- Email: [email protected]
Cardiac systolic time intervals in fetal monkeys: Pre-ejection period
Cardiac systolic time intervals in fetal monkeys: Pre-ejec tion period YUJI
MURATA,
CHESTER
B.
TSUYOMU ROY
M.D. MARTIN,
JR.,
IKENOUE,
H.
PETRIE,
Los Angeles,
California
M.D.
M.D.*
M.D.**
The systolic time intervals of the fetal cardiac cycle were studied by means of simultaneous recordings of electrocardiogram (ECG) and uitrasound Doppler cardiogram (GCG) in chronic preparations of fetal rhesus monkeys. Recordings were made under physiologic conditions as well as during various experimental stresses. The pre-ejection period (PEP) showed no significant relationship with heart rate in the unstressed fetuses, but the acceleration of heart rate induced by epinephrine was accompanied by shortening of PEP. The PEP increased with advancing fetal age. The PEP was inversely correlated with left ventricular end-diastolic pressure and arterial pulse pressure, but showed a positive correlation with both systolic and diastolic arterial blood pressure. The PEP also exhibited strong negative correlation with arterial blood pH. The prolongation was essentially the same whether acidosis was of respiratory or metabolic origin. The PEP increased slightly but significantly during nonacidemic hypoxemia; however, there was no correlatii between Pao, and PEP. Epinephrine shortened the PEP significantly, whereas the effect of a&opine was inconsistent. Afteratlon of the plasma glucose level by injection of insulin or glucose did not affect PEP. These findings demonstrate that the PEP may be a useful indicator of fetal cardiac performance, reflecting both myocardial contractility and the hemodynamic state of the cardiovascular system. (AM. J. OBSTET. GYNECOL. 132: 285, 1978.)
THE SYSTOLIC time intervals (STI) of the cardiac cycle have been found to be useful indicators of myocardial function.’ More recently, these intervals have been investigated as possible indicators of the functional state of the fetal heart, with the use of the fetal electrocardiogram (FECG) in combination with
From the Department of Obstetrics and Gynecology, Angeles County-University of Southern California Medical Center.
Los
Supported in part by Grant HD-06406 from the National Institutes of Health, United States Public Health Service, and by the Rohi ScienfiJc Corporation, Santa Ana, Calijornifl. Received for publication
Augwt
Revised December
16, 1977.
AcceptedJanuq
23, 1978.
6, 1976.
Reprint requests: Yuj Murata, M.D., Department of Obstetrics and Gynecology, I240 N. Mission Rd., Los Angeles, California 90033. *Present address: Kagoshima Municipal Hospital, Department of Obstetrics and Gynecology, Kagoshima, Japan. **Present address: College of Phy+cians and Surgeons, Columbia University, Department of Obstetrics and G>Ttecology, New York, New York. OOOZ-9378178/190285+09$00.9010
@ 1978
The C. V. Mosby
Co.
the phonocatdiogramZ or ultrasonic Doppler cardiogram (DCG) to detect and measure the STI’s2* 4, 5 Prolongation of the pre-ejection period (PEP) was observed in five fetuses exhibiting heart rate patterns indicative of mild distress. In other studies,4* 5 changes in the STI’s were demonstrated during variable deceleration (cord compression) and late deceleration (hypoxic) FHR patterns. These findings provided evidence that the STI’s might yield clinically useful information regarding the status of the fetal heart and thus the condition of the fetus. In order to define more clearly the factors affecting the STI’s of the fetal heart, the studies described herein were carried out in fetal rhesus monkeys under relatively physiologic conditions, during controlled stress, induced fetal stress, or spontaneously occurring fetal distress in labor.
Materfals and methods These experiments were performed in pregnant rhesus monkeys obtained from the breeding colony at the Primate Center at the University of CaliforniaDavis. The duration of pregnancy was known from timed matings in 11 monkeys and in the remaining 285
286
Table
Murata
et al.
I. Summary
of experiments No. of
Type of experiment
High CO2 administration to mother Low O2 administration to mother 0.15N HCI infusion to fetus l!.LZ&: Oxytocin-induced Spontaneous following surgery Epinephrine
animals
No. of experiments 6 6 3
2 3
2 3
administratbn:
To unstressed fetus To distressed fetus Atropine administration to fetus 20% glucose administration to mother Insulin administration to mother and/or fetus
4 5 5 3 2
4
four it was estimated from rectal palpation early in pregnancy. ECG electrodes and carotid or femoral arterial catheters were implanted in rheses monkey fetuses between 115 days’ gestation and near term by means of techniques previously described.6 A catheter was placed in the left cardiac ventricle through the ascending aorta in one fetal monkey. The location was confirmed by the configuration of the pressure tracing and by subsequent autopsy. Intrauterine catheters were inserted in order to measure uterine contractions. A catheter was placed in the jugular vein of six fetuses. Following the operation, the mothers were placed in restraining chairs and permitted to recover from the anesthesia. Eleven animals became chronic preparations, and were studied at times ranging from 48 hours to 34 days postoperatively. Four other animals developed progressively increasing uterine contractions following the operation, and were studied between 12 and 48 hours postoperatively. The FECG, FHR, fetal blood pressure, and intrauterine pressure were recorded on a standard biomedical recorder and simultaneously on magnetic tape. Fetal arterial blood samples were obtained from the arterial catheters for determination of pH and respiratory gas tensions. The samples were collected anaerobically in heparinized plastic syringes. The pH was determined with a Radiometer microelectrode, and PoZ and Pcq were measured with Radiometer BMS-3 or IL micro blood-gas analyzers. For studies of the relationship between STI’s and fetal biochemical parameters, blood samples were obtained either immediately before or immediately after the DCG record was made. The fetal DCG was obtained with a Kretz (MGD-2) fetal pulse detector. The output of the instrument was recorded on a second tape recorder at 7% ‘or 15 i.p.s.,
along with the FECG, blood pressure, and intrauterine pressure. These tapes were later played back at an eightfold reduction in speed through appropriate filters into a high-frequency ink-writing rerordet (Elema Mingograf Model No. 24.B) in order to measure the STI’s manually. The tapes were also used to generate a continuous microfilm record of the STI‘s in successive cardiac cycles. These techniques have been described previously.“, I In these studies on intrauterine fetuses, it has not been possible to determine whether mitral or tricuspid, or aortic or pulmonic valve signals (or both in combination) were being recorded on the DCG. In this report, the first valve-motion signal of the FDCG has been designated MC, and the second and third, A0 and AC, respectively. Separate right and left ventricular STI’s could not be measured, since multiple intracardiac and intra-arterial catheters were not implanted. Four STI’s were measured from the combined ECG-DCG records: (1) from the beginning of the Q wave to the end of the atrioventricular valve closing signal (Q-MC = electromechanical latent time, EMLT): (2) from the onset of Q to the beginning of the semilunar valve opening signal (Q-Ao = pre-ejection period, PEP); (3) the difference of those two intervals (McAo = isovolumetric contraction time, ICI‘); and (4) from the beginning of the semilunar valve opening signa! to the end of the closing signal (Ao-AC = ventricular ejection time, VET). The findings relating to the first three of these intervals will be described in this report. The values for the STI’s used in most of the determinations were the mean values calculated from 20 consecutive cardiac cycles. The variability of the intervals within this number of cycles was small: for example, the mean coefficient of variation for Q-Ao was 4.5 per cent, with a range from 3.5 to 6.3 per cent. The values for the intervals within individual cardiac cycles were used in searching for a relationship between the STI’s and the R-R interval. In the chronic preparations, combined ECG-DCG records were obtained from fetuses under unstressed conditions. These records were repeated at intervals over periods ranging from a few days to 4% weeks. In most instances, the mother accepted the hand-held placement of the ultrasound transducer against the abdomen without protest, so that no element of fetal distress produced by maternal fear or struggling was likely to be present. Normal FHR variability plus accelerations and decelerations occurring with gross fetal movements and fetal breathing movements provided a range of heart rates from 120 to 240 b.p.m. In a few
Volume Number
Cardiac systolic time intervals
132 3
instances, transabdominal compression of the fetal head or flushing the fetal trachea catheter with a small volume of saline was employed to evoke transient FHR decelerations. In addition to these normal, unstressed observations, combined ECG-DCG recordings were obtained from fetuses stressed by a variety of techniques (Table I). The maternal inspiratory gas tensions were altered by administering mixtures of varying compositions to the mother by means of a lucite helmet. The approximate volume of the helmet was 2 L. Flow rates of 6 to 8 L. per minute were maintained to ensure a minimum of rebreathing. Acute respiratory acidosis was produced by the administration of a mixture of 20 per cent COz and 20 per cent O2 in nitrogen. The fetal arterial blood PO, in these experiments did not decrease in any instance and the maximum increase was 16 mm. Hg. The fetal arterial PO, was lowered by administering low oxygen mixtures to the mother-as low as 8 per cent maternal FI,. The fetal arterial blood pH remained relatively constant in most of these experiments. In one experiment in which the hypoxia was prolonged, the fetal pH decreased by 0.33 unit. In other instances, fetal acidemia was induced by the administration of 0.15N HCl to the fetus at rates up to 0.4 ml. per minute. The endpoint in these acute stress experiments was slowing of the FHR to the range of 90 to 120 b.p.m. The acute stress experiments generally lasted between 10 and 30 minutes, the ST1 records being made near the end of the stress period. The single prolonged hypoxic experiment continued for 90 minutes, with multiple ST1 recbrds being obtained. The combination of fetal hypoxemia and acidosis resulting from uterine contractions was studied in two animals during oxytocin-induced uterine activity and in three animals during labor following the operation. In two of these latter animals, the fetal arterial blood PO, was increased without significant change in pH by increasing the maternal Fro*; in one of them the pH was altered without significant PO, change by the administration of bicarbonate to the fetus. Epinephrine was administered to four fetuses (nine experiments). Two of these fetuses (five experiments) were severely distressed. Atropine was also infused in five experiments into three unstressed fetuses. Sixteen determinations of the STI’s were performed on three established chronic fetal monkey preparations to investigate any relationship between the systolic time intervals and fetal blood glucose level. Acute hyper- or hypoglycemia was induced-by infusion of 20 per cent glucose solution to the mother or of insulin to the mother and/or fetus. Fetal plasma glucose values were
1 110
I 120
I 130
I 140
I 150
110
I 120
I 130
I 140
1 150
GESTATIONAL
AGE
287
[days]
Fig. 1. Relationships between systolic time intervals and gestational age in normal fetal rhesus monkeys. A, Electromechanical latent time (Q-MC): B, pre-ejection period (Q-Ao); C, QRS duration. elevated as high as 296 mg. per 100 ml. and decreased as low as 6 mg. per 100 ml.
Results Heart rate. In the unstressed fetal monkeys, there was no sigtificant correlation between the duration of the Q-Ao interval (PEP) and heart rate. This was true both for mean values and for individual cycles recorded during decelerations or accelerations. Gestational a&. The durations of both the Q-MC and Q-Ao intervals increased with advancing gestational age in the unstressed fetuses (p < 0.0 l)* (Fig. 1). The slope of the regression line for Q-Ao and gestational age is somewhat steeper than that for Q-MC, but *Pearson Product-moment
correlation
coefficient.
288
Murata et al.
I_ B
50 $
.
*
40
/
I
.40
SYSTOLIC
Fig. 2. Correlation pressure in normal
50
60
BLOOD mmHg
70
80
between the pre-ejection fetal rhesus monkeys.
.
. I.*. .. * *. .: -. . . . 1 *.. * .
20
-
BLOOD mmHg
(Q-Ao) and systolic
period
. . *.
30
DIASTOLIC
PRESSURE
. S’
.
2 % a
PtP=O25iDBP,+35: 4 51 n=49 P<“O1 .
(Ai and
.
40
PRESSURE
diastolic
(B) blood
80
60
12
14
ARTERIAL Fig. 3. Relationship
between
the pre-ejection
PULSE period
20
18
16
PRESSURE (Q-Ao)
(mmHg) and pulse
pressure
in a distressed
fetal
monkey. the difference is not statistically significant. The lengthening of these intervals with advancing gestational age appeared to result entirely from prolongation of the time required for electrical depolarization of the ventricles (QRS duration, Fig. 1, C). The quantity (Q-A0 interval minus QRS duration} was constant throughout the range of gestational ages studied, and varied narrowly in the normal fetuses between 12.6 and 15.6 msec. Blood pressure. The Q-Ao interval was positively correlated with both systolic and diastolic blood pressure (Fig. 2, A and B). The intervals Q-MC and MC-Ao were not significantly related to either blood pressure value. In the one fetus studied by means of a left ventricular catheter, an inverse relationship was observed between Q-Ao and left ventricular end-
diastolic pressure over the observed range I to 9 mm. Hg (r = 0.82, n = 33, p < 0.01). This fetus appeared to be stable with regard to heart rate, systolic blood pressure, and arterial blood PI-I values, which were within the normal range throughout the observation period. The range of aortic pulse pressures recorded in the normal monkeys was rather small, and no relationship could lx detected between PEP and pulse pressure. In one distressed fetus, however, there was a clear inverse correlation between Q-Ao duration and aortic QUke pressure (Fig. 3, p < O.Ol).* Both systolic and diastolic blood pressure showed a slight but significant rise with decreasing arterial blood *Pearson
Product-moment
correlation
coefficient.
Volume
Cardiac systolic time intervals
132
289
Number3
. .
20
.
k.
.
:
i
. . .. .* I * !‘.iri.ew .
1:
-.
0:.
.
. .
.
.
.
l
l
.-
.*. .
l
l *r.
OMc=-15(pH)+130 r=-0 31 n=85
60-
.
’ .
.
.
l .
50-
,
CIAO=-36(pH)+300 r=-0 72 ~86 I
70
7.1
30-
I
7.2
7.3
7.4
l
.
lo-
.
.
.
McAo=2llpH)+170 r=O 38 n=85
710
.
7:2
7'1
:
.
l
7'4
713
ARTERIAL BLOOD pH Fig. 4. Correlation Electromechanical time (MC-Ao).
between latent
systolic
time intervals
under physiologic (systolic blood pressure = -4OOpH + conditions 2,996, r = -0.48, n = 45, p < 0.01; diastolic blood pressure = -263 pH + 1,963, r = -0.42, n = 45, p < O.Ol).” Arterial blood pH. The Q-MC, Q-Ao, and MC-Ao intervals became prolonged with fetal acidemia or acidosis (Fig. 4). The relationship between the duration of these intervals and the pH of fetal arterial blood was linear in the range from pH 6.99 to pH 7.48 (p < 0.01) in all three cases.* The slope of the regression line for the Q-Ao interval is steeper than that for Q-MC, dem-
pH
in
the unstressed
fetal
*Pearson Proclua-moment
and fetal
time (Q-MC); B, pre-ejection
monkeys
correlation
coefficient.
arterial
blood
pH in rhesus
monkeys.
A,
period (Q-Ao); C, isovolumetric contraction
onstrating that both components of the PEP (EMLT, measured as Q-MC, and ICT, derived) become prolonged during acidosis. The duration of the QRS complex did not change significantly over the observed range of pH. The prolongation of PEP observed during fetal acidemialacidosis did not seem to be influenced by the mechanisms producing the lowered pH. Thus, the data points from the experiments in which acute acidemia was induced in healthy fetuses by administration of COs to the mother were undistinguishable from those observed during infusions of HCI into the fetus. Further, the values from the acute acidemia experiments were similar, at equivalent pH levels, to those
290
et al.
Murata
October
Am. J. Obstet.
Fig. 5. Continuous
55
: m
1 I
.
45
E 2
‘0.
. 35
.
:
.a.*
.
.**
.*
l s:l
.
I
display of the systolic time intervals during maternal hypercapnea.
*
. .- .a-. .f <‘f
*
.
. . *. . . f.
a
.
.
. -.
.
. .
.#.
25 15
20
ARTERIAL
25
30
35
40
BLOOD
pO2
mmHg
45
Fig. 6. Correlation between the pre-ejection period (Q-Ao) and fetal arterial blood PO, in rhesus monkeys. obtained from fetuses who had developed metabolic or mixed metabolic-respiratory acidosis during labor. Ten sets of paired observations were available from experiments in which the fetal arterial pH decreased significantly (p < 0.0 l)* with no accompanying consistent change in PO, (p > 0.05).* In these experiments, interval increased significantly also, the Q-Ao (p < 0.05)* with increasing fetal acidosis. Fig. 5 illustrates the changes in the systolic time intervals with acidosis during one of the COz inhalation experiments, displayed by means of the continuous microfilm recording technique.5 Arterial blood oxygen tension. For the entire group of observations taken together, there was no correlation? between the Q-Ao interval and the oxygen tension (Pam) of fetal arterial blood within the observed range (14 to 43 mm. Hg) (Fig. 6). When the paired values (control, hypoxemia) of individual animals were *The p values by paired t test. tPearson Product-moment correlation
1, 1978 Gynecol
coefficient.
analyzed, however, a small but significant (p < 0.05)* prolongation of Q-Ao was found to occur during nonacidemic hypoxemia. The mean increase in Q-.40 was 2.4 msec. for a mean decrease in Paon of 8.5 mm. Hg. The systolic and diastolic blood pressures averaged 2.4 and 2.6 mm. Hg, respectively, below control values at the time the STI’% were measured during acute hypoxemia. Hypoxemia, acidosis, and combined hypoxeqia and acidosis. Seventy-two determinations of the ST.I’s were divided into four groups according to the results of arterial blood chemistry, as shown in Table II. The change in the Q-Ao interval followed most closely the change in the arterial blood pH. There was a slight but statistically significant prolongation (p < 0.04)” of the Q-Ao interval in the hypoxemia group(B) as compared with the normal group (A). despite the absence of a significant pH difference. There was, however, no difference in Q-Ao intervals between the acidosis group (C) and hypoxemia plus acidosis group(D) (p > 0.05).* A significant difference in MC-Ao intervals was found only between the normal group c<4j and the group with combined acidosis and hypoxemia (0) (p < 0.01)” Epinephrine and atropine. Infusion of epinephrine into the fetus resulted in significant shortening of the Q-Ao and MC-Ao intervals in all experiments (p < O.OOlT for each interval). The change in the Q-MC interval caused by epinephrine was not as consistent as those in the Q-Ao or MC-Ao intervals, for statistically significant shortening was observed only in three of the seven cases in which Q-MC could be measured (p > 0.05).7 *Student’s t test. tPaired
t test
Volume
132
Number
3
Table
Cardiac systolic time intervals
II. Changes
in systolic time intervals
during
A. Norm&* No. PH: Mean f 1 SD. Range PO, (min. Hg): Mean + S.D. Range
Q-MC (msec.): mean + 1 SD. Q-Ao (msec.): mean + 1 S.D. MC-Ao (msec.): mean f 1 S.D. *A: tB: SC: $D:
pH pH pH pH
> > 5 s
7.25, Pop > 25 mm. 7.25, PO* 5 25 mm. 7.25, PO* > 25 mm. 7.25, PO* 5 25 mm.
acidosis and/or
B. Hy$oxemia
37 7.36 7.26 29.7 25.1
2 -
.06 7.48
k 3.6 - 43.0
7.35 7.31 21.6 17.0
Product-moment
k -
.05 7.43
AZ 2.8 - 25.0
21.45 2 4.05 43.11 * 2.91
19.36
21.45
5.43
zk
alone*
D. Hrpoxnia
4.59
7.20 + 7.08 34.0 25.3 26.39 48.79 22.40
and addcsisl 9
.06 7.25
+ 5.8 - 42.0 + 9.77 + 5.16 + 9.79
7.17 rt 6.99 19.4 14 21.21 46.87 26.00
.08 7.25
-c 4.2 - 25 f 4.5 f 6.25 f 7.08
Hg. Hg. Hg. Hg.
-.’
Heart rate. Many investigators have found a slight but significant shortening of PEP with increasing heart rate in infants, children, and adults. Organ and COworkers3, 7 have described a similar relationship between PEP and heart rate in human3 and ovine’ fetuses. On the other hand, our fetal monkeys exhibited no rate-related alteration in PEP under unstressed conditions, nor was such a relationship found in our earlier studies of normal human fetuses.j In phonocardiographic studies, also, the Q-&interval has been found to be constant despite variations in the FHR.* The explanation of this descrepancy may lie in the selection of fetuses and cardiac cycles for study. Adrenergic stimulation increases both heart rate and myocardial contractility simultaneously, and produces shortening of PEP.* Changes in heart rate resulting from vagal stimulationg* lo or cardiac pacing,g however, have been found not to be accompanied by changes in PEP. The observations described in this report of shortening of the fetal PEP in response to epinephrine, but no significant change after atropine, are in accord with these findings. In the present study as in our previous study of human fetuses during labor, the record segments selected for analysis of the PEP-heart rate *Paired t test. “TPearson
C. Acidosis 6
22.10 f 5.36 41.47 + 2.79 +
hypoxemia
20
Atropine shortened all three intervals slightly in four out of five cases; however, these shortenings were not statistically significant (p > 0.05 in all cases).* Blood glucose level. The PEP did not show any significant relationship to fetal blood glucose level over the range from 6 to 296 mg. per 100 ml. (p > 0.057 for each individual experiment and for pooled data).
Comment
alunst
291
correlation
coefficient.
relationship were those in which there appeared to be minimal likelihood of hypoxia, acidosis, or other fetal stress. Our observations of lack of change in fetal PEP with spontaneous heart rate variability, during FHR accelerations accompanyiing fetal movements, and during the initial slowing of variable decelerations suggest that these alterations in FHR result from changes in vagal tone and that the adrenergic contribution may be minimal. Indeed, Friedman” has demonstrated that the sympathetic innervation of the fetal heart is partially deficient in sheep but increases rapidly during the 48 hours following birth. Fetal age. Lengthening of the PEP with advancing chronologic age has been observed in human fetuses,4 neonates and infants. The nearly identical rates of increase in Q-MC and Q-Ao intervals and QRS duration with gestational age (Fig. 1) in fetal monkeys suggest that prolongation of the time required for electrical activation is responsible for lengthening in both Q-MC and Q-Ao intervals. The increase in QRS duration is probably due to increasing myocardial volume. The value {Q-A0 interval minus QRS duration} did not change during the last third of pregnancy in fetal monkeys, however, and this may prove helpful in the application of PEP determinations to those human pregnancies in which the fetal age is uncertain. Preload. The Q-Ao interval was found to have significant inverse correlation with left ventricular end-diastolic pressure (LVEDP). In the absence of cardiac failure, an increase in LVEDP implies an increase in ventricular filling and, thus, in the initial tension upon the myocardial fibers (preload). Within the limits of the Frank-Starling law, increasing preload evokes increasing force of contraction and rate of development of tension. A related phenomenon has been observed during arrhythmias in human fetuses: the
292
Murata et al.
PEP was prolonged following an abnormally short R-R interval and a shortened PEP followed the longer R-R intervals. Shortening of PEP has also been observed during transfusion of hypovelemic newborn human fetuses. Blood pressure. The interrelations between aortic blood pressure, myocardial contractility, and PEP are complex and as yet somewhat controversial. If contractility remains unchanged, the greater the diastolic blood pressure, the greater the time required after the onset of contraction for intraventricular pressure to exceed intra-aortic pressure and effect opening of the aortic valves. Thus, the longer will be the PEP. On the other hand, widening of pulse pressure implies an increase in stroke volume and ventricular preload,12 other factors being equal, and should be accompanied by shortening of PEP. In the present study, we observed a positive correlation between PEP and both systolic and diastolic blood pressure, a finding in accord with the reports of Harris, Schoenfeld, and Weisslers and Tally, Meyer, and McNay.8 The relationships were not especially strong (r = 0.49 and 0.51, respectively, for systolic and diastolic blood pressure). The components of PEP, Q-MC and MC-Ao, were not significantly correlated with blood pressure, but this may only reflect the smaller absolute variation of the component intervals and the less precise detection of the end point of MC. The narrow range of arterial pulse pressure in the normal fetal monkeys did not permit detection of a relationship between PEP and pulse pressure under physiologic conditions. The marked prolongation of PEP with narrowing of pulse pressure in the distressed fetus (Fig. 3) probably represents marked impairment of myocardial contractility. Acidosis and hypoxia. The most interesting of these observations, and one of potential clinical significance, was the prolongation of PEP with a decrease in fetal arterial blood pH. The prolongation of PEP with pH decrease was essentially similar whether the acidosis was produced by administration of high COP mixtures to the mother, by metabolic or mixed acidosis occurring during labor or fetal distress due to other causes, or by direct infusion of dilute hydrochloric acid into the fetus. The change in PEP cannot be explained by changes in arterial blood pressure, for the blood pressure of the distressed fetuses was usually lower than that of the healthy ones, whereas blood pressure increased slightly during the acute acidemia experiments. Fig. 4 demonstrates that the mean PEP duration increases 3.6 msec. for a 0.1 unit decrease in pH. The regression slope of PEP on diastolic blood pressure (Fig. 2) indicates that a pressure rise of approximately
14 mm. Hg would be required to effect a similar prolongation of PEP, were this the sole mechanism opemt.ing. Even in the acute hypercapnia experiments, the maximum increase (above control levels) in diastolic blood pressure at the time the PEP measurements were made was 2 mm. Hg, and the average change was a 2.3 mm. Hg decrease. We have been unable to find any previous studies in which the relationship between pH and cardiac systolic time intervals was studied directly. Studies employing isolated, perfused myocardial preparations have demonstrated a decrease in contractility with acidosis.‘” Cingolani and co-workers ‘* showed that this was not a function of extracellular pH, and suggested that either Pco? or intracellular pH could be the major determinant of the change in myocardial contractility with acidbase changes. Acid-base changes produced smaller changes in myocardial contractility in experiments in which the heart was studied in situ’” than in those investigations on isolated, perfused myocardial strips. Goodyear and associates I6 demonstrated a decrease in cardiac output and in coronary blood flow during metabolic acidosis, but found no significant changes in left ventricular efficiency. Downing, Talner. and Gardner’” observed a decrease in myocardial contractility during combined hypoxemia and addemia, but either small or insignificant changes in contractility when either of these factors existed alone. In the experiments reported here, the PEP was prolonged by either acidosis alone or by hypoxemia alone. The effect of acidosis appeared to be the greater, for PEP exhibited a strong linear correlation with pH, whereas no such relationship existed with Paoz Significant prolongation of both PEP and ICT (MC-Ao) was observed in fetuses with acidosis alone and in those with hypoxemia alone. If prolongation of ICT indicates depression of myocardial contractility, then our findings suggest that the contractile force of the fetal myocardium is depressed by acidosis but not by mild to moderate hypoxemia. Our results thus differ somewhat from those of Downing and co-workers.‘” but do not appear incompatible with the earlier findings. Electromechanical latent time (Q-MC) also increased during acidosis, indicating that a delay between electrical activation and the onset of mechanical contraction also contributes to prolongation of PEP with acidosis. The rate of spread of depolarization (QRS duration) did not change with the degree of acidosis achieved in these experiments. Our findings differ significantly from those of Organ and co-worker?, 7 with regard to the changes in PEP during hypoxia. These latter investigators observed shortening of PEP during acute hypoxia in both the
Cardiac systolic time intervals
Volume 132 Number3
human
fetus3
dency
and
toward
stances in discrepancy factory
and
hypoxia,
or
of
siveness
of
the
additional
Blood
whereas of
of interspecies
sympathetic
await
lamb,’
we found
PEP in
similar
a tencircum-
glucose
differences fetal
differences
parasympathetic in
heart.
the Further
in the
rela-
responses adrenergic
to
hypo-
and
produced did
ied. In functions even
not
by the change
previously seemed during
served
period
The Budy,
authors D.V.M.,
administration any
of the
healthy fetuses, to be maintained
extreme
hypoglycemia
of insulin cardiac fetal
or
intervals
glustud-
cardiovascular at normal levels within
the
ob-
of 60 to 90 minutes.
respon-
resolution
must
investigations.
levels. Acute
mia cose
fetal monkeys. It is difficult to explain the in results except on the somewhat unsatis-
grounds
tive
fetal
prolongation
293
thank M. J. Donahue, B.S., for their technical assistance.
and
N. M.
hyperglyce-
REFERENCES 1. Weissler, A. M.: Current concepts in cardiology-systolic time intervals, N. Engl. J. Med. 296: 321-235, 1977. 2. Goodlin, R. C., Girard, J., and Hollmen, A.: Systolic time intervals in the fetus and neonate, Obstet. Gynecol. 39: 295, 1972. 3. Organ, L. W., Bernstein, A., and Smith, K. C.: The preejection period of tbe fetal heart: Patterns of change duringlabor, AM.J.OBSTET. GYNECOL. 140: 49, 1974. 4. Murata, Y., and Martin, C. B., Jr.: Systolic time intervals of the fetal cardiac cycle, Obstet. Gynecol. 44: 224, 1974. 5. Hon, E. H., Murata, Y., Zanini, B., Martin, C. B., Jr., and Lewis, D.: Continuous microfilm display of the electromecbanical intervals of the cardiac cycle, Obstet. Gynecol. 43: 722, 1974. C. B., Jr., Murata, Y., and Parer, J. T.: A method 6. Martin, for obtained biochemical and biophysical measurements from rhesus monkey fetuses for prolonged periods, AM. J.OBSTET.GYNECOL. 117: 377, 1973. I. L.: Evaluation of 8. Tally, R. C.. Meyer., . .1. F., and McNav, the pre-ejection period as an estimate-of myocardial contractility in dogs, Am. J. Cardiol. 27: 384, 1971. 9. Harris, W. C., Schoenfeld, C. D., and Weissler, A. M.: Effect of adrenergic receptor activation and blockade on
10.
11. 12. 13.
14.
15.
16.
the systolic pre-ejection period, heart rate and arterial oressure in man. 1. Clin. Invest. 46: 1704. 1967. frank, M. N., and Kinlaw, W. B.: Indirect measurement of isovolumetric contraction time and tension period in normal subjects, Am. J. Cardiol. 10: 800, 1962. Friedman. W. F.: Neuropharmacologic studies of perinatal myocardium, Cardiovasc. Clin.& 44, 1972. Wallence, A. G., Skinner, N. S., Jr,, and Mitchell, J. H.: Hemodynamic determinations of the maximal rate of rise of left ventricular pressure, Am. J. Physiol. 205: 30, 1963. Onie, L. H.: Effect of extracellular nH on function and metabolism of isolated perfused rat beart, Am. J. Physiol. 209: 1075, 1965. Cingolani, H. E., Mattiazzi, A. R., Blesa, E. S., and Gonzales, N. C.: Contracility in isolated mammalian heart muscle after acid-base changes, Circ. Res. 26: 269, 1970. Downing, S. E., Talner, N. S., and Gardner, T. H.: Influence of hvooxemia and acidemia on left ventricular function, Am. y. Physiol. 210: 1327, 1966. Goodvear. A. V. N.. Eckhardt. W. F.. Obtberg:. R. H.. and Goodkin, ‘M. J.: Effect of metabolic acidosis and alkalosis on coronary blood flow and myocardial metabolism in the intact dog, Am. J. Physiol. 200: 628, 1961.
MURATA,
CHESTER
B.
TSUYOMU ROY
M.D. MARTIN,
JR.,
IKENOUE,
H.
PETRIE,
Los Angeles,
California
M.D.
M.D.*
M.D.**
The systolic time intervals of the fetal cardiac cycle were studied by means of simultaneous recordings of electrocardiogram (ECG) and uitrasound Doppler cardiogram (GCG) in chronic preparations of fetal rhesus monkeys. Recordings were made under physiologic conditions as well as during various experimental stresses. The pre-ejection period (PEP) showed no significant relationship with heart rate in the unstressed fetuses, but the acceleration of heart rate induced by epinephrine was accompanied by shortening of PEP. The PEP increased with advancing fetal age. The PEP was inversely correlated with left ventricular end-diastolic pressure and arterial pulse pressure, but showed a positive correlation with both systolic and diastolic arterial blood pressure. The PEP also exhibited strong negative correlation with arterial blood pH. The prolongation was essentially the same whether acidosis was of respiratory or metabolic origin. The PEP increased slightly but significantly during nonacidemic hypoxemia; however, there was no correlatii between Pao, and PEP. Epinephrine shortened the PEP significantly, whereas the effect of a&opine was inconsistent. Afteratlon of the plasma glucose level by injection of insulin or glucose did not affect PEP. These findings demonstrate that the PEP may be a useful indicator of fetal cardiac performance, reflecting both myocardial contractility and the hemodynamic state of the cardiovascular system. (AM. J. OBSTET. GYNECOL. 132: 285, 1978.)
THE SYSTOLIC time intervals (STI) of the cardiac cycle have been found to be useful indicators of myocardial function.’ More recently, these intervals have been investigated as possible indicators of the functional state of the fetal heart, with the use of the fetal electrocardiogram (FECG) in combination with
From the Department of Obstetrics and Gynecology, Angeles County-University of Southern California Medical Center.
Los
Supported in part by Grant HD-06406 from the National Institutes of Health, United States Public Health Service, and by the Rohi ScienfiJc Corporation, Santa Ana, Calijornifl. Received for publication
Augwt
Revised December
16, 1977.
AcceptedJanuq
23, 1978.
6, 1976.
Reprint requests: Yuj Murata, M.D., Department of Obstetrics and Gynecology, I240 N. Mission Rd., Los Angeles, California 90033. *Present address: Kagoshima Municipal Hospital, Department of Obstetrics and Gynecology, Kagoshima, Japan. **Present address: College of Phy+cians and Surgeons, Columbia University, Department of Obstetrics and G>Ttecology, New York, New York. OOOZ-9378178/190285+09$00.9010
@ 1978
The C. V. Mosby
Co.
the phonocatdiogramZ or ultrasonic Doppler cardiogram (DCG) to detect and measure the STI’s2* 4, 5 Prolongation of the pre-ejection period (PEP) was observed in five fetuses exhibiting heart rate patterns indicative of mild distress. In other studies,4* 5 changes in the STI’s were demonstrated during variable deceleration (cord compression) and late deceleration (hypoxic) FHR patterns. These findings provided evidence that the STI’s might yield clinically useful information regarding the status of the fetal heart and thus the condition of the fetus. In order to define more clearly the factors affecting the STI’s of the fetal heart, the studies described herein were carried out in fetal rhesus monkeys under relatively physiologic conditions, during controlled stress, induced fetal stress, or spontaneously occurring fetal distress in labor.
Materfals and methods These experiments were performed in pregnant rhesus monkeys obtained from the breeding colony at the Primate Center at the University of CaliforniaDavis. The duration of pregnancy was known from timed matings in 11 monkeys and in the remaining 285
286
Table
Murata
et al.
I. Summary
of experiments No. of
Type of experiment
High CO2 administration to mother Low O2 administration to mother 0.15N HCI infusion to fetus l!.LZ&: Oxytocin-induced Spontaneous following surgery Epinephrine
animals
No. of experiments 6 6 3
2 3
2 3
administratbn:
To unstressed fetus To distressed fetus Atropine administration to fetus 20% glucose administration to mother Insulin administration to mother and/or fetus
4 5 5 3 2
4
four it was estimated from rectal palpation early in pregnancy. ECG electrodes and carotid or femoral arterial catheters were implanted in rheses monkey fetuses between 115 days’ gestation and near term by means of techniques previously described.6 A catheter was placed in the left cardiac ventricle through the ascending aorta in one fetal monkey. The location was confirmed by the configuration of the pressure tracing and by subsequent autopsy. Intrauterine catheters were inserted in order to measure uterine contractions. A catheter was placed in the jugular vein of six fetuses. Following the operation, the mothers were placed in restraining chairs and permitted to recover from the anesthesia. Eleven animals became chronic preparations, and were studied at times ranging from 48 hours to 34 days postoperatively. Four other animals developed progressively increasing uterine contractions following the operation, and were studied between 12 and 48 hours postoperatively. The FECG, FHR, fetal blood pressure, and intrauterine pressure were recorded on a standard biomedical recorder and simultaneously on magnetic tape. Fetal arterial blood samples were obtained from the arterial catheters for determination of pH and respiratory gas tensions. The samples were collected anaerobically in heparinized plastic syringes. The pH was determined with a Radiometer microelectrode, and PoZ and Pcq were measured with Radiometer BMS-3 or IL micro blood-gas analyzers. For studies of the relationship between STI’s and fetal biochemical parameters, blood samples were obtained either immediately before or immediately after the DCG record was made. The fetal DCG was obtained with a Kretz (MGD-2) fetal pulse detector. The output of the instrument was recorded on a second tape recorder at 7% ‘or 15 i.p.s.,
along with the FECG, blood pressure, and intrauterine pressure. These tapes were later played back at an eightfold reduction in speed through appropriate filters into a high-frequency ink-writing rerordet (Elema Mingograf Model No. 24.B) in order to measure the STI’s manually. The tapes were also used to generate a continuous microfilm record of the STI‘s in successive cardiac cycles. These techniques have been described previously.“, I In these studies on intrauterine fetuses, it has not been possible to determine whether mitral or tricuspid, or aortic or pulmonic valve signals (or both in combination) were being recorded on the DCG. In this report, the first valve-motion signal of the FDCG has been designated MC, and the second and third, A0 and AC, respectively. Separate right and left ventricular STI’s could not be measured, since multiple intracardiac and intra-arterial catheters were not implanted. Four STI’s were measured from the combined ECG-DCG records: (1) from the beginning of the Q wave to the end of the atrioventricular valve closing signal (Q-MC = electromechanical latent time, EMLT): (2) from the onset of Q to the beginning of the semilunar valve opening signal (Q-Ao = pre-ejection period, PEP); (3) the difference of those two intervals (McAo = isovolumetric contraction time, ICI‘); and (4) from the beginning of the semilunar valve opening signa! to the end of the closing signal (Ao-AC = ventricular ejection time, VET). The findings relating to the first three of these intervals will be described in this report. The values for the STI’s used in most of the determinations were the mean values calculated from 20 consecutive cardiac cycles. The variability of the intervals within this number of cycles was small: for example, the mean coefficient of variation for Q-Ao was 4.5 per cent, with a range from 3.5 to 6.3 per cent. The values for the intervals within individual cardiac cycles were used in searching for a relationship between the STI’s and the R-R interval. In the chronic preparations, combined ECG-DCG records were obtained from fetuses under unstressed conditions. These records were repeated at intervals over periods ranging from a few days to 4% weeks. In most instances, the mother accepted the hand-held placement of the ultrasound transducer against the abdomen without protest, so that no element of fetal distress produced by maternal fear or struggling was likely to be present. Normal FHR variability plus accelerations and decelerations occurring with gross fetal movements and fetal breathing movements provided a range of heart rates from 120 to 240 b.p.m. In a few
Volume Number
Cardiac systolic time intervals
132 3
instances, transabdominal compression of the fetal head or flushing the fetal trachea catheter with a small volume of saline was employed to evoke transient FHR decelerations. In addition to these normal, unstressed observations, combined ECG-DCG recordings were obtained from fetuses stressed by a variety of techniques (Table I). The maternal inspiratory gas tensions were altered by administering mixtures of varying compositions to the mother by means of a lucite helmet. The approximate volume of the helmet was 2 L. Flow rates of 6 to 8 L. per minute were maintained to ensure a minimum of rebreathing. Acute respiratory acidosis was produced by the administration of a mixture of 20 per cent COz and 20 per cent O2 in nitrogen. The fetal arterial blood PO, in these experiments did not decrease in any instance and the maximum increase was 16 mm. Hg. The fetal arterial PO, was lowered by administering low oxygen mixtures to the mother-as low as 8 per cent maternal FI,. The fetal arterial blood pH remained relatively constant in most of these experiments. In one experiment in which the hypoxia was prolonged, the fetal pH decreased by 0.33 unit. In other instances, fetal acidemia was induced by the administration of 0.15N HCl to the fetus at rates up to 0.4 ml. per minute. The endpoint in these acute stress experiments was slowing of the FHR to the range of 90 to 120 b.p.m. The acute stress experiments generally lasted between 10 and 30 minutes, the ST1 records being made near the end of the stress period. The single prolonged hypoxic experiment continued for 90 minutes, with multiple ST1 recbrds being obtained. The combination of fetal hypoxemia and acidosis resulting from uterine contractions was studied in two animals during oxytocin-induced uterine activity and in three animals during labor following the operation. In two of these latter animals, the fetal arterial blood PO, was increased without significant change in pH by increasing the maternal Fro*; in one of them the pH was altered without significant PO, change by the administration of bicarbonate to the fetus. Epinephrine was administered to four fetuses (nine experiments). Two of these fetuses (five experiments) were severely distressed. Atropine was also infused in five experiments into three unstressed fetuses. Sixteen determinations of the STI’s were performed on three established chronic fetal monkey preparations to investigate any relationship between the systolic time intervals and fetal blood glucose level. Acute hyper- or hypoglycemia was induced-by infusion of 20 per cent glucose solution to the mother or of insulin to the mother and/or fetus. Fetal plasma glucose values were
1 110
I 120
I 130
I 140
I 150
110
I 120
I 130
I 140
1 150
GESTATIONAL
AGE
287
[days]
Fig. 1. Relationships between systolic time intervals and gestational age in normal fetal rhesus monkeys. A, Electromechanical latent time (Q-MC): B, pre-ejection period (Q-Ao); C, QRS duration. elevated as high as 296 mg. per 100 ml. and decreased as low as 6 mg. per 100 ml.
Results Heart rate. In the unstressed fetal monkeys, there was no sigtificant correlation between the duration of the Q-Ao interval (PEP) and heart rate. This was true both for mean values and for individual cycles recorded during decelerations or accelerations. Gestational a&. The durations of both the Q-MC and Q-Ao intervals increased with advancing gestational age in the unstressed fetuses (p < 0.0 l)* (Fig. 1). The slope of the regression line for Q-Ao and gestational age is somewhat steeper than that for Q-MC, but *Pearson Product-moment
correlation
coefficient.
288
Murata et al.
I_ B
50 $
.
*
40
/
I
.40
SYSTOLIC
Fig. 2. Correlation pressure in normal
50
60
BLOOD mmHg
70
80
between the pre-ejection fetal rhesus monkeys.
.
. I.*. .. * *. .: -. . . . 1 *.. * .
20
-
BLOOD mmHg
(Q-Ao) and systolic
period
. . *.
30
DIASTOLIC
PRESSURE
. S’
.
2 % a
PtP=O25iDBP,+35: 4 51 n=49 P<“O1 .
(Ai and
.
40
PRESSURE
diastolic
(B) blood
80
60
12
14
ARTERIAL Fig. 3. Relationship
between
the pre-ejection
PULSE period
20
18
16
PRESSURE (Q-Ao)
(mmHg) and pulse
pressure
in a distressed
fetal
monkey. the difference is not statistically significant. The lengthening of these intervals with advancing gestational age appeared to result entirely from prolongation of the time required for electrical depolarization of the ventricles (QRS duration, Fig. 1, C). The quantity (Q-A0 interval minus QRS duration} was constant throughout the range of gestational ages studied, and varied narrowly in the normal fetuses between 12.6 and 15.6 msec. Blood pressure. The Q-Ao interval was positively correlated with both systolic and diastolic blood pressure (Fig. 2, A and B). The intervals Q-MC and MC-Ao were not significantly related to either blood pressure value. In the one fetus studied by means of a left ventricular catheter, an inverse relationship was observed between Q-Ao and left ventricular end-
diastolic pressure over the observed range I to 9 mm. Hg (r = 0.82, n = 33, p < 0.01). This fetus appeared to be stable with regard to heart rate, systolic blood pressure, and arterial blood PI-I values, which were within the normal range throughout the observation period. The range of aortic pulse pressures recorded in the normal monkeys was rather small, and no relationship could lx detected between PEP and pulse pressure. In one distressed fetus, however, there was a clear inverse correlation between Q-Ao duration and aortic QUke pressure (Fig. 3, p < O.Ol).* Both systolic and diastolic blood pressure showed a slight but significant rise with decreasing arterial blood *Pearson
Product-moment
correlation
coefficient.
Volume
Cardiac systolic time intervals
132
289
Number3
. .
20
.
k.
.
:
i
. . .. .* I * !‘.iri.ew .
1:
-.
0:.
.
. .
.
.
.
l
l
.-
.*. .
l
l *r.
OMc=-15(pH)+130 r=-0 31 n=85
60-
.
’ .
.
.
l .
50-
,
CIAO=-36(pH)+300 r=-0 72 ~86 I
70
7.1
30-
I
7.2
7.3
7.4
l
.
lo-
.
.
.
McAo=2llpH)+170 r=O 38 n=85
710
.
7:2
7'1
:
.
l
7'4
713
ARTERIAL BLOOD pH Fig. 4. Correlation Electromechanical time (MC-Ao).
between latent
systolic
time intervals
under physiologic (systolic blood pressure = -4OOpH + conditions 2,996, r = -0.48, n = 45, p < 0.01; diastolic blood pressure = -263 pH + 1,963, r = -0.42, n = 45, p < O.Ol).” Arterial blood pH. The Q-MC, Q-Ao, and MC-Ao intervals became prolonged with fetal acidemia or acidosis (Fig. 4). The relationship between the duration of these intervals and the pH of fetal arterial blood was linear in the range from pH 6.99 to pH 7.48 (p < 0.01) in all three cases.* The slope of the regression line for the Q-Ao interval is steeper than that for Q-MC, dem-
pH
in
the unstressed
fetal
*Pearson Proclua-moment
and fetal
time (Q-MC); B, pre-ejection
monkeys
correlation
coefficient.
arterial
blood
pH in rhesus
monkeys.
A,
period (Q-Ao); C, isovolumetric contraction
onstrating that both components of the PEP (EMLT, measured as Q-MC, and ICT, derived) become prolonged during acidosis. The duration of the QRS complex did not change significantly over the observed range of pH. The prolongation of PEP observed during fetal acidemialacidosis did not seem to be influenced by the mechanisms producing the lowered pH. Thus, the data points from the experiments in which acute acidemia was induced in healthy fetuses by administration of COs to the mother were undistinguishable from those observed during infusions of HCI into the fetus. Further, the values from the acute acidemia experiments were similar, at equivalent pH levels, to those
290
et al.
Murata
October
Am. J. Obstet.
Fig. 5. Continuous
55
: m
1 I
.
45
E 2
‘0.
. 35
.
:
.a.*
.
.**
.*
l s:l
.
I
display of the systolic time intervals during maternal hypercapnea.
*
. .- .a-. .f <‘f
*
.
. . *. . . f.
a
.
.
. -.
.
. .
.#.
25 15
20
ARTERIAL
25
30
35
40
BLOOD
pO2
mmHg
45
Fig. 6. Correlation between the pre-ejection period (Q-Ao) and fetal arterial blood PO, in rhesus monkeys. obtained from fetuses who had developed metabolic or mixed metabolic-respiratory acidosis during labor. Ten sets of paired observations were available from experiments in which the fetal arterial pH decreased significantly (p < 0.0 l)* with no accompanying consistent change in PO, (p > 0.05).* In these experiments, interval increased significantly also, the Q-Ao (p < 0.05)* with increasing fetal acidosis. Fig. 5 illustrates the changes in the systolic time intervals with acidosis during one of the COz inhalation experiments, displayed by means of the continuous microfilm recording technique.5 Arterial blood oxygen tension. For the entire group of observations taken together, there was no correlation? between the Q-Ao interval and the oxygen tension (Pam) of fetal arterial blood within the observed range (14 to 43 mm. Hg) (Fig. 6). When the paired values (control, hypoxemia) of individual animals were *The p values by paired t test. tPearson Product-moment correlation
1, 1978 Gynecol
coefficient.
analyzed, however, a small but significant (p < 0.05)* prolongation of Q-Ao was found to occur during nonacidemic hypoxemia. The mean increase in Q-.40 was 2.4 msec. for a mean decrease in Paon of 8.5 mm. Hg. The systolic and diastolic blood pressures averaged 2.4 and 2.6 mm. Hg, respectively, below control values at the time the STI’% were measured during acute hypoxemia. Hypoxemia, acidosis, and combined hypoxeqia and acidosis. Seventy-two determinations of the ST.I’s were divided into four groups according to the results of arterial blood chemistry, as shown in Table II. The change in the Q-Ao interval followed most closely the change in the arterial blood pH. There was a slight but statistically significant prolongation (p < 0.04)” of the Q-Ao interval in the hypoxemia group(B) as compared with the normal group (A). despite the absence of a significant pH difference. There was, however, no difference in Q-Ao intervals between the acidosis group (C) and hypoxemia plus acidosis group(D) (p > 0.05).* A significant difference in MC-Ao intervals was found only between the normal group c<4j and the group with combined acidosis and hypoxemia (0) (p < 0.01)” Epinephrine and atropine. Infusion of epinephrine into the fetus resulted in significant shortening of the Q-Ao and MC-Ao intervals in all experiments (p < O.OOlT for each interval). The change in the Q-MC interval caused by epinephrine was not as consistent as those in the Q-Ao or MC-Ao intervals, for statistically significant shortening was observed only in three of the seven cases in which Q-MC could be measured (p > 0.05).7 *Student’s t test. tPaired
t test
Volume
132
Number
3
Table
Cardiac systolic time intervals
II. Changes
in systolic time intervals
during
A. Norm&* No. PH: Mean f 1 SD. Range PO, (min. Hg): Mean + S.D. Range
Q-MC (msec.): mean + 1 SD. Q-Ao (msec.): mean + 1 S.D. MC-Ao (msec.): mean f 1 S.D. *A: tB: SC: $D:
pH pH pH pH
> > 5 s
7.25, Pop > 25 mm. 7.25, PO* 5 25 mm. 7.25, PO* > 25 mm. 7.25, PO* 5 25 mm.
acidosis and/or
B. Hy$oxemia
37 7.36 7.26 29.7 25.1
2 -
.06 7.48
k 3.6 - 43.0
7.35 7.31 21.6 17.0
Product-moment
k -
.05 7.43
AZ 2.8 - 25.0
21.45 2 4.05 43.11 * 2.91
19.36
21.45
5.43
zk
alone*
D. Hrpoxnia
4.59
7.20 + 7.08 34.0 25.3 26.39 48.79 22.40
and addcsisl 9
.06 7.25
+ 5.8 - 42.0 + 9.77 + 5.16 + 9.79
7.17 rt 6.99 19.4 14 21.21 46.87 26.00
.08 7.25
-c 4.2 - 25 f 4.5 f 6.25 f 7.08
Hg. Hg. Hg. Hg.
-.’
Heart rate. Many investigators have found a slight but significant shortening of PEP with increasing heart rate in infants, children, and adults. Organ and COworkers3, 7 have described a similar relationship between PEP and heart rate in human3 and ovine’ fetuses. On the other hand, our fetal monkeys exhibited no rate-related alteration in PEP under unstressed conditions, nor was such a relationship found in our earlier studies of normal human fetuses.j In phonocardiographic studies, also, the Q-&interval has been found to be constant despite variations in the FHR.* The explanation of this descrepancy may lie in the selection of fetuses and cardiac cycles for study. Adrenergic stimulation increases both heart rate and myocardial contractility simultaneously, and produces shortening of PEP.* Changes in heart rate resulting from vagal stimulationg* lo or cardiac pacing,g however, have been found not to be accompanied by changes in PEP. The observations described in this report of shortening of the fetal PEP in response to epinephrine, but no significant change after atropine, are in accord with these findings. In the present study as in our previous study of human fetuses during labor, the record segments selected for analysis of the PEP-heart rate *Paired t test. “TPearson
C. Acidosis 6
22.10 f 5.36 41.47 + 2.79 +
hypoxemia
20
Atropine shortened all three intervals slightly in four out of five cases; however, these shortenings were not statistically significant (p > 0.05 in all cases).* Blood glucose level. The PEP did not show any significant relationship to fetal blood glucose level over the range from 6 to 296 mg. per 100 ml. (p > 0.057 for each individual experiment and for pooled data).
Comment
alunst
291
correlation
coefficient.
relationship were those in which there appeared to be minimal likelihood of hypoxia, acidosis, or other fetal stress. Our observations of lack of change in fetal PEP with spontaneous heart rate variability, during FHR accelerations accompanyiing fetal movements, and during the initial slowing of variable decelerations suggest that these alterations in FHR result from changes in vagal tone and that the adrenergic contribution may be minimal. Indeed, Friedman” has demonstrated that the sympathetic innervation of the fetal heart is partially deficient in sheep but increases rapidly during the 48 hours following birth. Fetal age. Lengthening of the PEP with advancing chronologic age has been observed in human fetuses,4 neonates and infants. The nearly identical rates of increase in Q-MC and Q-Ao intervals and QRS duration with gestational age (Fig. 1) in fetal monkeys suggest that prolongation of the time required for electrical activation is responsible for lengthening in both Q-MC and Q-Ao intervals. The increase in QRS duration is probably due to increasing myocardial volume. The value {Q-A0 interval minus QRS duration} did not change during the last third of pregnancy in fetal monkeys, however, and this may prove helpful in the application of PEP determinations to those human pregnancies in which the fetal age is uncertain. Preload. The Q-Ao interval was found to have significant inverse correlation with left ventricular end-diastolic pressure (LVEDP). In the absence of cardiac failure, an increase in LVEDP implies an increase in ventricular filling and, thus, in the initial tension upon the myocardial fibers (preload). Within the limits of the Frank-Starling law, increasing preload evokes increasing force of contraction and rate of development of tension. A related phenomenon has been observed during arrhythmias in human fetuses: the
292
Murata et al.
PEP was prolonged following an abnormally short R-R interval and a shortened PEP followed the longer R-R intervals. Shortening of PEP has also been observed during transfusion of hypovelemic newborn human fetuses. Blood pressure. The interrelations between aortic blood pressure, myocardial contractility, and PEP are complex and as yet somewhat controversial. If contractility remains unchanged, the greater the diastolic blood pressure, the greater the time required after the onset of contraction for intraventricular pressure to exceed intra-aortic pressure and effect opening of the aortic valves. Thus, the longer will be the PEP. On the other hand, widening of pulse pressure implies an increase in stroke volume and ventricular preload,12 other factors being equal, and should be accompanied by shortening of PEP. In the present study, we observed a positive correlation between PEP and both systolic and diastolic blood pressure, a finding in accord with the reports of Harris, Schoenfeld, and Weisslers and Tally, Meyer, and McNay.8 The relationships were not especially strong (r = 0.49 and 0.51, respectively, for systolic and diastolic blood pressure). The components of PEP, Q-MC and MC-Ao, were not significantly correlated with blood pressure, but this may only reflect the smaller absolute variation of the component intervals and the less precise detection of the end point of MC. The narrow range of arterial pulse pressure in the normal fetal monkeys did not permit detection of a relationship between PEP and pulse pressure under physiologic conditions. The marked prolongation of PEP with narrowing of pulse pressure in the distressed fetus (Fig. 3) probably represents marked impairment of myocardial contractility. Acidosis and hypoxia. The most interesting of these observations, and one of potential clinical significance, was the prolongation of PEP with a decrease in fetal arterial blood pH. The prolongation of PEP with pH decrease was essentially similar whether the acidosis was produced by administration of high COP mixtures to the mother, by metabolic or mixed acidosis occurring during labor or fetal distress due to other causes, or by direct infusion of dilute hydrochloric acid into the fetus. The change in PEP cannot be explained by changes in arterial blood pressure, for the blood pressure of the distressed fetuses was usually lower than that of the healthy ones, whereas blood pressure increased slightly during the acute acidemia experiments. Fig. 4 demonstrates that the mean PEP duration increases 3.6 msec. for a 0.1 unit decrease in pH. The regression slope of PEP on diastolic blood pressure (Fig. 2) indicates that a pressure rise of approximately
14 mm. Hg would be required to effect a similar prolongation of PEP, were this the sole mechanism opemt.ing. Even in the acute hypercapnia experiments, the maximum increase (above control levels) in diastolic blood pressure at the time the PEP measurements were made was 2 mm. Hg, and the average change was a 2.3 mm. Hg decrease. We have been unable to find any previous studies in which the relationship between pH and cardiac systolic time intervals was studied directly. Studies employing isolated, perfused myocardial preparations have demonstrated a decrease in contractility with acidosis.‘” Cingolani and co-workers ‘* showed that this was not a function of extracellular pH, and suggested that either Pco? or intracellular pH could be the major determinant of the change in myocardial contractility with acidbase changes. Acid-base changes produced smaller changes in myocardial contractility in experiments in which the heart was studied in situ’” than in those investigations on isolated, perfused myocardial strips. Goodyear and associates I6 demonstrated a decrease in cardiac output and in coronary blood flow during metabolic acidosis, but found no significant changes in left ventricular efficiency. Downing, Talner. and Gardner’” observed a decrease in myocardial contractility during combined hypoxemia and addemia, but either small or insignificant changes in contractility when either of these factors existed alone. In the experiments reported here, the PEP was prolonged by either acidosis alone or by hypoxemia alone. The effect of acidosis appeared to be the greater, for PEP exhibited a strong linear correlation with pH, whereas no such relationship existed with Paoz Significant prolongation of both PEP and ICT (MC-Ao) was observed in fetuses with acidosis alone and in those with hypoxemia alone. If prolongation of ICT indicates depression of myocardial contractility, then our findings suggest that the contractile force of the fetal myocardium is depressed by acidosis but not by mild to moderate hypoxemia. Our results thus differ somewhat from those of Downing and co-workers.‘” but do not appear incompatible with the earlier findings. Electromechanical latent time (Q-MC) also increased during acidosis, indicating that a delay between electrical activation and the onset of mechanical contraction also contributes to prolongation of PEP with acidosis. The rate of spread of depolarization (QRS duration) did not change with the degree of acidosis achieved in these experiments. Our findings differ significantly from those of Organ and co-worker?, 7 with regard to the changes in PEP during hypoxia. These latter investigators observed shortening of PEP during acute hypoxia in both the
Cardiac systolic time intervals
Volume 132 Number3
human
fetus3
dency
and
toward
stances in discrepancy factory
and
hypoxia,
or
of
siveness
of
the
additional
Blood
whereas of
of interspecies
sympathetic
await
lamb,’
we found
PEP in
similar
a tencircum-
glucose
differences fetal
differences
parasympathetic in
heart.
the Further
in the
rela-
responses adrenergic
to
hypo-
and
produced did
ied. In functions even
not
by the change
previously seemed during
served
period
The Budy,
authors D.V.M.,
administration any
of the
healthy fetuses, to be maintained
extreme
hypoglycemia
of insulin cardiac fetal
or
intervals
glustud-
cardiovascular at normal levels within
the
ob-
of 60 to 90 minutes.
respon-
resolution
must
investigations.
levels. Acute
mia cose
fetal monkeys. It is difficult to explain the in results except on the somewhat unsatis-
grounds
tive
fetal
prolongation
293
thank M. J. Donahue, B.S., for their technical assistance.
and
N. M.
hyperglyce-
REFERENCES 1. Weissler, A. M.: Current concepts in cardiology-systolic time intervals, N. Engl. J. Med. 296: 321-235, 1977. 2. Goodlin, R. C., Girard, J., and Hollmen, A.: Systolic time intervals in the fetus and neonate, Obstet. Gynecol. 39: 295, 1972. 3. Organ, L. W., Bernstein, A., and Smith, K. C.: The preejection period of tbe fetal heart: Patterns of change duringlabor, AM.J.OBSTET. GYNECOL. 140: 49, 1974. 4. Murata, Y., and Martin, C. B., Jr.: Systolic time intervals of the fetal cardiac cycle, Obstet. Gynecol. 44: 224, 1974. 5. Hon, E. H., Murata, Y., Zanini, B., Martin, C. B., Jr., and Lewis, D.: Continuous microfilm display of the electromecbanical intervals of the cardiac cycle, Obstet. Gynecol. 43: 722, 1974. C. B., Jr., Murata, Y., and Parer, J. T.: A method 6. Martin, for obtained biochemical and biophysical measurements from rhesus monkey fetuses for prolonged periods, AM. J.OBSTET.GYNECOL. 117: 377, 1973. I. L.: Evaluation of 8. Tally, R. C.. Meyer., . .1. F., and McNav, the pre-ejection period as an estimate-of myocardial contractility in dogs, Am. J. Cardiol. 27: 384, 1971. 9. Harris, W. C., Schoenfeld, C. D., and Weissler, A. M.: Effect of adrenergic receptor activation and blockade on
10.
11. 12. 13.
14.
15.
16.
the systolic pre-ejection period, heart rate and arterial oressure in man. 1. Clin. Invest. 46: 1704. 1967. frank, M. N., and Kinlaw, W. B.: Indirect measurement of isovolumetric contraction time and tension period in normal subjects, Am. J. Cardiol. 10: 800, 1962. Friedman. W. F.: Neuropharmacologic studies of perinatal myocardium, Cardiovasc. Clin.& 44, 1972. Wallence, A. G., Skinner, N. S., Jr,, and Mitchell, J. H.: Hemodynamic determinations of the maximal rate of rise of left ventricular pressure, Am. J. Physiol. 205: 30, 1963. Onie, L. H.: Effect of extracellular nH on function and metabolism of isolated perfused rat beart, Am. J. Physiol. 209: 1075, 1965. Cingolani, H. E., Mattiazzi, A. R., Blesa, E. S., and Gonzales, N. C.: Contracility in isolated mammalian heart muscle after acid-base changes, Circ. Res. 26: 269, 1970. Downing, S. E., Talner, N. S., and Gardner, T. H.: Influence of hvooxemia and acidemia on left ventricular function, Am. y. Physiol. 210: 1327, 1966. Goodvear. A. V. N.. Eckhardt. W. F.. Obtberg:. R. H.. and Goodkin, ‘M. J.: Effect of metabolic acidosis and alkalosis on coronary blood flow and myocardial metabolism in the intact dog, Am. J. Physiol. 200: 628, 1961.