in newborn lambs and goats, the pHcs~ is less efficient to control breathing than the adult pHcsF. Third, it is observed that the peripheral chemoreceptors’ drive of breathing has l not reached full maturity in this age group. Although the failure in the regulation of pHcs~ seems obvious during metabolic acidosis and during its therapy with bicarbonate infusion, during resting control conditions, the acid-base status relationship between blood and CSF in newborns is similar to that of adults. In this study, the control acid-base status of blood and CSF of newborn lambs and goats is similar to that of previous studies in adults (10, 14) (Fig. 1). Cisternal pH is about 0.06 pH units lower than pH,, PcsFcoZ is about 10 Torr higher than pH, and [HCO&SF/[HCO& is 1.06, which is the expected ratio according to the level of [HCO& (6). The Pace, is very similar to that of Purves (15-17) in unanesthetized newborn lambs although ventilation is slightly higher in our animals than in those of Purves; this difference can be explained by variations in methods. In Purves’ study, the newborn lambs had a tracheostomy that, by decreasing dead space, lowered VE, whereas the face ma sk used here caused a slight increase in VE bY increasing the dead space. During each of the different phases of metabolic acidosis (being the induction, the steady state, the therapy -with base infusion, and the recovery) the acid-base status
1 A’ 0’
#) P
FIG. 4. Comparison between mean (&SE) ratios of bicarbonate concentration in cerebrospinal fluid and in arterial blood in newborn lambs and kids (d) and adult dogs (---(I - -) during HCl infusion in (left panel) and during bicarbonate therapy (right panel).
0’ 0’ 0’ 0’ 0’ 0’
u’ P
ONTRW
1
I
1
1
4
0
2
4
6
8
HOURS
OF
ML
ACIDOSIS
I
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 12, 2019.
1216 the newborn lambs and goats than in adult dogs (Fig. 4). During the early phase of metabolic acidosis, the differences are small between the two groups; however, as the metabolic acidosis is maintained beyond the first 2 h, clear differences can be seen between adult dogs and newborn lambs and goats. At hour 8.~00 of acidosis, in the adults, the A[HCO&&A[HCO& is 33%, whereas in the newborns it reaches 56%, suggesting thereby a greater permeability of the blood-brain barrier in the neonatal period. Furthermore, after the bicarbonate infusion is started, clear differences between adult and newborn are again observed. Indeed, during the 1st h after rapid bicarbonate infusion, while in the adult no change of [HCO&F occurred, in the newborn [HCO&F increased within 15 min (Fig. 1) to A[HCO&F/ A[HCO& of 58%, whereas in adults a maximal ratio of 17% is achieved after 24 h. Thus, on the contrary to observations of adult mammals, in newborn lambs and kids, the changes in arterial bicarbonate concentration are immediately followed by a large change in bicarbonate concentration at the CSF level. The mechanisms whereby this takes place are at the present time unknown and are beyond the scope of the present study but this results in a large pH variation in the CSF of newborn. In adult mammals, such pH variations would considerably alter the drive of breathing, since the pHcs~ would stimulate medullary chemoreceptors, reported to be exquisitely sensitive to minimal pHcs~ variation (6). This expected increase in VE under pHcs~ stimulation was not observed in our newborns. In our newborn lambs and kids, progressive acidjfication of the CSF did not change \j~ as can be seen in Fig. 2. Indeed, as is shown in Fig. 3, despite progressive acidification of CSF from hour 3:oQ to hour 6:OOduring metabolic acidosis, no change in VE or in Pace, is observed. In adults, such CSF acidosis combined with arterial acidosis has been shown to induce a synergistic stimulation of ventilation to produce large changes in \j~ and Pace, (3). In the newborn animals of this study, the maximal increase of ventilation is achieved within the fast 3 h of acidosis. This occurs under the stimulation of a large fall in pH, (0.295 pH unit). Whether the concomitant small fall in the pH csF (0.040 pH unit) has contributed to the initial hyperventilation cannot be elucidated in this study but it is believed that such drive from the pHcs~ was minimal inasmuch as further acidification of the CSF did not further increase VE. It appears that the H-+-sensitive medullary chemoreceptors in the neonatal period are not yet sufficiently mature to contribute further to the control of ventilation after the initial systemic metabolic acidosis. Such observations had been suggested by previous investigators (5, 6), but their experimental conditions, which included general anesthesia and surgical procedures on the animal, precluded firm conclusions on the role of central pH in the control of breathing (7) in the neonates; in this study, neonates were in nearphysiological conditions and no major maneuver was used that could influence the interpretation of the results. Besides the immaturity of the drive of breathing coming from the H+ medullary receptors, the data presented here indicate that there is, in newborns, an immaturity
BUREAU,
BaGIN,
AND
BERTHIAUME
in that of the arterial H+ chemoreceptors. This is suggested by the limitation in the ventilatory response to acidosis. With mild-to-moderate acidosis (pH, > ?7.20) the response of hyperventilation is appropriate in the newborns but the response falls off with further decrease in pH,. This phenomenon of the limited increase of ventilation is consistent with neurophysiological studies showing limitations in the increase of neural impulses generated at the level of the peripheral chemoreceptors during progressive acidosis. In the adult cat (2), the rate of neural impulses has a hyperbolic configuration and tends to plateau with the decrease of arterial pH, below 7.15. This would suggest a limitation in the hyperventilation uniquely under the drive of peripheral chemoreceptors. In the newborns of this study the ventilatory response to a drop in arterial pH tends to plateau at a pH, of 7.20, thereby suggesting an incomplete maturity of the peripheral chemosensitivity at this age. Alternative explanations for the observed blunted increase of VE during progressive acidosis must also be considered. First it may be argued that the central acidosis of the pH csF may depress the neuronal activity of the respiratory centers that would fail to sustain the hyperventilation under adequate chemical drive from the chemoreceptors. This central acidosis would depress the respiratory centers in the same manner as has been suggested that hypoxia causes central depression of ventilation in neonates (18). The main argument against such a hypothesis is shown by the fast reactivity of the central neurons occurring at hour 895 after the bicarbonate infusion. Indeed, following a rapid return of the pH, towards normal, the neural output is then in sufficiently good physiological condition to very rapidly adjust the ventilation to the new level of chemical drive. It is thus unlikely that central depression causes a blunted increase in VE from hour 3:00 to hour 8:00 and then allows central neurons to perform so well from hour 8:OO to hour 8:15. In the second alternative hypothesis, the limitation of hyperventilation could be attributed to the limit achieved in the capacity of the effector, this being the respiratory apparatus. Such an explanation is unlikely since, in the present study, VE increased by only 25% in the first 3 h of acidosis, whereas it has been shown by previous investigators that unanesthetized newborn lambs could double their resting ventilation under chemical drive of respiration induced by hypercapnia or hypoxia or both (15, 17). Third, one may postulate that the respiratory muscles may be exhausted after 3 h of acidosis and respiratory muscle fatigue prevented further hyperventilation in these newborns. Although such a hypothesis cannot be excluded from the present study we believe that the level of hyperventilation (less than 25% above basal level) is not sufficient in healthy newborn to cause exhaustion of the muscles of respiration. In conclusion, the present study of the ventilatory response to acidosis in newborn lambs or kids, has suggested an immaturity in the drive of breathing coming from the H+-sensitive medullary and peripheral chemoreceptors. Similar observations of an immaturity of the chemical drive of breathing have been recognized in
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 12, 2019.
REGULATION
OF
RESPIRATION
IN
TERM
1217
NEWBORN
newborns who acquire further postnatal maturation in the ventilatory response to CO2 stimulation (19) and in the 02 drive of breathing (9). The authors
acknowledge
the capable
technical
assistance
of Mr.
J.
Labbe. This work was supported by Medical Research Council of Canada, Grant 5726, and by Le Conseil de Recherche en Sante du Quebec. Received
1 May
1979; accepted
in fmal form
19 July
1979
REFERENCES 1. BERGER, A. J., R. A. MITCHELL, AND J. W. SEVERINGHAUS. Regulation of respiration. A?. EngL J. Med. 297: 194-201, 1977. 2. BRISCOE, T. J., M. J. PURVES, AND S. R. SAMPSON. The frequency of nerve impulses in single carotid body chemoreceptor afferent fibers recorded in vivo with intact circulation. J. Physiol. London 208: 121-131,197O. 3. BUREAU, M. A., G. OUELLET, R. B&GIN, N. GAGNON, L. GEOFFROY, AND Y. BERTHIAUME. Dynamics of the control of ventilation during metabolic acidosis and its correction. Am. Rev. Respir. Dis. 119: 933-939,1979. 4. BUREAU, M. A., AND P. BOUVEROT. Blood and CSF acid-base.. changes and rate of ventilatory acclimatization of awake dogs to 3,550 m. Respir. Physiol. 24: 203-216, 1975. 5. HERRINGTON, R. T., H. S. HARNED, J. I. FERRIERO, AND C. A. GRIFFIN. The role of the central nervous system in perinatal respiration: studies of chemoregulatory mechanism in the term lamb. Pediatrics 47: 857-864, 1971. 6. HODSON, A. W., A. FENNER, G. BRUMLEY, V. CHERNICK, AND M. E. AVERY. Cerebrospinal fluid and blood acid-base relationships in fetal and neonatal lambs and pregnant ewes. Respir. Physiol. 4: 322-332,1968. 7. JANSEN, A. H. Central chemoreceptor function in the fetus. Semin. Perinatol. 1: 323-327, 1977. 8. KRAUSS, A. N., D. W. THIBEAULT, AND P. A. M. AULD. Acid-base balance in cerebrospinal fluid of newborn infants. BioZ. Neonate 21: 25-34, 1972. 9. LAHIRI, S., J. S. BRODY, E. K. MOTOYAMA, AND T. M. VELASQUEZ. Regulation of breathing in newborns at high altitude. J. AppZ. Physiol.: Respirat. Environ. Exercise Physiol. 44: 673-678, 1978. 10. LEUSEN, I. Regulation of cerebrospinal fluid composition with
reference to breathing. Physiol. Rev. 52: l-56, 1972. 11. MITCHELL, R. A., C. T. CA&AN, J. W. SEVERINGHAUS, B. W. RICHARDSON, M. M. SINGER, AND S. SHNIDER. Stability of cerebrospinal fluid pH in chronic acid-base disturbances in blood. J. AppZ. Physiol. 20: 443-452, 1965. 12. MITCHELL, R. A., AND M. M. SINGER. Respiration and cerebrospinal fluid pH in metabolic acidosis and alkalosis. J. AppZ. Physiol. 20: 905-911,1965. 13. PAPPENHEIMER, J. R. The ionic composition of cerebral extracellular fluid and its relation to control of breathing. Harvey Lect. 61: 71-94, 1967. 14. PLUM, F., AND B. K. SIESJ~. Recent advances in CSF physiology. Anesthesiology 42: 708-729, 1975. 15. PURVES, M. J. Respiratory and circulatory effects of breathing 100% oxygen in newborn lamb before and after denervation of the carotid body. J. Physiol. London 185: 42-59,1966. 16. PURVES, M. J. The effects of hypoxia in newborn lambs before and after denervation of the carotid body chemoreceptors. J. Physiol. London 185: 60-77, 1966. 17. PURVES, M. J. The respiratory response of the newborn lamb to inhaled CO:! with and without accompanying hypoxia. J. Physiol. London 185: 78-94,1966. 18. RIGATTO, H. Ventilatory response to hypoxia. Semin. Perinatol. 1: 357-363, 1977. 19. RIGATTO, H., J. P. BRADY, AND B. R. DE LA TORRE. Chemoreceptor reflexes in preterm infants. II. The effect of gestational and postnatal age on the ventilatory response to inhaled COZ. Pediatrics 55: 614-640, 1975. 20. SNEDECOR, G. W., AND W. G. COCHRAN. Statistical Methods. Ames, IA: Iowa State Univ. Press, 1967.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 12, 2019.
regulation
of respiration
M. A. BUREAU, R. BEGIN, AND Y. BERTHIAUME Unite’ de Recherche Pulmonaire, Faculte’ de Midecine, Sherbrooke, Quebec JlH 5N4, Canada
BUREAU, M. A., R. B&GIN, AND Y. BERTHIAUME. Central chemical regulation of respiration in term newborn. J. Appl. Physiol.: Respirat. Environ. Exercise Physiol. 47(6): 1212-1217, 1979.-The role of the medullary H+-sensitive chemoreceptors on the drive of breathing was studied in 10 unanesthetized newborn animals (8 lambs and 2 kids). The experiment consisted of sequential measurements of ventilation (VE) during a progressive change in the arterial pH (pH,) and in the pH of the cisternal cerebrospinal fluid (pH&, induced by intravenous infusion of hydrochloric acid (HCl) followed after an 8-h steady state of acidosis by rapid bicarbonate [HCO:] infusion. It is shown that a rapid change in [HCO&F occurs during the infusion of HCl or NaHC03. As a consequence both CSF and arterial pH change in the same direction and large changes in pHcsF (from 7.331 to 7.227) were observed. Such CSF acidosis did not contribute to further increase VE beyond the level by hyperventilation induced by the initial fall of pH,. The ventilatory response to the decrease in pH, was found to fall off with moderate to severe acidosis (pH, < 7.20). In conclusion, this study demonstrates an instability of the pHcs* during neonatal metabolic acidosis and it suggests an immaturity of both the H’-sensitive medullary and peripheral chemoreceptors in the &day-old newborns. metabolic acidosis; medulky chemoreceptor; peripheral chemoreceptor; cerebrospinal fluid acid base; newborn lamb
OF THE PH of the cerebrospinal fluid (pHcsF) in the chemical control of breathing of newborn mammals is largely unknown. In adults, the blood-brain barrier maintains the pH csF within a narrow range despite induction of severe systemic metabolic acidosis (10, 14). This stability of the pH csF during metabolic acidosis is caused by a relatively low permeability of the bloodbrain barrier that tends to maintain a stable CSF bicarbonate concentration [HCO&F despite decrease in arterial bicarbonate [HCO&. Because of such a barrier, paradoxical changes (i.e., in opposite direction) occurring in pH, and in pHcs~ are observed during transient changes in pH, induced by H+ or HCOi infusion (1, 3,
THE ROLE
12)
Indeed, following a decrease in pH, occurring with the induction of systemic metabolic acidosis, the pHcs~ becomes alkaline for a few hours (3, 12). This is explained by the fall of pH, that increases the drive for breathing at the peripheral chemoreceptor level and causes a rapid decrease in the intravascular carbon dioxide tension (Pco2); carbon dioxide diffuses rapidly out of the CSF 1212
Universite’
de Sherbrooke,
space increasing the pHcs~, since the [HCO&sF does not decrease rapidly enough to compensate the initial central respiratory alkalosis. After a few hours of systemic acidosis, the [HCO&F falls, pHcs~ returns to normal and remains at the lower normal level during a steady state of metabolic acidosis (3, 12). The same phenomenon occurs with infusion of base in the correction of metabolic acidosis: the pH, returns towards normal whereas the pH csF drops rapidly because of infused HCO: does not penetrate the CSF as quickly as Pco2 accumulates in CSF due to a return of ventilation to lower level. Knowledge of the time course of these pHcs~ and pH, changes in metabolic acidosis have led to better comprehension of the dynamics of the control of breathing and have shown that ventilation changes result from the opposite or synergistic interactions of pH, and pHcs~ (3, 12), stimulating (or inhibiting), respectively, the peripheral chemoreceptors and the medullary chemoreceptors. Thus the biphasic increase of ventilation that was described years ago (11) in subjects undergoing a systemic metabolic acidosis (or during its recovery) has been explained by such interaction in both groups of chemoreceptors (11). Because of their locations in the blood compartment and beyond the blood-brain barrier, at any moment of metabolic acidosis these chemoreceptors receive different signals during systemic acidosis; these signals may be different qualitatively (i.e., acidosis versus alkalosis) or quantitatively (i.e., the degree of pH changes) (3, 12). Besides the difference in signals, the response in ventilation to a given pH change seems to differ: the ventilatory response induced by pH change at the level of the medullary chemoreceptors appears much more sensitive to pH variation than that of the peripheral chemoreceptors (1). In newborns, however, the efficiency of the regulation of the pHcs~ in the course of acidosis is not known and consequently the role of the medullary chemoreceptors on the newborn respiration is largely unknown. So far, very few studies have been carried out on the regulation of the pHcs~ in the neonatal age and the maturation of the H’sensitive medullary receptors has not been studied. Although an impermeability of the blood-brain barrier to H+ or HCO; has been suggested in studies of newborn lambs (5,6) and human newborns (8)) additional work is clearly needed to better understand the role of the blood-brain barrier in the regulation of the pHcs~
OlSl-7567/79/0000-0000$01.25
Copyright
0 1979 the American
Physiological
Society
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 12, 2019.
REGULATION
OF
RESPIRATION
IN
TERM
1213
NEWBORN
and consequently the effects of the pHcs~ on the drive of respiration during the neonatal period. It is the purpose of this work to analyze, in unanesthetized newborn lambs and goats, the relationships between the arterial and CSF acid-base status in the induction, the maintenance, the therapy, and the recovery of metabolic acidosis. An additional goal was to evaluate the contribution of the medullary receptors and the peripheral chemoreceptors to the control of breathing in this age group. MATERIALS
AND
METHODS
csf Y 0’ 0 1 Pa
- -
0 lambs and goats were chosen because their size permitted repeated CSF samplings 4c from the cisterna magna and because some knowledge of neonatal CSF acid-base physiology has already been established in these species (7, 8). The eight newborn f lambs and the two newborn goats of the study were born x 3e healthy by spontaneous delivery. At the time of the study 2 their mean age was 8.1 t 1.6 (SE) days, and their mean 3c weight 4.4 t 0.5 kg. They were left with their mothers beforethe experiments and during the recovery period of the experiments. Procedures. In each unanesthetized animal breathing 2E room air for the whole experiment, a hydrochloric acid (HCl) metabolic acidosis was induced with a 5 mmol/kg 7.40 -+ iv infusion over 1 h and the metabolic acidosis was maintained with a 2-mmol/kge h infusion, a total of 8 h of acidosis. The objective was to decrease the [HCO& . a -6to a level below 12 mmol/l. The HCl was infused in a 5% /i0 n 7.30 dextrose, 0.45% NaCl solution and the fluid was infused I Q at a constant rate of 4 ml/kg h. No oral fluids were given throughout the period of acidosis. After 8 h of acidosis, 7.20 a l5-min iv infusion of bicarbonate was given with a mean amount of 22.5 t 2.2 mm01 of bicarbonate. The bicarbonate given to return the arterial pH towards 7.10 _ normal has been calculated based on preliminary experiments as body weight ( kg) x base deficit x 0.5 factor of 0 2 4 6 8 10 24 HOURS distribution. At the end of the infusion the animals were ONTROll ACIDOSIS # RECOVERY 1 returned to their mothers to recover. FIG. 1. Comparison between acid-base status of arterial blood and Sampling and measurements. During the acidosis and cerebrospinal fluid in newborn lambs and kids in control conditions, during the recovery, 2 ml of arterial blood and 1 ml of during acidosis, during bicarbonate therapy, and during recovery. BroCSF were sampled un .der stric t anaerobic con iditions from ken Line refers to mean * SE for pH, Pcoz, and HCO: of cerebrospinal fluid and unbroken line to arterial blood. the femoral artery and from the cisterna magna by percutaneous punction after local subcutaneous instillation of 1% lidocaine (Xylocaine, Astra) according to techniques previously detailed (3, 4). Briefly, for CSF samdiffered by 0.005 pH unit or less. Samples were obtained pling, a BD-22G (Becton-Dickinson, NJ), disposable immediately before the induction of acidosis (time 0.40) spinal tap needle was introduced in the cisterna magna and at hours l:OO, 3:00, 6.90, and 8:OOof acidosis mainand, while CSF was allowed to flow under its natural tenance, at hour 8:15 which was at the end of the 15-min pressure, CSF was withdrawn into a l-ml Hamilton gas- bicarbonate infusion and during the spontaneous recovtight syringe (model 1001 LT, Hamilton, Beno, NV) ery from acidosis, hour 11:OO and hour 24:O0. equipped with a 22.gauge needle introduced at the outlet Ten minutes before the period of sampling, ventilation was recorded during 5 min to obtain respiratory freof the spinal needle shaft. The dead space of the syringe had been filled with mercury and immediately after quency (f), tidal volume (VT), and expired minute ventilation (VE) as well. Each animal was wearing a soft sampling the syringe was sealed with mercury. Immediately after sampling, blood and CSF were an- plastic cone mask attached to a pneumotachograph volalyzed for pH, Pco~, and POZ with a Corning 165 bloodume integrator system (Hewlett-Packard). The dead space of the mask was approximately 10 ml. gas apparatus. The pH csF was measured repeatedly without rinsing the electrode until two consecutive readings StatisticaL anaZysis. In the presentation of the data, AnimaZs. The newborn
l
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 12, 2019.
1214
BUREAU,
mean values are followed by the standard error of the mean (SEM) as an index of dispersion. The data was analyzed statistically by the paired Student t test for acid-base values of arterial blood, for acid-base value of CSF samples, and for minute ventilation values obtained at different times throughout the experiment. Differences with P < 0.05 were considered significant (20).
BaGIN,
AND
BERTHIAUME
-1c
,E w
RESULTS
Acid-base status. Figure 1 presents the time course of the changes occurring in the blood and CSF acid-base values during induction, steady-state, rapid correction, and recovery of metabolic acidosis. pH. In base-line control condition, the pH, was 7.397 and the pHosF 7.331. During HCl infusion, both arterial and pHos~ moved in the same direction but pHcs* was more stable. After 1 h of acidosis, pH, was 7.259 and pHcsF 7.318. At hour &OO, pH, had dropped to 7.145 and the pHcsF to 7.227. With infusion of bicarbonate, the pH, returned to 7.268 and the pHcsF to 7.258. At the end of the recovery period, hour 24:00, pH, was 7.326 and pHcsF
A a H’a
nMol
7.306. Pcoz. Carbon dioxide tension in both compartments changes in the same direction. Control values of Pace, AI H:sf nMol and Pcs~coz were 34.6 and 42.7 Torr. After 1 h of acidosis, 20 Pace, was 30.8 Torr and PcsFC02was 40.4 Torr, and after 3 h of acidosis, they were 27.6 and 38.5 Torr. Then, Pco2 did not change in blood or in CSF being, at hour 8:00, 10 27.1 Torr in arterial blood and 37.2 Torr in CSF. The infusion of bicarbonate caused an abrupt increase of both Pace, and Pcsfc02 to reach 30.1 Torr in blood and 41.0 0 tn B nmdm n n nn Torr in CSF. Thereafter both PCO~ slowly returned to normal. Bicarbonate. Control [HCO;] were 20.4 mmol/l for blood and 22.1 mmol for CSF. Within 1 h of acidosis, [HCOT], was 13.7 mmol but [HCO&SF was 20.3 mmol. After 8 h, they were, respectively, 9.3 and 15.5 mmol. The infusion of bicarbonate increased the [HCOi], to 13.6 FIG. 3. Relationship between change in arterial PCO~ in response to mmol and [HCO-]3 CSFto 18 mmol. After twenty-four h, changes in arterial and cerebrospinal hydrogen concentration. they were 16.9 and 20.1 mmol. Po2. Arterial POT was 66.0 t 1.5 Torr in control conditions and increased to 77.4 -t 2.3 Torr after 1 h of acidosis, then it remained high during acidosis (82.2 t 2.3 Torr at at hour 8:OO)but it decreased after bicarbonate infusion hour 3:OO; 80.4 t 2.7 Torr at hour 6:00, 80.9 t 1.9 Torr to 74.7 t 2.2 Torr at hour 9:00, 71.6 t 1.6 Torr at hour ll:OO, and 69.9 t 3.2 Torr at hour 24:OO. VentiZation. The time course for development of resVE 1- liter piratory compensation of acidosis and its correction is Kgamin presented in Fig. 2. Resting control ventilation was 0.487 0.6 1 ( BTPS) .kg-’ min. With acidosis VE increased in 1 h to 121% and reached its maximum at hour 3:O0. Ventilation increased in a monophasic manner and from hour 3100 0.5 no further increase of VE was observed despite continu+ ous rise in HL and pH+csF (Fig. 3). Following bicarbonate Z 0.4 therapy VE returned towards control levels (Fig. 2). Respiratory frequency and tidal volume. The mean (*SE) tidal volume (ml kg-’ emin-‘) in control period I I I I J was 8.7 t 0.7; it was 8.6 to 0.9 at hour l:OO, 8.9 t 0.8 at 0 2 4 6 8 10 HOURS 24 hour 3:00, 9.25 t 0.9 at hour 6:00, 9.2 t 0.9 at hour 8:00, lCONTROLl ACIDOSI S # RECOVERY 1 8.8 t 0.6 at hour 8:15, 8.5 t 0.7 at hour 1l:OO; 8.0 t 0.8 at hour 24:OO. The respiratory frequency (min-‘) was 58 FIG. 2. Minute ventilation (mean sfr SE) in course of metabolic + acidosis and its therapy with bicarbonate infusion in 10 newborn lambs - 6.1 in control period; it was 68 t 7.5 at hour l:OO, 68 and kids. t 7.3 at hour 3:00, 60 t 5.5 at hour 6:00, 62 t 6 at hour
t
1 1
NaHC03
l
l
NaHC03
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 12, 2019.
REGULATION
OF
RESPIRATION
IN
TERM
1215
NEWBORN
8:00,61 t 4.1 at hour 8:15,58 t 4.0 at hour ll:OO, and 55 t 6 at hour 2490.
of the CSF, presented in this study, was clearly different in newborns than that reported in adults but the difference is more quantitative than qualitative. First, the CSF paradoxical alkalosis of the early hours (induction) of metabolic acidosis as reported in studies in adult subjects (3, 12) does not exist in our newborn lambs and goats (Fig. 1). On the contrary, the infusion of HCl in blood is followed by an immediate fall of 0.013 pH unit in the pHcsF (hour 1~00).Then, with the steady state of acidosis (from hour 190 to 8.90) the pHcsF did not remain near normal but continued to decrease to a value (7.227) that is unusually low for CSF (11, 13, 14). Finally, the paradoxical CSF acidosis usually following the rapid infusion of bicarbonate (3, 11) is not found in our newborns, but base infusion caused similar pH changes toward normal in both blood and CSF. The absence of the “adult type” of paradoxical changes in pH, and pH csF in newborn is explained by the limitation (or immaturity) in mechanisms responsible for the stability of pH csF despite changes in pH,. Consequently, either acidosis in the arterial compartment or correction of arterial acidosis resulted in bicarbonate changes A[HC03] that are in CSF as high as slightly more than 50% of the A[HCOy] occurring in the arterial blood. As shown in Fig. 4, the ratios of A[HCO&F to A[HCO& reach 50% during H+ infusion and 57% following bicarbonate infusion. Also, as observed by others (3, 6), it appears that the rate of bicarbonate change from blood to CSF (occurring with intravenous infusion of bicarbonate) is faster than the change from CSF to blood (occurring during induction of acidosis). Indeed, after 6 h of acidosis A[HCO&F/A[HCO& reached a stable level whereas, after bicarbonate infusion, the stable level is achieved almost immediately (Fig. 4). Comparing the A[HCO&F/A[HCO& obtained in this study of newborns to those obtained in a previous study of metabolic acidosis of adult dogs using an identical method (3), the change in such ratio is greater in
DISCUSSION
The present data establish three particularities of the neonatal physiol .%Y related to the regulation of the CSF acid-base status and also related to the chemical control of breathing in this age group. First, it is shown that the mechanisms responsible for the stability of the pHcsF during metabolic acidosis are at this age inadequate to cope with metabolic acidosis. Second, it is found that in newborn lambs and goats, the pHcs~ is less efficient to control breathing than the adult pHcsF. Third, it is observed that the peripheral chemoreceptors’ drive of breathing has l not reached full maturity in this age group. Although the failure in the regulation of pHcs~ seems obvious during metabolic acidosis and during its therapy with bicarbonate infusion, during resting control conditions, the acid-base status relationship between blood and CSF in newborns is similar to that of adults. In this study, the control acid-base status of blood and CSF of newborn lambs and goats is similar to that of previous studies in adults (10, 14) (Fig. 1). Cisternal pH is about 0.06 pH units lower than pH,, PcsFcoZ is about 10 Torr higher than pH, and [HCO&SF/[HCO& is 1.06, which is the expected ratio according to the level of [HCO& (6). The Pace, is very similar to that of Purves (15-17) in unanesthetized newborn lambs although ventilation is slightly higher in our animals than in those of Purves; this difference can be explained by variations in methods. In Purves’ study, the newborn lambs had a tracheostomy that, by decreasing dead space, lowered VE, whereas the face ma sk used here caused a slight increase in VE bY increasing the dead space. During each of the different phases of metabolic acidosis (being the induction, the steady state, the therapy -with base infusion, and the recovery) the acid-base status
1 A’ 0’
#) P
FIG. 4. Comparison between mean (&SE) ratios of bicarbonate concentration in cerebrospinal fluid and in arterial blood in newborn lambs and kids (d) and adult dogs (---(I - -) during HCl infusion in (left panel) and during bicarbonate therapy (right panel).
0’ 0’ 0’ 0’ 0’ 0’
u’ P
ONTRW
1
I
1
1
4
0
2
4
6
8
HOURS
OF
ML
ACIDOSIS
I
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 12, 2019.
1216 the newborn lambs and goats than in adult dogs (Fig. 4). During the early phase of metabolic acidosis, the differences are small between the two groups; however, as the metabolic acidosis is maintained beyond the first 2 h, clear differences can be seen between adult dogs and newborn lambs and goats. At hour 8.~00 of acidosis, in the adults, the A[HCO&&A[HCO& is 33%, whereas in the newborns it reaches 56%, suggesting thereby a greater permeability of the blood-brain barrier in the neonatal period. Furthermore, after the bicarbonate infusion is started, clear differences between adult and newborn are again observed. Indeed, during the 1st h after rapid bicarbonate infusion, while in the adult no change of [HCO&F occurred, in the newborn [HCO&F increased within 15 min (Fig. 1) to A[HCO&F/ A[HCO& of 58%, whereas in adults a maximal ratio of 17% is achieved after 24 h. Thus, on the contrary to observations of adult mammals, in newborn lambs and kids, the changes in arterial bicarbonate concentration are immediately followed by a large change in bicarbonate concentration at the CSF level. The mechanisms whereby this takes place are at the present time unknown and are beyond the scope of the present study but this results in a large pH variation in the CSF of newborn. In adult mammals, such pH variations would considerably alter the drive of breathing, since the pHcs~ would stimulate medullary chemoreceptors, reported to be exquisitely sensitive to minimal pHcs~ variation (6). This expected increase in VE under pHcs~ stimulation was not observed in our newborns. In our newborn lambs and kids, progressive acidjfication of the CSF did not change \j~ as can be seen in Fig. 2. Indeed, as is shown in Fig. 3, despite progressive acidification of CSF from hour 3:oQ to hour 6:OOduring metabolic acidosis, no change in VE or in Pace, is observed. In adults, such CSF acidosis combined with arterial acidosis has been shown to induce a synergistic stimulation of ventilation to produce large changes in \j~ and Pace, (3). In the newborn animals of this study, the maximal increase of ventilation is achieved within the fast 3 h of acidosis. This occurs under the stimulation of a large fall in pH, (0.295 pH unit). Whether the concomitant small fall in the pH csF (0.040 pH unit) has contributed to the initial hyperventilation cannot be elucidated in this study but it is believed that such drive from the pHcs~ was minimal inasmuch as further acidification of the CSF did not further increase VE. It appears that the H-+-sensitive medullary chemoreceptors in the neonatal period are not yet sufficiently mature to contribute further to the control of ventilation after the initial systemic metabolic acidosis. Such observations had been suggested by previous investigators (5, 6), but their experimental conditions, which included general anesthesia and surgical procedures on the animal, precluded firm conclusions on the role of central pH in the control of breathing (7) in the neonates; in this study, neonates were in nearphysiological conditions and no major maneuver was used that could influence the interpretation of the results. Besides the immaturity of the drive of breathing coming from the H+ medullary receptors, the data presented here indicate that there is, in newborns, an immaturity
BUREAU,
BaGIN,
AND
BERTHIAUME
in that of the arterial H+ chemoreceptors. This is suggested by the limitation in the ventilatory response to acidosis. With mild-to-moderate acidosis (pH, > ?7.20) the response of hyperventilation is appropriate in the newborns but the response falls off with further decrease in pH,. This phenomenon of the limited increase of ventilation is consistent with neurophysiological studies showing limitations in the increase of neural impulses generated at the level of the peripheral chemoreceptors during progressive acidosis. In the adult cat (2), the rate of neural impulses has a hyperbolic configuration and tends to plateau with the decrease of arterial pH, below 7.15. This would suggest a limitation in the hyperventilation uniquely under the drive of peripheral chemoreceptors. In the newborns of this study the ventilatory response to a drop in arterial pH tends to plateau at a pH, of 7.20, thereby suggesting an incomplete maturity of the peripheral chemosensitivity at this age. Alternative explanations for the observed blunted increase of VE during progressive acidosis must also be considered. First it may be argued that the central acidosis of the pH csF may depress the neuronal activity of the respiratory centers that would fail to sustain the hyperventilation under adequate chemical drive from the chemoreceptors. This central acidosis would depress the respiratory centers in the same manner as has been suggested that hypoxia causes central depression of ventilation in neonates (18). The main argument against such a hypothesis is shown by the fast reactivity of the central neurons occurring at hour 895 after the bicarbonate infusion. Indeed, following a rapid return of the pH, towards normal, the neural output is then in sufficiently good physiological condition to very rapidly adjust the ventilation to the new level of chemical drive. It is thus unlikely that central depression causes a blunted increase in VE from hour 3:00 to hour 8:00 and then allows central neurons to perform so well from hour 8:OO to hour 8:15. In the second alternative hypothesis, the limitation of hyperventilation could be attributed to the limit achieved in the capacity of the effector, this being the respiratory apparatus. Such an explanation is unlikely since, in the present study, VE increased by only 25% in the first 3 h of acidosis, whereas it has been shown by previous investigators that unanesthetized newborn lambs could double their resting ventilation under chemical drive of respiration induced by hypercapnia or hypoxia or both (15, 17). Third, one may postulate that the respiratory muscles may be exhausted after 3 h of acidosis and respiratory muscle fatigue prevented further hyperventilation in these newborns. Although such a hypothesis cannot be excluded from the present study we believe that the level of hyperventilation (less than 25% above basal level) is not sufficient in healthy newborn to cause exhaustion of the muscles of respiration. In conclusion, the present study of the ventilatory response to acidosis in newborn lambs or kids, has suggested an immaturity in the drive of breathing coming from the H+-sensitive medullary and peripheral chemoreceptors. Similar observations of an immaturity of the chemical drive of breathing have been recognized in
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 12, 2019.
REGULATION
OF
RESPIRATION
IN
TERM
1217
NEWBORN
newborns who acquire further postnatal maturation in the ventilatory response to CO2 stimulation (19) and in the 02 drive of breathing (9). The authors
acknowledge
the capable
technical
assistance
of Mr.
J.
Labbe. This work was supported by Medical Research Council of Canada, Grant 5726, and by Le Conseil de Recherche en Sante du Quebec. Received
1 May
1979; accepted
in fmal form
19 July
1979
REFERENCES 1. BERGER, A. J., R. A. MITCHELL, AND J. W. SEVERINGHAUS. Regulation of respiration. A?. EngL J. Med. 297: 194-201, 1977. 2. BRISCOE, T. J., M. J. PURVES, AND S. R. SAMPSON. The frequency of nerve impulses in single carotid body chemoreceptor afferent fibers recorded in vivo with intact circulation. J. Physiol. London 208: 121-131,197O. 3. BUREAU, M. A., G. OUELLET, R. B&GIN, N. GAGNON, L. GEOFFROY, AND Y. BERTHIAUME. Dynamics of the control of ventilation during metabolic acidosis and its correction. Am. Rev. Respir. Dis. 119: 933-939,1979. 4. BUREAU, M. A., AND P. BOUVEROT. Blood and CSF acid-base.. changes and rate of ventilatory acclimatization of awake dogs to 3,550 m. Respir. Physiol. 24: 203-216, 1975. 5. HERRINGTON, R. T., H. S. HARNED, J. I. FERRIERO, AND C. A. GRIFFIN. The role of the central nervous system in perinatal respiration: studies of chemoregulatory mechanism in the term lamb. Pediatrics 47: 857-864, 1971. 6. HODSON, A. W., A. FENNER, G. BRUMLEY, V. CHERNICK, AND M. E. AVERY. Cerebrospinal fluid and blood acid-base relationships in fetal and neonatal lambs and pregnant ewes. Respir. Physiol. 4: 322-332,1968. 7. JANSEN, A. H. Central chemoreceptor function in the fetus. Semin. Perinatol. 1: 323-327, 1977. 8. KRAUSS, A. N., D. W. THIBEAULT, AND P. A. M. AULD. Acid-base balance in cerebrospinal fluid of newborn infants. BioZ. Neonate 21: 25-34, 1972. 9. LAHIRI, S., J. S. BRODY, E. K. MOTOYAMA, AND T. M. VELASQUEZ. Regulation of breathing in newborns at high altitude. J. AppZ. Physiol.: Respirat. Environ. Exercise Physiol. 44: 673-678, 1978. 10. LEUSEN, I. Regulation of cerebrospinal fluid composition with
reference to breathing. Physiol. Rev. 52: l-56, 1972. 11. MITCHELL, R. A., C. T. CA&AN, J. W. SEVERINGHAUS, B. W. RICHARDSON, M. M. SINGER, AND S. SHNIDER. Stability of cerebrospinal fluid pH in chronic acid-base disturbances in blood. J. AppZ. Physiol. 20: 443-452, 1965. 12. MITCHELL, R. A., AND M. M. SINGER. Respiration and cerebrospinal fluid pH in metabolic acidosis and alkalosis. J. AppZ. Physiol. 20: 905-911,1965. 13. PAPPENHEIMER, J. R. The ionic composition of cerebral extracellular fluid and its relation to control of breathing. Harvey Lect. 61: 71-94, 1967. 14. PLUM, F., AND B. K. SIESJ~. Recent advances in CSF physiology. Anesthesiology 42: 708-729, 1975. 15. PURVES, M. J. Respiratory and circulatory effects of breathing 100% oxygen in newborn lamb before and after denervation of the carotid body. J. Physiol. London 185: 42-59,1966. 16. PURVES, M. J. The effects of hypoxia in newborn lambs before and after denervation of the carotid body chemoreceptors. J. Physiol. London 185: 60-77, 1966. 17. PURVES, M. J. The respiratory response of the newborn lamb to inhaled CO:! with and without accompanying hypoxia. J. Physiol. London 185: 78-94,1966. 18. RIGATTO, H. Ventilatory response to hypoxia. Semin. Perinatol. 1: 357-363, 1977. 19. RIGATTO, H., J. P. BRADY, AND B. R. DE LA TORRE. Chemoreceptor reflexes in preterm infants. II. The effect of gestational and postnatal age on the ventilatory response to inhaled COZ. Pediatrics 55: 614-640, 1975. 20. SNEDECOR, G. W., AND W. G. COCHRAN. Statistical Methods. Ames, IA: Iowa State Univ. Press, 1967.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 12, 2019.
