173
Morgenstern et al., Systolic time intervals
Revfew article J. Perinat. Med. 6 (1978)173
Systolic time intervals of the fetal cardiac cycle J. Morgenstern, H. Czerny, H. Schmidt, J. Schulz*, K. Wernicke* Abt. für Biomedizinische Technik, Universitäts-Frauenklinik Düsseldorf, Germany *Universitäts-Frauenklinik Frankfurt/Main, Germany
l Introduction Fetal heart rate (FHR) monitoring offers a tremendous amount of Information about the state of the fetus. The visual Interpretation of FHR traces is based on descriptive criteria [22]. Many attempts have been made to associate particular FHR patterns with certain types of fetal stress, though it is common clinical knowledge that head compression, umbilical cord clamping and hypoxia can give rise to similar FHR patterns. Some of these investigators introduced new terminology and so muddled the FHR nomenclature thus making precise communication of research reports nearly impossible. Since there :is no means of associating different fetal stress situations with typical FHR patterns, the application of more recent Computer methods and techniques [38] seems to be a promising area in which to achieve further Information on the physiological condition of the fetus. Therefore correlation öf the mechanical and electrical events of the fetal cardiac cycle seems to be a most promising direction in which to investigate. The measurement of the fetal systolic time interval (STI) shows a relationship with myocardial contractility. The knowledge of changes in myocardial contractility may Supplement the Interpretation of FHR patterns äs an indicator of the presence or absence of fetal distress. It has been over 70 years since the introduction of a method of measuring STI. In 1906, CREMER [8] reported the first successful fetal electro-
cardiogram (EKG) and in 1908 HOFBAUER and WEISS [27] recorded the first phonocardiogram of the fetal heart sounds. The first consecutive phases of the cardiac cycles of animal studies were published by WIGGERS in 1920 [71]. Two years later, in 1922, KATZ and FEIL [30] carried out human studies. In 1941, the fetal EKG and the fetal phonocardiogram were studied simultaneously for the first time by PUETZ [51]. In 1951 and ten years later there were two other reports about simultaneously recording electro- and phonocardiogram of the fetus in a few cases of normal pregnancies. In 1967, KAMMACHER [21] had already established that the STI expressed in percent of the heart cycles offers further information about the FHR concerning the fetus' condition in utero. In 1968, WEISSLER and coworkers [68] described non-invasive determination of STI in cases of heart failure in adults, and pointed out the clinical relevance of STI of the cardiac cycle. These fmdingsled many investigators to the conclusion that STI — especially the preejection period (PEP) and the left ventricular ejection time (VET) - are sensitive indicators of cardiac performance. 2 STI-Definitions The STI varies with the method used. The time period from the onset of ventricular depolarization (QRS complex or deflection of the R0300-5577/78/0006-0173$02.00 © by Walter de Gruyter & Co.
Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
Morgenstern et al., Systolic time intervals
174
VET
wave from the isoelectric line) to the onset of ejection from the left ventricle is known äs PEP, and that to the first heart sound äs Q-S1. Q-S1 is also termed the electromechanical interval of the left ventricle. Between the onset of electrical depolarisation and the beginning of mechanical contraction there exists a time gap, which corresponds very closely to the interval Q-Mc, where MC Stands for the closure of the atrioventricular (mitral) valve. The opening of the aortic valve is called Ao. During the Mc-Ao interval the myocardium builds up tension isometrically. For this reason and because the pressure within the ventricle increases until the intraventricular pressure equals intra-aortic pressure, this interval is referred to äs isovolumetric contraction time (IVC). Thus PEP is made up of two time intervals:
Ac
PEP = (Q - MC) + (Mc - Ao) The remaining part of the STI is VET. It is the interval between semilunar valve opening and closure: VET = (Ao—Ac). VET is usually measured from the beginning of the rise to the dicrotic notch in the pulse wave. The mechanical systole is the interval between the first heart sound (Sl) and the second heart sound (S2). With these definitions other intervals of the fetal heart cycle e.g. R-S1, R-S2 are easy to understand. The time from the QRS complex to a Doppier detected placental pulse (Pl) and to a pulse wave detected at the fetal scalp are (R-P1) and (R-Sc) respectively [15]. Fig. l illustrates the position of some of these intervals. This figure is a reproduction of digitized data from acute animal experiments. Within these experiments no ultrasound Dopplerprobe has been used. Therefore the positions of MC, Mo and Ao, Ac are marked arbitrarily. Since the PEP consists of two parts and the interval Q-Mc is fairly constant in the fetal cardiac cycle, PEP will vary with the IVC. As myocardial contractility (dp/dt)max increases, PEP will decrease. The total interval PEP is inversely proportional to the rate of rise of ventricular pressure [47]. In other words, PEP is an index of left ventricular function and reflects changes in myocardial contractility, left ventricular enddiastolic volume and aortic diastolic pressure [36, 44]. Therefore the continuousmeasurement of STI,
u-*v~
QRS
PEP
Fig. 1. A schematic presentation of the systolic time intervals of the fetal sheep cardiac cycle. PEP: pre-ejection period, VET: ventricular ejection period, R^Sl, R-S2: time from the QRS-complex to the first (I) and second (II) heart sound, respectively, Ao, Ac: aortic opening and closure, Mo, MC: mitral opening and closure.
especially PEP, may provide the first direct clinical assessment of fetal myocardial function. Several investigators [14, 17, 32, 34, 40, 47] includüig ourselves [37, 39, 61] have found that continuous observations of changes in STI should enable discrimination between different fetal stress situations, especially those due to uteroplattental insufficiency. 3 Methodsof determiningSTI The general technique for determining STI is based on at least two simultaneously recorded Signals. One of these Signals, the EKG, is obtained J. Perihat. Med. 6 (1978)
Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
175
Morgenstern et al., Systolic time intervals
by the use of electrodes applied to either the maternal abdomen or the fetal scalp. This signal has the best signal-to-noise ratio and it is quite easy to obtain a trigger circuit that responds to the slope of the R-wave. Since the Q-wave is often difficult to discern in the fetal EKG, most of the STI measurements contain the systematic error derived from the ventricular activation time (Q-R-interval). The results from different investigators are therefore not stricfly comparable with one another. The second signal for determining STI could be the detection of the heart sounds, the valvular motions or the pulse propagations. The most common technique for this second signal is based on the Doppier frequency shift. With these techniques, an ultrasonic wave, generated by a transducer placed on the mother's abdomen, is directed towards the fetal heart. Since the incident ultrasonic beam usually has an angle of more than ~5° and because it is nearly impossible to direct it at the aortic valvular motion manually — though clarity of the Doppier signal was enhanced by means of bandpass füters [40, 46] — reflections take place from each of several surfaces äs well äs from the heart. Fundamental with this method is the fact that, since the velocity of the valvular structures is greater than that of myocardial wall and other motions, the resultant Doppier frequency shift should be different. ORGAN and colleagues have described the ultrasound Doppier technique for use with the human fetus. They used an averaging technique and a bandpass filter of 750-1000 Hz [46]. With this experimental arrangement they found it possible to record sounds which were generated only from the opening of the aortic and pulmonary valves and were able to determine the time of the opening of the aortic valve (Ao) in 77% of the fetuses of women in labour. Compared to ouf animal experiments in 1976 we performed 220 events on 12 sheep fetuses. We were able to monitor good PEP traces during 53% of the trials. As mentioned, the time sequences of the Doppier signal within the cardiac cycle apparently depend on the angle of incidence in accordance with fetal or maternal movements. These circumstances make this method — äs it does for FHR monitoring —
unsuitable for clinical routine. MURATA [40] tried to define the variability of the time delay of the Doppier signal due to differences in the angle between ultrasonic beam and direction of valve motion, which can cause different trigger points and consequent uncertainty of the time intervals by about lOmsec [7]. These effects are well known from the typical ultrasound Doppier FHR-jitterpattern. HON et al. [28] developed a technique for microfilm display of the STI. They used the R-wave of the EKG äs a trigger and displayed mechanical events of every cardiac cycle äs a sequence of curves using either the ultrasound Doppier cardiogram from the fetus or the arterial pressure tracing from the newborn infant. The pulse wave transmission methods for the second signal are very different because of their physical background. The probes for detecting the peripheral fetal pulse were tested and listed by GOODLIN and LÖWE [16]. One of the most interesting ones is the scalp probe which consists of an infra-red light source and a photoconductive cell which converts the reflected light into electrical pbtential. Since the same technique is used to determine-02-saturation, this transducer could become a very iirfportant one in fetal monitoring. GOODLIN et al. were able to record a satisfactory scalp pulse — which commonly is abbreviated BVP from blood volume pulse - in 61% of the intrapartus fetuses. The same authors used a Piezo crystal, thermistor, Doppier scalp, impedance plethsmography, an iiitrauterine microphone and aphysmographic receiver to record fetal pulse propagation. In order to use the fetal heart sound äs the second signal to determine STI, a sensitive microphone (condenser, Piezo crystal or dynamic) is attached to the maternal abdomen. Since the fetal S l is often prolonged and diffuse, with several components, the R-S1 interval is difficult to measure. With these methods unavoidable variables have to be taken into account, which depend on the stage of gestation and on the size and portion of the fetus. Furthermore the heart sounds are attenuated in their passage from the fetal heart by the volume of the amniotic fluid, the thickness of the abdominal wall and other factors. Since we found that the intensity of neither the first heart sound nor the second was consistenüy higher than the
J Perinat. Med. 6 (1978)
Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
176
Morgenstern et al., Systolic time intervals
other during acute experiments with slow or rapid heart rate, but that there was sometimes a significant difference in intensity, the Situation became more complicated. To overcome this problem the first and second heart sounds were controlled in different electronic circuits [9]. Because of these problems several investigators have used the R-S2 interval, assuming that any Prolongation that may occur is most likely to be a result of lengthening of R-S1 (including PEP) rather than S1-S2 [16]. This is amajor assumption and might well be an over-simplification. 4
Factors affecting STI
Recently reported results about STI of the fetal cardiac cycle have created a Stimulus for investigating the behaviour of these parameters in ahimal experiments under a variety of Stresses. For several reasons the sheep fetus has beeh the Standard research preparation. Numerous investigators have studied STI in the mature lamb fetus exteriorized by caesarean section with the umbilical circulation intact. As pointed out by GOODLIN [18] the saline filled glove placed over the fetus' head, during acute experiments, to prevent breathing has an apneic effect. On the other hand DAWES [11] has noted an arrest of breathing or apnea in fetal sheep during hypoxemia. Though these changes need not be reversible, the circumstances surrounding the fetal preparation might affect STI. HEYMANN and RUDOLPH have demonstrated that umbilical blood flow falls after exteriorization of the fetal lamb. This increase in umbilical-placental vascular resistance could markedly alter the distribution of the cardiac Output [24]. Thus there is no doubt that the fetal sheep in an acute and experimentally anesthetized and exteriorized Situation may have altered physiological functions. In this article no attempt will be made to distinguish between the usefulness and degree of precision of data obtained from acute, or on the other hand from chronic, fetal sheep preparations. The great clinical advantage in the study of hemodynamic changes of the fetal cardiovascular System using STI measurements lies in the non-invasiveness of the methods. Since the PEP is defined äs PEP = (Q-S2) - VET there is no easy and direct way to
measure the true PEP values. One systematic error occurs because of the difficulty in detecting the beginning of the QRS complex, and another uncertainty occurs in the impossibility of defining the upstroke and incisura of the arterial blood pressure wave in a non-invasive way. The three ways of detecting these points are with ultrasound, phono or pulse propagation methods. In our acute experiments with fetal lambs we used a sensitive microphone to detect the heart sounds S l and S2. In addition to some of these problems, GOODLIN et al. [19] found no agreement between PEP obtained with R-S2 minus ejection time (scalp blood volume pulse) and R-Ao. To explain this discrepancy, they employed RUDOLPH'S Suggestion that the fetal ventricles may not always act in synchrony, implying that R-Ao may be different for the right and left ventricles. These findings are interesting, when compared with our findings that PEP and Q-S1 are very closely correlated äs illustrated in Fig. 2. In this scatterdiagram, data from 8 animals in 75 trials are included, namely: 3 of hypoxemia on 3 fetuses by replacing the maternal oxygen, 3 blood transfusions (3 ml) into the umbilical vein, 4 compressions of the maternal aorta and 3 of the vena cava, 26 umbilical cord
PEP [msec] 140 ·
R = 0.89
100
60
20 o CM
Q-S1 [msec] Fig. 2. Scatterdiagram of the time intervals PEP and Q-S1 derived from the maximum of the Q-wave to the onset of the arterial blood pressure and to a defined phase shift point within the first heart sound Sl, respectively. For further explanation refer to the text. J. Perinat. Med. 6 (1978)
Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
177
Morgenstern et al., Systolic time intervals
occlusions, 14 head compressions, 11 umbilical artery and 11 vein occlusions. The correlation has been built up by determining Q-Ao (Ao upstroke of the aortic pressure) and Q-S1 via a Cursor visually on a storage scope. During the selection of the time points to establish PEP only the fetal EKG and the pressure wave were displayed and during the fixation of Q-S1 only the trace of the phonosignal together with the EKG were displayed. After the first component within the S l signal a clearly visible stage, a kind of phase shift, could always be identified äs a trigger. Each point in Fig. 2 is the mean value of at least 6 consecutive heart cycles either before, during or after acute experiments.
The correlation (R=0.89) leads us to the conclusion that it is satisfactory to measure Q-S1 instead of PEP. The Tabs. II and III give a summary of results from different investigators. In Tab. II the durations of cardiac intervals observed in human and sheep fetuses are put together. There is no marked difference in the results obtained from human and those obtained from sheep fetuses. Since the results depend on the different methods applied, especially on the QRS trigger, refer to the authors concerned for further Information. In Tab. III the regression equations for the PEP intervals are shown äs a function of the R-R intervals. Except in the last two lines, which refer to
Tab. II. Duration of the systolic time intervals observed in human (h) and sheep (s) fetuses PEP [msec]
Q-S1 [msecj
PEP/VET Ref.
315-750
35.0 ±15. +30. +30. + 16. +17. 64.0-23. 75.0-15. 280.0-25. 220.0-12. 65.0± 8. 73.0±10. 72.0 62.3 67.0± 9. 188.0±10. 29.4±5.61 41.3±3.31 56.7±3.71 70.0± 2.38 65.4± 18.4 »65.2± 18.6 218.3±27 .8171.5 ±29.310.38 ±0.02 74.2±27.9 71.7± 1.3
330-670 220-900 220-550
Ac-Mo [msec]
VET [msec]
Mc-Ao [msec]
240-340 520-670
Q-Mc [msec]
Q-S2 [msec]
Range (msecj
[32] h [16] h [47] s [46] h l 7]h 143] h [41] h [40] h + ) s + ) s [34] h
+) present investjgation Tab. III. Regression equation for the PEP intervals äs a function of RR intervals Sheep-No
PEP = a + b * RR[msec) interception a slopeb
Mean [msec]
Range [msec]
Number of heart cycles
Correlation Coeff.
72 73 75 77 79 80
59.4(0.9) 31.7(0.5) 21.8(0.6) 20.5(0.7) 22.0(1.4) 24.0(0.3)
0.155(0.002) 0.105(0.002) 0.120(0.002) 0.041(0.002) 0.165(0.004) 0.146(0.001)
71.2(27.2) 65 .4( 7.4) 61.9(12.1) 36.7( 6.6) 76.7( 7.2) 73.9(15.3)
290-550 240-420 240-500 300-520 260-400 220-550
9963 18495 15389 7912 4770 8282
0.568 +) 0.445 +) 0.482 +) 0.255 +) 0.492 +) 0.888 *)
Total
14.5(1.0)
0.169(0.003)
74.2(27.9)
220-550
18648
0.398 +)
[61] [61]
20.0 29.5
0.132 0.130
220-460 220-460
3000 500
[40] [26]
36.14 33.8(1.4)
0.295 0.108(0.005)
330-530
800
.
70.0( 2.4) 74.7(19.1)
220-380
*) present investigation J. Perinat. Med. 6 (1978)
Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
0.85 0.71
178
Morgenstern et al., Systolic time intervals
adults during uninterrupted exercise [26], and to human fetuses [40] respectively, all results are derived from acute animal experiments. The TOTAL' line and the line next to it, have been built up by selecting at random about 3000 heart cycles from each of the above 6 experiments and from experiments in 1975 [61], respectively. Within all results the slopes and interceptions are in fairly good agreement. The correlation between PEP and R-R is highly significant, PEP increasing with increasing R-R. Fig. 3. shows the relationship between the PEP mean values and the slopes or the interceptions from Table . From the values of Tab. III we tried to find a method to evaluate a common but individual regression equation between PEP intervals and R-R intervals. Therefore we calculated the two regression equations
MEAN [msec]
80·)
eV •M °2
60
MEAN [msec]
PEP 10
20
50
30
60
'
80 6«
,
.060
,
.100
r
.140
,
PEP
.180 slope
Fig. 3..Scatterdiagram of the values from Tab. HL The different values for the PEP interceptions and PEP slopes with the corresponding mean PEP values are shown. The numbers refer to the lines, T to the line 'Total', M to line no. 10 (Ref. 40) and V to line no. 11 (Ref. 26). The PEP dope value 0.295 (Ref. 40) is not presented.
(l)PEP(interception) = -310 + 5 * mean PEP value [msec]; R~0.2 (2) PEP (slope) -0.52 + 0.01 * mean PEP value [msec]; R=0.57 With these equations it is easy to built up a common but individual regression equation between PEP and R-R intervals. In a clinical applicatipn a mean PEP value of about 65 msec might have been measured. Placing this value in equation (1) and (2) the calculated regression equation for this example is: (3) PEP = 15 + 0.13 * R-R [msec] To continue the clinical example, if a FHR deceleration from the baseline of 150 to 90 bpm (400 to 667 msec) occurs, it is possible to calculate the change in PEP due the FHR change. For this special case the PEP before and after the deceleration using equation (3) is 67 and 102 msec, respectively. Thus the difference between the actual measured and calculated PEP might give further Information of the fetus' condition. Since our results were evaluäted from acute animal experiments, we analysed VET for increasing R-R intervals in order to compare it with results estimated from human fetuses. Our findings are very close to those of MURATA and MARTIN [40]. They obtained for the relation between the duration of Ao-Ac and R-R interval from 15 normal fetuses a linear increase in Ao-Ac äs R-R increases from 330 to 530 msec. We have found that VET increases äs R-R increases from 225 to 525 msec (heart rate decreases from 267 to 114 bpm). No further Prolongation of VET was observed with an R-R interval greater than 525 msec, äs demonstrated in Fig. 4. The studies done by KELLY [32] on fifty women between their twenty-fourth and fortieth week showed that the mechanical systole is shortened during tachycardia and that there is no further Prolongation below a heart rate of about 115 bpm. Since VAN DER HOEVEN [26] published a regression line with tK&heart rate äs the variable, we transformed their data and obtained the following: VET = (36.14) + (0.295) * R-R [msec]
[40]
J. Perinat. Med. 6 (1978) Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
179
Morgenstern et al., Systolic time intervals
VET
[msec] N=1013 R=Q97
250
150
50 200
600
600
800
R R-INTERVAL [msec] Fig. 4. Relation between duration of VET intervals and R-R intervals from 8 fetal sheep. VET is the interval from Q to the onset of arteiial bloodpressure. VET increases äs R-R incieases from 225 to 525 msec. (Heart rate decreases from 267 to 114 bpm). No further Prolongation of VET was observed with R-R intervals greater than 525 msec. Mean ± SD
LVET = (147.0 ± 6.6) + (0.18 ± 0.01) * R-R [msec] [26] VET = (73.8 ± 5.8) + (0.20 ± 0.01) * R-R [msec] [this publ] The inverse relationship between VET and FHR is not in contrast to GOODLIN et al. [18], who found all intrapartum fetal STI (Q-S2, S1-S2, VET) were constant even with ranges of 60 to 180 beats/min in the same fetus. These results are in contrast with those of WEISSLER [67] who, carried out measurements in 121 male adult subjects in the supine position and with those of WILLEMS et al. [72] who carried out measurements in 72 normals in the supine position at rest. The notable differences between the results of the latter two obtained in the supine position at rest, and those of VAN DER HoEVEN et al. [26] obtained in the course of uninterrupted exercise, and ours obtained from values before and during very different events on fetal lambs, might be due to Variation in stress.
The VET chiefly reflects the isotonic phase of cardiac work. This time interval is closely related to heart rate, stroke volume and, indirectly, to afterload and contractility. Taking into account these facts, Fig. 4 might reflect the findings of RUDOLPH and HEYMANN [59], that within the ränge of 125 and 180 bpm the SV is relatively constant and with further increase in heart rate it falls rapidly. On the other hand KIR KPATRICK et al. [33] found constant left ventricular Output over the ränge of 114 to 180 bpm. In Fig. 5 PEP and VET are shown äs a proportion (%) of the heart cycle (%PEP and %VET). The %VET increases with increasing heart rate and reaches 50% of the heart cycle at about 220 bpm (275 msec). We estimate the same trend for %PEP, which increases from about 8% to about 25% of the heart cycle äs R-R intervals decrease from 800 to 200 bpm. Very high tachycardia, where the systolic reaches the diastolic, reflects that the fetal heart does not work economically any more. HAMMACHER [21] suggests that %VET compared with physiological values increases due to tachycardia with hypoxic metabolism and decreases due to bradycardia with 02-lack. We tested HAMMACHER'S idea by selecting 11 events with FHR bradycardia due to 02-lack and 9 bradycardias without. These results are shown in Fig. 6. The open circles refer to those with 02-lack, while the dots refer to those without 02-lack. The mean values for %VET are: 30.6 ±4.6% and 25.8 ±5.2%
The difference between these mean values is statistically significant, p < 0.05. These findings are in agreement with those of GOODLIN et al. [16] in human fetuses. They found a relative consistency in the duration of S1-S2 interval and a markedly prolonged diastolic interval during periods of fetal bradycardia associated with uterine contractions. While SMYTH and FARROW [65] believed it was associated with decreased venous return, and PERSIANINOV and colleagues [49] thought it was due to incomplete vagal effects, GOODLIN et al. had no explanation for its etiology. From these findings they speculated that the constant S1-S2 interval during distinct FHR decelerations is benign.
J. Perinat. Med. 6 (1978)
Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
Morgenstern et al., Systolic time intervals
180 PEP
VET
N=1167 R=-Q96
N=1170 40 R=-Q97
60 50
30
40 30
10
20
200
400
600
200
800
400
600
800
RR INTERVAL [msec]
RR INTERVAL [msec]
Fig. 5. Relation between VET and PEP intervals äs percentage of the heart cycle and R-R intervals. VET [%] increases from about 20% to 50%, whereas PEP [%] increases from about 8% to 25% äs R-R decreases from 800 to 200 msec. The linear regression equations are: %PEP = (29.8 ± 1.0) - (0.031 ± 0.002) * RR [msec] %VET = (58.2 ± 1.3) - (0.046 ± 0.00) * RR [msec]
VET [%] 40
o° OO
30
o o o
20
10
600
700
800
900
R R - I N T E R V A L [msec] Fig. 6. The diagram shows the relationship between VET, äs percentage of the heart cycle, and the R-R interval. The mean values of 11 events with FHR bradycardia due to O2-lack (open circles) together with 9 events with FHR bradycardia without O2-lack (dots) are depicted. The mean values of the two groups are different (p < 0.05). J Perinat. Med. 6 (1978) Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
181
Morgenstern et al., Systolic time intervals
PEP/VET N=1168 R=Q97
1JO
.6
200
400
600
800
RR-INTERVAL [msec] Fig. 7. Relation between the PEP/VET quotient and R-R intervals. The linear regression equation is: PEP/VET = (0.61 ± 0.03) - (0.00036 ± 0.0006) * RR[msec]
The quotient PEP/VET is suggested to be a sensitive Parameter indicating short term hypoxia [54]. In Fig. 7 this quotient is shown äs a function of R-R interval PEP/VET decreased with increasing R-R interval. Because of the positive correlation found between the PEP/LVET ratio and (max dp/dt)/p PUST et al. [52] used this ratio äs an index of myocardial contractility. As pointed out, there are still many problems in interpretating fetal heartrate patterns. Therefore it is necessary that we attempt to understand the various mechanisms that control the FHR in utero. The FHR response to stress can vary and depends upon the interplay of numerous factors. In any case, fetal bradycardia seems to be a protective medianism for the fetus in response to stress.
PAUL et al- [48] studied the effects of fetal head compression on term fetal lambs in 1964. They noted an elevation in blood pressure and a decrease in the fetal heart rate and carotid artery flow. They tried to explain their fmdings by relating them to increased vagal tone and to an increase in the Stimulation of the baroreceptors in the aortic arch, and postulated that the occasionally persisting bradycardia after the release of external pressure could be due to central hypoxia. These findings are in agreement with our results. In Fig. 8 typical responses to two head compressions are shown. The deceleration in FHR pattern is accompanied by a Prolongation in PEP. A small part of our data concerning head compression and umbilical cord clamping are evaluated and presented in Tab. IV. In this table in the first seven and in the last five rows the values are given for PEP, Q-S2, VET, PEP/VET and the FHR of 14 cord clampings and 10 head compressions, respect-
' -7 }
i r t;
· "::v:r ·'» t-p:·: *'' -|·:·—-"-»·-»
::
Morgenstern et al., Systolic time intervals
Revfew article J. Perinat. Med. 6 (1978)173
Systolic time intervals of the fetal cardiac cycle J. Morgenstern, H. Czerny, H. Schmidt, J. Schulz*, K. Wernicke* Abt. für Biomedizinische Technik, Universitäts-Frauenklinik Düsseldorf, Germany *Universitäts-Frauenklinik Frankfurt/Main, Germany
l Introduction Fetal heart rate (FHR) monitoring offers a tremendous amount of Information about the state of the fetus. The visual Interpretation of FHR traces is based on descriptive criteria [22]. Many attempts have been made to associate particular FHR patterns with certain types of fetal stress, though it is common clinical knowledge that head compression, umbilical cord clamping and hypoxia can give rise to similar FHR patterns. Some of these investigators introduced new terminology and so muddled the FHR nomenclature thus making precise communication of research reports nearly impossible. Since there :is no means of associating different fetal stress situations with typical FHR patterns, the application of more recent Computer methods and techniques [38] seems to be a promising area in which to achieve further Information on the physiological condition of the fetus. Therefore correlation öf the mechanical and electrical events of the fetal cardiac cycle seems to be a most promising direction in which to investigate. The measurement of the fetal systolic time interval (STI) shows a relationship with myocardial contractility. The knowledge of changes in myocardial contractility may Supplement the Interpretation of FHR patterns äs an indicator of the presence or absence of fetal distress. It has been over 70 years since the introduction of a method of measuring STI. In 1906, CREMER [8] reported the first successful fetal electro-
cardiogram (EKG) and in 1908 HOFBAUER and WEISS [27] recorded the first phonocardiogram of the fetal heart sounds. The first consecutive phases of the cardiac cycles of animal studies were published by WIGGERS in 1920 [71]. Two years later, in 1922, KATZ and FEIL [30] carried out human studies. In 1941, the fetal EKG and the fetal phonocardiogram were studied simultaneously for the first time by PUETZ [51]. In 1951 and ten years later there were two other reports about simultaneously recording electro- and phonocardiogram of the fetus in a few cases of normal pregnancies. In 1967, KAMMACHER [21] had already established that the STI expressed in percent of the heart cycles offers further information about the FHR concerning the fetus' condition in utero. In 1968, WEISSLER and coworkers [68] described non-invasive determination of STI in cases of heart failure in adults, and pointed out the clinical relevance of STI of the cardiac cycle. These fmdingsled many investigators to the conclusion that STI — especially the preejection period (PEP) and the left ventricular ejection time (VET) - are sensitive indicators of cardiac performance. 2 STI-Definitions The STI varies with the method used. The time period from the onset of ventricular depolarization (QRS complex or deflection of the R0300-5577/78/0006-0173$02.00 © by Walter de Gruyter & Co.
Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
Morgenstern et al., Systolic time intervals
174
VET
wave from the isoelectric line) to the onset of ejection from the left ventricle is known äs PEP, and that to the first heart sound äs Q-S1. Q-S1 is also termed the electromechanical interval of the left ventricle. Between the onset of electrical depolarisation and the beginning of mechanical contraction there exists a time gap, which corresponds very closely to the interval Q-Mc, where MC Stands for the closure of the atrioventricular (mitral) valve. The opening of the aortic valve is called Ao. During the Mc-Ao interval the myocardium builds up tension isometrically. For this reason and because the pressure within the ventricle increases until the intraventricular pressure equals intra-aortic pressure, this interval is referred to äs isovolumetric contraction time (IVC). Thus PEP is made up of two time intervals:
Ac
PEP = (Q - MC) + (Mc - Ao) The remaining part of the STI is VET. It is the interval between semilunar valve opening and closure: VET = (Ao—Ac). VET is usually measured from the beginning of the rise to the dicrotic notch in the pulse wave. The mechanical systole is the interval between the first heart sound (Sl) and the second heart sound (S2). With these definitions other intervals of the fetal heart cycle e.g. R-S1, R-S2 are easy to understand. The time from the QRS complex to a Doppier detected placental pulse (Pl) and to a pulse wave detected at the fetal scalp are (R-P1) and (R-Sc) respectively [15]. Fig. l illustrates the position of some of these intervals. This figure is a reproduction of digitized data from acute animal experiments. Within these experiments no ultrasound Dopplerprobe has been used. Therefore the positions of MC, Mo and Ao, Ac are marked arbitrarily. Since the PEP consists of two parts and the interval Q-Mc is fairly constant in the fetal cardiac cycle, PEP will vary with the IVC. As myocardial contractility (dp/dt)max increases, PEP will decrease. The total interval PEP is inversely proportional to the rate of rise of ventricular pressure [47]. In other words, PEP is an index of left ventricular function and reflects changes in myocardial contractility, left ventricular enddiastolic volume and aortic diastolic pressure [36, 44]. Therefore the continuousmeasurement of STI,
u-*v~
QRS
PEP
Fig. 1. A schematic presentation of the systolic time intervals of the fetal sheep cardiac cycle. PEP: pre-ejection period, VET: ventricular ejection period, R^Sl, R-S2: time from the QRS-complex to the first (I) and second (II) heart sound, respectively, Ao, Ac: aortic opening and closure, Mo, MC: mitral opening and closure.
especially PEP, may provide the first direct clinical assessment of fetal myocardial function. Several investigators [14, 17, 32, 34, 40, 47] includüig ourselves [37, 39, 61] have found that continuous observations of changes in STI should enable discrimination between different fetal stress situations, especially those due to uteroplattental insufficiency. 3 Methodsof determiningSTI The general technique for determining STI is based on at least two simultaneously recorded Signals. One of these Signals, the EKG, is obtained J. Perihat. Med. 6 (1978)
Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
175
Morgenstern et al., Systolic time intervals
by the use of electrodes applied to either the maternal abdomen or the fetal scalp. This signal has the best signal-to-noise ratio and it is quite easy to obtain a trigger circuit that responds to the slope of the R-wave. Since the Q-wave is often difficult to discern in the fetal EKG, most of the STI measurements contain the systematic error derived from the ventricular activation time (Q-R-interval). The results from different investigators are therefore not stricfly comparable with one another. The second signal for determining STI could be the detection of the heart sounds, the valvular motions or the pulse propagations. The most common technique for this second signal is based on the Doppier frequency shift. With these techniques, an ultrasonic wave, generated by a transducer placed on the mother's abdomen, is directed towards the fetal heart. Since the incident ultrasonic beam usually has an angle of more than ~5° and because it is nearly impossible to direct it at the aortic valvular motion manually — though clarity of the Doppier signal was enhanced by means of bandpass füters [40, 46] — reflections take place from each of several surfaces äs well äs from the heart. Fundamental with this method is the fact that, since the velocity of the valvular structures is greater than that of myocardial wall and other motions, the resultant Doppier frequency shift should be different. ORGAN and colleagues have described the ultrasound Doppier technique for use with the human fetus. They used an averaging technique and a bandpass filter of 750-1000 Hz [46]. With this experimental arrangement they found it possible to record sounds which were generated only from the opening of the aortic and pulmonary valves and were able to determine the time of the opening of the aortic valve (Ao) in 77% of the fetuses of women in labour. Compared to ouf animal experiments in 1976 we performed 220 events on 12 sheep fetuses. We were able to monitor good PEP traces during 53% of the trials. As mentioned, the time sequences of the Doppier signal within the cardiac cycle apparently depend on the angle of incidence in accordance with fetal or maternal movements. These circumstances make this method — äs it does for FHR monitoring —
unsuitable for clinical routine. MURATA [40] tried to define the variability of the time delay of the Doppier signal due to differences in the angle between ultrasonic beam and direction of valve motion, which can cause different trigger points and consequent uncertainty of the time intervals by about lOmsec [7]. These effects are well known from the typical ultrasound Doppier FHR-jitterpattern. HON et al. [28] developed a technique for microfilm display of the STI. They used the R-wave of the EKG äs a trigger and displayed mechanical events of every cardiac cycle äs a sequence of curves using either the ultrasound Doppier cardiogram from the fetus or the arterial pressure tracing from the newborn infant. The pulse wave transmission methods for the second signal are very different because of their physical background. The probes for detecting the peripheral fetal pulse were tested and listed by GOODLIN and LÖWE [16]. One of the most interesting ones is the scalp probe which consists of an infra-red light source and a photoconductive cell which converts the reflected light into electrical pbtential. Since the same technique is used to determine-02-saturation, this transducer could become a very iirfportant one in fetal monitoring. GOODLIN et al. were able to record a satisfactory scalp pulse — which commonly is abbreviated BVP from blood volume pulse - in 61% of the intrapartus fetuses. The same authors used a Piezo crystal, thermistor, Doppier scalp, impedance plethsmography, an iiitrauterine microphone and aphysmographic receiver to record fetal pulse propagation. In order to use the fetal heart sound äs the second signal to determine STI, a sensitive microphone (condenser, Piezo crystal or dynamic) is attached to the maternal abdomen. Since the fetal S l is often prolonged and diffuse, with several components, the R-S1 interval is difficult to measure. With these methods unavoidable variables have to be taken into account, which depend on the stage of gestation and on the size and portion of the fetus. Furthermore the heart sounds are attenuated in their passage from the fetal heart by the volume of the amniotic fluid, the thickness of the abdominal wall and other factors. Since we found that the intensity of neither the first heart sound nor the second was consistenüy higher than the
J Perinat. Med. 6 (1978)
Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
176
Morgenstern et al., Systolic time intervals
other during acute experiments with slow or rapid heart rate, but that there was sometimes a significant difference in intensity, the Situation became more complicated. To overcome this problem the first and second heart sounds were controlled in different electronic circuits [9]. Because of these problems several investigators have used the R-S2 interval, assuming that any Prolongation that may occur is most likely to be a result of lengthening of R-S1 (including PEP) rather than S1-S2 [16]. This is amajor assumption and might well be an over-simplification. 4
Factors affecting STI
Recently reported results about STI of the fetal cardiac cycle have created a Stimulus for investigating the behaviour of these parameters in ahimal experiments under a variety of Stresses. For several reasons the sheep fetus has beeh the Standard research preparation. Numerous investigators have studied STI in the mature lamb fetus exteriorized by caesarean section with the umbilical circulation intact. As pointed out by GOODLIN [18] the saline filled glove placed over the fetus' head, during acute experiments, to prevent breathing has an apneic effect. On the other hand DAWES [11] has noted an arrest of breathing or apnea in fetal sheep during hypoxemia. Though these changes need not be reversible, the circumstances surrounding the fetal preparation might affect STI. HEYMANN and RUDOLPH have demonstrated that umbilical blood flow falls after exteriorization of the fetal lamb. This increase in umbilical-placental vascular resistance could markedly alter the distribution of the cardiac Output [24]. Thus there is no doubt that the fetal sheep in an acute and experimentally anesthetized and exteriorized Situation may have altered physiological functions. In this article no attempt will be made to distinguish between the usefulness and degree of precision of data obtained from acute, or on the other hand from chronic, fetal sheep preparations. The great clinical advantage in the study of hemodynamic changes of the fetal cardiovascular System using STI measurements lies in the non-invasiveness of the methods. Since the PEP is defined äs PEP = (Q-S2) - VET there is no easy and direct way to
measure the true PEP values. One systematic error occurs because of the difficulty in detecting the beginning of the QRS complex, and another uncertainty occurs in the impossibility of defining the upstroke and incisura of the arterial blood pressure wave in a non-invasive way. The three ways of detecting these points are with ultrasound, phono or pulse propagation methods. In our acute experiments with fetal lambs we used a sensitive microphone to detect the heart sounds S l and S2. In addition to some of these problems, GOODLIN et al. [19] found no agreement between PEP obtained with R-S2 minus ejection time (scalp blood volume pulse) and R-Ao. To explain this discrepancy, they employed RUDOLPH'S Suggestion that the fetal ventricles may not always act in synchrony, implying that R-Ao may be different for the right and left ventricles. These findings are interesting, when compared with our findings that PEP and Q-S1 are very closely correlated äs illustrated in Fig. 2. In this scatterdiagram, data from 8 animals in 75 trials are included, namely: 3 of hypoxemia on 3 fetuses by replacing the maternal oxygen, 3 blood transfusions (3 ml) into the umbilical vein, 4 compressions of the maternal aorta and 3 of the vena cava, 26 umbilical cord
PEP [msec] 140 ·
R = 0.89
100
60
20 o CM
Q-S1 [msec] Fig. 2. Scatterdiagram of the time intervals PEP and Q-S1 derived from the maximum of the Q-wave to the onset of the arterial blood pressure and to a defined phase shift point within the first heart sound Sl, respectively. For further explanation refer to the text. J. Perinat. Med. 6 (1978)
Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
177
Morgenstern et al., Systolic time intervals
occlusions, 14 head compressions, 11 umbilical artery and 11 vein occlusions. The correlation has been built up by determining Q-Ao (Ao upstroke of the aortic pressure) and Q-S1 via a Cursor visually on a storage scope. During the selection of the time points to establish PEP only the fetal EKG and the pressure wave were displayed and during the fixation of Q-S1 only the trace of the phonosignal together with the EKG were displayed. After the first component within the S l signal a clearly visible stage, a kind of phase shift, could always be identified äs a trigger. Each point in Fig. 2 is the mean value of at least 6 consecutive heart cycles either before, during or after acute experiments.
The correlation (R=0.89) leads us to the conclusion that it is satisfactory to measure Q-S1 instead of PEP. The Tabs. II and III give a summary of results from different investigators. In Tab. II the durations of cardiac intervals observed in human and sheep fetuses are put together. There is no marked difference in the results obtained from human and those obtained from sheep fetuses. Since the results depend on the different methods applied, especially on the QRS trigger, refer to the authors concerned for further Information. In Tab. III the regression equations for the PEP intervals are shown äs a function of the R-R intervals. Except in the last two lines, which refer to
Tab. II. Duration of the systolic time intervals observed in human (h) and sheep (s) fetuses PEP [msec]
Q-S1 [msecj
PEP/VET Ref.
315-750
35.0 ±15. +30. +30. + 16. +17. 64.0-23. 75.0-15. 280.0-25. 220.0-12. 65.0± 8. 73.0±10. 72.0 62.3 67.0± 9. 188.0±10. 29.4±5.61 41.3±3.31 56.7±3.71 70.0± 2.38 65.4± 18.4 »65.2± 18.6 218.3±27 .8171.5 ±29.310.38 ±0.02 74.2±27.9 71.7± 1.3
330-670 220-900 220-550
Ac-Mo [msec]
VET [msec]
Mc-Ao [msec]
240-340 520-670
Q-Mc [msec]
Q-S2 [msec]
Range (msecj
[32] h [16] h [47] s [46] h l 7]h 143] h [41] h [40] h + ) s + ) s [34] h
+) present investjgation Tab. III. Regression equation for the PEP intervals äs a function of RR intervals Sheep-No
PEP = a + b * RR[msec) interception a slopeb
Mean [msec]
Range [msec]
Number of heart cycles
Correlation Coeff.
72 73 75 77 79 80
59.4(0.9) 31.7(0.5) 21.8(0.6) 20.5(0.7) 22.0(1.4) 24.0(0.3)
0.155(0.002) 0.105(0.002) 0.120(0.002) 0.041(0.002) 0.165(0.004) 0.146(0.001)
71.2(27.2) 65 .4( 7.4) 61.9(12.1) 36.7( 6.6) 76.7( 7.2) 73.9(15.3)
290-550 240-420 240-500 300-520 260-400 220-550
9963 18495 15389 7912 4770 8282
0.568 +) 0.445 +) 0.482 +) 0.255 +) 0.492 +) 0.888 *)
Total
14.5(1.0)
0.169(0.003)
74.2(27.9)
220-550
18648
0.398 +)
[61] [61]
20.0 29.5
0.132 0.130
220-460 220-460
3000 500
[40] [26]
36.14 33.8(1.4)
0.295 0.108(0.005)
330-530
800
.
70.0( 2.4) 74.7(19.1)
220-380
*) present investigation J. Perinat. Med. 6 (1978)
Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
0.85 0.71
178
Morgenstern et al., Systolic time intervals
adults during uninterrupted exercise [26], and to human fetuses [40] respectively, all results are derived from acute animal experiments. The TOTAL' line and the line next to it, have been built up by selecting at random about 3000 heart cycles from each of the above 6 experiments and from experiments in 1975 [61], respectively. Within all results the slopes and interceptions are in fairly good agreement. The correlation between PEP and R-R is highly significant, PEP increasing with increasing R-R. Fig. 3. shows the relationship between the PEP mean values and the slopes or the interceptions from Table . From the values of Tab. III we tried to find a method to evaluate a common but individual regression equation between PEP intervals and R-R intervals. Therefore we calculated the two regression equations
MEAN [msec]
80·)
eV •M °2
60
MEAN [msec]
PEP 10
20
50
30
60
'
80 6«
,
.060
,
.100
r
.140
,
PEP
.180 slope
Fig. 3..Scatterdiagram of the values from Tab. HL The different values for the PEP interceptions and PEP slopes with the corresponding mean PEP values are shown. The numbers refer to the lines, T to the line 'Total', M to line no. 10 (Ref. 40) and V to line no. 11 (Ref. 26). The PEP dope value 0.295 (Ref. 40) is not presented.
(l)PEP(interception) = -310 + 5 * mean PEP value [msec]; R~0.2 (2) PEP (slope) -0.52 + 0.01 * mean PEP value [msec]; R=0.57 With these equations it is easy to built up a common but individual regression equation between PEP and R-R intervals. In a clinical applicatipn a mean PEP value of about 65 msec might have been measured. Placing this value in equation (1) and (2) the calculated regression equation for this example is: (3) PEP = 15 + 0.13 * R-R [msec] To continue the clinical example, if a FHR deceleration from the baseline of 150 to 90 bpm (400 to 667 msec) occurs, it is possible to calculate the change in PEP due the FHR change. For this special case the PEP before and after the deceleration using equation (3) is 67 and 102 msec, respectively. Thus the difference between the actual measured and calculated PEP might give further Information of the fetus' condition. Since our results were evaluäted from acute animal experiments, we analysed VET for increasing R-R intervals in order to compare it with results estimated from human fetuses. Our findings are very close to those of MURATA and MARTIN [40]. They obtained for the relation between the duration of Ao-Ac and R-R interval from 15 normal fetuses a linear increase in Ao-Ac äs R-R increases from 330 to 530 msec. We have found that VET increases äs R-R increases from 225 to 525 msec (heart rate decreases from 267 to 114 bpm). No further Prolongation of VET was observed with an R-R interval greater than 525 msec, äs demonstrated in Fig. 4. The studies done by KELLY [32] on fifty women between their twenty-fourth and fortieth week showed that the mechanical systole is shortened during tachycardia and that there is no further Prolongation below a heart rate of about 115 bpm. Since VAN DER HOEVEN [26] published a regression line with tK&heart rate äs the variable, we transformed their data and obtained the following: VET = (36.14) + (0.295) * R-R [msec]
[40]
J. Perinat. Med. 6 (1978) Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
179
Morgenstern et al., Systolic time intervals
VET
[msec] N=1013 R=Q97
250
150
50 200
600
600
800
R R-INTERVAL [msec] Fig. 4. Relation between duration of VET intervals and R-R intervals from 8 fetal sheep. VET is the interval from Q to the onset of arteiial bloodpressure. VET increases äs R-R incieases from 225 to 525 msec. (Heart rate decreases from 267 to 114 bpm). No further Prolongation of VET was observed with R-R intervals greater than 525 msec. Mean ± SD
LVET = (147.0 ± 6.6) + (0.18 ± 0.01) * R-R [msec] [26] VET = (73.8 ± 5.8) + (0.20 ± 0.01) * R-R [msec] [this publ] The inverse relationship between VET and FHR is not in contrast to GOODLIN et al. [18], who found all intrapartum fetal STI (Q-S2, S1-S2, VET) were constant even with ranges of 60 to 180 beats/min in the same fetus. These results are in contrast with those of WEISSLER [67] who, carried out measurements in 121 male adult subjects in the supine position and with those of WILLEMS et al. [72] who carried out measurements in 72 normals in the supine position at rest. The notable differences between the results of the latter two obtained in the supine position at rest, and those of VAN DER HoEVEN et al. [26] obtained in the course of uninterrupted exercise, and ours obtained from values before and during very different events on fetal lambs, might be due to Variation in stress.
The VET chiefly reflects the isotonic phase of cardiac work. This time interval is closely related to heart rate, stroke volume and, indirectly, to afterload and contractility. Taking into account these facts, Fig. 4 might reflect the findings of RUDOLPH and HEYMANN [59], that within the ränge of 125 and 180 bpm the SV is relatively constant and with further increase in heart rate it falls rapidly. On the other hand KIR KPATRICK et al. [33] found constant left ventricular Output over the ränge of 114 to 180 bpm. In Fig. 5 PEP and VET are shown äs a proportion (%) of the heart cycle (%PEP and %VET). The %VET increases with increasing heart rate and reaches 50% of the heart cycle at about 220 bpm (275 msec). We estimate the same trend for %PEP, which increases from about 8% to about 25% of the heart cycle äs R-R intervals decrease from 800 to 200 bpm. Very high tachycardia, where the systolic reaches the diastolic, reflects that the fetal heart does not work economically any more. HAMMACHER [21] suggests that %VET compared with physiological values increases due to tachycardia with hypoxic metabolism and decreases due to bradycardia with 02-lack. We tested HAMMACHER'S idea by selecting 11 events with FHR bradycardia due to 02-lack and 9 bradycardias without. These results are shown in Fig. 6. The open circles refer to those with 02-lack, while the dots refer to those without 02-lack. The mean values for %VET are: 30.6 ±4.6% and 25.8 ±5.2%
The difference between these mean values is statistically significant, p < 0.05. These findings are in agreement with those of GOODLIN et al. [16] in human fetuses. They found a relative consistency in the duration of S1-S2 interval and a markedly prolonged diastolic interval during periods of fetal bradycardia associated with uterine contractions. While SMYTH and FARROW [65] believed it was associated with decreased venous return, and PERSIANINOV and colleagues [49] thought it was due to incomplete vagal effects, GOODLIN et al. had no explanation for its etiology. From these findings they speculated that the constant S1-S2 interval during distinct FHR decelerations is benign.
J. Perinat. Med. 6 (1978)
Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
Morgenstern et al., Systolic time intervals
180 PEP
VET
N=1167 R=-Q96
N=1170 40 R=-Q97
60 50
30
40 30
10
20
200
400
600
200
800
400
600
800
RR INTERVAL [msec]
RR INTERVAL [msec]
Fig. 5. Relation between VET and PEP intervals äs percentage of the heart cycle and R-R intervals. VET [%] increases from about 20% to 50%, whereas PEP [%] increases from about 8% to 25% äs R-R decreases from 800 to 200 msec. The linear regression equations are: %PEP = (29.8 ± 1.0) - (0.031 ± 0.002) * RR [msec] %VET = (58.2 ± 1.3) - (0.046 ± 0.00) * RR [msec]
VET [%] 40
o° OO
30
o o o
20
10
600
700
800
900
R R - I N T E R V A L [msec] Fig. 6. The diagram shows the relationship between VET, äs percentage of the heart cycle, and the R-R interval. The mean values of 11 events with FHR bradycardia due to O2-lack (open circles) together with 9 events with FHR bradycardia without O2-lack (dots) are depicted. The mean values of the two groups are different (p < 0.05). J Perinat. Med. 6 (1978) Brought to you by | University of Queensland - UQ Library Authenticated Download Date | 6/18/15 10:25 PM
181
Morgenstern et al., Systolic time intervals
PEP/VET N=1168 R=Q97
1JO
.6
200
400
600
800
RR-INTERVAL [msec] Fig. 7. Relation between the PEP/VET quotient and R-R intervals. The linear regression equation is: PEP/VET = (0.61 ± 0.03) - (0.00036 ± 0.0006) * RR[msec]
The quotient PEP/VET is suggested to be a sensitive Parameter indicating short term hypoxia [54]. In Fig. 7 this quotient is shown äs a function of R-R interval PEP/VET decreased with increasing R-R interval. Because of the positive correlation found between the PEP/LVET ratio and (max dp/dt)/p PUST et al. [52] used this ratio äs an index of myocardial contractility. As pointed out, there are still many problems in interpretating fetal heartrate patterns. Therefore it is necessary that we attempt to understand the various mechanisms that control the FHR in utero. The FHR response to stress can vary and depends upon the interplay of numerous factors. In any case, fetal bradycardia seems to be a protective medianism for the fetus in response to stress.
PAUL et al- [48] studied the effects of fetal head compression on term fetal lambs in 1964. They noted an elevation in blood pressure and a decrease in the fetal heart rate and carotid artery flow. They tried to explain their fmdings by relating them to increased vagal tone and to an increase in the Stimulation of the baroreceptors in the aortic arch, and postulated that the occasionally persisting bradycardia after the release of external pressure could be due to central hypoxia. These findings are in agreement with our results. In Fig. 8 typical responses to two head compressions are shown. The deceleration in FHR pattern is accompanied by a Prolongation in PEP. A small part of our data concerning head compression and umbilical cord clamping are evaluated and presented in Tab. IV. In this table in the first seven and in the last five rows the values are given for PEP, Q-S2, VET, PEP/VET and the FHR of 14 cord clampings and 10 head compressions, respect-
' -7 }
i r t;
· "::v:r ·'» t-p:·: *'' -|·:·—-"-»·-»
::
