- Email: [email protected]
Variable heart rate decelerations and transcutaneous Po2 during umbilical cord occlusion in fetal monkeys
Variable heart rate decelerations and transcutaneous PO, during umbilical cord occlusion in fetal monkeys EBERHARD ANTHONY Pittsburgh,
MUELLER-HEUBACH, F. BATTELLI,
M.D. B.S.
Pennsylz~ania
In five near-term rhesus monkeys under ketamine anesthesia, fetal heart rate, blood pressure, intrauterine pressure, and transcutaneous oxygen tension (tcPo,) were recorded continuously while the umbilical cord was ocduded for 15 or 30 seconds. Fetal heart rate decreased 55 + 19 bpm. (mean ? SD) and 78 2 20 bpm, respectively, with 15 and 30 second occlusions, while tcPo, declined 6.0 2 1.6 and 11.5 t 3.3 torr, respectively. Fetal hypertension started promptly with cord occlusion and reached a maximum within 5 to 16 seconds; fetal blood pressure decreased thereafter and, in most instances, fell below the baseline until fetal heart rate began to recover. Fetal heart rate decreased within 1 to 2 seconds of cord occlusion. tcPo, changes were the same, while heart rate changes were attenuated when the fetus was pretreated with 0.2 mg of atropine. These findings support a rapid and marked initial baroreceptor response, followed by a less marked chemoreceptor response. Thirty seconds following cord occlusion, Paon had returned to baseline, while tcPo, was still 3 to 7 torr below baseline, indicating cutaneous vasoconstriction even though systemic blood pressure was normal. Relative heat output required to maintain the tcPo2 electrode temperature at 44” C did not reflect alterations in skin perfusion during periods of cardiovascular change. (AM. J. OBSTET. GYNECOL. 144:796, 1982.)
VARIABLE DECELERATION of the fetal heart rate during labor is the most frequently observed deceleration pattern in human labor. As the name implies, it is somewhat variable in temporal relationship to uterine contractions and is the result of umbilical cord occlusion.’ It has been known since the work of Barcrof? that a vascular reflex involving the vagus nerve in combination with the effect of asphyxia on the cardiac muscle is involved in the fetal heart rate response to umbilical cord occlusion. Fetal capillary blood sampling in human fetuses during labor has revealed an increasing risk of fetal hypoxia and acidosis when variable decelerations of fetal heart rate become severe.3 Clini-
From the Department of Obstetrics and Gynecology, University of Pittsburgh School of Medicine, Magee- Wornens Hospital. Supported by Magee- Wornens Hospital Grant No. 292-414. Received for publication
Februay
Research
Fund
23, 1982.
Revised June 4, 1982. Accepted July 13, 1982. Reprint requests: Eberhard Mueller-Heubach, M.D., Magee- Womens Hospital, Forbes Ave. and Halket St., Pittsburgh, Pennsylvania 15213. 796
cal investigation of the relationship between fetal heart rate change during umbilical cord occlusion, vagally mediated vascular reflex, and oxygenation is complicated by the fact that degree and duration of umbilical cord occlusion during successive uterine contractions in human labor cannot be quantitated and may vary. Therefore, an animal model in which repetitive, wellquantitated umbilical cord occlusions can be produced experimentally is necessary. Alterations in fetal oxygenation during the rapid changes in fetal cardiovascular performance accompanying umbilical cord occlusion cannot be properly assessed by intermittent capillary blood sampling from the presenting part in the human fetus or intermittent arterial blood sampling from a catheter in the experimental animal fetus. The advent of a transcutaneous electrode for continuous measurement of oxygen tension (Po2) has provided a noninvasive means of recording PO, during labor in the human fetus.“’ 5 Studies in the human fetus6 as well as in the animal fetus7 indicate a close relationship between arterial PO* and transcutaneous Po2 (tcPo,). The present study was designed to investigate the relationship between fetal heart changes and tcPo, during well-quantitated periods of umbilical cord occlusion in the near-term rhesus monkey. The aim of the study 0002-9378/82/230796+07$00.70/O
0
1982 The C. V. Mosbv Co.
Volume Number
144 7
Variable
decelerations
and tcPo,
during
cord occlusion
797
was to delineate the contributions of the vascular vagal reflex and fetal hypoxia to the pattern of variable decelerations of fetal heart rate. In addition, the usefulness of the tcPo? electrode during the changes in fetal oxygenation accompanying umbilical cord occlusion was studied.
Material and methods Five pregnant rhesus monkeys from the Department of Obstetrics and Gynecology at the University of Pittsburgh School of Medicine were used. Their gestational ages at the time of the experiments were 158 to 162 days. (Term is about 168 days). Before the experiments, the animals were not fed for 12 hours overnight but were allowed water ad libitum. Anesthesia was established by intramuscular injection of 100 mg of ketamine. Throughout the experiment, anesthesia and hydration were maintained by infusion of ketamine in isotonic saline solution (I mg/ml) at a rate of 1 ml/min. Laparotomy and hysterotomy through a low-segment transverse incision were done. The fetal head and neck were delivered from the uterus, and a polyvinyl catheter (0.015 inside diameter) was inserted into a carotid artery. Silver/silver chloride disk electrodes were fixed subcutaneously at both fetal shoulders. A plastic inflatable occlusion device (Rhodes Medical Instruments, Inc., Woodland Hills, California) with an internal diameter of 10 mm was placed around the umbilical cord close to the fetal abdomen. Care was taken that the occlusion device did not kink the umbilical cord. The upper chest of the fetus was then shaved, and the tcPoz electrode was glued with Eastman 910 tissue glue to the upper fetal chest over an intercostal space slightly lateral to the sternum. In preliminary studies, the tcPo, electrode had been glued to the shaved scalp of the monkey, fetus; however, in this location the tcPo2 electrode did not yield any tcPo, readings. The partially delivered fetus was then gently returned into the uterus. A polyvinyl catheter (0.050 inside diameter) was placed into the uterine cavity. Uterine and abdominal incisions were closed. Loss of amniotic fluid was estimated and replaced with isotonic saline solution through the intrauterine catheter. Studies were done after a l-hour recovery period following surgery. Intermittent total occlusions of the umbilical cord were produced by complete inflation of the occlusion device for periods of 15 and 30 seconds in random sequence. This was accomplished by timed injections of 2 ml of isotonic saline solution through the catheter of the device. It had been established before placement of the occlusion device that a 2 ml volume completely obliterated the internal opening of the occludrr. Injection and removal of the 2 ml of saline so-
Fig. 1. tcPo,, fetal heart rate (FHR), fetal blood pressure (FBP) and intrauterine pressure (IUP) during a 30-second umbilical cord occlusion at a paper speed of 3 cmimin.
lution at beginning and end of cord occlusions took no more than 2 seconds. Fetal carotid arterial blood samples were taken intermittently. Umbilical cord occlusions were done only when carotid arterial blood pH values were above 7.20. In cases of spontaneous uterine activity, umbilical cord occlusions were done at least 90 seconds after the end of the previous uterine contraction. instances of coincident umbilical cord occlusion and uterine contraction were not included in the analysis. Study periods did not exceed 3 hours, since previous observations in fetal lambs and juvenile rhesus monkeys6 have indicated that tcPo2 readings become unreliable after this time period as a result of skin changes due to the heat of the electrode. At the end of the study period, cesarean section was done, the fetus was delivered, and mother and newborn were returned to their cage. Measurement techniques. Carotid arterial and in-
798
Mueller-Heubach
and Battelli
December 1, IYX’L .%m. J. Obstet. Gynecrd.
Fig. 2. tcPo,, electrocardiogram (KG), intrauterine relative heat output (RNO). and fetal heart rate (FHRI speed of 15 cmimin
Table I. Relationship between fetal heart rate and tcPo* during umbilical cord compression Cwd
com-
pession
(set) 15 30
AFHR M-4
J.
55 -t 19 78 k 20
&cPo, I ctmr)
AFHRLI AtcPo,J. (bpmlto7-7)
6.0 ” 1.6 11.5 t 3.3
9 -e 3.6 7 -e 2.4
~/A&PO,?. (secltm) 10 2 2.3 8 2 2.8
All values are mean 2 W. trauterine pressures were measured with Statham P23 Db pressure transducers. Fetal blood samples were taken intermittently from the carotid arterial catheter; pH, PO,, and base deficit were determined with a Corning 165 blood gas analyzer. Fetal heart rate was obtained with a cardiotachometer triggered by the R wave of the electrocardiogram signal. The tcPo, electrode was connected to an electronic signal processor, from which signals of tcPo, and relative heat output or “flow” were recorded on an eight-channel Beckman R611 recorder together with fetal heart rate, arterial
pressure (ICP). fetal blood pressure (FBP). during a ?&second cord occlusion at a papel-
blood pressure, and intrauterine pressure. The relative heat output signal was amplified over the range of 5 V/mm to 0.1 mV/mm. Speeds of the recording paper ranged from 3 to 15 cm/min. The tcPo2 electrode was heated to a temperature of 14” C only during the actual study periods. Before attachment of the electrode, calibration was carried out by blowing nitrogen or 5.79% oxygen
over
the
electrode
surface.
which
yielded
ox”-
gen tensions of 0 and about 40 torr. respectively, depending on barometric pressure. Ninety percent ofthe electrode response was achieved within 10 seconds (time constant). At the end of the studies, the tcPo* electrode was recalibrated as described. Experiments in which tcPo, values at calibration and recalibration had drifted more than & 1 torr were excluded from analysis. Results Comparisons of tcPo, readings and intermittently obtained Paot values of carotid arterial blood were made when the tcPo, readings were stable for 2 minutes. Only studies where tcPo, and Paor values were
Volume
144
Numbw
7
Variable decelerations and tcPq during cord occlusion
799
F%P tow : 4%-.
m 15
SEC.
CORD
@XLUSlON
Fig. 3. tcPoz, fetal heart rate (FHR), fetal blood pressure relative heat output (RHO) during a 15-second umbilical contraction at a paper speed of 6 cm/min. within % 1 torr have been included. Uterine activity was present during part of or rhe total duration of the studies. Fetal baseline values before umbilical cord occlusions had the following ranges: pH 7.29 to 7.20; Paoz 24 to 12 torr: systolic blood pressures, 59 to 78 torr: diastolic blood pressures, 34 to 44 tort-; and fetal heart rate, 138 to 210 bpm. A\ in human labor, sequential pH values during the studies gradually declined, while PO, fluctuated. The effects of a 30-second occlusion of the umbilical cord on fetal tcPoz, heart rate, and blood pressure are illustrated in Fig. 1. Fetal heart rate decreases and recovers rapidly, the characteristic pattern of a “variable deceleration” in clinical monitoring. Fetal blood pressure increases initially and then declines toward the end of the umbilical cord occlusion. The tcPo* begins to decrease shortly after onset of the cord occlusion and recovers slowlv after the end of the occlusion, while fetal heart rate and blood pressure have returned to the values before occlusion. The temporal relationships between umbilical cord occlusion, fetal heart rate, fetal blood pressure, and tcPo, are clearly delineated in a recording obtained at a paper speed of 15 cmimin (Fig. 2). Fetal blood pressure increases 1 to
(FBP), intrauterine cord occlusion and
pressure (ZUP), and a prolonged uterine
1.5 seconds after start of umbilical cord compression, followed within 1 to 2 seconds by a progressive decline in fetal heart rate. The tcPo, starts to decrease 12 seconds after the onset of umbilical cord compression and does not return to its value before occlusion until 75 seconds after the cord occlusion has been terminated. To characterize the relationships between fetal tcPoZ and fetal heart rate more precisely, data from 24 umbilical cord occlusions of 15 or 30 seconds duration were summarized (Table I). In three animals, four occlusions were done; in the remaining two animals, five and seven occlusions, respectively, were performed. Maximal fetal heart rate decrease (AFHRJ) was rapid during a 15-second cord occlusion. A lesser decrease was seen during the second 15 seconds of the 30-second cord occlusions. This is compatible with the concept of a vagalfy mediated fetal heart rate decrease due to baroreceptor stimulation during the initial increase in fetal blood pressure. The time from onset of cord occlusion to maximal fetal blood pressure elevation (i.e., maximal baroreceptor stimulation) ranged from 5 to 16 seconds. However, this initial blood pressure increase is followed by a decrease, resulting in return of
800
Mueller-Heubach
December 1, 1982 Am. J. Obstet. Gynecol.
and Battelli
the blood pressure to the preocclusion value 19 seconds after beginning of the occlusion and subsequent hypotension throughout the remainder of the period of cord occlusion (Fig. 2). Although fetal blood pressure decreased after reaching a maximum, this decrease went beyond baseline blood pressure in only 19 instances (Fig. 2), while in five observations fetal blood pressure decrease did not exceed baseline fetal blood pressure before cord occlusion (Fig. 1). The tcPo, decrease (AtcPo,i) during 15- and 30-second cord occlusions is progressive and linear. TcPo, decrease was measured by subtracting the tcPo2 value at the nadir of tcPo, decrease from the tcPoz value at the moment when cord occIusion was started. The fall in fetal heart rate for each torr decrease in tcPo, (AFHRJlAtcPo,&) is more marked during the first 15 seconds of an umbilical cord occlusion. We have compared the tcPo* value at the beginning of umbilical cord occlusion (x) to the fall in fetal heart rate during variable deceleration (y). There was no correlation for 15-second cord occlusions (y = -0.61x + 71.67, r = O.lO), whereas there was an apparent relationship for the 30-second occlusions (y = -3.73x + 146.19, r = 0.63). Recovery of tcPo, after cord occlusion (t/AtcPo,J) is very slow, taking 8 to 10 seconds for each torr tcPoz. In 10 instances, carotid arterial blood samples were taken 30 seconds after umbilical cord occlusion and Paoz was determined. Although tcPoz was still 3 to 7 torr below the value before cord occlusion, Pa,,* was found to be within 2 1 torr of the preocclusion tcPoB value. Relative heat output of the tcPo, electrode, i.e., the heat required to maintain electrode temperature, is thought to reflect perfusion of the skin under the electrode. Fig. 3 illustrates tcPo, and relative heat output in a hypoxic fetus subjected to a 15-second occlusion of the umbilical cord, followed by a prolonged uterine contraction. Characteristic “variable” and “late” deceleration of fetal heart rate is observed, with concomitant declines in fetal tcPo,. Amplification of the relative heat output signal is x 1,000 greater in Fig. 3 than in Fig. 2. There are inconsistent changes in relative heat output which do not allow a definitive conclusion about possible changes in skin perfusion. The pH value of carotid arterial blood during a period of stable tcPo, reading just before umbilical cord occlusion in this case was 7.21. In two experiments, after several cord occlusions, 0.2 mg of atropine was administered to the fetus via the carotid catheter and cord occlusions were repeated. In a total of three cord occlusions, changes in tcPo, were of the same magnitude as those seen with cord occlusions before atropine was given to the fetus; however, the observed decreases in fetal heart rate were delayed and diminished in amplitude.
Comment In the present study, we have examined changes in fetal heart rate, fetal blood pressure, continuously recorded tcPo,, and relative heat output of the tcPo2 electrode during well-quantitated umbilical cord occlusions in the subhuman primate fetus. During the experiments, there were uterine contractions during part of or the total duration of the studies, with a gradual decline in fetal blood pH and fluctuating PO, values in each animal. This makes the observations comparable to the situation in the human because variable decelerations of fetal heart rate can occur throughout the progression of labor, which is known to be accompanied by a gradual decrease in fetal pH and changing PO, values. During cord occlusions of 15 and 30 seconds’ duration, there was a linear decrease in tcPo,, as evidenced by a doubling in the tcPo, decrease when 15-and 30-second occlusions were compared. The delay in decrease of tcPo, after onset of umbilical cord occlusion (12 seconds, Fig. 2) is due to the time constant of the tcPo, electrode (90% response in 10 seconds) and the circulation time. Complete occlusion of the umbilical cord, as performed in this study, will lead to a sudden marked increase in peripheral vascular resistance because normally about 40% of fetal cardiac output is umbilical blood flow.” This leads to an immediate rise in arterial blood pressure, followed within 1 to 2 seconds by a decrease in fetal heart rate as a result of carotid baroreceptor stimulation. However, the initial rise in blood pressure is short lived, followed by a decrease in blood pressure which crosses the baseline blood pressure present before cord occlusion in most instances about 19 seconds after beginning of the cord occlusion (Fig. 2). Thus, the initial baroreceptor stimulation is of limited duration, and the continued decline in fetal heart rate while fetal blood pressure decreases from a peak has to be attributed to chemoreceptor stimulation in view of the continuing linear fall of tcPo2. When declines in fetal heart rate during 15- and 30-second umbilical cord occlusions are compared to decreases in tcPo, and changes in blood pressure, it is apparent that the response to baroreceptor stimulation is more pronounced than the response to chemoreceptor stimulation, as fetal heart rate declines 55 bpm during a 15-second occlusion and 78 bpm during a 30second occlusion (Table I). The difference in decline of fetal heart rate (23 bpm) occurring during the second 15 seconds of cord occlusion is the predominant result of chemoreceptor stimulation. This contention is supported by the fact that there is no relationship between fetal heart rate decrease and baseline tcPo2 during a 15-second cord occlusion; whereas, there is an apparent slight correlation during a 30-second cord occlusion. We attribute the latter to chemoreceptor stimula-
Volume Xumber
144 7
a situation which is similar to previously reported relationships in late decelerations of fetal heart rate in the monkey. lo Vagal blockade with atropine attenuates the decrease in fetal heart rate with umbilical cord compression. This is mainly the result of blocking the response to baroreceptor stimulation. However, the lesser contribution of c hemoreceptor stimulation to fetal heart rate decrease is also likely to be diminished because baroreceptor as well as chemoreceptor stimulation produces bradycardia via the vagus nerve. The role of a chemoreceptor reflex in fetal heart rate changes with umbilical cord compression has been pointed out in work in fetal lambs.” We did not observe any myocardial conduction defects on electrocardiogram tracings in association with umbilical cord occlusion, in contrast to a previous study in fetal baboons.” However, in the latter study, only first-degree atrioventricular block was observed after 20 seconds of umbilical cord occlusion. more severe conduction defects required umbilical cord occlusions for durations longer than in our stud\-. We have previously reported a close relationship between tcPoz and arterial Pop in fetal lambs when the tcPo2 recording has been stable for at least 2 minutes.’ The results of the present study indicate discrepancies between tcPoz and arterial PO, when fetal oxygenation changes rapidly. The lower tcPoB values during the period after cord occlusion when arterial Paz has already returned to its value before cord occlusion indicate an alteration in the distribution of cardiac output with a decrease in skin perfusion. In contrast to our observations with fetal tcPoz changes due to interruption of uterine blood flow in the near-term pregnant ewe,‘the fetal blood pressure is normal during the period of slow tcPo? recovery after umbilical cord occlusion in the present stuclv. During the gradual fetal tion,
Variable
decelerations
and tcPo2 during
cord occlusion
801
tcPo, recovery after interruption of uterine blood flow in the pregnant ewe, fetal blood pressure remains elevated until tcPo* has returned to normal.‘, ” This has been correlated with elevated blood concentrations of norepinephrine. I4 Cutaneous vasoconstriction due to catecholamines cannot be excluded in our observations with umbilical cord occlusion. Nevertheless, local factors regulating skin perfusion are more likely, considering the absence of systemic hypertension. Peripheral vasoconstriction has been reported during fetal hypoxia in the subhuman primate.‘j In fetal lambs, lower tcPo, values have been observed with low Pa,,, values; whereas, with normal or high Pa,, values, comparable tcPo, values were found.“j The energy required to heat the tcPoZ electrode is thought to be a reflection of relative skin perfusion under the electrode. Thus, significant changes in relative heat output of the tcPo, electrode would be expected, with marked cardiovascular changes in the fetus. However, in our observations, relative heat output changes were inconsistent and did not give an indication of skin perfusion (Fig. 3). This is in agreement with the observations of Jansen and associates.‘” In summary, our findings suggest that fetal heart rate changes during umbilical cord compression are mediated by a rapid baroreceptor component due to transient hypertension, followed by a decrease in blood pressure with continuing linear tcPo* fall resulting in a secondary chemoreceptor-controlled decline of fetal heart rate with lesser magnitude. The tcPoz decreases linearly during cord compression, with prolonged recovery after cord compression indicating decreased skin perfusion in view of Paoz values which are back to baseline. Heat output of the electrode does not indicate relative skin perfusion.
REFERENCES
1. Lee, S. T., and Hon, E. H.: Fetal hemodynamic response to umbilical cord compression, Obstet. Gynecol. 22:553, 1963. 2. Barcroft, J.: Researches on Pre-Natal Life, Oxford, 1946, vol 1, Blackwell Scientific Publications, Ltd., D. 123. 3. Kubli, F. W., Hon, E. H., Khazin, A. i., and’Takemura, H.: Observations on heart rate and pH in the human fetus during labor, AM. J. OBSTET. GYNECOL. 104: 1190,
7. Mueller-Heubach, E., Caritis, S. N., Edelstone, D. I., Huch, R., and Huch, A.: Comparison of continuous transcutaneous PO? measurements with intermittent ar8.
1969.
4. Huch, A., Huch, R., Schneider, H., and Rooth, G.: Continuous transcutaneous monitoring of fetal oxygen tension during labour, Br. J. Obstet. Gynaecol. (Suppl. 1)
9.
84: 1, 1975.
5. Willcourt, R. J., King, J. C., Indyk, L., and Queenan, J. T.: The relationship of fetal heart rate patterns to the fetal transcutaneous PO,, AM. J. OBSTET. GYNECOL. 140: 760. 1981. 6. Schneider, H., Huch, R., Schachinger, H., and Huch, A.: Correlation between fetal scalp tcPo, and microblood samples, Birth Defects 15:235, 1979.
10.
11.
terial PO, determinations in fetal lambs, Obstet. Gynecol. 57:248, 1981. Fall, O., Johnsson, M., Nilsson, B. A., Rooth, G., and Willdeck-Lund, G.: Continuous transcutaneous oxygen tension measurements in experiments with Rhesus monkeys and pregnant ewes, J. Ferinat. Med. 7:39, 1979. Cohn. H. E.. Sacks. A. I.. Hevmann, M. A., and RudolDh. A. MI: Cardiovas&lar”r~spdnses to hypoxemia and acidemia in fetal lambs, AM. J. OBSTET. GYNECOL. 120:817, 1974. Myers, R. E., Mueller-Heubach, E., and Adamsons, K.: Predictability of the state of fetal oxygenation from quantitative analysis of components of late deceleration, AM. J. OBSTET. GYNECOL. 115:1083, 1973. Siassi, B., Wu, P. Y. K., Blanco, C., and Martin, C. B.: Baroreceptor and chemoreceptor responses to umbilical
cord occlusion in fetal lambs, Biol. Neonate 3.5~66, 1979.
802
Mueller-Heubach
and Battelli
12. Yeh, M. N., Morishima, H. 0.. Nieman. U’. H., and James, L. S.: Myocardial conduction defects in association with comprewon of the umbilical cord. Experimental observations in fetal baboons, AM. J. OBSTET. GYNECOI.. 121:951. 1975. 13. Kiinzel. W.. Kastendieck. E., Kurr. C-S., and Paulik. R.: Transcutaneous PO, and cardiovascular observations in the sheep fetus following the reduction of uterine blood flow. I. Perinat. Med. 8:85. 1980. 14. Jensen, A., Kiinrel, W.. and Kastendieck, E.: Shock indication hp transcutaneous Pet, measurements in the sheep fetus and catecholamine release, in Proceedings of the
Second Annual Meeting, Sot. Perinat. Ohstet.. I!$x~. (Ahst. 31.) 1.5. Hehrman. R. E., Lees, M. H.. Peterson, E. N.. deLCclhli(t’r. C. W., and Seeds, .4. E.: Distribution of the circulaticm ITI the normal and asphyxiated fetal primate, AM, J. ~hSTFT. <;YNECOL. 108:956, 1970. 16. Jansen. C. A. M., Bass, F. CC.. Lorve. h. C-1.. and ‘Nathanielv, P. W.: Comparison of‘ continclc)uc trans. cutaneous and continuous intravascular Po? measurement in tctal sheep. AM. ,J. OBSTET. GYNECW.. 138:670, 19x0.
MUELLER-HEUBACH, F. BATTELLI,
M.D. B.S.
Pennsylz~ania
In five near-term rhesus monkeys under ketamine anesthesia, fetal heart rate, blood pressure, intrauterine pressure, and transcutaneous oxygen tension (tcPo,) were recorded continuously while the umbilical cord was ocduded for 15 or 30 seconds. Fetal heart rate decreased 55 + 19 bpm. (mean ? SD) and 78 2 20 bpm, respectively, with 15 and 30 second occlusions, while tcPo, declined 6.0 2 1.6 and 11.5 t 3.3 torr, respectively. Fetal hypertension started promptly with cord occlusion and reached a maximum within 5 to 16 seconds; fetal blood pressure decreased thereafter and, in most instances, fell below the baseline until fetal heart rate began to recover. Fetal heart rate decreased within 1 to 2 seconds of cord occlusion. tcPo, changes were the same, while heart rate changes were attenuated when the fetus was pretreated with 0.2 mg of atropine. These findings support a rapid and marked initial baroreceptor response, followed by a less marked chemoreceptor response. Thirty seconds following cord occlusion, Paon had returned to baseline, while tcPo, was still 3 to 7 torr below baseline, indicating cutaneous vasoconstriction even though systemic blood pressure was normal. Relative heat output required to maintain the tcPo2 electrode temperature at 44” C did not reflect alterations in skin perfusion during periods of cardiovascular change. (AM. J. OBSTET. GYNECOL. 144:796, 1982.)
VARIABLE DECELERATION of the fetal heart rate during labor is the most frequently observed deceleration pattern in human labor. As the name implies, it is somewhat variable in temporal relationship to uterine contractions and is the result of umbilical cord occlusion.’ It has been known since the work of Barcrof? that a vascular reflex involving the vagus nerve in combination with the effect of asphyxia on the cardiac muscle is involved in the fetal heart rate response to umbilical cord occlusion. Fetal capillary blood sampling in human fetuses during labor has revealed an increasing risk of fetal hypoxia and acidosis when variable decelerations of fetal heart rate become severe.3 Clini-
From the Department of Obstetrics and Gynecology, University of Pittsburgh School of Medicine, Magee- Wornens Hospital. Supported by Magee- Wornens Hospital Grant No. 292-414. Received for publication
Februay
Research
Fund
23, 1982.
Revised June 4, 1982. Accepted July 13, 1982. Reprint requests: Eberhard Mueller-Heubach, M.D., Magee- Womens Hospital, Forbes Ave. and Halket St., Pittsburgh, Pennsylvania 15213. 796
cal investigation of the relationship between fetal heart rate change during umbilical cord occlusion, vagally mediated vascular reflex, and oxygenation is complicated by the fact that degree and duration of umbilical cord occlusion during successive uterine contractions in human labor cannot be quantitated and may vary. Therefore, an animal model in which repetitive, wellquantitated umbilical cord occlusions can be produced experimentally is necessary. Alterations in fetal oxygenation during the rapid changes in fetal cardiovascular performance accompanying umbilical cord occlusion cannot be properly assessed by intermittent capillary blood sampling from the presenting part in the human fetus or intermittent arterial blood sampling from a catheter in the experimental animal fetus. The advent of a transcutaneous electrode for continuous measurement of oxygen tension (Po2) has provided a noninvasive means of recording PO, during labor in the human fetus.“’ 5 Studies in the human fetus6 as well as in the animal fetus7 indicate a close relationship between arterial PO* and transcutaneous Po2 (tcPo,). The present study was designed to investigate the relationship between fetal heart changes and tcPo, during well-quantitated periods of umbilical cord occlusion in the near-term rhesus monkey. The aim of the study 0002-9378/82/230796+07$00.70/O
0
1982 The C. V. Mosbv Co.
Volume Number
144 7
Variable
decelerations
and tcPo,
during
cord occlusion
797
was to delineate the contributions of the vascular vagal reflex and fetal hypoxia to the pattern of variable decelerations of fetal heart rate. In addition, the usefulness of the tcPo? electrode during the changes in fetal oxygenation accompanying umbilical cord occlusion was studied.
Material and methods Five pregnant rhesus monkeys from the Department of Obstetrics and Gynecology at the University of Pittsburgh School of Medicine were used. Their gestational ages at the time of the experiments were 158 to 162 days. (Term is about 168 days). Before the experiments, the animals were not fed for 12 hours overnight but were allowed water ad libitum. Anesthesia was established by intramuscular injection of 100 mg of ketamine. Throughout the experiment, anesthesia and hydration were maintained by infusion of ketamine in isotonic saline solution (I mg/ml) at a rate of 1 ml/min. Laparotomy and hysterotomy through a low-segment transverse incision were done. The fetal head and neck were delivered from the uterus, and a polyvinyl catheter (0.015 inside diameter) was inserted into a carotid artery. Silver/silver chloride disk electrodes were fixed subcutaneously at both fetal shoulders. A plastic inflatable occlusion device (Rhodes Medical Instruments, Inc., Woodland Hills, California) with an internal diameter of 10 mm was placed around the umbilical cord close to the fetal abdomen. Care was taken that the occlusion device did not kink the umbilical cord. The upper chest of the fetus was then shaved, and the tcPoz electrode was glued with Eastman 910 tissue glue to the upper fetal chest over an intercostal space slightly lateral to the sternum. In preliminary studies, the tcPo, electrode had been glued to the shaved scalp of the monkey, fetus; however, in this location the tcPo2 electrode did not yield any tcPo, readings. The partially delivered fetus was then gently returned into the uterus. A polyvinyl catheter (0.050 inside diameter) was placed into the uterine cavity. Uterine and abdominal incisions were closed. Loss of amniotic fluid was estimated and replaced with isotonic saline solution through the intrauterine catheter. Studies were done after a l-hour recovery period following surgery. Intermittent total occlusions of the umbilical cord were produced by complete inflation of the occlusion device for periods of 15 and 30 seconds in random sequence. This was accomplished by timed injections of 2 ml of isotonic saline solution through the catheter of the device. It had been established before placement of the occlusion device that a 2 ml volume completely obliterated the internal opening of the occludrr. Injection and removal of the 2 ml of saline so-
Fig. 1. tcPo,, fetal heart rate (FHR), fetal blood pressure (FBP) and intrauterine pressure (IUP) during a 30-second umbilical cord occlusion at a paper speed of 3 cmimin.
lution at beginning and end of cord occlusions took no more than 2 seconds. Fetal carotid arterial blood samples were taken intermittently. Umbilical cord occlusions were done only when carotid arterial blood pH values were above 7.20. In cases of spontaneous uterine activity, umbilical cord occlusions were done at least 90 seconds after the end of the previous uterine contraction. instances of coincident umbilical cord occlusion and uterine contraction were not included in the analysis. Study periods did not exceed 3 hours, since previous observations in fetal lambs and juvenile rhesus monkeys6 have indicated that tcPo2 readings become unreliable after this time period as a result of skin changes due to the heat of the electrode. At the end of the study period, cesarean section was done, the fetus was delivered, and mother and newborn were returned to their cage. Measurement techniques. Carotid arterial and in-
798
Mueller-Heubach
and Battelli
December 1, IYX’L .%m. J. Obstet. Gynecrd.
Fig. 2. tcPo,, electrocardiogram (KG), intrauterine relative heat output (RNO). and fetal heart rate (FHRI speed of 15 cmimin
Table I. Relationship between fetal heart rate and tcPo* during umbilical cord compression Cwd
com-
pession
(set) 15 30
AFHR M-4
J.
55 -t 19 78 k 20
&cPo, I ctmr)
AFHRLI AtcPo,J. (bpmlto7-7)
6.0 ” 1.6 11.5 t 3.3
9 -e 3.6 7 -e 2.4
~/A&PO,?. (secltm) 10 2 2.3 8 2 2.8
All values are mean 2 W. trauterine pressures were measured with Statham P23 Db pressure transducers. Fetal blood samples were taken intermittently from the carotid arterial catheter; pH, PO,, and base deficit were determined with a Corning 165 blood gas analyzer. Fetal heart rate was obtained with a cardiotachometer triggered by the R wave of the electrocardiogram signal. The tcPo, electrode was connected to an electronic signal processor, from which signals of tcPo, and relative heat output or “flow” were recorded on an eight-channel Beckman R611 recorder together with fetal heart rate, arterial
pressure (ICP). fetal blood pressure (FBP). during a ?&second cord occlusion at a papel-
blood pressure, and intrauterine pressure. The relative heat output signal was amplified over the range of 5 V/mm to 0.1 mV/mm. Speeds of the recording paper ranged from 3 to 15 cm/min. The tcPo2 electrode was heated to a temperature of 14” C only during the actual study periods. Before attachment of the electrode, calibration was carried out by blowing nitrogen or 5.79% oxygen
over
the
electrode
surface.
which
yielded
ox”-
gen tensions of 0 and about 40 torr. respectively, depending on barometric pressure. Ninety percent ofthe electrode response was achieved within 10 seconds (time constant). At the end of the studies, the tcPo* electrode was recalibrated as described. Experiments in which tcPo, values at calibration and recalibration had drifted more than & 1 torr were excluded from analysis. Results Comparisons of tcPo, readings and intermittently obtained Paot values of carotid arterial blood were made when the tcPo, readings were stable for 2 minutes. Only studies where tcPo, and Paor values were
Volume
144
Numbw
7
Variable decelerations and tcPq during cord occlusion
799
F%P tow : 4%-.
m 15
SEC.
CORD
@XLUSlON
Fig. 3. tcPoz, fetal heart rate (FHR), fetal blood pressure relative heat output (RHO) during a 15-second umbilical contraction at a paper speed of 6 cm/min. within % 1 torr have been included. Uterine activity was present during part of or rhe total duration of the studies. Fetal baseline values before umbilical cord occlusions had the following ranges: pH 7.29 to 7.20; Paoz 24 to 12 torr: systolic blood pressures, 59 to 78 torr: diastolic blood pressures, 34 to 44 tort-; and fetal heart rate, 138 to 210 bpm. A\ in human labor, sequential pH values during the studies gradually declined, while PO, fluctuated. The effects of a 30-second occlusion of the umbilical cord on fetal tcPoz, heart rate, and blood pressure are illustrated in Fig. 1. Fetal heart rate decreases and recovers rapidly, the characteristic pattern of a “variable deceleration” in clinical monitoring. Fetal blood pressure increases initially and then declines toward the end of the umbilical cord occlusion. The tcPo* begins to decrease shortly after onset of the cord occlusion and recovers slowlv after the end of the occlusion, while fetal heart rate and blood pressure have returned to the values before occlusion. The temporal relationships between umbilical cord occlusion, fetal heart rate, fetal blood pressure, and tcPo, are clearly delineated in a recording obtained at a paper speed of 15 cmimin (Fig. 2). Fetal blood pressure increases 1 to
(FBP), intrauterine cord occlusion and
pressure (ZUP), and a prolonged uterine
1.5 seconds after start of umbilical cord compression, followed within 1 to 2 seconds by a progressive decline in fetal heart rate. The tcPo, starts to decrease 12 seconds after the onset of umbilical cord compression and does not return to its value before occlusion until 75 seconds after the cord occlusion has been terminated. To characterize the relationships between fetal tcPoZ and fetal heart rate more precisely, data from 24 umbilical cord occlusions of 15 or 30 seconds duration were summarized (Table I). In three animals, four occlusions were done; in the remaining two animals, five and seven occlusions, respectively, were performed. Maximal fetal heart rate decrease (AFHRJ) was rapid during a 15-second cord occlusion. A lesser decrease was seen during the second 15 seconds of the 30-second cord occlusions. This is compatible with the concept of a vagalfy mediated fetal heart rate decrease due to baroreceptor stimulation during the initial increase in fetal blood pressure. The time from onset of cord occlusion to maximal fetal blood pressure elevation (i.e., maximal baroreceptor stimulation) ranged from 5 to 16 seconds. However, this initial blood pressure increase is followed by a decrease, resulting in return of
800
Mueller-Heubach
December 1, 1982 Am. J. Obstet. Gynecol.
and Battelli
the blood pressure to the preocclusion value 19 seconds after beginning of the occlusion and subsequent hypotension throughout the remainder of the period of cord occlusion (Fig. 2). Although fetal blood pressure decreased after reaching a maximum, this decrease went beyond baseline blood pressure in only 19 instances (Fig. 2), while in five observations fetal blood pressure decrease did not exceed baseline fetal blood pressure before cord occlusion (Fig. 1). The tcPo, decrease (AtcPo,i) during 15- and 30-second cord occlusions is progressive and linear. TcPo, decrease was measured by subtracting the tcPo2 value at the nadir of tcPo, decrease from the tcPoz value at the moment when cord occIusion was started. The fall in fetal heart rate for each torr decrease in tcPo, (AFHRJlAtcPo,&) is more marked during the first 15 seconds of an umbilical cord occlusion. We have compared the tcPo* value at the beginning of umbilical cord occlusion (x) to the fall in fetal heart rate during variable deceleration (y). There was no correlation for 15-second cord occlusions (y = -0.61x + 71.67, r = O.lO), whereas there was an apparent relationship for the 30-second occlusions (y = -3.73x + 146.19, r = 0.63). Recovery of tcPo, after cord occlusion (t/AtcPo,J) is very slow, taking 8 to 10 seconds for each torr tcPoz. In 10 instances, carotid arterial blood samples were taken 30 seconds after umbilical cord occlusion and Paoz was determined. Although tcPoz was still 3 to 7 torr below the value before cord occlusion, Pa,,* was found to be within 2 1 torr of the preocclusion tcPoB value. Relative heat output of the tcPo, electrode, i.e., the heat required to maintain electrode temperature, is thought to reflect perfusion of the skin under the electrode. Fig. 3 illustrates tcPo, and relative heat output in a hypoxic fetus subjected to a 15-second occlusion of the umbilical cord, followed by a prolonged uterine contraction. Characteristic “variable” and “late” deceleration of fetal heart rate is observed, with concomitant declines in fetal tcPo,. Amplification of the relative heat output signal is x 1,000 greater in Fig. 3 than in Fig. 2. There are inconsistent changes in relative heat output which do not allow a definitive conclusion about possible changes in skin perfusion. The pH value of carotid arterial blood during a period of stable tcPo, reading just before umbilical cord occlusion in this case was 7.21. In two experiments, after several cord occlusions, 0.2 mg of atropine was administered to the fetus via the carotid catheter and cord occlusions were repeated. In a total of three cord occlusions, changes in tcPo, were of the same magnitude as those seen with cord occlusions before atropine was given to the fetus; however, the observed decreases in fetal heart rate were delayed and diminished in amplitude.
Comment In the present study, we have examined changes in fetal heart rate, fetal blood pressure, continuously recorded tcPo,, and relative heat output of the tcPo2 electrode during well-quantitated umbilical cord occlusions in the subhuman primate fetus. During the experiments, there were uterine contractions during part of or the total duration of the studies, with a gradual decline in fetal blood pH and fluctuating PO, values in each animal. This makes the observations comparable to the situation in the human because variable decelerations of fetal heart rate can occur throughout the progression of labor, which is known to be accompanied by a gradual decrease in fetal pH and changing PO, values. During cord occlusions of 15 and 30 seconds’ duration, there was a linear decrease in tcPo,, as evidenced by a doubling in the tcPo, decrease when 15-and 30-second occlusions were compared. The delay in decrease of tcPo, after onset of umbilical cord occlusion (12 seconds, Fig. 2) is due to the time constant of the tcPo, electrode (90% response in 10 seconds) and the circulation time. Complete occlusion of the umbilical cord, as performed in this study, will lead to a sudden marked increase in peripheral vascular resistance because normally about 40% of fetal cardiac output is umbilical blood flow.” This leads to an immediate rise in arterial blood pressure, followed within 1 to 2 seconds by a decrease in fetal heart rate as a result of carotid baroreceptor stimulation. However, the initial rise in blood pressure is short lived, followed by a decrease in blood pressure which crosses the baseline blood pressure present before cord occlusion in most instances about 19 seconds after beginning of the cord occlusion (Fig. 2). Thus, the initial baroreceptor stimulation is of limited duration, and the continued decline in fetal heart rate while fetal blood pressure decreases from a peak has to be attributed to chemoreceptor stimulation in view of the continuing linear fall of tcPo2. When declines in fetal heart rate during 15- and 30-second umbilical cord occlusions are compared to decreases in tcPo, and changes in blood pressure, it is apparent that the response to baroreceptor stimulation is more pronounced than the response to chemoreceptor stimulation, as fetal heart rate declines 55 bpm during a 15-second occlusion and 78 bpm during a 30second occlusion (Table I). The difference in decline of fetal heart rate (23 bpm) occurring during the second 15 seconds of cord occlusion is the predominant result of chemoreceptor stimulation. This contention is supported by the fact that there is no relationship between fetal heart rate decrease and baseline tcPo2 during a 15-second cord occlusion; whereas, there is an apparent slight correlation during a 30-second cord occlusion. We attribute the latter to chemoreceptor stimula-
Volume Xumber
144 7
a situation which is similar to previously reported relationships in late decelerations of fetal heart rate in the monkey. lo Vagal blockade with atropine attenuates the decrease in fetal heart rate with umbilical cord compression. This is mainly the result of blocking the response to baroreceptor stimulation. However, the lesser contribution of c hemoreceptor stimulation to fetal heart rate decrease is also likely to be diminished because baroreceptor as well as chemoreceptor stimulation produces bradycardia via the vagus nerve. The role of a chemoreceptor reflex in fetal heart rate changes with umbilical cord compression has been pointed out in work in fetal lambs.” We did not observe any myocardial conduction defects on electrocardiogram tracings in association with umbilical cord occlusion, in contrast to a previous study in fetal baboons.” However, in the latter study, only first-degree atrioventricular block was observed after 20 seconds of umbilical cord occlusion. more severe conduction defects required umbilical cord occlusions for durations longer than in our stud\-. We have previously reported a close relationship between tcPoz and arterial Pop in fetal lambs when the tcPo2 recording has been stable for at least 2 minutes.’ The results of the present study indicate discrepancies between tcPoz and arterial PO, when fetal oxygenation changes rapidly. The lower tcPoB values during the period after cord occlusion when arterial Paz has already returned to its value before cord occlusion indicate an alteration in the distribution of cardiac output with a decrease in skin perfusion. In contrast to our observations with fetal tcPoz changes due to interruption of uterine blood flow in the near-term pregnant ewe,‘the fetal blood pressure is normal during the period of slow tcPo? recovery after umbilical cord occlusion in the present stuclv. During the gradual fetal tion,
Variable
decelerations
and tcPo2 during
cord occlusion
801
tcPo, recovery after interruption of uterine blood flow in the pregnant ewe, fetal blood pressure remains elevated until tcPo* has returned to normal.‘, ” This has been correlated with elevated blood concentrations of norepinephrine. I4 Cutaneous vasoconstriction due to catecholamines cannot be excluded in our observations with umbilical cord occlusion. Nevertheless, local factors regulating skin perfusion are more likely, considering the absence of systemic hypertension. Peripheral vasoconstriction has been reported during fetal hypoxia in the subhuman primate.‘j In fetal lambs, lower tcPo, values have been observed with low Pa,,, values; whereas, with normal or high Pa,, values, comparable tcPo, values were found.“j The energy required to heat the tcPoZ electrode is thought to be a reflection of relative skin perfusion under the electrode. Thus, significant changes in relative heat output of the tcPo, electrode would be expected, with marked cardiovascular changes in the fetus. However, in our observations, relative heat output changes were inconsistent and did not give an indication of skin perfusion (Fig. 3). This is in agreement with the observations of Jansen and associates.‘” In summary, our findings suggest that fetal heart rate changes during umbilical cord compression are mediated by a rapid baroreceptor component due to transient hypertension, followed by a decrease in blood pressure with continuing linear tcPo* fall resulting in a secondary chemoreceptor-controlled decline of fetal heart rate with lesser magnitude. The tcPoz decreases linearly during cord compression, with prolonged recovery after cord compression indicating decreased skin perfusion in view of Paoz values which are back to baseline. Heat output of the electrode does not indicate relative skin perfusion.
REFERENCES
1. Lee, S. T., and Hon, E. H.: Fetal hemodynamic response to umbilical cord compression, Obstet. Gynecol. 22:553, 1963. 2. Barcroft, J.: Researches on Pre-Natal Life, Oxford, 1946, vol 1, Blackwell Scientific Publications, Ltd., D. 123. 3. Kubli, F. W., Hon, E. H., Khazin, A. i., and’Takemura, H.: Observations on heart rate and pH in the human fetus during labor, AM. J. OBSTET. GYNECOL. 104: 1190,
7. Mueller-Heubach, E., Caritis, S. N., Edelstone, D. I., Huch, R., and Huch, A.: Comparison of continuous transcutaneous PO? measurements with intermittent ar8.
1969.
4. Huch, A., Huch, R., Schneider, H., and Rooth, G.: Continuous transcutaneous monitoring of fetal oxygen tension during labour, Br. J. Obstet. Gynaecol. (Suppl. 1)
9.
84: 1, 1975.
5. Willcourt, R. J., King, J. C., Indyk, L., and Queenan, J. T.: The relationship of fetal heart rate patterns to the fetal transcutaneous PO,, AM. J. OBSTET. GYNECOL. 140: 760. 1981. 6. Schneider, H., Huch, R., Schachinger, H., and Huch, A.: Correlation between fetal scalp tcPo, and microblood samples, Birth Defects 15:235, 1979.
10.
11.
terial PO, determinations in fetal lambs, Obstet. Gynecol. 57:248, 1981. Fall, O., Johnsson, M., Nilsson, B. A., Rooth, G., and Willdeck-Lund, G.: Continuous transcutaneous oxygen tension measurements in experiments with Rhesus monkeys and pregnant ewes, J. Ferinat. Med. 7:39, 1979. Cohn. H. E.. Sacks. A. I.. Hevmann, M. A., and RudolDh. A. MI: Cardiovas&lar”r~spdnses to hypoxemia and acidemia in fetal lambs, AM. J. OBSTET. GYNECOL. 120:817, 1974. Myers, R. E., Mueller-Heubach, E., and Adamsons, K.: Predictability of the state of fetal oxygenation from quantitative analysis of components of late deceleration, AM. J. OBSTET. GYNECOL. 115:1083, 1973. Siassi, B., Wu, P. Y. K., Blanco, C., and Martin, C. B.: Baroreceptor and chemoreceptor responses to umbilical
cord occlusion in fetal lambs, Biol. Neonate 3.5~66, 1979.
802
Mueller-Heubach
and Battelli
12. Yeh, M. N., Morishima, H. 0.. Nieman. U’. H., and James, L. S.: Myocardial conduction defects in association with comprewon of the umbilical cord. Experimental observations in fetal baboons, AM. J. OBSTET. GYNECOI.. 121:951. 1975. 13. Kiinzel. W.. Kastendieck. E., Kurr. C-S., and Paulik. R.: Transcutaneous PO, and cardiovascular observations in the sheep fetus following the reduction of uterine blood flow. I. Perinat. Med. 8:85. 1980. 14. Jensen, A., Kiinrel, W.. and Kastendieck, E.: Shock indication hp transcutaneous Pet, measurements in the sheep fetus and catecholamine release, in Proceedings of the
Second Annual Meeting, Sot. Perinat. Ohstet.. I!$x~. (Ahst. 31.) 1.5. Hehrman. R. E., Lees, M. H.. Peterson, E. N.. deLCclhli(t’r. C. W., and Seeds, .4. E.: Distribution of the circulaticm ITI the normal and asphyxiated fetal primate, AM, J. ~hSTFT. <;YNECOL. 108:956, 1970. 16. Jansen. C. A. M., Bass, F. CC.. Lorve. h. C-1.. and ‘Nathanielv, P. W.: Comparison of‘ continclc)uc trans. cutaneous and continuous intravascular Po? measurement in tctal sheep. AM. ,J. OBSTET. GYNECW.. 138:670, 19x0.