- Email: [email protected]
Mechanisms of late decelerations in the fetal heart rate
EUROP. J. OBSTET. GYNEC. REPROD. BIOL., 1979,9/6, o Elsevier/North-Holland Biomedical Press
361-373
Mechanisms of late decelerations in the fetal heart rate. A study with autonomic blocking agents in fetal lambs
C.B. Martin, Jr., J. de Haan *, B. van der Wildt, H.W. Jongsma, A. Dieleman and T.H.M. Arts Department
of Obstetrics and Gynecology,
Catholic University and Sint Radboudziekenhuis,
Nijmegen, The Netherlands
Accepted for publication 11 June 1979 MARTIN, C.B., Jr., de HAAN, J., van der WILDT, B., JONGSMA, H.W., DIELEMAN, A. and ARTS, T.H.M. (1979): Mechanisms of late decelerations in the fetal heart rate. A study with autonomic blocking agents in fetal lambs. Europ. J. Ubstet. Gynec. reprod. Biol., 9/6,361-373.
Fetal heart rate decelerations resembling the late deceleration FHR pattern were produced in fetal sheep by periodic occlusion of the maternal common hypogastric artery for 30-60 sec. Transient fetal hypertension also occurred during the occlusions. Alpha-adrenergic blockade with phentolamine eliminated or markedly reduced the hypertensive response. FHR decelerations still occurred intermittently with some occlusions; however, their character was greatly altered. After parasympathetic blockade with atropine, the decelerations were replaced by periodic FHR accelerations during the occlusions. These accelerations were, in turn, eliminated by the beta-adrenergic blocking agent, propranolol. In the presence of combined parasympathetic, alpha- and beta-adrenergic blockade, the FHR remained essentially constant during the hypogastric artery occlusions in non-acidemic fetuses. FHR decelerations persisted after parasympathetic or total autonomic blockade when the fetuses were significantly hypoxic, as judged by depressed arterial blood pH and base excess values. Beat-to-beat variability of the baseline FHR persisted in the face of severe hypoxia and acidosis. These observations demonstrate that reflex mechanisms are involved importantly in the genesis of late deceleration FHR patterns in the acutely hypoxemic fetus, but that direct depression of myocardial rhythmicity becomes a factor as hypoxic acidosis develops. uterine blood flow; adrenergic blockade; cholinergic blockade; beat-to-beat phentolamine;
The
Introduction
FHR The late deceleration
FHR pattern
type II dip (Caldeyro-Barcia accepted
variability; fetal hypoxia, fetal acidosis, fetal pH,
propranolol; atropine
as indicating
(Hon,
1968) or
changes
episodes result
of hypoxia from
precipitating
periodic
decreases
the in ma-
ternal placental blood flow caused by uterine contractions. A background of chronically impaired placental
et al., 1966)‘is generally
the presence
periodic
exchange
of fetal hypoxia.
is present
in some instances,
there is an acute reduction flow as a result of, e.g., uterine hypertonus.
* Present address: Department of Obstetrics and Gynecology, State University of South Limburg and Sint Annadalziekenhuis, Maastricht, the Netherlands.
With respect 361
while in others
in baseline uterine blood maternal hypotension or
to the mechanism
producing
the FHR
362
deceleration, most emphasis has been placed on direct hypoxic depression of myocardial rhythmicity (Hon, 1962, 1968; James et al., 1972; Myers et al., 1973). On the other hand, both Hon (1962) and MendezBauer et al. (1963) have published FHR tracings showing modification (but not elimination) of late FHR decelerations in human subjects following parasympathetic blockade with atropine. Studies of chronically catheterized fetal laboratory animals have demonstrated the important involvement of reflex mechanisms in the fetal cardiovascular response to acute hypoxia (Boddy et al., 1974; Evers, 1978). Systematic study of the mechanisms producing late FHR decelerations has not been carried out, however, perhaps because of the difficulty in inducing controlled rhythmic uterine activity sufficient to cause fetal hypoxia in a stabilized chronic animal preparation. This paper describes a method for simulating late FHR decelerations in the fetal lamb under chronic study conditions, and presents preliminary observations demonstrating the role of reflex mechanisms in the genesis of late FHR decelerations.
Material and methods
Fetal vascular catheters (polyvinyl), ECG electrodes, and an intra-amniotic catheter were implanted in gravid ewes (2 Texel, 1 Dorset) between 122 and 144 days of pregnancy. A single fetus was present in each instance. Standard fetal surgical techniques were employed under general, halothane anesthesia. One catheter was passed into the fetal abdominal aorta via a femoral artery, and another was inserted into the upper abdominal or lower thoracic segment of the fetal inferior vena cava via a femoral vein. In addition, an inflatable cuff occluder (Rhodes Medical) was placed around the maternal common hypogastric artery trunk, and a non-cannulating electromagnetic flow probe (Skalar Instruments) was placed on the hypogastric artery ipsilateral to the uterine horn predominantly containing the fetus. Postoperatively, the animals were placed in a movable stall and permitted to recover. They received food and water ad libitum. The experimental observations were carried out 48 h to 9 days after surgery (at 127-148 days of pregnancy). Each animal was studied 3 times, for a
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
total of 9 experiments. The animals were brought in their stalls to the laboratory; the catheters, electrodes and flow probe were connected to appropriate transducers and couplers (Hewlett-Packard, 8800 series) and flow meter (Model 601, Skalar Instruments). The analog signals representing fetal ECG, FHR, fetal arterial blood pressure, intra-amniotic pressure and hypogastric artery blood flow were recorded on a multichannel strip-chart recorder (Model 1065, Hewlett-Packard), and simultaneously on magnetic tape. After a stable baseline recording period of at least 30 min, a fetal arterial blood sample was drawn for pH and respiratory gas tension analysis, and the simulation of uterine contractions was begun. The effect of uterine contractions in reducing maternal placental blood flow, and thus oxygen delivery to the fetoplacental unit, was simulated by periodically inflating the occluder around the maternal common hypogastric artery trunk, thus interrupting the major source of blood flow to the uterus. Occlusion was maintained for periods of from 30 to 60 set, and was repeated at an interval of 2.5 min from the beginning of one occlusion to the start of the following compression. To increase the degree of hypoxia in some experiments, the occluder was left partially filled between the total occlusions. Additional, fetal arterial blood samples were obtained at intervals to monitor the fetal acid-base status. These samples were taken either during the last seconds of the occlusion, or at 1 min following release of the occlusion. Samples were withdrawn into heparinized plastic syringes and refrigerated until analyzed, within 2 h of withdrawal. Previous observations in this laboratory have shown that only minimal changes occur in the pH and gas tensions of samples treated in this fashion. The pH and gas tensions were measured with micro blood analysis equipment (Model 165 pH and blood gas analyser, Corning Instruments). To investigate the role of reflex mechanisms in the fetal cardiovascular changes accompanying the periodic maternal artery occlusions, the various divisions of the fetal autonomic nervous system were blocked by pharmacological means. Parasympathetic postganglionic impulses wer~eblocked by atropine, 1 mg/ kg estimated fetal weight (EFW); alpha-adrenergic effects were antagonized with phentolamine, 2.5-5
C.B. Martin, Jr. et al.: Mechanisms of latecieceierations
mg/kg EFW; and beta-adrenergic influences, by propranolol, 1 mg/kg EFW. The completeness of blockade was tested by means of acetylcholine, 10 /&kg EFW, noradrenaline, 5-15 I.cg/kg EFW, and isoproterenol, 0.1 pg/kg EFW, respectively. The adorlomic agonists and antagonists were administered to the fetus directly via the venous catheter. Flockade of one or more autonomic divisions was established early in the course of 5 trials, after a short series of preliminary occlusions (usually 5-7) had produced repetitive decelerations, and before the development of fetal acidosis (pH > 7.30, base excess > -6.0 mmol in each instance). Autonomic blockade was established on 2 occasions in acidotic fetuses: cholinergic blockade alone in one fetus with pH 6.96 after a long serie? of occlusions, and triple blockade in another fetus with pH 7.23. Experimental results were analyzed by inspection of the analog records. Quantitative data regarding heart rate and pressure were read directly from the records. Fetal blood pressure values were corrected by subtracting the simultaneous amniotic fluid pressure readings. The beat-to-beat variability of the FHR was determined from the tape-recorded FECG signal, using the interval difference index (IDI) calculation of Jongsma et al. (1978).
Results The periodic occlusions of the common hypogastric artery produced decelerations in the FHR (Fig. 1A and B). FHR decelerations occurred with the first occlusion in 6 of the 9 experiments, and were consistently present from the 4th occlusion onward in 8 experiments. Transient accelerations of the FHR accompanied the initial occlusions in all 3 instances in which the appearance of FHR decelerations was delayed. This sequence was observed twice in one animal and once in another. In each animal, the beginning fetal PO, value was higher in the study sessions in which the initial FHR response was acceleration than in those sessions in which FHR deceleration
363
accompanied the first occlusion. A similar pattern was not observed for pH and Pco,. Consistent FHR decelerations could not be produced in one experiment despite 2 h of repetitive periodic occlusions of the maternal hypogastric trunk, although marked beat-to-beet arrhythmia and intermittent clecelerations did occur with the occlusions. Consistent FHR decelerations were produced on two other occasions in this animal, in association with lower initial and post-occlusion PO2 levels. The FHR decelerations produced by occlusion of the common hypogastric artery in the other experiments had the shape of a wide V or U in non-acidemic fetuses (Fig. 1). Onset of the deceleration occurred usually 15-25 set after the start of the occlusion, although there was sometimes considerable variation between successive occlusions in the interval to onset of deceleration. The lowest FHR was reached 5-12 set after the end of the occlusion. Recovery of the FHR began a few seconds later, with return to the baseline usually 35-l 00 set after release of the occlusion. Transient acceleration of the FHR during recovery occurred inconsistently after lo-15% of the occlusions. Fetal blood pressure rose transiently during the occlusions (Fig. 1). The onset of detectable hypertension and the appearance of cardiac slowing most often coincided within l-2 sec. With some occlusions, however, these two events differed by as much as 15-20 set and occurred in either sequence. In nonacidemic fetuses, the blood pressure rise was progressive, reaching a maximum level shortly after the end of the occlusion. The times at which the highest blood pressure and lowest FHR levels were attain&l corresponded within 4 set (2 mm on the record) in 60% of the occlusions in non-acidemic fetuses. In the remaining instances, greater deviations of as much as 30 set were observed. In 2 of the 3 individual fetal lambs, and in the pooled data from all 3 fetuses, there was a significant (P< 0.01) linear correlation between the amount of increase in systolic blood pressure and the amount of decrease in the FHR for the occlusions carried out
Fig. 1. Periodic FHR decelerations and hypertension in response to periodic occlusion of the maternal common hypogastric artery trunk in 2 fetal lambs. In panel A, the sustained hypertension and tachycardia following occlusion no. 4 suggest catecholamine release by the fetal adrenal at this point. In this and subsequent figures, the negative intrauterine pressure reflects placement of the pressure transducers at a higher level than the uterus.
PH
PO2 pC0, BE
7.40 1.7 kPa 5.6 kPa l1.3 mmol/l
fetal arterial pressure (mm Hg)
fetal heart frequency (bpm)
v-h-a utenne pressure (InrnHg)
flow uterine artery (ml/min)
FECG
ewe 58178 3 days postop.
:
gest age 128 days periodic total occlusion
,
f time(mln)
PH POP PC4 BE
0 fetal arterial pressure (mm Hg)
738 1.8 kPa 3.3 kPa -7.7 m molll
I
100,
1
fetal heart frequency (b.pm 1
intra uterine pressure (mm Hg)
flow uterine artery (ml/mm)
FECG I,
ewe 92-70 8 days postop.
I
I
: :
I
gest. age: 129 days periodic total occlusion
I
1
I time (min)
1
1
I
I
1
365
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
prior to the development of acidemia or the institution of autonomic blockade. This was true both for the absolute data, expressed in mm Hg and beats/min, and for the proportionate data calculated as percent change. The systolic blood pressure and average baseline FHR immediately preceding each occlusion were used as the reference values for the calculations of absolute and percent change during the occlusion. The regression line for the pooled absolute data was AFHR=-12.6 -3.1 (ASBP), rc0.74, n=60. The slopes for the regression lines for the individual fetuses also indicated a decrease of approximately 3 beats/min in FHR for each mm Hg increase in systolic blood pressure. Alpha-adrenergic blockade with phentolamine prevented or largely eliminated the progressive hypertensive response of the fetus to periodic interruption of the uterine blood flow (Fig. 2). Although FHR
decelerations still occurred with some of the occlusions in the absence of hypertension, the pattern of these decelerations was markedly different from that of the decelerations before alpha-blockade (Fig. 2, occlusion 6). In non-acidemic fetuses, cholinergic blockade with atropine eliminated the FHR deceleration response to periodic interruption of the uterine blood flow of 60 set duration. Instead, periodic acceleration of the FHR occurred, beginning usually 15-30 set after the start of occlusion (Fig. 3). These FHR accelerations occurred both in the presence and absence of concomitant alpha-adrenergic blockade. Beta-adrenergic blockade with propranolol prevented or markedly reduced the FHR acceleration response of the atropinized fetus to the occlusions (Fig. 4). Small (5-20 bpm) periodic increases in the FHR persisted in the experiment illustrated, presumPH PO2 pC02 BE
t&al artermt 100
735 14 6.1 -07
kPa kPa mmol/l
1
pressure (mm Hg) 50
fetal heart trequency
250
(bpm)
150
lntra
uterine
pressure (mmHg)
flow uterine artery (mlfmcn)
0
20
500
0
FECG
ewe 56-76 3 days postop.
,
gest age 126 days periodic total oc&sion
a-adrenergic
blockade
time(mln)
Fig. 2. Modification of the fetal response pattern following alpha-adrenergic blockade with phentolamine. Compare with Fig. 1A. The phasic fetal hypertension is absent. A V-shaped deceleration occurred with fetal movement during occlusion 3, and brief decelerations occurred after occlusion 4 and during occlusion 6.
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
366
PH
7.33 1.2 kPa PC02 5.1 kPa BE -48 mmol/I PO2
fetal
arterial 100 7
1
pressure (mm
Hg)
fetal heart frequency (bp.m)
50
250 -7
M
I
150-
I I
50 -
intra uterine PWSS"re (mmHg)
flow uterine artery
1
0 2.
500 -
(ml/min)
FECG
ewe 58-78 3 days wstop.
1
gest age 120 days periodic total occlusion
a-adrenergic , cholinergic blockade
I
time (min) Fig. 3. Combined alphaadrenergic and cholinergic blockade in the same fetus shown in Figs. 1A and 2. The FHR now responds to the occlusions with periodic acceleration.
ably reflecting incomplete blockade of locally liberated catecholamines. In other experiments, however, the accelerations were eliminated completely by betaadrenergic blockade. Changes occurred in the fetal blood pressure and, to a lesser extent, in the FHR responses to periodic hypoxemia when the series of occlusions was extended to produce fetal acidosis (Fig. SA-C). As the pH decreased to below 7.30, the fetal hypertensive response lost its progressive character and tended to reach a plateau early after the beginning of the occlusion (Fig. 5A). The FHRcontinued to decelerate throughout the period of hypoxemia, however, with the nadir occurring at or after the end of the occlusion. With increasing acidosis, there was further attenuation of the fetal blood pressure response (Fig.
5B), and the eventual appearance of a biphasic pattern with slight hypotension following an initial increase in pressure (Fig. SC). The FHR decelerations became broader and more U-shaped, and the low point of the deceleration tended to occur earlier during the occlusion period (Fig. 5B and C). Administration of atropine to the severely acidotic fetus (pH 6.96) resulted in an increase in the baseline FHR, but no change in the amount of the deceleration (Fig. 6). The interval to the onset of deceleration decreased slightly. In the other acidotic fetus, administration of atropine after combined blockade of both adrenergic systems resulted in a transient decrease in the amount of deceleration, but this returned to the previous magnitude by the third occlusion after atropine. The fetal pH decreased
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
361 7.33 PH 1.1 kPa PO2 pCOz 50 kPa BE -5.5 mmol/l 1
fetal arterr pressure (mm Hg)
loo 50 0
fetal heart frequency (bpm)
250 150 50
lntra uterlnc pressure (mm+@)
I
0 1
/
flow utenne artery (mllmln)
500 n
FECG ewe 58-78 3 days postop
I
:
gest. age 128 days perlodu total occIwon
0. and P adrenergic
, chollnerglc
blockade
I
time (mln)
Fig. 4. Fetal response to periodic occlusion of the maternal common hypogastric artery trunk in the presence of combined alphaand beta-adrenergic and cholinergic blockade. There remains a slight FHR acceleration during the occlusions, and a small hypertensive response following some of them, probably reflecting incomplete blockade.
rapidly in the face of triple autonomic blockade, making further interpretation uncertain. Beat-to-beat variability of the FHR, as measured by the interval difference index (IDI) (Jongsma et al., 1978) persisted in the face of increasing hypoxemic acidosis in the fetus subjected to the prolonged series of repeated occlusions (Figs. 1B and 5A-C). The median ID1 during the 30-min control period immediately before this experiment was 19 units. In the experimental period, the median IDI, calculated from the 30-set segments immediately preceding each of the 91 occlusions, was 34. During the last 30 min of this experiment, the fetal pH decreased from 7.10 to 6.96. The median ID1 during this period was 25. The
ID1 decreased markedly, to 3.3, following administration of atropine to the fetus (Fig. 6). Discussion
The categorization of transient periodic decreases in the FHR as early,variable or late decelerations (Hon, 1968) or as type I or type II dips (Caldeyro-Barcia et al., 1966), was based in large part on the temporal relationship between the FHR decelerations and the uterine contractions provoking them. In the present study, actual uterine contractions were not produced; however, one of the important effects of uterine contractions, the transient reduction of the
Method
7.27 1.1 kPa pCOn 51 kPa BE -BBmmolll
1
fetal arterfal pressure (mm Hg)
fetal heart frequency (b.o.m )
intra uterine pressure (mm Hg)
flow uterfne artery (mllmln)
FECG I
,
1
ewe 92-78 8 days postop.
I
I
Y ,‘I,
I...,,
I..,
,
,
‘1
I..,
: gest. age: 129 days : continuing partial occlusion, periodic total occlusion
1
1
,
!
I
,
I
,
I
time (min) 7.22 PH 1.2 kPa PO2 pC02 5.6 kPa -10.3 mmolll BE
B fetal arterial pressure (mm Hg)
100 1
50 0 fetal heart frequency (b.p.m )
250
uterme pressure (mml-fg) intra
25
flow uterine artery (mllmin)
FECG
I:
1
C.B. Martin, Jr. et al.: Mechanisms of Iate decelerations
369
PH PO, PC4 BE
7.14 1.5 kPa 6.3 kPa -131 mmolll
1
fetal heart frequency (b.pm)
lntra uterine pressure (mm Hg) 25
flow uterine artery (ml/mln)
600
FECG ewe 92-78 8 days postop. I
: gest. age: 129 days : continuing partial occlusion, periodic total occlusion
I
,
!
I
time (min)
Fig. 5. Fetal response to an extended series of periodic total occlusions of the maternal common hypogastric artery trunk combined with continuing partial occlusion of this vessel. A-C: sections of the record at increasing degrees of fetal acidosis. Note the progressive disappearance of the fetal hypertensive response, and the occurrence of slight diastolic hypotension in C. The FHR decelerations become wider and more symmetrical with increasing acidosis. Compare with Fig. 1B.
maternal placental blood flow, was mimicked. The term ‘late decelerations’, as used in this report, refers therefore to the temporal relationship between the FHR decelerations and the periodic interruptions of the uteroplacental blood flow. The role played by uterine contractions in the production of late periodic FHR decelerations is the periodic reduction of maternal placental blood flow, and thus fetal oxygenation, during the contractions. If this thesis is accepted, as seems generally to be the case, then one must also accept that FHR decelerations resulting from the periodic reduction of uteroplacental blood flow by another method, specifically in these experiments periodic occlusion of the major
arterial supply to the uterus, are equivalent in their pathophysiologic mechanisms to ‘natural’ late decelerations. Chronic reduction in resting uteroplacental blood flow occurring as a result of placental or uteroplacental vascular pathology, which often sets the stage for late decelerations in labor, can be mimicked to some extent by continuing partial occlusion of the uterine arterial supply ~between the periodic episodes of greater occlusion. In this latter case, however, the simulation in the present experiments was only partial; for the partial occlusion was maintained only for l-2 h, whereas the condition may endure several weeks in complicated human pregnancies. The present methods, therefore, provide
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
370
PH
6.96
fetal arterial pressure (mm Hg)
fetal heart frequency (b.p.m )
irkra uterine pressure (mm Hg)
flow uterine artery (mllmtn)
FECG
ewe 92
I
gest. age : 129 days continuing partial occlusion, periodic total occlusion
-70
1
I
I
I
9
I
I
I
I
time (min)
Fig. 6. Fetal response to cholinergic blockade during severe acidosis. The baseline FHR increases following atropine, but the amount of FHR deceleration is unchanged. Beat-to-beat variability is present at the beginning of the panel and disappears after atropine. Note that the FHR scale has been compressed in this figure. a means for studying the pathophysiologic mechanisms involved in late decelerations occurring with acute fetal distress, both before and after the development of fetal acidosis. Myers et al. (1973) employed occlusion of the maternal abdominal aorta as a means of producing FHR decelerations in fetal monkeys. However, they do not seem to have used the technique in a repetitive, periodic fashion, as in the present study. Moreover, their experiments were carried out immediately after surgery and under conditions of maternal anesthesia. There is a possible source of discrepancy between the mechanisms activated by periodic occlusion of the uterine arterial ,supply and those involved when uterine contractions interrupt maternal placental blood flow. Power and Longo (1973) have postulated
the occurrence of ‘sluice flow’ in the placenta. According to this concept, there should be increased resistance to flow in the fetal placental circulation during contractions as a result of compression of fetal capillaries by an increasing volume of blood trapped in the maternal placental blood spaces. The reverse would occur, of course, if only the maternal arterial supply were occluded. However, studies specifically directed toward detection of a sluice flow mechanism in chronically instrumented pregnant sheep failed to confirm its presence (Berman et al., 1976). Moreover, if sluice flow were a significant feature of placental circulation in sheep, the fetuses with triple autonomic blockade in the present experiments should have demonstrated hypotension during the maternal arterial occlusions. This was not observed (Fig. 4).
371
C.B. Martin, Jr. et al.: Mechanisms of hte decelerations
Although it has Mechanisms of late deceleration been recognized that elements of both reflex cardiac deceleration and direct depression of myocardial rhythmicity contribute to the late deceleration FHR pattern, there seems to be little agreement among authorities as to the relative importance of each mechanism. Hon (1962, 1968) has placed major emphasis on hypoxic depression of the myocardium, whereas Caldeyro-Barcia and associates (cf. CaldeyroBarcia et al., 1966; Mendez-Bauer et al., 1963) concluded that their type II dips resulted largely from transient increases in vagal cardiodecelerator tone. The former viewpoint has received support from observations carried out on fetal monkeys under acute experimental conditions, in which late decelerations were accompanied by hypotension and were quantitatively related to the degree of fetal hypoxemia (James et al., 1972; Myers et al., 1973). In the chronically catheterized fetal lamb, on the other hand, the initial bradycardia associated with acute hypoxemia can be converted to a tachycardiac response by vagotomy (Boddy et al., 1974) or cholinergic blockade (Parer, 1977; Evers, 1978). The bradycardia response to acute hypoxemia can also be delayed substantially by atropine in fetal monkeys, when the observations are carried out under conditions of chronic study (Martin, Murata and Ikenoue, unpublished observations). Su and Friedman (1973) have shown that strips of fetal myocardium are relatively resistant to chronotropic depression by hypoxia, in comparison to adult cardiac muscle. These and similar observations have led to the suggestion (Martin and Gingerich, 1976; Martin, 1978) that late FHR decelerations might be produced by either reflex or direct mechanisms, and more often by both acting together, with the relative contribution of each mechanism at a particular time in a specific fetus depending on the degree and duration of fetal oxygen lack. The observations described in this report appear to confirm that suggestion. The non-acidemic fetuses responded to periodic reductions of uterine blood flow by periodic hypertension and FHR decelerations. Elimination of the fetal hypertensive response by alpha-adrenergic blockade markedly modified the FHR decelerations, but did not eliminate them completely. Cholinergic blockade converted the periodic decelerations to accelerations, which in turn were blocked by propranolol. When all 3 autonomic mech-
anisms were blocked, the non-acidemic fetuses showed essentially no heart rate response to periodic hypoxemia. In contrast, blockade of the efferent limb of the cardiodecelerator reflex did not alter the degree of FHR deceleration in response to periodic occlusion of the maternal hypogastric artery trunk in fetuses made severely acidotic by a long series of such occlusions. The alpha-adrenergic blocker used in these experiments, phentolamine, has a variety of additional actions including beta-adrenergic, cholinergic and histaminergic effects (Goodman and Gilman, 1975). Since during the occlusions after phentolamine blockade there was not only a virtual absence of FHR decelerations, but also absence of phasic fetal hypertension, it seems most likely that these changes were related: that the absence of FHR decelerations was the result of absence of phasic blood pressure increases during the hypoxic episode. The possibility remains, however, that one or more of the other actions of phentolamine may have played a role in altering the fetal response to transient interruptions of the uterine blood flow. The reflex pathways presumably involved in ‘acute’ late decelerations are summarized in Fig. 7. The
hypoxaemia /\
J ch
7
sympathetic
l/
\
yre\
\
center
2l 4
-- --a
blockode
vasoconstrlctlon _.__-6
I
blockade
I + blood
pressure
baro
receptor
fetal
heart
3)
4/
I
!
t
rate
Fig. 7. Schematic diagram of the reflex pathways presumed to be invdlved in the production of late decelerations in the non-acidemic fetus.
312
dominant mechanism would appear to be chemoreceptor-mediated reflex vasoconstriction (Dawes et al., 1968, 1969) producing hypertension and, in turn, baroreflex-induced cardiac slowing (Shinebourne et al., 1972). That chemoreceptor (peripheral or central) stimulation by hypoxia can also activate a cardiodecelerator reflex primarily, reinforcing the indirect vasoconstriction-baroreceptor mechanism, is suggested by the frequent onset of heart rate deceleration several seconds before the appearance of any blood pressure increase in the unblocked fetuses. This is also demonstrated by the occurrence of some cardiac slowing with occlusions after the hypertension response had been blocked by phentolamine. The reflex nature of these latter decelerations is apparent from their disappearance following cholinergic blockade. A cardiodecelerator response to aortic chemoreceptor stimulation has been previously demonstrated in the fetal lamb by Goodlin and Rudolph (1972). Although cardiodecelerator mechanisms predominate in the response of the non-acidemic fetal lamb to hypoxemia, an adrenergic cardioaccelerator component is also present. This is most clearly demonstrated by the periodic accelerations accompanying interruption of the uterine blood flow following cholinergic blockade. It is also shown by the acceleration response of 2 fetuses to the initial occlusions. The cardioaccelerator response to acute hypoxemia predominates in immature fetallambs (e.g. 97-102 days) (Boddy et al., 1974), and was also observed in older fetal lambs studied acutely under conditions of maternal systemic anesthesia(e.g. Dawes et al., 1969). Involvement of beta-adrenergic mechanisms in the accelerator response was confirmed in the present investigation by the elimination or reduction of the accelerations by propranolol. The present experiments substantiate the role of direct hypoxic depression of myocardial chronotropism in late decelerations under conditions of severe fetal acidosis. Here the reflex component seems to be absent or minimal, for choline@ blockade did not alter the amount of the decelerations in the severely acidotic fetal lamb. The absence of a reflex element in these decelerations was not due to paralysis of the parasympathetic system, however; for persisting vagal activity was demonstrated by elevation of the baseline FHR following atropine
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
administration, and by the continued presence of beat-to-beat variability in the baseline FHR which disappeared with cholinergic blockade. Perhaps the absence of phasic hypertension with the occlusions during severe acidosis accounts for the absence of a reflex component in those late decelerations. Our findings thus indicate that reflex mechanisms accounted for all of the ‘late decelerations’ produced by periodic reduction of uteroplacental blood flow in fetuses with normal pH, and essentially none of the deceleration during severe acidosis. The point at which direct hypoxic depression of cardiac rhythmicity begins to play a role was not defined; however, our observations suggest that this may occur quite early in the course of fetal distress - at pH values between 7.30 and 7.25. In this range, the fetal blood pressure response to periodic hypoxemia lost its progressive character, although the FHR continued to decrease progressively throughout the occlusion. Thus, ‘pure reflex’ late decelerations may be restricted to situations of short-term, acute hypoxemia in previously unstressed fetuses. Further definition of the relative contributions of reflex and direct mechanisms to late decelerations between the extremes of fetal condition would require the institution of autonomic blockade at successively lower pH values in an extensive series of experiments; for the inhibition of circulatory reflexes, by interfering with the preferential perfusion of the heart and brain during hypoxemia (Cohn et al., 1974; Peeters, 1978), might result in an erroneous conclusion should the fetus simply be made progressively acidotic by repeated uterine artery occlusions during pre-existing blockade. In addition, the point of appearance of direct hypoxic depression of cardiac chronotropism may vary considerably among individual fetuses depending, for example, on the relative duration of hypoxemia and recovery, the lowest oxygen content reached during the occlusion, and perhaps also such factors as cardiac glycogen stores and blood glucose levels. The persistence of beat-to-beat arrhythmia in the FHR of our severely acidotic fetal lamb is at variance with the usual observation in clinical FHR monitoring, that beat-to-beat variability is diminished in the presence of significant hypoxemic fetal acidosis. Our observation is, however, in general agreement with the finding of Dalton et al. (1977) that beat-to-beat
373
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
variability increased during hypoxemic acidosis in the fetal lamb, even when the hypoxemia was prolonged to produce increasing acidosis, and did not fall below the pre-hypoxemic value until the pH was 6.9. That vagal mechanisms still account for the beat-to-beat arrhythmia during severe hypoxemic acidosis was demonstrated by the prompt disappearance of baseline variability following administration of atropine to the fetus in the present study (Fig. 6). Whether the apparent difference between human and ovine fetuses in the course of FHR variability during hypoxemic acidosis represents differences in the reactions of the cardiac control mechanisms, or in such factors as resistance to hypoxic depression or degree of maturity achieved by late gestation, remains to be defined. Moreover, the induction of fetal hypoxia and acidosis by relatively acute interference with the uterine blood flow or reduction of the O2 content of the gas mixture breathed by the mother may well be an incomplete simulation of the large proportion of human fetal distress that is preceded by several days or weeks of diminishing placental function.
References Berman, W., Goodlin, R.C., Heymann, M.A. and Rudolph, M.A. (1976): Relationships between pressure and flow in the umbilical and uterine circulations of the sheep. Cir&at. Res., 38, 262. Boddy, K., Dawes, G.S., Fisher, R., Pinter, S. and Robinson, J.S. (1974): Fetal respiratory movements, electrocortical and cardiovascular responses to hypoxaemia and hypercapnea in sheep. J. Physiol. (Lond.) 243, 599. Caldeyro-Barcia, R., Mendez-Bauer, C., Poseiro, J.J., Escarcena, L.A., Pose, S.V., Bienarz, J., Arnt, I., Gulin, L. and Althabe, 0. (1966): Control of human fetal heart rate during labor. In: IIre Heart and Circulation in the Newborn infant, p. 7. Editor: D.E. Cassells. Grune and Stratton, New York. Cohn, H.E., Sacks, E.J., Heymann, M.A. and Rudolph, M.A. (1974): Cardiovascular responses to hypoxemia and acidemia in fetal lambs. Amer. J. Obstet. Gynec., 120, 817.
Dalton, K.J., Dawes, G.S. and Patrick, J.E. (1977): Diurnal, respiratory and other rhythms of fetal heart rate in lambs. Amer. J. Obstet. Gynec., 127, 414.
Dawes, G.S., Lewis, B.V., Milligan, J.E., Roach, M.L. and Talner, N.S. (1968): Vasomotor responses in the hind limbs of foetal and newborn lambs to asphyxia and aortic chemoreceptor stimulation. J. Physiol. (Lond.), 195, 55.
Dawes, G.S., Duncan, S.L.B., Lewis, B.V., Merlet, C.L., Owen-Thomas, J.B. and Reeves, J.T. (1969): Hypoxemia and aortic body chemoreceptor function in foetal lambs. J. Physiol. (Lond.).
201, 105.
Evers, J.L.H. (1978): The Cardiac Pre-Ejection Period During Prenatal Life. Thesis, Faculty of Medicine, Catholic University, Nijmegen, The Netherlands. Goodlin, R.C. and Rudolph, A.M. (1972): Factors Associated with the Onset of Breathing. In: Proc. Int. Symp. Physiological Biochemistry of the Fetus. Editors: A.A. Hodari and F.G. Mariona. Thomas, Springfield. Goodman, L.S. and Gilman, A. (1975): 7’he Pharmacological Basics of Therapeutics. Macmillan, New York. Hon, E.H. (1962): Electronic evaluation of the fetal heart rate. VI. Fetal distress - a working hypothesis. Amer. J. Obstet. Gynec., 83, 333.
Hon, E.H. (1968): An Atlas of Fetal Heart Rate Patterns. Harty Press, New Haven, Conn. James, L.S., Morishima, H.O., Daniel, S.S., Bowe, E.I., Cohen, H. and Nieman, W.H. (1972): Mechanisms of late deceleration of the fetal heart rate. Amer. J. Obstet. Gynec., 113, 578.
Jongsma, H.W., van Geijn, H.P. and de Haan, J. (1978): The analysis of heart rate variability in the perinatal period. In: Computerdiagnostik in den Geburtshilfe, Proc. 1st. Symposium, FriederichSchiller Universitat, Jena, p. 249. Martin, C.B. and Gingerich, B.H. (1976): Factors affecting the fetal heart rate: genesis of fetal heart rate patterns. J. Obstet. Gynec. Nursing, 5, 30s. Martin, C.B. (1978): Regulation of the fetal heart rate and genesis of FHR patterns. Sem. Perinatol., 2, 131. Mendez-Bauer, C., Poseiro, J.J., Arellano-HernLndez, G., Zambrana, M.A. and Caldeyro-Barcia, R. (1963): Effects of atropine on the heart rate of the human fetus during labor. Amer. J. Obstet. Gynec., 85, 1033. Myers, R.E., Mueller-Heubach, E. and Adamsons, K. (1973): Predictability of the state of fetal oxygenation from a quantitative analysis of the components of late deceleration. Amer. J. Obstet. Gynec., 115, 1089. Parer, J.T. (1977): Effect of atropine on heart rate and oxygen consumption of the hypoxic fetus. Gynec. Invest., 8, 50. Peeters, L.L.H. (1978): Fetal Blood Flow at Various Levels of Oxygen. Thesis, Faculty of Medicine, Catholic University, Nijmegen, The Netherlands. Power, G.G. and Longo, L.D. (1973): Sluice flow in the placenta: maternal vascular pressure effects on fetal circulation. Amer. J. Physiol., 225, 1490. Shinebourne, E.A., Vapaavouri, E.K., Williams, R.L., Heymann, M.A. and Rudolph, M.A. (1972): The development of baroreflex activity in unanesthetized fetal and neonatal lambs. Circulat. Res., 31, 710. Su, J.Y. and Friedman, W.F. (1973): Comparison of the responses of fetal and adult cardiac muscle to hypoxia. Amer. J. Physiol., 224, 1249.
361-373
Mechanisms of late decelerations in the fetal heart rate. A study with autonomic blocking agents in fetal lambs
C.B. Martin, Jr., J. de Haan *, B. van der Wildt, H.W. Jongsma, A. Dieleman and T.H.M. Arts Department
of Obstetrics and Gynecology,
Catholic University and Sint Radboudziekenhuis,
Nijmegen, The Netherlands
Accepted for publication 11 June 1979 MARTIN, C.B., Jr., de HAAN, J., van der WILDT, B., JONGSMA, H.W., DIELEMAN, A. and ARTS, T.H.M. (1979): Mechanisms of late decelerations in the fetal heart rate. A study with autonomic blocking agents in fetal lambs. Europ. J. Ubstet. Gynec. reprod. Biol., 9/6,361-373.
Fetal heart rate decelerations resembling the late deceleration FHR pattern were produced in fetal sheep by periodic occlusion of the maternal common hypogastric artery for 30-60 sec. Transient fetal hypertension also occurred during the occlusions. Alpha-adrenergic blockade with phentolamine eliminated or markedly reduced the hypertensive response. FHR decelerations still occurred intermittently with some occlusions; however, their character was greatly altered. After parasympathetic blockade with atropine, the decelerations were replaced by periodic FHR accelerations during the occlusions. These accelerations were, in turn, eliminated by the beta-adrenergic blocking agent, propranolol. In the presence of combined parasympathetic, alpha- and beta-adrenergic blockade, the FHR remained essentially constant during the hypogastric artery occlusions in non-acidemic fetuses. FHR decelerations persisted after parasympathetic or total autonomic blockade when the fetuses were significantly hypoxic, as judged by depressed arterial blood pH and base excess values. Beat-to-beat variability of the baseline FHR persisted in the face of severe hypoxia and acidosis. These observations demonstrate that reflex mechanisms are involved importantly in the genesis of late deceleration FHR patterns in the acutely hypoxemic fetus, but that direct depression of myocardial rhythmicity becomes a factor as hypoxic acidosis develops. uterine blood flow; adrenergic blockade; cholinergic blockade; beat-to-beat phentolamine;
The
Introduction
FHR The late deceleration
FHR pattern
type II dip (Caldeyro-Barcia accepted
variability; fetal hypoxia, fetal acidosis, fetal pH,
propranolol; atropine
as indicating
(Hon,
1968) or
changes
episodes result
of hypoxia from
precipitating
periodic
decreases
the in ma-
ternal placental blood flow caused by uterine contractions. A background of chronically impaired placental
et al., 1966)‘is generally
the presence
periodic
exchange
of fetal hypoxia.
is present
in some instances,
there is an acute reduction flow as a result of, e.g., uterine hypertonus.
* Present address: Department of Obstetrics and Gynecology, State University of South Limburg and Sint Annadalziekenhuis, Maastricht, the Netherlands.
With respect 361
while in others
in baseline uterine blood maternal hypotension or
to the mechanism
producing
the FHR
362
deceleration, most emphasis has been placed on direct hypoxic depression of myocardial rhythmicity (Hon, 1962, 1968; James et al., 1972; Myers et al., 1973). On the other hand, both Hon (1962) and MendezBauer et al. (1963) have published FHR tracings showing modification (but not elimination) of late FHR decelerations in human subjects following parasympathetic blockade with atropine. Studies of chronically catheterized fetal laboratory animals have demonstrated the important involvement of reflex mechanisms in the fetal cardiovascular response to acute hypoxia (Boddy et al., 1974; Evers, 1978). Systematic study of the mechanisms producing late FHR decelerations has not been carried out, however, perhaps because of the difficulty in inducing controlled rhythmic uterine activity sufficient to cause fetal hypoxia in a stabilized chronic animal preparation. This paper describes a method for simulating late FHR decelerations in the fetal lamb under chronic study conditions, and presents preliminary observations demonstrating the role of reflex mechanisms in the genesis of late FHR decelerations.
Material and methods
Fetal vascular catheters (polyvinyl), ECG electrodes, and an intra-amniotic catheter were implanted in gravid ewes (2 Texel, 1 Dorset) between 122 and 144 days of pregnancy. A single fetus was present in each instance. Standard fetal surgical techniques were employed under general, halothane anesthesia. One catheter was passed into the fetal abdominal aorta via a femoral artery, and another was inserted into the upper abdominal or lower thoracic segment of the fetal inferior vena cava via a femoral vein. In addition, an inflatable cuff occluder (Rhodes Medical) was placed around the maternal common hypogastric artery trunk, and a non-cannulating electromagnetic flow probe (Skalar Instruments) was placed on the hypogastric artery ipsilateral to the uterine horn predominantly containing the fetus. Postoperatively, the animals were placed in a movable stall and permitted to recover. They received food and water ad libitum. The experimental observations were carried out 48 h to 9 days after surgery (at 127-148 days of pregnancy). Each animal was studied 3 times, for a
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
total of 9 experiments. The animals were brought in their stalls to the laboratory; the catheters, electrodes and flow probe were connected to appropriate transducers and couplers (Hewlett-Packard, 8800 series) and flow meter (Model 601, Skalar Instruments). The analog signals representing fetal ECG, FHR, fetal arterial blood pressure, intra-amniotic pressure and hypogastric artery blood flow were recorded on a multichannel strip-chart recorder (Model 1065, Hewlett-Packard), and simultaneously on magnetic tape. After a stable baseline recording period of at least 30 min, a fetal arterial blood sample was drawn for pH and respiratory gas tension analysis, and the simulation of uterine contractions was begun. The effect of uterine contractions in reducing maternal placental blood flow, and thus oxygen delivery to the fetoplacental unit, was simulated by periodically inflating the occluder around the maternal common hypogastric artery trunk, thus interrupting the major source of blood flow to the uterus. Occlusion was maintained for periods of from 30 to 60 set, and was repeated at an interval of 2.5 min from the beginning of one occlusion to the start of the following compression. To increase the degree of hypoxia in some experiments, the occluder was left partially filled between the total occlusions. Additional, fetal arterial blood samples were obtained at intervals to monitor the fetal acid-base status. These samples were taken either during the last seconds of the occlusion, or at 1 min following release of the occlusion. Samples were withdrawn into heparinized plastic syringes and refrigerated until analyzed, within 2 h of withdrawal. Previous observations in this laboratory have shown that only minimal changes occur in the pH and gas tensions of samples treated in this fashion. The pH and gas tensions were measured with micro blood analysis equipment (Model 165 pH and blood gas analyser, Corning Instruments). To investigate the role of reflex mechanisms in the fetal cardiovascular changes accompanying the periodic maternal artery occlusions, the various divisions of the fetal autonomic nervous system were blocked by pharmacological means. Parasympathetic postganglionic impulses wer~eblocked by atropine, 1 mg/ kg estimated fetal weight (EFW); alpha-adrenergic effects were antagonized with phentolamine, 2.5-5
C.B. Martin, Jr. et al.: Mechanisms of latecieceierations
mg/kg EFW; and beta-adrenergic influences, by propranolol, 1 mg/kg EFW. The completeness of blockade was tested by means of acetylcholine, 10 /&kg EFW, noradrenaline, 5-15 I.cg/kg EFW, and isoproterenol, 0.1 pg/kg EFW, respectively. The adorlomic agonists and antagonists were administered to the fetus directly via the venous catheter. Flockade of one or more autonomic divisions was established early in the course of 5 trials, after a short series of preliminary occlusions (usually 5-7) had produced repetitive decelerations, and before the development of fetal acidosis (pH > 7.30, base excess > -6.0 mmol in each instance). Autonomic blockade was established on 2 occasions in acidotic fetuses: cholinergic blockade alone in one fetus with pH 6.96 after a long serie? of occlusions, and triple blockade in another fetus with pH 7.23. Experimental results were analyzed by inspection of the analog records. Quantitative data regarding heart rate and pressure were read directly from the records. Fetal blood pressure values were corrected by subtracting the simultaneous amniotic fluid pressure readings. The beat-to-beat variability of the FHR was determined from the tape-recorded FECG signal, using the interval difference index (IDI) calculation of Jongsma et al. (1978).
Results The periodic occlusions of the common hypogastric artery produced decelerations in the FHR (Fig. 1A and B). FHR decelerations occurred with the first occlusion in 6 of the 9 experiments, and were consistently present from the 4th occlusion onward in 8 experiments. Transient accelerations of the FHR accompanied the initial occlusions in all 3 instances in which the appearance of FHR decelerations was delayed. This sequence was observed twice in one animal and once in another. In each animal, the beginning fetal PO, value was higher in the study sessions in which the initial FHR response was acceleration than in those sessions in which FHR deceleration
363
accompanied the first occlusion. A similar pattern was not observed for pH and Pco,. Consistent FHR decelerations could not be produced in one experiment despite 2 h of repetitive periodic occlusions of the maternal hypogastric trunk, although marked beat-to-beet arrhythmia and intermittent clecelerations did occur with the occlusions. Consistent FHR decelerations were produced on two other occasions in this animal, in association with lower initial and post-occlusion PO2 levels. The FHR decelerations produced by occlusion of the common hypogastric artery in the other experiments had the shape of a wide V or U in non-acidemic fetuses (Fig. 1). Onset of the deceleration occurred usually 15-25 set after the start of the occlusion, although there was sometimes considerable variation between successive occlusions in the interval to onset of deceleration. The lowest FHR was reached 5-12 set after the end of the occlusion. Recovery of the FHR began a few seconds later, with return to the baseline usually 35-l 00 set after release of the occlusion. Transient acceleration of the FHR during recovery occurred inconsistently after lo-15% of the occlusions. Fetal blood pressure rose transiently during the occlusions (Fig. 1). The onset of detectable hypertension and the appearance of cardiac slowing most often coincided within l-2 sec. With some occlusions, however, these two events differed by as much as 15-20 set and occurred in either sequence. In nonacidemic fetuses, the blood pressure rise was progressive, reaching a maximum level shortly after the end of the occlusion. The times at which the highest blood pressure and lowest FHR levels were attain&l corresponded within 4 set (2 mm on the record) in 60% of the occlusions in non-acidemic fetuses. In the remaining instances, greater deviations of as much as 30 set were observed. In 2 of the 3 individual fetal lambs, and in the pooled data from all 3 fetuses, there was a significant (P< 0.01) linear correlation between the amount of increase in systolic blood pressure and the amount of decrease in the FHR for the occlusions carried out
Fig. 1. Periodic FHR decelerations and hypertension in response to periodic occlusion of the maternal common hypogastric artery trunk in 2 fetal lambs. In panel A, the sustained hypertension and tachycardia following occlusion no. 4 suggest catecholamine release by the fetal adrenal at this point. In this and subsequent figures, the negative intrauterine pressure reflects placement of the pressure transducers at a higher level than the uterus.
PH
PO2 pC0, BE
7.40 1.7 kPa 5.6 kPa l1.3 mmol/l
fetal arterial pressure (mm Hg)
fetal heart frequency (bpm)
v-h-a utenne pressure (InrnHg)
flow uterine artery (ml/min)
FECG
ewe 58178 3 days postop.
:
gest age 128 days periodic total occlusion
,
f time(mln)
PH POP PC4 BE
0 fetal arterial pressure (mm Hg)
738 1.8 kPa 3.3 kPa -7.7 m molll
I
100,
1
fetal heart frequency (b.pm 1
intra uterine pressure (mm Hg)
flow uterine artery (ml/mm)
FECG I,
ewe 92-70 8 days postop.
I
I
: :
I
gest. age: 129 days periodic total occlusion
I
1
I time (min)
1
1
I
I
1
365
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
prior to the development of acidemia or the institution of autonomic blockade. This was true both for the absolute data, expressed in mm Hg and beats/min, and for the proportionate data calculated as percent change. The systolic blood pressure and average baseline FHR immediately preceding each occlusion were used as the reference values for the calculations of absolute and percent change during the occlusion. The regression line for the pooled absolute data was AFHR=-12.6 -3.1 (ASBP), rc0.74, n=60. The slopes for the regression lines for the individual fetuses also indicated a decrease of approximately 3 beats/min in FHR for each mm Hg increase in systolic blood pressure. Alpha-adrenergic blockade with phentolamine prevented or largely eliminated the progressive hypertensive response of the fetus to periodic interruption of the uterine blood flow (Fig. 2). Although FHR
decelerations still occurred with some of the occlusions in the absence of hypertension, the pattern of these decelerations was markedly different from that of the decelerations before alpha-blockade (Fig. 2, occlusion 6). In non-acidemic fetuses, cholinergic blockade with atropine eliminated the FHR deceleration response to periodic interruption of the uterine blood flow of 60 set duration. Instead, periodic acceleration of the FHR occurred, beginning usually 15-30 set after the start of occlusion (Fig. 3). These FHR accelerations occurred both in the presence and absence of concomitant alpha-adrenergic blockade. Beta-adrenergic blockade with propranolol prevented or markedly reduced the FHR acceleration response of the atropinized fetus to the occlusions (Fig. 4). Small (5-20 bpm) periodic increases in the FHR persisted in the experiment illustrated, presumPH PO2 pC02 BE
t&al artermt 100
735 14 6.1 -07
kPa kPa mmol/l
1
pressure (mm Hg) 50
fetal heart trequency
250
(bpm)
150
lntra
uterine
pressure (mmHg)
flow uterine artery (mlfmcn)
0
20
500
0
FECG
ewe 56-76 3 days postop.
,
gest age 126 days periodic total oc&sion
a-adrenergic
blockade
time(mln)
Fig. 2. Modification of the fetal response pattern following alpha-adrenergic blockade with phentolamine. Compare with Fig. 1A. The phasic fetal hypertension is absent. A V-shaped deceleration occurred with fetal movement during occlusion 3, and brief decelerations occurred after occlusion 4 and during occlusion 6.
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
366
PH
7.33 1.2 kPa PC02 5.1 kPa BE -48 mmol/I PO2
fetal
arterial 100 7
1
pressure (mm
Hg)
fetal heart frequency (bp.m)
50
250 -7
M
I
150-
I I
50 -
intra uterine PWSS"re (mmHg)
flow uterine artery
1
0 2.
500 -
(ml/min)
FECG
ewe 58-78 3 days wstop.
1
gest age 120 days periodic total occlusion
a-adrenergic , cholinergic blockade
I
time (min) Fig. 3. Combined alphaadrenergic and cholinergic blockade in the same fetus shown in Figs. 1A and 2. The FHR now responds to the occlusions with periodic acceleration.
ably reflecting incomplete blockade of locally liberated catecholamines. In other experiments, however, the accelerations were eliminated completely by betaadrenergic blockade. Changes occurred in the fetal blood pressure and, to a lesser extent, in the FHR responses to periodic hypoxemia when the series of occlusions was extended to produce fetal acidosis (Fig. SA-C). As the pH decreased to below 7.30, the fetal hypertensive response lost its progressive character and tended to reach a plateau early after the beginning of the occlusion (Fig. 5A). The FHRcontinued to decelerate throughout the period of hypoxemia, however, with the nadir occurring at or after the end of the occlusion. With increasing acidosis, there was further attenuation of the fetal blood pressure response (Fig.
5B), and the eventual appearance of a biphasic pattern with slight hypotension following an initial increase in pressure (Fig. SC). The FHR decelerations became broader and more U-shaped, and the low point of the deceleration tended to occur earlier during the occlusion period (Fig. 5B and C). Administration of atropine to the severely acidotic fetus (pH 6.96) resulted in an increase in the baseline FHR, but no change in the amount of the deceleration (Fig. 6). The interval to the onset of deceleration decreased slightly. In the other acidotic fetus, administration of atropine after combined blockade of both adrenergic systems resulted in a transient decrease in the amount of deceleration, but this returned to the previous magnitude by the third occlusion after atropine. The fetal pH decreased
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
361 7.33 PH 1.1 kPa PO2 pCOz 50 kPa BE -5.5 mmol/l 1
fetal arterr pressure (mm Hg)
loo 50 0
fetal heart frequency (bpm)
250 150 50
lntra uterlnc pressure (mm+@)
I
0 1
/
flow utenne artery (mllmln)
500 n
FECG ewe 58-78 3 days postop
I
:
gest. age 128 days perlodu total occIwon
0. and P adrenergic
, chollnerglc
blockade
I
time (mln)
Fig. 4. Fetal response to periodic occlusion of the maternal common hypogastric artery trunk in the presence of combined alphaand beta-adrenergic and cholinergic blockade. There remains a slight FHR acceleration during the occlusions, and a small hypertensive response following some of them, probably reflecting incomplete blockade.
rapidly in the face of triple autonomic blockade, making further interpretation uncertain. Beat-to-beat variability of the FHR, as measured by the interval difference index (IDI) (Jongsma et al., 1978) persisted in the face of increasing hypoxemic acidosis in the fetus subjected to the prolonged series of repeated occlusions (Figs. 1B and 5A-C). The median ID1 during the 30-min control period immediately before this experiment was 19 units. In the experimental period, the median IDI, calculated from the 30-set segments immediately preceding each of the 91 occlusions, was 34. During the last 30 min of this experiment, the fetal pH decreased from 7.10 to 6.96. The median ID1 during this period was 25. The
ID1 decreased markedly, to 3.3, following administration of atropine to the fetus (Fig. 6). Discussion
The categorization of transient periodic decreases in the FHR as early,variable or late decelerations (Hon, 1968) or as type I or type II dips (Caldeyro-Barcia et al., 1966), was based in large part on the temporal relationship between the FHR decelerations and the uterine contractions provoking them. In the present study, actual uterine contractions were not produced; however, one of the important effects of uterine contractions, the transient reduction of the
Method
7.27 1.1 kPa pCOn 51 kPa BE -BBmmolll
1
fetal arterfal pressure (mm Hg)
fetal heart frequency (b.o.m )
intra uterine pressure (mm Hg)
flow uterfne artery (mllmln)
FECG I
,
1
ewe 92-78 8 days postop.
I
I
Y ,‘I,
I...,,
I..,
,
,
‘1
I..,
: gest. age: 129 days : continuing partial occlusion, periodic total occlusion
1
1
,
!
I
,
I
,
I
time (min) 7.22 PH 1.2 kPa PO2 pC02 5.6 kPa -10.3 mmolll BE
B fetal arterial pressure (mm Hg)
100 1
50 0 fetal heart frequency (b.p.m )
250
uterme pressure (mml-fg) intra
25
flow uterine artery (mllmin)
FECG
I:
1
C.B. Martin, Jr. et al.: Mechanisms of Iate decelerations
369
PH PO, PC4 BE
7.14 1.5 kPa 6.3 kPa -131 mmolll
1
fetal heart frequency (b.pm)
lntra uterine pressure (mm Hg) 25
flow uterine artery (ml/mln)
600
FECG ewe 92-78 8 days postop. I
: gest. age: 129 days : continuing partial occlusion, periodic total occlusion
I
,
!
I
time (min)
Fig. 5. Fetal response to an extended series of periodic total occlusions of the maternal common hypogastric artery trunk combined with continuing partial occlusion of this vessel. A-C: sections of the record at increasing degrees of fetal acidosis. Note the progressive disappearance of the fetal hypertensive response, and the occurrence of slight diastolic hypotension in C. The FHR decelerations become wider and more symmetrical with increasing acidosis. Compare with Fig. 1B.
maternal placental blood flow, was mimicked. The term ‘late decelerations’, as used in this report, refers therefore to the temporal relationship between the FHR decelerations and the periodic interruptions of the uteroplacental blood flow. The role played by uterine contractions in the production of late periodic FHR decelerations is the periodic reduction of maternal placental blood flow, and thus fetal oxygenation, during the contractions. If this thesis is accepted, as seems generally to be the case, then one must also accept that FHR decelerations resulting from the periodic reduction of uteroplacental blood flow by another method, specifically in these experiments periodic occlusion of the major
arterial supply to the uterus, are equivalent in their pathophysiologic mechanisms to ‘natural’ late decelerations. Chronic reduction in resting uteroplacental blood flow occurring as a result of placental or uteroplacental vascular pathology, which often sets the stage for late decelerations in labor, can be mimicked to some extent by continuing partial occlusion of the uterine arterial supply ~between the periodic episodes of greater occlusion. In this latter case, however, the simulation in the present experiments was only partial; for the partial occlusion was maintained only for l-2 h, whereas the condition may endure several weeks in complicated human pregnancies. The present methods, therefore, provide
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
370
PH
6.96
fetal arterial pressure (mm Hg)
fetal heart frequency (b.p.m )
irkra uterine pressure (mm Hg)
flow uterine artery (mllmtn)
FECG
ewe 92
I
gest. age : 129 days continuing partial occlusion, periodic total occlusion
-70
1
I
I
I
9
I
I
I
I
time (min)
Fig. 6. Fetal response to cholinergic blockade during severe acidosis. The baseline FHR increases following atropine, but the amount of FHR deceleration is unchanged. Beat-to-beat variability is present at the beginning of the panel and disappears after atropine. Note that the FHR scale has been compressed in this figure. a means for studying the pathophysiologic mechanisms involved in late decelerations occurring with acute fetal distress, both before and after the development of fetal acidosis. Myers et al. (1973) employed occlusion of the maternal abdominal aorta as a means of producing FHR decelerations in fetal monkeys. However, they do not seem to have used the technique in a repetitive, periodic fashion, as in the present study. Moreover, their experiments were carried out immediately after surgery and under conditions of maternal anesthesia. There is a possible source of discrepancy between the mechanisms activated by periodic occlusion of the uterine arterial ,supply and those involved when uterine contractions interrupt maternal placental blood flow. Power and Longo (1973) have postulated
the occurrence of ‘sluice flow’ in the placenta. According to this concept, there should be increased resistance to flow in the fetal placental circulation during contractions as a result of compression of fetal capillaries by an increasing volume of blood trapped in the maternal placental blood spaces. The reverse would occur, of course, if only the maternal arterial supply were occluded. However, studies specifically directed toward detection of a sluice flow mechanism in chronically instrumented pregnant sheep failed to confirm its presence (Berman et al., 1976). Moreover, if sluice flow were a significant feature of placental circulation in sheep, the fetuses with triple autonomic blockade in the present experiments should have demonstrated hypotension during the maternal arterial occlusions. This was not observed (Fig. 4).
371
C.B. Martin, Jr. et al.: Mechanisms of hte decelerations
Although it has Mechanisms of late deceleration been recognized that elements of both reflex cardiac deceleration and direct depression of myocardial rhythmicity contribute to the late deceleration FHR pattern, there seems to be little agreement among authorities as to the relative importance of each mechanism. Hon (1962, 1968) has placed major emphasis on hypoxic depression of the myocardium, whereas Caldeyro-Barcia and associates (cf. CaldeyroBarcia et al., 1966; Mendez-Bauer et al., 1963) concluded that their type II dips resulted largely from transient increases in vagal cardiodecelerator tone. The former viewpoint has received support from observations carried out on fetal monkeys under acute experimental conditions, in which late decelerations were accompanied by hypotension and were quantitatively related to the degree of fetal hypoxemia (James et al., 1972; Myers et al., 1973). In the chronically catheterized fetal lamb, on the other hand, the initial bradycardia associated with acute hypoxemia can be converted to a tachycardiac response by vagotomy (Boddy et al., 1974) or cholinergic blockade (Parer, 1977; Evers, 1978). The bradycardia response to acute hypoxemia can also be delayed substantially by atropine in fetal monkeys, when the observations are carried out under conditions of chronic study (Martin, Murata and Ikenoue, unpublished observations). Su and Friedman (1973) have shown that strips of fetal myocardium are relatively resistant to chronotropic depression by hypoxia, in comparison to adult cardiac muscle. These and similar observations have led to the suggestion (Martin and Gingerich, 1976; Martin, 1978) that late FHR decelerations might be produced by either reflex or direct mechanisms, and more often by both acting together, with the relative contribution of each mechanism at a particular time in a specific fetus depending on the degree and duration of fetal oxygen lack. The observations described in this report appear to confirm that suggestion. The non-acidemic fetuses responded to periodic reductions of uterine blood flow by periodic hypertension and FHR decelerations. Elimination of the fetal hypertensive response by alpha-adrenergic blockade markedly modified the FHR decelerations, but did not eliminate them completely. Cholinergic blockade converted the periodic decelerations to accelerations, which in turn were blocked by propranolol. When all 3 autonomic mech-
anisms were blocked, the non-acidemic fetuses showed essentially no heart rate response to periodic hypoxemia. In contrast, blockade of the efferent limb of the cardiodecelerator reflex did not alter the degree of FHR deceleration in response to periodic occlusion of the maternal hypogastric artery trunk in fetuses made severely acidotic by a long series of such occlusions. The alpha-adrenergic blocker used in these experiments, phentolamine, has a variety of additional actions including beta-adrenergic, cholinergic and histaminergic effects (Goodman and Gilman, 1975). Since during the occlusions after phentolamine blockade there was not only a virtual absence of FHR decelerations, but also absence of phasic fetal hypertension, it seems most likely that these changes were related: that the absence of FHR decelerations was the result of absence of phasic blood pressure increases during the hypoxic episode. The possibility remains, however, that one or more of the other actions of phentolamine may have played a role in altering the fetal response to transient interruptions of the uterine blood flow. The reflex pathways presumably involved in ‘acute’ late decelerations are summarized in Fig. 7. The
hypoxaemia /\
J ch
7
sympathetic
l/
\
yre\
\
center
2l 4
-- --a
blockode
vasoconstrlctlon _.__-6
I
blockade
I + blood
pressure
baro
receptor
fetal
heart
3)
4/
I
!
t
rate
Fig. 7. Schematic diagram of the reflex pathways presumed to be invdlved in the production of late decelerations in the non-acidemic fetus.
312
dominant mechanism would appear to be chemoreceptor-mediated reflex vasoconstriction (Dawes et al., 1968, 1969) producing hypertension and, in turn, baroreflex-induced cardiac slowing (Shinebourne et al., 1972). That chemoreceptor (peripheral or central) stimulation by hypoxia can also activate a cardiodecelerator reflex primarily, reinforcing the indirect vasoconstriction-baroreceptor mechanism, is suggested by the frequent onset of heart rate deceleration several seconds before the appearance of any blood pressure increase in the unblocked fetuses. This is also demonstrated by the occurrence of some cardiac slowing with occlusions after the hypertension response had been blocked by phentolamine. The reflex nature of these latter decelerations is apparent from their disappearance following cholinergic blockade. A cardiodecelerator response to aortic chemoreceptor stimulation has been previously demonstrated in the fetal lamb by Goodlin and Rudolph (1972). Although cardiodecelerator mechanisms predominate in the response of the non-acidemic fetal lamb to hypoxemia, an adrenergic cardioaccelerator component is also present. This is most clearly demonstrated by the periodic accelerations accompanying interruption of the uterine blood flow following cholinergic blockade. It is also shown by the acceleration response of 2 fetuses to the initial occlusions. The cardioaccelerator response to acute hypoxemia predominates in immature fetallambs (e.g. 97-102 days) (Boddy et al., 1974), and was also observed in older fetal lambs studied acutely under conditions of maternal systemic anesthesia(e.g. Dawes et al., 1969). Involvement of beta-adrenergic mechanisms in the accelerator response was confirmed in the present investigation by the elimination or reduction of the accelerations by propranolol. The present experiments substantiate the role of direct hypoxic depression of myocardial chronotropism in late decelerations under conditions of severe fetal acidosis. Here the reflex component seems to be absent or minimal, for choline@ blockade did not alter the amount of the decelerations in the severely acidotic fetal lamb. The absence of a reflex element in these decelerations was not due to paralysis of the parasympathetic system, however; for persisting vagal activity was demonstrated by elevation of the baseline FHR following atropine
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
administration, and by the continued presence of beat-to-beat variability in the baseline FHR which disappeared with cholinergic blockade. Perhaps the absence of phasic hypertension with the occlusions during severe acidosis accounts for the absence of a reflex component in those late decelerations. Our findings thus indicate that reflex mechanisms accounted for all of the ‘late decelerations’ produced by periodic reduction of uteroplacental blood flow in fetuses with normal pH, and essentially none of the deceleration during severe acidosis. The point at which direct hypoxic depression of cardiac rhythmicity begins to play a role was not defined; however, our observations suggest that this may occur quite early in the course of fetal distress - at pH values between 7.30 and 7.25. In this range, the fetal blood pressure response to periodic hypoxemia lost its progressive character, although the FHR continued to decrease progressively throughout the occlusion. Thus, ‘pure reflex’ late decelerations may be restricted to situations of short-term, acute hypoxemia in previously unstressed fetuses. Further definition of the relative contributions of reflex and direct mechanisms to late decelerations between the extremes of fetal condition would require the institution of autonomic blockade at successively lower pH values in an extensive series of experiments; for the inhibition of circulatory reflexes, by interfering with the preferential perfusion of the heart and brain during hypoxemia (Cohn et al., 1974; Peeters, 1978), might result in an erroneous conclusion should the fetus simply be made progressively acidotic by repeated uterine artery occlusions during pre-existing blockade. In addition, the point of appearance of direct hypoxic depression of cardiac chronotropism may vary considerably among individual fetuses depending, for example, on the relative duration of hypoxemia and recovery, the lowest oxygen content reached during the occlusion, and perhaps also such factors as cardiac glycogen stores and blood glucose levels. The persistence of beat-to-beat arrhythmia in the FHR of our severely acidotic fetal lamb is at variance with the usual observation in clinical FHR monitoring, that beat-to-beat variability is diminished in the presence of significant hypoxemic fetal acidosis. Our observation is, however, in general agreement with the finding of Dalton et al. (1977) that beat-to-beat
373
C.B. Martin, Jr. et al.: Mechanisms of late decelerations
variability increased during hypoxemic acidosis in the fetal lamb, even when the hypoxemia was prolonged to produce increasing acidosis, and did not fall below the pre-hypoxemic value until the pH was 6.9. That vagal mechanisms still account for the beat-to-beat arrhythmia during severe hypoxemic acidosis was demonstrated by the prompt disappearance of baseline variability following administration of atropine to the fetus in the present study (Fig. 6). Whether the apparent difference between human and ovine fetuses in the course of FHR variability during hypoxemic acidosis represents differences in the reactions of the cardiac control mechanisms, or in such factors as resistance to hypoxic depression or degree of maturity achieved by late gestation, remains to be defined. Moreover, the induction of fetal hypoxia and acidosis by relatively acute interference with the uterine blood flow or reduction of the O2 content of the gas mixture breathed by the mother may well be an incomplete simulation of the large proportion of human fetal distress that is preceded by several days or weeks of diminishing placental function.
References Berman, W., Goodlin, R.C., Heymann, M.A. and Rudolph, M.A. (1976): Relationships between pressure and flow in the umbilical and uterine circulations of the sheep. Cir&at. Res., 38, 262. Boddy, K., Dawes, G.S., Fisher, R., Pinter, S. and Robinson, J.S. (1974): Fetal respiratory movements, electrocortical and cardiovascular responses to hypoxaemia and hypercapnea in sheep. J. Physiol. (Lond.) 243, 599. Caldeyro-Barcia, R., Mendez-Bauer, C., Poseiro, J.J., Escarcena, L.A., Pose, S.V., Bienarz, J., Arnt, I., Gulin, L. and Althabe, 0. (1966): Control of human fetal heart rate during labor. In: IIre Heart and Circulation in the Newborn infant, p. 7. Editor: D.E. Cassells. Grune and Stratton, New York. Cohn, H.E., Sacks, E.J., Heymann, M.A. and Rudolph, M.A. (1974): Cardiovascular responses to hypoxemia and acidemia in fetal lambs. Amer. J. Obstet. Gynec., 120, 817.
Dalton, K.J., Dawes, G.S. and Patrick, J.E. (1977): Diurnal, respiratory and other rhythms of fetal heart rate in lambs. Amer. J. Obstet. Gynec., 127, 414.
Dawes, G.S., Lewis, B.V., Milligan, J.E., Roach, M.L. and Talner, N.S. (1968): Vasomotor responses in the hind limbs of foetal and newborn lambs to asphyxia and aortic chemoreceptor stimulation. J. Physiol. (Lond.), 195, 55.
Dawes, G.S., Duncan, S.L.B., Lewis, B.V., Merlet, C.L., Owen-Thomas, J.B. and Reeves, J.T. (1969): Hypoxemia and aortic body chemoreceptor function in foetal lambs. J. Physiol. (Lond.).
201, 105.
Evers, J.L.H. (1978): The Cardiac Pre-Ejection Period During Prenatal Life. Thesis, Faculty of Medicine, Catholic University, Nijmegen, The Netherlands. Goodlin, R.C. and Rudolph, A.M. (1972): Factors Associated with the Onset of Breathing. In: Proc. Int. Symp. Physiological Biochemistry of the Fetus. Editors: A.A. Hodari and F.G. Mariona. Thomas, Springfield. Goodman, L.S. and Gilman, A. (1975): 7’he Pharmacological Basics of Therapeutics. Macmillan, New York. Hon, E.H. (1962): Electronic evaluation of the fetal heart rate. VI. Fetal distress - a working hypothesis. Amer. J. Obstet. Gynec., 83, 333.
Hon, E.H. (1968): An Atlas of Fetal Heart Rate Patterns. Harty Press, New Haven, Conn. James, L.S., Morishima, H.O., Daniel, S.S., Bowe, E.I., Cohen, H. and Nieman, W.H. (1972): Mechanisms of late deceleration of the fetal heart rate. Amer. J. Obstet. Gynec., 113, 578.
Jongsma, H.W., van Geijn, H.P. and de Haan, J. (1978): The analysis of heart rate variability in the perinatal period. In: Computerdiagnostik in den Geburtshilfe, Proc. 1st. Symposium, FriederichSchiller Universitat, Jena, p. 249. Martin, C.B. and Gingerich, B.H. (1976): Factors affecting the fetal heart rate: genesis of fetal heart rate patterns. J. Obstet. Gynec. Nursing, 5, 30s. Martin, C.B. (1978): Regulation of the fetal heart rate and genesis of FHR patterns. Sem. Perinatol., 2, 131. Mendez-Bauer, C., Poseiro, J.J., Arellano-HernLndez, G., Zambrana, M.A. and Caldeyro-Barcia, R. (1963): Effects of atropine on the heart rate of the human fetus during labor. Amer. J. Obstet. Gynec., 85, 1033. Myers, R.E., Mueller-Heubach, E. and Adamsons, K. (1973): Predictability of the state of fetal oxygenation from a quantitative analysis of the components of late deceleration. Amer. J. Obstet. Gynec., 115, 1089. Parer, J.T. (1977): Effect of atropine on heart rate and oxygen consumption of the hypoxic fetus. Gynec. Invest., 8, 50. Peeters, L.L.H. (1978): Fetal Blood Flow at Various Levels of Oxygen. Thesis, Faculty of Medicine, Catholic University, Nijmegen, The Netherlands. Power, G.G. and Longo, L.D. (1973): Sluice flow in the placenta: maternal vascular pressure effects on fetal circulation. Amer. J. Physiol., 225, 1490. Shinebourne, E.A., Vapaavouri, E.K., Williams, R.L., Heymann, M.A. and Rudolph, M.A. (1972): The development of baroreflex activity in unanesthetized fetal and neonatal lambs. Circulat. Res., 31, 710. Su, J.Y. and Friedman, W.F. (1973): Comparison of the responses of fetal and adult cardiac muscle to hypoxia. Amer. J. Physiol., 224, 1249.