Fetal breathing and adaptation to maternal hemorrhage in the sheep

Fetal breathing and adaptation to maternal hemorrhage in the sheep P.

L.

M.

TOUBAS,

M.D.

MONSET-COUCHARD,

P.

REY,

J.

PREDINE,

M.

M.D.

PH.D. PH.D.

VERBRUGGE.

M.D.

J.

LEANDRI.

C.

TCHOBROUTSKY.

Parr.\,

M.D. M.D.

Frcrntc

The effect of maternal hemorrhage in chronic preparations was studied on fital lambs in the last month of gestation. Fourteen to 20 per cent of maternal blood wasestimated to ha-c~ been remoiled within 30 minutes, which resulted in a drop of 30 per cent of mean matc~mal arterial pressure. A fetal bradycardia started 28 k 13 minutes after the beginning oj maternal hemorrhage. It lasted 30 2 15 minutes and was concomitant with a rise in fetal arterial pressure. It u1a.s followed by a long-lasting fetal tachycardia of 130 + 38 minutes and was corrected only by reinfusion of blood to the mother. The fetal blood gases demon.rtrated a mild asphyxia with a persistent metabolic acidemia until reinfusion of blood to the mother. Maternal and fetal plasma cortisol levels rose significantly at the end of the hemorrhage. Trachealjuidjow did not change. Fetal breathing recorded 20 hours before and 24 hours after the experiment did not shorn consistent changes, but during fetal bradycardia there was no fetal breath,ing. Recent clinical investigations in thisjield have been made in the human fetus to estimate standards of.fetal well being. These peculiar animat experiments do not show any significant improvement by recording fetal breathing over the recording oj‘ prelabor fetal heart rate. (AM. J. OBSTET. GYNECOL. 127: 505, 1977.)

RESPIRATORY movements in the normal fetus were generally thought to be absent until the demonstration was made by our laboratory’, ’ and by Dawes and coworker?, ’ of intrauterine breathing movements in healthy fetal lambs. Fetal breathing was subsequently described in the rabbit,’ in the human species by Boddy and Robinson,’ and in the rhesus monkey by Martin From the Laboratoire du CNRS. H+tal Broussais and Department of Obstetrics and Department of Neonatalogy, Mater&e de Port Royal, Universiti Reni Descartes. Supported bv grant ATP 7.7-1.28 No. I7 and ATP 23.75.46,from the Institut National de la Sante et de la Recherrhr Medicale. Recrirwd

for publicntion

Accepted October

July

9, 1976.

5, 1976.

Reprint quests: C. Tchobroutsky, Royal, CHU Cochin-Port Roval, Roval, Paris 14e, France.

Materniti de Port 123, boulevard de Port

and associates.6 There were no differences in pH and respiratory gas tensions of fetal arterial blood between breathing and nonbreathing periods.lm3 However. hypoxemia induced by giving the pregnant ewe lowoxygen gas tensions mixture to breathe for 1 hour reduced the incidence of breathing during hypoxia and during the next 6 hours.7 It is now generally admitted that the presence of fetal breathing is a symptom of fetal well being. ‘, ” Using the technique of the in met-o fetal lamb chronic preparation, we studied the changes induced by acute maternal hemorrhage on the fetal breathing movements, heart rate, blood pressure, and biologic variables as blood gases and plasma cortisol. Materials

and methods

Sixteen observations were made on 10 Prealpes fetal lambs of 116 to 13 1 days of gestation. After 24 hours of fasting, the pregnant ewes were anesthetized with in-

506

Toubas

et al

RECORD MATERNAL

FETAL

B

BPIT P IAP

1

r 0 H R B P I T P I A P

: Heort : Blood : Intro : Intro

rote pressure thorocic amniotic

120

30

210

min

pressure pressure

Fig. 1. Timing of the experiment travenous injection of 200 mg. of Ketamine.* Anesthesia was maintained with a perfusion of 1 Gm. of ketamine which had been dissolved in 500 ml. of 5 per cent dextrose. The abdominal wall was opened by a midline incision, the ewe lying supine. Catheters were implanted into a fetal carotid artery, the fetal trachea, and the amniotic fluid by a midline incision of the fetal neck without exteriorization of the fetal head. An electrode was inserted beneath the fetal skin to record instantaneous fetal heart rate.? The catheters were brought out through the maternal flank. A catheter was inserted into the carotid artery of the ewe. In order to collect tracheal fluid in five ewes. the fetal trachea was ligated around the largest fitting cannula inserted 2 cm. toward the lower part of the trachea. The cannula was connected to the input of a micropack plastic bag (volume 250 ml.) which remained inside the amniotic cavity. The output of the bag was inserted into a catheter which was also brought out through the flank of the mother with the other catheters. A third catheter positioned in the thoracic segment of the trachea passed through the whole system in order to measure the tracheal pressure. The ewes were allowed to recover for at least 3 days after surgery before any experiment was performed. Thereafter, an interval of at least 3 days was left before *Ketalar, Parke-Davis. WZardiotachymeter Roche 540.

any further experiments. A continuous rrcord of fetal instantaneous heart rate and of arterial, tracheal, and amniotic pressure was made 20 hours before, during, and 24 hours after the experiment. Maternal arterial pressure was recorded only during the 3% hours of the experiment. Pressures were measured with Statham P.23 pressure transducers connected to a Brush Mark 260 recorder. After the control period a maternal hemorrhage of 800 ml. was induced within 30 minutes. The weight of the ewes was hetlceen 45 and 65 kilograms. and the percentage of blood removal was estimated to be between 14 and 20 per cent of total blood volume.Y The animal was allowed to remain at this level of bleeding for 90 minutes. A reinfusion was carried out during which the blood was reinfused to the mother at the same rate at which it was removed. Maternal and fetal arterial blood samples were collected anaerobically in syringes in which the dead space was filled with heparin. Blood Po2, Pco2, and pH were immediately measured at 37” <:. with Radiometer BMS 3 and corrected to the colonic temperature of the mother. Hematocrit was also measured. The total plasma cortisol levels were determined in the fetus and the mother with the following technique: extraction of’ 0.5 ml. of plasma was performed M.ith 5 ml. of carbon tetrachloride to remove most of the competitive steroids (progesterone and its hydroxylated derivatives and corticosterone). A second extraction was made

Volume

127

Number

5

Fetal

breathing

and adaptation

to maternal

hemorrhage

507

Maternal arterial pressure 50 1 basls

/

m,n

200 Fetal rafe

heart

150 100 1

Mean arterial

fetal pressure

min Fig.

2. Mean

maternal

and fetal

blood

pressure

and heart

rate plotted

as mean

k2

SD.

Table I. Mean maternal arterial pressure, mean fetal arterial pressure, and heart rate during the control period, at the end of the maternal bleeding period, before, during, and 1 hour after reinfusion of the blood to the mother (mean = 2 S.D., paired t test) End of maternal bkeding period (jr0 min.)

Control (0 min.) Maternal: Mean arterial pressure (mm. Hg) Fetal: Mean arterial pressure (mm. Hg) Heart rate

982

14

47 2 10

167 f

16

Interval (75 min.)

Before reinfusion (120 min.)

During reinfusion (135 min.)

1 hr. after winfusion (210 min.)

69 T 12t

82 2 15*

962

13

992

16

98 r 16

57 + 9*

51 zk 11

51 -t 13

472

11

47 k 11

221 f 27t

180 “_ 26t

186 + 25*

173 -c 16

103 AZ 20t

*p < 0.05. tp < 0.01.

with 5 ml. of methylene chloride. The cortisol was determined by the competitive protein-binding method of Murphy” on the dried extract. ‘The blood samples were drawn: (1) immediately before the maternal hemorrhage, (2) at the end of the maternal hemorrhage, (3) before the reinfusion of the blood to the mother and (4) 1 hour after the end of the reinfusion of the blood to the mother. Fig. 1 illustrates the timing of the experiment.

Results Mean maternal arterial pressure (Table I, Fig. 2) (mm./Hg). The pressure dropped gradually to 70 per cent of its initial value (from 98 +- 14 to 69 -I- 12) at the end of the bleeding period (p < 0.01) then slowly came back to control levels within 124 k 21 minutes from the beginning of hemorrhage. Reinfusion of blood to the mother did not significantly change the mean maternal arterial pressure.

508

Toubas

etal.

-18

-12

-6

Fig. 3. Incidence per hour)

0

6

12

18

of fetal rapid irregular breathing (expressed in per cent of respirator)20 hours before and 24 hours after maternal hemorrhage.

2L

haul

s

movements

Table II. Fetal and maternal pH. blood gases, and hematocrit during the control period, after maternal hemorrhage, before reinfusion, and after reinfusion of the blood to the mother (mean + 2 SD., paired

t test)

Control prrid (0 rni?l.)

F&d: PH P(X), (tot-r) PO? (torr) Hte

7.36 42

19

k t

0.05 5

37

k-4 -t 1.5

7.53 32

2 0.04 +-4

84 28

28 25

lMntPr?KLl:

PH P(X), (torr) PO, (torr) Hte

*p < 0.05. Fetal heart rate (beats per minute). The heart rate dropped significantly (p < 0.01) from 167 2 16 to 103 k 20. The bradycardia started 28 2 13.4 minutes after the onset of the maternal hemorrhage. The maximum of deceleration was reached within 8.5 ? 6.3 minutes after its onset. Its total duration was 29.8 k 15.8 minutes. During the ascending phase of fetal heart rate, large beat-to-beat variations were observed. In two fetuses. the heart rate already slowly decreased 15 minutes after the beginning of the maternal hemorrhage. In all cases, fetal bradycardia was immediately followed by tachycardia. Maximum

tachycardia (22 1 t 27) was teat-hed 36 2 16.5 minutes after the maximum of bradycardia. Duration of the tachycardia was 130 i 38 minutes. Fetal heart rate rc’turned to control values 90 minutes after the beginning of reinfusion of blood to the mother (Table I, Fig. 2). Mean fetal arterial blood pressure (mm./Hg). The arterial blood pressure increased from 47 k 10 to 57 2 9 (p < 0.05) at the end of the maternal hemorrhage. In all but two experiments the rise in blood pressure occurred simultaneously with the onset of bradycardia. In two cases the increase of arterial pressure preceeded by a few minutes the onset oi

Volume 127 Number

Fetal

breathing

and adaptation

to maternal

hemorrhage

509

5

FETAL MATERNAL

PLASMA

PLASMA

CORTISOL

CORTISOL cp I loon

w I100 In 6

5

4

4

. 3

3

2

2

.

b .

t

l

0 0

1

1

. b .

b

.

b

i

210

O”p

a

8

120

210

e

min

12L

min

R*i"t"‘,r"

Fig.

4. Maternal

plasma

curtisol.

bradycardia. When the fetal tachycardia reached its peak value, the fetal blood pressure returned to control levels: 51 f 13 (N.S.). No further changes of mean fetal arterial pressure were observed (Table 1, Fig. 2). Blood gases and pH (Table II). Maternal hemorrhage did not significantly change the maternal carotid arterial Paoz Pacop, or pH. The hematocrit fell from 28 ? 5 to 25 ZL 3.5 (p < 0.05) and remained low until blood was reinfused. The fetus exhibited a moderate but significant asphyxia at the end of the maternal hemorrhage. PO? fell from 19 & 4 to 14 ? 3 torr (p < 0.05). pH from 7.36 k 0.05 to 7.29 ? 0.07 (p < 0.05). and Pco2 rose from 42 2 5 to 48 * 5 torr (p < 0.05). Just before reinfusion of maternal blood, Pcoz and Pop were already back to control levels. pH was still significantly low: 7.30 * 0.09 (p < 0.05). One hour after reinfusion of blood to the mother, fetal pH was normal. No significant changes were observed in fetal hematocrit values. Fetal breathing (Fig. 3). T71lenty hours hjorc maternal

h~morrhagr. Rapid irregular breathing lasting from a few seconds to more than an hour were observed up to 40 per cent of the unit time (an hour) in the 20 hour control period.

Fig.

5. Fetal

plasma

co&d.

However, the percentage of this kind of fetal breathing varied in a high proportion from one animal to another and in the same animal from 1 hour to another. No circadian rhythm was observed. Irregular breathing periods were not accompanied by changes in amniotic fluid pressure. No significant differences were found in blood gases and pH obtained during the control period between breathing or apneic animals. Sometimes gasping was also observed but was not quantified. During thr’ +wrimuk During maternal hemorrhage and during the fetal bradycardia which persisted after bleeding, no respiratory movements were observed in 14 out of 16 experiments. During this apneic period, the blood gases and pH demonstrated a moderate asphyxia. The blood gases of the two fetuses which maintained some irregular breathing during the bradycardie episode showed some differences: both maintained a normal pH of 7.44 and 7.35. Pop and Pcoz were in the same range as the other fetuses which became apneic. Twntyjiur hoztr.s ujtu matrrnal hrmorrhag~. During the 24 hours following the experiment, no significant changes could be noticed compared to the 24 hours of the control period. No change could either be elicited if

510

Toubas et al

the comparison were done in the 6 hours preceding or the 6 hours following maternal hemorrhage. Cortisol (Fig. 4 and 5). The mean concentration of cortisol in the mutwnnl plasma was 0.62 ? 0.60 pg per 100 milliliters. The level rose sharply to 3.14 t 1.35 at the end of the hemorrhage which was 300 to 3,300 per cent of the control values. The mean difference between control and posthemorrhagic samples was +2.5 pg per 100 milliliters (paired t test, p < 0.01). Thereafter, maternal plasma cortisol values fell back to control levels. The concentration in the fetal plasma at the control sample was 0.43 2 0.38 pg per 100 milliliters; after maternal hemorrhage it rose to 1.4 * 0.86 pg per 100 milliliters. This represents an increase of 100 to 2,200 per cent of the control values. The mean difference between control and posthemorrhage cortisol was 0.97 pg per 100 milliliters (p < 0.01 paired t test). Thereafter, the mean fetal plasma cortisol remained at the same level, without statistical significance, because of a large standard deviation. In the fourth sample, the level was back to control values. In the oldest fetus (141 days) there was a rise from 3.12 to 3.44 pg per 100 milliliters. This case was not included in the calculations. Tracheal fluid flow. In seven experiments on five ewes, tracheal fluid flow ranged between 7.7 to 18.3 ml. per hour over a period of 24 hours, in fetuses whose gestational ages varied from 113 to 137 days. No significant difference appeared after fetal asphyxia, in the periods ~3, +6, +9, +24, +48, and +72 hours after maternal hemorrhage.

Comments Cardiovascular adjustments. Adjustments to maternal hemorrhage have been described by Romney, Gabel, and Takeda,” Boba, Plotz, and Linkie,‘* Mofid, and Greiss,‘” and more recently by Brinkman, Assali.‘” It is generally admitted that maternal heniorrhage induces fetal asphyxia by a fall of uterine blood flow and a decrease in uteroplacental oxygen transfer. The amount of blood removed in our experiment was comparable to that of Brinkman, Mofid, and Assali.” We chose empirically to withdraw 800 ml. of maternal blood during a period of 30 minutes because in preliminary experiments we only observed a drop in fetal heart rate after the withdrawal of 600 to 700 ml. of blood. Progressive removal of approximately 14 to 20 per cent of maternal blood volume produced significant slowing of fetal heart rate, which was maintained for half an hour and accompanied by a rise in arterial pressure. Thereafter, the fetal arterial pressure

came back to control levels, but a significant tachycardia persisted until blood was reinfused to the mother. These results are in contrast with those observed h> previous workers who did not find neither tachycardia nor hypertension. Their studies were made during acute experiments, the ewe being anesthetized and al.tificially ventilated. I4 All these factors may explain the. different adjustments of the fetus. They are more in agreement wrth the cardiovascular adjustment observed in fetal lambs near term during hypoxia illduced by giving a low-oxygen gas mixture to the ewe to breathe in chronic experiments7 although the bradycardia observed by us was of greater duration. It was shown that fetal hypoxia acts on the aortic chemoreceptors to induce a rise in fetal arterial pressure and bradycardia which, in turn,‘” induces a diminution of cardiac output and a simultaneous rcdistribution of regional blood flow towards the upper part of the body (brain and heart) with a maintained umhilical circulation. Fetal tachycardia. The tachycardia which lbllowed may be considered as a period of adaptation perhaps due to an increase of plasma catecholamines. At the time of tachycardia, maternal arterial pressure rose progressively to control levels, and the fetal blood gas tensions returned to control levels. However, a persistent low pH reflected a metabolic acidemia which corrected itself during the reinfusion of maternal blood. The fetal tachycardia persisted as long as the metabolic acidemia. As could be expected. there was a significant increase in maternal plasma cortCsrtl after the withdrawal of blood. The same phenomenon was observed in the fetus, although the rise in fetal cortisol was not as striking as that of the mother. A smaller change in plasma corticosteroids was observed by Boddy and associates’” during anoxia. At the time of blood reinfusion to tllr mother, fetal cortisol levels returned to control levels in all but two animals. It is not likely that the increase in fetal plasma cortisol could be due to a transplacental transfer because of the relative impermeability of the ovine placenta to cortisol in the absence of anesthesia. Moreover, we observed the same rise in fetal cortisol levels after partial cord clamping during chronic- preparations (unpublished observations). Tracheal fluid flow. The fluid flow was in the same order of magnitude as the data published by Merlet and associates17 and Adamson and associates.” The interesting fact was that it did not appear modified by the moderate asphyxia although a decrease in pulmonary blood flow could have been expected. Fetal breathing. Boddy and associates7 were first to

Volume

127

Number

5

demonstrate that fetal breathing movements are abolished by hypoxemia in sheep. Hypoxemia in the fetus was induced by giving the ewe a low-oxygen mixture (9 per cent 02 with 3 per cent COB in N2) to breathe. Not only were fetal breathing movements abolished during the 60 minutes of hypoxia, but control levels were not reached before 6 hours after the end of hypoxia. In three of these 22 experiments where a small rise in Pacoz (5 mm. Hg) was present, the fetal breathing did not cease during hypoxernia. Hypercapnia alone increased the proportion of time occupied by breathing movements. Our experiments are not comparable to those of these authors because changes in fetal blood gases and pH were induced by maternal hemorrhage. The fall of Paop and pH was, however, similar to their changes, but moderate hypercapnia was also present (mean 38 + 5 Torr). Despite this, fetal breathing ceased during fetal asphyxia as with hypoxemia alone, but returned to control levels very quickly. When fetal breathing resumed,a normal pattern in our experiments, fetal blood pH remained low for more than 1 hour, although in the experiments of Boddy and associates7 where the pH values were quite similar after hypoxemia, fetal breathing remained scarce during 6 hours. This discrepancy is difficult to explain.

Fetal

breathing

and adaptation

to maternal

hemorrhage

511

Although we have also shown a lack of fetal breathing during moderate fetal asphyxia, no more alterations of the breathing pattern were observed in the “recuperation” phase when fetal tachycardia and lowered pH were still present. In our experiment, instantaneous heart rate was a better index of fetal well being than fetal breathing. The interest of fetal heart rate monitoring before labor have recently been stressed.” Only careful clinical work, recording in the same period fetal breathing movements and instantaneous heart rate, may settle the discrepancy. It was striking to observe the fetal lamb’s good toleration of a maternal hemorrhage comprising 14 to 20 per cent of her blood volume, although it resulted in a fall of 0.07 pH units in fetal arterial blood. However, the pH remained in the normal range (7.29 + 0.07) by clinical standards. In accordance with these minor changes, we observed no changes in the levels of fetal plasma arginin vasopressin through the experiment.” It is generally accepted that maternal hypotension is poorly tolerated by the human fetus. However, under the conditions of our experiments, the fetal lamb was able to quickly recover after a 30 per cent drop in mean maternal blood pressure. We wish to thank Miss Nathalie lent technical assistance.

Noyer for her excel-

REFERENCES 1. Merlet, C., Hoerter, J., Devilleneuve, C., and Tchobroutsky, C.: Mise en e’vidence de mouvementS respiratoires chez Ie foetus d’agneau in u&o au tours du dernier mois de la gestation, C. R. Acad. Sci. (Paris) 270: 2462, 1970. 2. Merlet, C., Hoerter, J., and Tchobroutsky. C.: Respiratory movements in fetal lambs and rabbits in utero, Twenty-fifth International Congress of Physiological Sciences 9: 1146, 1971. 3. Dawes. G. S., Fox, H. E., Leduc, B. M., et al.: Respiratory movements and paradoxical sleep in the foetal lamb, J. Physiol. 210: 47, 1970. 4. Dawes, G. S., Fox, H. E., Leduc, B. M.. et al.: Respiratory movements and rapid eye movement sleep in the foetal lamb, J. Physiol. 2iO: 110, 1972. K., and Robinson. 1. S.: External method for de5. Boddv, tect& of fetal breathing in utero, Lancet 4: 123 1, 197 1. 6. Martin, C. B.. Murata. Y.. Petrie, R. H., et al.: Respiratory movements in fetal rhesus monkeys, AM. J. OBSTET. GYNECOL. 119: 7, 939, 1974. 7. Boddy. K.. Dawes. G. S., Fisher R., et al.: Foetal respiratory movements, electrocortical and cardiovascular responses to hypoxaemia and hypercapnia in sheep, J. Physiol. 243: 599, 1974. 8. Gennser, G., Marsal, K., and Brantmark, B.: Maternal smoking and fetal breathing movements, AM. J. OBSTET. GYNECOL. 123: 861, 1975. 9. Blunt, M. H.: The blood of sheep. Composition and function, Berlin, 1975, Springer-Verlag. 10. Murphy, B. E. P.: Some studies of the protein binding of

steroids

and

ultramicro

their

by competitive crinol. Metab. 11.

to the routine

micro

and

protein binding 27: 973, 1967.

radioassay,

J. Clin.

Endo-

Romney, S. L., Gabel, P. V., and Takeda, Y.: Experimental hemorrhage in late pregnancy, AM. J. OBSTET. GYNECOL.

12.

applications

measurement of various steroids in body fluids

87: 5, 636,

1963.

Boba, A., Plotz, E., and Linkie, D. M.: Effect of atropine

on fetal bradycardia and arterial oxygenation: Experimental study in the dog during graded hemorrhage and following vasopressor administration, Surgery 58: I, 267, 1965. 13. Greiss. F. C.: Uterine vascular response to hemorrhage during pregnancy, Obstet. Gynecol. 27: 4, 549, 1966. 14. Brinkman, III, C. R.. Mofid, M., and Assali, N. S.: Circulatory shock in pregnant sheep, AM. J. OBSTET. GYNECOL. 118: 1, 77, 1974. 15. Rudolph, A. M., and Heymann, M. A.: Cardiac output in the fetal iamb: The effects of spontaneous and induced changes of heart rate on right and left ventricular output, AM. J. OBSTET. GYNECOL. 124: 2, 183, 1976. 16. Boddy, K., Jones, C. T., Mantell, C., et al.: Changes in plasma ACTH and corticosteroid of rhe maternal and fetal sheep during hypoxia. Endocrinol. 94, 588, 1974. 17. Merlet, C., Hoerter, J., Devilleneuve, Ch.. and Tchobroutsky, C.: Mise en Pvidence de mouvements re-

spiratoires chez le foetus d’agneau in ut&o, J. Physiol. 62(Supp. 18. Adamson,

3): 416, 1970. T. M., Brodecky.

V., Lambert,

T. F., et al.:

512

Toubas et al.

The production and composition of lung liquid in the in utero foetal lamb. Foetal and Neonatal Physiology, Proceedings of the Sir Joseph Barcroft Centenary Svmposium, Cambridge University Press 208: 1973. 19. Goupil. F.. and Sureau. C.: Rythme cardiaque toetal pendant la grosscsse. Journkes Parisiennes de PPdiatrie. Flammarion, Paris, 197.5, p. 1 18.

20. Czernichow, P., Basmaciogullari. A.. ‘l‘ouhas, P., VI al.: .4rginine vasopressine (AVP) secretion and placental transfer in maternal and fetal sheep, Rotterdam, 10’76. accepted by European Soriet~ of Pediatric Rcwarch. abstract to be published in Pedlatr. Res.