Ulmsoundm Med. & Biol. Vol. 17, No. 5, pp. 471-478, Printed in the U.S.A.
0301-5629/91 $3.00 + .OO 0 1991 Pergamon Press PIG
1991
*Original Contribution AORTIC DIAMETER PULSE WAVES AND BLOOD FLOW VELOCITY IN THE SMALL, FOR GESTATIONAL AGE, FETUS
+ Department of Obstetrics and Gynaecology, Lund University, Malmij General Hospital, S-2 14 0 1 Malm6, Sweden, and *Department of Electrical Measurements, Lund Institute of Technology, Lund, Sweden Abstract-An ultrasound phase-locked, echo-tracking system was used for noninvasive measurements of pulsatile diameter changes in the descending aorta of 60 small, for gestational age @GA), fetuses and of 60 fetuses appropriate for gestational age (AGA). Pulsed Doppler ultrasound was used for the recording of blood flow velocity in the aorta and in the umbilical artery of the SGA fetuses. In the SGA fetuses, a weight-related higher end-diastolic diameter and a lower relative pulse amplitude suggest that diastolic blood pressure was increased, a less steep rise of the initial ascending part of the pulse wave and a lower relative pulse amplitude suggest that the absolute stroke volume was decreased. Except for a positive correlation between relative pulse amplitude and mean velocity in the aorta, no correlation was found between diameter pulse waves and blood flow velocity. Aortic diameter pulse waves apparently yield no unequivocal information as to peripheral resistance, for which purpose blood flow velocity waveform analysis would seem, at least at present, to be the only available method. Key Words: Ultrasound, Ultrasonic echo-tracking, Doppler ultrasound, Blood flow velocity, Fetal aorta, Aortic diameter, Blood pressure, Compliance, Intrauterine growth retardation.
INTRODUCTION
increase in PI indicating increased resistance in the placental vascular bed and other parts of the vascular bed supplied by the descending aorta (Lam-in et al. 1987a). Moreover, in a pilot study it was shown that characteristics of the aortic diameter pulse wave differ in the growth-retarded fetus from those in the normal fetus (Gennser and Stale 1988). These findings indicate that profound circulatory changes occur in growth-retarded fetuses, and therefore it is of interest to study the aortic haemodynamics in this high-risk group in order to follow the pathophysiological processes. Possibly, the information on fetal aortic pulse wave and blood velocity might be used for clinical assessment of a growth-retarded fetus. However, low birth weight in relation to gestational age does not necessarily imply IUGR, as it might equally be due to genetic factors. Here, therefore, the more proper term, small for gestational age (SGA), has been used to designate age-related low fetal weight.
Intrauterine growth retardation (IUGR)is a major cause of perinatal mortality and morbidity (Laurin et al. 1987b; Schauseil-Zipf et al. 1989). Accurate diagnosis and proper management continue to pose a challenge even to the experienced obstetrician. According to Grunewald (1963), chronic fetal distress due to placental insufficiency is the single most important cause of IUGR, although causes other than fetal malhutrition must be sought. During intrauterine asphyxia, e.g., in the severely growth-retarded fetus, the fetal blood flow is redistributed to ensure preferential blood supply to vital organs such as the brain, myocardium and adrenals (Creasy et al. 1973). In the human fetus, the brainsparing effect is reflected in a low PI in the cerebral vessels, and a concomitant increase of PI in the fetal aorta and umbilical artery (Wladimiroff et al. 1986). The blood velocity waveform recorded from the descending aorta of the growth-retarded fetus shows a reduction in diastolic velocity and a corresponding
METHODS In a prospective study, pulsatile diameter changes were measured on the descending aorta of 60 SGA fetuses. SGA was defined as an ultrasound-esti-
Address all correspondence to HHkan Stale. 471
472
Ultrasound in Medicine and Biology
mated fetal weight below the normal age-related mean - 2 SD, the mean weight being obtained from ultrasound weight-curves for a reference population of the city of Malmii (Persson and Weldner 1986). The gestational age of all fetuses had been ascertained by early ultrasound fetometry and, at diameter pulse wave measurement, ranged from 25 completed weeks + 6 days to 41 completed weeks + 6 days. One measurement from each of 60 fetuses, truly SGA at birth, were included in the analysis. The results from each measurement were compared with those pulsatile diameter changes measured on the descending aorta in 60 fetuses in uncomplicated pregnancies, matched for gestational age and with birth weights appropriate for gestational age (AGA). The median age of the women in the SGA group was 26 years (range 17-45) and, in the control group, 29 years (range 20-39). In the SGA group, 38 women were nulliparae, 17 primiparae and 5 multiparae, while the control group comprised 27 nulliparae, 23 primiparae and 10 multiparae. There were four twin pregnancies in the SGA group, but only one twin in each pregnancy was SGA, and thus included in the study. The control group comprised singleton pregnancies only, There were 20 smokers among mothers in the SGA group and 28 in the control group. In the SGA group, 25 women had symptoms of preeclampsia (blood pressure 2 140/90 mm Hg and proteinuria); one woman with diabetes mellitus also had hypertension, and isolated hypertension was observed in another seven women. One woman in the control group had mild symptoms of preeclampsia. Median gestational age at delivery was 37 completed weeks + 4 days (range 28 weeks + 2 days to 42 weeks + 1 day) in the SGA group, and 40 completed weeks (range 35 weeks + 5 days to 42 weeks + 3 days) in the control group. Median birth weight among the SGA fetuses was 2.06 kg (range 545-2.99 kg), the median deviation from the expected weight at delivery being -35% (range -22% to -59%). All fetuses in the control group had birth weights appropriate for gestational age, the median being 3.64 kg (range 2.50- 4.4 1 kg). There were 25 male infants and 35 female infants in the SGA group, as compared with 26 and 34, respectively, among the controls. In the SGA group, 20 infants were delivered vaginally (two by vacuum extraction), and the remaining 40 by Caesarean section. In the control group, 55 infants were delivered vaginally (four by vacuum extraction), and the remaining five by Caesarean section. There were 26 cases of ODFD (operative delivery for fetal distress) and/or fetal distress at birth in the SGA group, and seven in the control group. ODFD comprised Caesarean section or vaginal instrumental delivery because of cardiotoco-
Volume 17, Number 5, 1991
graphic patterns consistent with imminent asphyxia, and fetal distress at birth was defined as an Apgar score < 7 at 1 or 5 min, or pH I 7.10 in the umbilical artery or pH I 7.20 in the umbilical vein. There were three perinatal deaths in the SGA group; one antepartal death of unknown cause and two postpartal deaths, one in connection with streptococcal septicaemia, the other as a consequence of multiple malformation. No other major malformations were seen either in the SGA group or in the control group. For noninvasive monitoring of pulsatile diameter changes in the fetal descending aorta, an electronic echo-tracking instrument (Diamove@, Teltec, Lund, Sweden) was used, interfaced with a real-time ultrasound scanner (ADR, Tempe, AZ) fitted with a 3.0MHz linear array transducer (Lindstriim et al. 1987). For echo-tracking, phase-locked loop circuits were used, which restore the position of an electronic gate relative to the moving echo. The discrete compensatory steps of the gate are measured, yielding an output signal representing the echo movement per unit time. The instrument is equipped with dual echo-tracking loops, which enable two separate echoes from opposite vessel walls to be tracked simultaneously, the differential signals between them representing any instantaneous change in vessel diameter. In the system used, the smallest detectable movement is 14 pm, which far exceeds the resolution capacity of the realtime scanner for static objects. The fetal descending aorta was visualised in longitudinal section. Generally, the fetus was insonated at an oblique antero-posterior angle. Two horizontal electronic markers, each representing one tracking gate, were aligned with and locked to the luminal interfaces of the echo image of the proximal and the distal vessel wall. The echo-tracker measured the distance between the vessel walls at a frequency of 850 Hz in a direction perpendicular to the longitudinal axis of the vessel. Although tilting the transducer away from the longitudinal axis will falsely increase the measured diameter and its changes, in practice this error is minim&d as, owing to the continuity of vessel visualisation, the transducer can be kept parallel to the course of the aorta throughout the recording. Two pairs of dual echo-tracking loops allowed simultaneous measurements of the aortic diameter at two different levels: (a) just above the diaphragm, and (b) at a preset distance down the abdominal aorta. Thus, the propagation velocity of the pulse could be studied together with the waveform of the diameter pulse curve. Whenever possible, each measurement included three independent recordings, each of 10 consecutive pulse cycles. Only recordings obtained during periods
Aortic pulses and blood velocity 0 H.
without fetal gross movement or fetal breathing movement were accepted for analysis. The analog output signals from the tracking instrument were stored on magnetic tape for subsequent playback on a chart recorder with a paper speed of 200 mm/s. A digitiser (HIPAD, Houston Instruments, Houston, TX), feeding a microcomputer, was used for off-line analog-to-digital conversion of the pulse curves. Mean values of pulse-wave characteristics from each measurement were calculated. In Fig. 1, a typical pulse wave recorded in the fetal descending aorta is shown schematically. The following characteristics of the pulse curve were measured and evaluated: 1) (a) End-diastolic diameter (Ddiast), i.e., the minimum diameter at the foot of the pulse wave, expressed in millimeters; (b) relative pulse amplitude (AD,,,), i.e., the pulse amplitude expressed in percent of the end-diastolic diameter. 2) (a) Maximum incremental velocity (MIV), i.e., the maximum slope of the initial steep rising part of the pulse wave, expressed in millimeters per second; (b) late decremental velocity (LDV), i.e., the slope of the late descending part of the pulse wave, expressed in millimeters per second. 3) Pulse wave velocity (PWV), expressed in meters per second, and calculated by dividing the electronically given distance between the two sites of measurement, by the time lag. 4) (a) Pulse duration, measured between the onset of two consecutive pulses, expressed in milliseconds; (b) relative crest time ( Tcrestre,),i.e., the time from the onset of the pulse-wave to its peak, expressed in percent of the pulse duration. Nine values of PWV in the SGA group and three in the control group were missing. Recordings of fetal bloodjlow velocity had been performed in 56 fetuses in the SGA group. To measure the blood flow velocity, a combination of a ~-MHZ pulsed Doppler ultrasound instrument (Alfred@, Vingmed, Hot-ten, Norway) and a ~-MHZ real-time linear array scanner (ADR, Tempe, AZ) was used, as described previously by Eik-Nes et al. (I 984). The Doppler transducer was attached to the real-time transducer at a fixed angle of 45”, thereby enabling correction of the measured blood flow velocity for the angle. The fetal aorta was visualised in longitudinal section with the real-time transducer kept parallel to the vessel, the sample volume positioned to cover the lumen of the vessel, the angle of insonation being 45”. The umbilical cord was localised in the amniotic fluid, and the sample volume positioned appropriately. The Doppler shift signals were analysed on-line
STALE ef al.
413
Fig. 1. The fetal aortic diameter pulse wave is the correct and accepted expression (see also Fig. 2). Ddias,= enddiastolit diameter; AD = pulse amplitude; MIV = maximum incremental velocity; LDV = late decremental velocity; T,,,, = crest time. by a spectrum analyser (Daisy@, Vingmed, Hot-ten, Norway). The maximum blood flow velocity in the aorta in the umbilical artery, and the mean blood flow velocity in the aorta were estimated automatically online. The pulsatility index (PI), as defined by Gosling et al. (I 97 l), was used to character& the velocity waveform in the aorta in the umbilical artery. In addition, the velocity waveform in the aorta and in the umbilical artery was described by blood flow classes (BFC) as defined by Lam-in et al. (1987a): BFC 0 (normal PI), BFC I (PI > mean + 2 SO), BFC II (absence of end-diastolic flow) and BFC III (complete absence of diastolic flow or presence of reversed flow). Blood flow velocity measurements performed more than one week before or after the corresponding diameter pulse wave measurement were excluded, at which information of aortic blood flow classes remained in 50 SGA fetuses, and of umbilical artery blood flow classes in 45 SGA fetuses. Values of aortic mean blood flow velocity, of aortic PI, and of umbilical artery PI were available for analysis in 48, 49 and 45 SGA fetuses, respectively. All the women in both groups gave their informed consent before any measurements were made. The use of noninvasive ultrasonic recordings of pulsatile diameter changes and Doppler measurements of blood flow velocity in the fetal aorta has been approved by the Ethics Committee, University of Lund. Statistics The paired t test was used to compare pulse wave characteristics in the SGA group with those in the control group. Linear correlation analysis was used to test for correlation between pulse wave characteristics (AD,, , MIV and LDV) and blood flow velocity variables (mean velocity and PI). Fisher’s exact probability test was used to evaluate fetal outcome and the distribution of pathological blood flow velocity wave-
474
Ultrasound in Medicine and Biology
forms in the SGA group, when the AD,, value was either above or below that of the corresponding matched control. To illustrate the relationship between the pulsewave characteristics in the SGA group and those in the control group, the results were aggregated for twoweek periods and presented as means + 1 SD. Fig. 2 shows the number of fetuses for which results were obtained in each such period.
Volume 17, Number 5, 1991
b-
I f “,
d p
5.
Pp 1
4 .o 3. 0 2. I.
RESULTS
0
.,.,.,.,.,.,.,.,.,
24
Mean (+ 1 SD) fetal heart rate (FHR) calculated from the mean pulse duration was 138 & 10 beats/ min in the SGA group, and 140 + 8 beats/min in the control group. The difference was not significant. There was no significant difference in PWV between the two groups. The mean PWV was 2.34 + 0.6 m/s in the SGA group and 2.46 + 0.6 m/s in the control group. Ddhst was significantly lower (P < 0.02) in the SGA group than in the control group (Fig. 3). However, the weight-related Ddiastof the SGA group exceeded that of the control group. AD,,, was significantly lower (P < 0.0001) in the SGA group than in the control group (Fig. 4). MIV (Fig. 5) and LDV (Fig. 6) were also significantly lower in the SGA group than in the control group (P < 0.000 1). Tcrestre,was significantly shorter in the SGA group than in the control group (P < 0.000 1). In the SGA group, the average aortic mean blood flow velocity was 25.2 + 5.6 cm/s, the mean aortic PI 2.53 4 0.58 and the mean umbilical artery PI 1.65 +- 0.69. Blood flow velocity waveforms in the descendn
26
26
30
32
34
36
38
40
42
gestationalweeks Fig. 3. Aggregated mean end-diastolic diameter for twoweek periods. Open circles: AGA; filled circles: SGA; vertical bars: 1 SD.
ing aorta were pathological in 26 of the 50 fetuses (5 1%) and, of the 26, 20 were BFC II, and six were BFC III. Increased PI (> mean + 2 SD) in the umbilical artery was seen in 30 of the 45 fetuses for which this datum was available (66.7%) and, of the 30, 13 were BFC II and four BFC III. A significant positive correlation (Y = 0.47; P < 0.001) was found between AD,, and aortic mean blood flow velocity (Fig. 7). No other significant correlations were found between AD,, , MIV or LDV, and aortic mean blood flow velocity, aortic PI or umbilical artery PI. Nineteen SGA fetuses differed from the majority in that they had a higher AD,, than their matched controls in six cases also associated with a higher MIV and/or a higher LDV. Three SGA fetuses had a higher MIV only, or a higher MIV in combination with a higher LDV. Table 1 shows the distribution of ODFD, of fetal distress at birth, and of pathological
3530. 25.
5 1
25
27
29
31
33
35
37
39
41
gestational weeks Fig. 2. Number of matched pairs in two-week periods. Open bar: diameter pulse wave measurement; cross-hatching bar: blood flow velocity measurement (SGA only).
04 ., 24
., 26
., 26
., 30
, 32
34
.,
., 36
., 36
. 40
‘
42
gestationalweeks Fig. 4. Aggregated mean relative pulse amplitude for twoweek periods. Open circles: AGA; filled circles; SGA; vertical bars: 1 SD.
Aortic pulses and blood velocity 0 H.
475
STALE et al.
Mean velocity
cm/s 45. .
40.
.
.
25.
35.
.
l
.
30.
. .
,"r..' 0. . l .P .F9 . l * .
. mm*08
25. 20. 15.
s
IO. 5.
04
.,
24
.,
., 28
26
, 30
, 32
*
., 34
,.,.,.( 36
0
38
40
42
gestationalweeks
.,
., 0
5
.,
., IO
15
.
20
AL& AGA
18/41 6119
5141 3119
0 0
5128 8/17
18/33 8117
12128 5117
Significance of difference
NS
NS
NS
NS
NS
NS = nonsignificant; Ao: fetal descending aorta; Umb. a.: umbilical artery.
hypoxia, the hypertensive response being related to increased peripheral resistance (Cohn et al. 1974). In several studies on human blood flow, peripheral resistance has been found to be increased in pregnancies complicated by IUGR (Griffin et al. 1984; Lingman et al. 1986). Increase of peripheral resistance in the placental vascular bed, in the vascular bed of the lower extremities and in the visceral vascular bed have all been suggested to promote the redistribution of blood to the fetal heart and to the cerebral vascular system; i.e., the so-called brain-sparing effect (Wladimiroff et al. 1986). A postulated increase in peripheral resistance distal to the site of the diameter pulse wave measurements might have been the cause of the increase of the diastolic blood pressure suggested among the present SGA fetuses. The net effect of the subsequent, increase in Ddiast would probably be a reduced vessel compliance, as reflected in a loss of AD,, during systole (Imura et al. 1986). This interpretation derives support from the significantly lower A& found in the present SGA group, as compared with the control group. In the present study, the mean aortic PI of the SGA fetuses exceeded the mean aortic PI found in normal pregnancies in the third trimester by Lingman and Mar%1 (1986b). As both aortic PI and umbilical artery PI are considered to increase with increasing peripheral resistance; e.g., in pregnancies complicated by IUGR (Laurin et al. 1987a; Gudmundsson and Ma&Q 1988) we now tested for correlation between A&, and PI in both vessels, but found no such correlations. Ruissen (1990) found a very weak and negative correlation between AD, and umbilical artery PI, a larger aortic systolic distension accompanying a lower PI in AGA fetuses between 23 and 33 gestational weeks. On the other hand, a significantly positive correlation was found between A&, and aortic mean blood flow velocity in the present study (Fig. 7) although, as the significance of this correlation is based on a small number of values, it must be inter-
preted with caution. Nevertheless, the correlation is noteworthy in view of the significantly lower aortic mean blood flow velocity among SGA fetuses previously reported by Laurin et al. (1987a), and with the findings of significant negative correlations between aortic mean blood flow velocity and fetal outcome (Hackett et al. 1987). As no significant difference in mean FHR was found in the present study between the SGA and the control groups, a decrease in A&r might theoretically explain both the observed decrease in MIV and LDV. was, in most cases, significantly However, as Tcrestre, shorter in the SGA than in the controls, this is unlikely-though a decrease in A&, and a shorter Tcresue,, might per se, have been responsible for the decrease in LDV. In our earlier studies in the nonpregnant cat (Gustafsson et al. 1989), quite high positive correlations were found both between MIV and aortic flow acceleration (a measure of cardiac inotropy), and between MIV and cardiac output (CO)/stroke volume (SV); a relatively high negative correlation was found between MIV and total peripheral resistance. Accordingly, the significantly lower MIV in the present SGA group might reflect reduced contractility, decreased absolute CO/SV, increased peripheral resistance, or any combination of these alternatives, but was probably also an effect of the reduced compliance mentioned above. The higher values of the rising slope (RS) of the blood flow velocity waveform found in the nonhypoxaemic, nonacidaemic SGA fetus, on the other hand, suggest an increase in cardiac contractility (Laurin et al. 1987a). In the range of physiological central venous pressures, CO can be expressed mathematically as the ratio between mean arterial pressure and total peripheral resistance. Accordingly, the lower MIV in the present SGA group might reflect both an increased peripheral resistance and a decreased absolute CO, and due to the unchanged FHR a decreased SV, in the SGA group as compared with the controls. The interpretation of a reduced absolute SV also de-
477
Aortic pulses and blood velocity 0 H. STALE et al.
rives support from the lower AD,, , which is a finding in accord with the quite high positive correlation between AD and SV found in the earlier cat study (Gustafsson et al. 1989). Using combined real-time and M-mode techniques, Vosters et al. (1983) demonstrated that the SGA fetus, as compared with the AGA fetus, has a reduced absolute left ventricular SV, but not in relation to fetal weight, especially late in pregnancy. However, the present interpretation of a decrease in the absolute SV, would imply any increase in blood pressure to be relatively smaller than the increase in peripheral resistance, which seems reasonable in view of the exceptional viscoelastic properties of the fetal aorta (Sindberg Eriksen 1984). No significant correlations were found between MIV and aortic mean blood flow velocity, aortic PI and umbilical artery PI, though the expected tendencies could be discerned. A weak correlation was found between MIV and umbilical artery PI (r = -0.28), showing an increasing PI to be accompanied by a decreasing MIV. No correlations were found between LDV and mean blood flow velocity, aortic PI and umbilical artery PI. This is in accord with our earlier findings in the nonpregnant cat (Gustafsson et al. 1989), where, except for a relatively high positive correlation between CO and LDV, there were no other unequivocal relationships between LDV and invasively recorded central haemodynamic variables. The propagation velocity of the pressure pulse in arterial vessels is dependent on blood pressure, increasing with increasing blood pressure and increasing stiffness of the vessel wall. Contrary to the earlier published data by Gennser et al. ( 1984), no significant difference in PWV between the SGA group and the control group was found in the present study; however, the calculation of PWV is susceptible to error. Thus, owing to the very short distance between the two sites of measurement, manual analog-to-digital conversion performed via the digitiser requires a very high degree of accuracy. Divergent results do not necessarily indicate biological differences but may also reflect the limitations of the method, even if the rather small spread of data found in the present study partly argues against this. Higher frequencies of both pathological blood flow velocity waveforms, ODFD, and fetal distress at birth are to be expected among truly IUGR fetuses. On the assumption that the majority of IUGR fetuses are to be found among SGA fetuses, where the pulse wave characteristics diverge from those of controls in the significant manner mentioned above, the SGA fetuses were divided into those where the AD,, value was above that in controls, and those where it was
below. However, no differences were found between these groups in the frequencies of either pathological velocity waveforms, ODFD or fetal distress at birth. CONCLUSION The present results are in accord with and confirm those of an earlier, but smaller study by our group (Gennser and Stale 1988). The pulsatile diameter changes of the fetal descending aorta found among the present SGA fetuses suggest increased diastolic blood pressure to be a possible pathological component of the complex fetal haemodynamic changes seen in connection with intrauterine growth retardation. Irrespective of pathophysiological mechanisms, AD,, would seem to be related to changes in the development of fetal weight and might be useful in the study of IUGR. To some extent, the diameter pulse waveform probably also reflects changes in cardiac performance, characterised in the SGA fetus by a decrease of the absolute SV as compared with the AGA fetus. However, the diameter pulse waveform does not apparently yield any direct information as to peripheral resistance, for which purpose waveform analysis of Doppler ultrasound blood flow velocity recordings would seem to be the only available method. Acknowledgements-Funding for this project was provided by the General Maternity Hospital Foundation, Swedish Medical Research Council (grant no. 05980) and the Faculty of Medicine, University of Lund. Expert assistance in collecting data was provided by Mr. Anders Sjbstrom.
REFERENCES Barker, D. J. P.; Osmond, C.; Golding, J.; Kuh, D.; Wadsworth, M. E. J. Growth in utero, blood pressure in childhood and adult life, and mortality from cardiovascular disease. Br. Med. J. 298564-567; 1989. Barker, D. J. P.; Bull, A. R.; Osmond, 0.; Simmonds, S. J. Fetal and placental size and risk of hypertension in adult life. Br. Med. J. 301:259-262; 1990. Bel, F. van; Bor, M. van de; Stijnen, T.; Ruys, J. H. Decreased cerebrovascular resistance in small for gestational age infants. Eur. J. Obstet. Gynecol. Reprod. Biol. 23:137-144; 1986. Cartier, M. S.; Davidoff, A.; Wameke, L. A.; Hirsh, M. P.; Bannon, S.; St. John Sutton, M.; Doubilet, P. The normal diameter ofthe fetal aorta and pulmonary artery: echocardiographic evaluation in utero. Am. J. Radiol. 149:1003-1007; 1987. Cartier, M. S.; Doubilet, P. Fetal aortic and pulmonary artery diameters: sonographic measurements in growth retarded fetuses. Am. J. Radiol. 151:991-993; 1988. Clarkson, P. M.; Brandt, P. W. T. Aortic diameters in infants and young children: normative angiographic data. Pediatr. Cardiol. 6:3-6; 1985. Cohn, H. E.; Sacks, E. J.; Heymann, M. A.; Rudolph, A. M. Cardiovascular responses to hypoxia and acidemia in fetal lambs. Am. J. Obstet. Gynecol. 120:8 17-824; 1974. Creasy, R. K.; De Swiet, M.; Kahanpti, K. V.; Young, W. P.; Rudolph, A. M. Pathophysiological changes in the foetal lambwith growth retardation. In: Comline, R. S.; Cross, K. W.; Dawes, G. S.; Nathanielz, P. W., eds. Foetal and neonatal physiology. Sir Joseph Barcroft Centenary Symposium. Cambridge: University Press; 1973:398-402.
478
Ultrasound in Medicine and Biology
Eik-Nes, S. H.; Ma&, K.; Kristoffersen, K. Methodology and basic problems related to blood flow studies in the human fetus. Ultrasound Med. Biol. l&329-337; 1984. Gennser, G.; Vetter, K.; Sindberg Eriksen, P.; Huch, R.; Huch, A. Ultrasonic evidence for altered pulsatile dynamics in the aorta of the growth retarded fetus. In: Proceedings of the 4th World Congress International Society Study of Hypertension in Pregnancy. Amsterdam: 1984: 173 (abstr). Gennser, G.; Rymark, P.; Isberg, P. E. Low birth weight and risk for high blood pressure in adulthood. Br. Med. J. 296: 1498-1500; 1988. Gennser, G.; Stale, H. Aortic pulse waves in growth retarded human fetuses. In: Jones, C. T., ed. Fetal and neonatal development. New York: Perinatology Press; 1988543-547. Gosling, R. G.; Dunbar, G.; King, D. H.; Newman, D. L.; Side, C. D.: Woodcock. J. P.: Fitzaerald. D. E.: Keates. J. S.: MacMillan D. The quantitative analysis occlusiveperipheral arterial disease by a non-intrusive ultrasonic technique. Angiology 2252-55; 197 I. Griffin, D.; Bilardo, K.; Masini, L.; Diazrecasens, J.; Pearce, J. M.; Willson, K.: Campbell, S. Doppler waveforms in the descending thoracic aorta of the human fetus. Br. J. Obstet. Gynaecol. 91:997-1006; 1984. Grunewald, P. Chronic fetal distress and placental insufficiency. Biol. Neonat. 5:2 15-265; 1963. Gudmundsson, S.; Matil, K. Umbilical and uteroplacental flow velocity waveforms in pregnancies with fetal growth retardation. Eur. J. Obstet. Gynecol. Reprod. Biol. 27: 187- 196; 1988. Gustafsson, D.; Stale, H.; Bjorkman, J-A.; Gennser, G. Derivation of haemodynamic information from ultrasonic recordings of aortic diameter changes. Ultrasound Med. Biol. 15: 189- 199; 1989. Hackett, G. A.; Campbell, S.; Gamsu, H.; Cohen-Overbeek, T.; Pierce, J. M. F. Doppler studies in the growth retarded fetus and prediction of neonatal necrotising enterocolitis, haemorrhage, and neonatal morbidity. Br. Med. J. 294: 13- 16; 1987. Imura, T.; Yamamoto, K.; Kanamori, K.; Mikami, T.; Yasuda, H. Non-invasive ultrasonic measurement of the elastic properties of the human abdominal aorta. Cardiovasc. Res. 20:208-2 14; 1986.
of
Volume 17, Number 5, 1991 Laurin, J.; Lingman, G.; Ma&l, K.; Persson, P-H. Fetal blood flow in pregnancies complicated by intrauterine growth retardation. Obstet. Gynecol. 69:895-902; 1987a. Laurin, J.; Persson, P-H.; Polberger, S. Perinatal outcome in growth retarded pregnancies dated by ultrasound. Acta Obstet. Gyneco]. Stand. 66:337-343; 1987b. Lindstrom, K.; Gennser, G.; Sindberg Eriksen, P.; Benthin, M.; Dahl, P. An improved echo-tracker for studies on pulse waves in the fetal aorta. In: Rolfe P., ed. Fetal and neonatal physiological measurements. London: Butterworths; 1987:2 17-226. Lingman, G.; Ma&, K. Fetal central blood circulation in the third trimester of normal pregnancy. A longitudinal study: I. Aortic and umbilical blood flow. Early Hum. Dev. 13: 137- 150; 1986a. Lingman, G.; Ma&l, K. Fetal central blood circulation in the third trimester of normal pregnancy. A longitudinal study. II. Aortic blood velocity waveform Early Hum. Dev. 13: 15 I- 159; 1986b. Lingman, G.; Laurin, J.; Ma&l, K. Circulatory changes in fetuses with imminent asphyxia. Biol. Neonate 45:66; 1986. Persson, P-H.; Weldner, B-M. Intra-uterine weight curves obtained by ultrasound. Acta Obstet. Gynecol. Stand. 65: 169- 173; 1986. Ruissen, C. J. Fetoplacental circulation (measurements techniques and applications). Thesis. Limburg University, Maastricht, The Netherlands; 1990. Schauseil-Zipf, U.; Hamm, W.; Stenzel, B.; Bolte, A.; Gladtke, E. Severe intrauterine growth retardation: obstetrical management and follow-up studies in children born between 1970 and 1985. Eur. J. Obstet. Gynecol. Reprod. Biol. 30:1-9; 1989. Sindberg Eriksen, P. Fetal dynamics and maternal smoking. Thesis. University of Lund, Malmo, Sweden; 1984. Stale, H.; Gennser, G. Aortic diameter pulse waves during fetal development. J. Fet. Mat. Invest. 1:41-45; 1991. Wiggers, C. J.; Wegria, R. Active changes in size and distensibility of the aorta during acute hypertension. Am. J. Physiol. 124:603-611; 1938. Vosters, R. P.; Wladimiroff, J. W.; Stewart, P. A.; Tonge, H. M. Cardiac ventricular geometry and function in the growth-retarded fetus. Early Hum. Dev. 8:209-2 15; 1983. Wladimiroff, J. W.; Tonge, H. M.; Stewart, P. A.; Reuss, A. Severe intrauterine growth retardation; assessment of its origin from fetal arterial flow velocitv waveforms. Eur. J. Obstet. Gvnecol. Reprod. Biol. 22:23-28; -1986.
0301-5629/91 $3.00 + .OO 0 1991 Pergamon Press PIG
1991
*Original Contribution AORTIC DIAMETER PULSE WAVES AND BLOOD FLOW VELOCITY IN THE SMALL, FOR GESTATIONAL AGE, FETUS
+ Department of Obstetrics and Gynaecology, Lund University, Malmij General Hospital, S-2 14 0 1 Malm6, Sweden, and *Department of Electrical Measurements, Lund Institute of Technology, Lund, Sweden Abstract-An ultrasound phase-locked, echo-tracking system was used for noninvasive measurements of pulsatile diameter changes in the descending aorta of 60 small, for gestational age @GA), fetuses and of 60 fetuses appropriate for gestational age (AGA). Pulsed Doppler ultrasound was used for the recording of blood flow velocity in the aorta and in the umbilical artery of the SGA fetuses. In the SGA fetuses, a weight-related higher end-diastolic diameter and a lower relative pulse amplitude suggest that diastolic blood pressure was increased, a less steep rise of the initial ascending part of the pulse wave and a lower relative pulse amplitude suggest that the absolute stroke volume was decreased. Except for a positive correlation between relative pulse amplitude and mean velocity in the aorta, no correlation was found between diameter pulse waves and blood flow velocity. Aortic diameter pulse waves apparently yield no unequivocal information as to peripheral resistance, for which purpose blood flow velocity waveform analysis would seem, at least at present, to be the only available method. Key Words: Ultrasound, Ultrasonic echo-tracking, Doppler ultrasound, Blood flow velocity, Fetal aorta, Aortic diameter, Blood pressure, Compliance, Intrauterine growth retardation.
INTRODUCTION
increase in PI indicating increased resistance in the placental vascular bed and other parts of the vascular bed supplied by the descending aorta (Lam-in et al. 1987a). Moreover, in a pilot study it was shown that characteristics of the aortic diameter pulse wave differ in the growth-retarded fetus from those in the normal fetus (Gennser and Stale 1988). These findings indicate that profound circulatory changes occur in growth-retarded fetuses, and therefore it is of interest to study the aortic haemodynamics in this high-risk group in order to follow the pathophysiological processes. Possibly, the information on fetal aortic pulse wave and blood velocity might be used for clinical assessment of a growth-retarded fetus. However, low birth weight in relation to gestational age does not necessarily imply IUGR, as it might equally be due to genetic factors. Here, therefore, the more proper term, small for gestational age (SGA), has been used to designate age-related low fetal weight.
Intrauterine growth retardation (IUGR)is a major cause of perinatal mortality and morbidity (Laurin et al. 1987b; Schauseil-Zipf et al. 1989). Accurate diagnosis and proper management continue to pose a challenge even to the experienced obstetrician. According to Grunewald (1963), chronic fetal distress due to placental insufficiency is the single most important cause of IUGR, although causes other than fetal malhutrition must be sought. During intrauterine asphyxia, e.g., in the severely growth-retarded fetus, the fetal blood flow is redistributed to ensure preferential blood supply to vital organs such as the brain, myocardium and adrenals (Creasy et al. 1973). In the human fetus, the brainsparing effect is reflected in a low PI in the cerebral vessels, and a concomitant increase of PI in the fetal aorta and umbilical artery (Wladimiroff et al. 1986). The blood velocity waveform recorded from the descending aorta of the growth-retarded fetus shows a reduction in diastolic velocity and a corresponding
METHODS In a prospective study, pulsatile diameter changes were measured on the descending aorta of 60 SGA fetuses. SGA was defined as an ultrasound-esti-
Address all correspondence to HHkan Stale. 471
472
Ultrasound in Medicine and Biology
mated fetal weight below the normal age-related mean - 2 SD, the mean weight being obtained from ultrasound weight-curves for a reference population of the city of Malmii (Persson and Weldner 1986). The gestational age of all fetuses had been ascertained by early ultrasound fetometry and, at diameter pulse wave measurement, ranged from 25 completed weeks + 6 days to 41 completed weeks + 6 days. One measurement from each of 60 fetuses, truly SGA at birth, were included in the analysis. The results from each measurement were compared with those pulsatile diameter changes measured on the descending aorta in 60 fetuses in uncomplicated pregnancies, matched for gestational age and with birth weights appropriate for gestational age (AGA). The median age of the women in the SGA group was 26 years (range 17-45) and, in the control group, 29 years (range 20-39). In the SGA group, 38 women were nulliparae, 17 primiparae and 5 multiparae, while the control group comprised 27 nulliparae, 23 primiparae and 10 multiparae. There were four twin pregnancies in the SGA group, but only one twin in each pregnancy was SGA, and thus included in the study. The control group comprised singleton pregnancies only, There were 20 smokers among mothers in the SGA group and 28 in the control group. In the SGA group, 25 women had symptoms of preeclampsia (blood pressure 2 140/90 mm Hg and proteinuria); one woman with diabetes mellitus also had hypertension, and isolated hypertension was observed in another seven women. One woman in the control group had mild symptoms of preeclampsia. Median gestational age at delivery was 37 completed weeks + 4 days (range 28 weeks + 2 days to 42 weeks + 1 day) in the SGA group, and 40 completed weeks (range 35 weeks + 5 days to 42 weeks + 3 days) in the control group. Median birth weight among the SGA fetuses was 2.06 kg (range 545-2.99 kg), the median deviation from the expected weight at delivery being -35% (range -22% to -59%). All fetuses in the control group had birth weights appropriate for gestational age, the median being 3.64 kg (range 2.50- 4.4 1 kg). There were 25 male infants and 35 female infants in the SGA group, as compared with 26 and 34, respectively, among the controls. In the SGA group, 20 infants were delivered vaginally (two by vacuum extraction), and the remaining 40 by Caesarean section. In the control group, 55 infants were delivered vaginally (four by vacuum extraction), and the remaining five by Caesarean section. There were 26 cases of ODFD (operative delivery for fetal distress) and/or fetal distress at birth in the SGA group, and seven in the control group. ODFD comprised Caesarean section or vaginal instrumental delivery because of cardiotoco-
Volume 17, Number 5, 1991
graphic patterns consistent with imminent asphyxia, and fetal distress at birth was defined as an Apgar score < 7 at 1 or 5 min, or pH I 7.10 in the umbilical artery or pH I 7.20 in the umbilical vein. There were three perinatal deaths in the SGA group; one antepartal death of unknown cause and two postpartal deaths, one in connection with streptococcal septicaemia, the other as a consequence of multiple malformation. No other major malformations were seen either in the SGA group or in the control group. For noninvasive monitoring of pulsatile diameter changes in the fetal descending aorta, an electronic echo-tracking instrument (Diamove@, Teltec, Lund, Sweden) was used, interfaced with a real-time ultrasound scanner (ADR, Tempe, AZ) fitted with a 3.0MHz linear array transducer (Lindstriim et al. 1987). For echo-tracking, phase-locked loop circuits were used, which restore the position of an electronic gate relative to the moving echo. The discrete compensatory steps of the gate are measured, yielding an output signal representing the echo movement per unit time. The instrument is equipped with dual echo-tracking loops, which enable two separate echoes from opposite vessel walls to be tracked simultaneously, the differential signals between them representing any instantaneous change in vessel diameter. In the system used, the smallest detectable movement is 14 pm, which far exceeds the resolution capacity of the realtime scanner for static objects. The fetal descending aorta was visualised in longitudinal section. Generally, the fetus was insonated at an oblique antero-posterior angle. Two horizontal electronic markers, each representing one tracking gate, were aligned with and locked to the luminal interfaces of the echo image of the proximal and the distal vessel wall. The echo-tracker measured the distance between the vessel walls at a frequency of 850 Hz in a direction perpendicular to the longitudinal axis of the vessel. Although tilting the transducer away from the longitudinal axis will falsely increase the measured diameter and its changes, in practice this error is minim&d as, owing to the continuity of vessel visualisation, the transducer can be kept parallel to the course of the aorta throughout the recording. Two pairs of dual echo-tracking loops allowed simultaneous measurements of the aortic diameter at two different levels: (a) just above the diaphragm, and (b) at a preset distance down the abdominal aorta. Thus, the propagation velocity of the pulse could be studied together with the waveform of the diameter pulse curve. Whenever possible, each measurement included three independent recordings, each of 10 consecutive pulse cycles. Only recordings obtained during periods
Aortic pulses and blood velocity 0 H.
without fetal gross movement or fetal breathing movement were accepted for analysis. The analog output signals from the tracking instrument were stored on magnetic tape for subsequent playback on a chart recorder with a paper speed of 200 mm/s. A digitiser (HIPAD, Houston Instruments, Houston, TX), feeding a microcomputer, was used for off-line analog-to-digital conversion of the pulse curves. Mean values of pulse-wave characteristics from each measurement were calculated. In Fig. 1, a typical pulse wave recorded in the fetal descending aorta is shown schematically. The following characteristics of the pulse curve were measured and evaluated: 1) (a) End-diastolic diameter (Ddiast), i.e., the minimum diameter at the foot of the pulse wave, expressed in millimeters; (b) relative pulse amplitude (AD,,,), i.e., the pulse amplitude expressed in percent of the end-diastolic diameter. 2) (a) Maximum incremental velocity (MIV), i.e., the maximum slope of the initial steep rising part of the pulse wave, expressed in millimeters per second; (b) late decremental velocity (LDV), i.e., the slope of the late descending part of the pulse wave, expressed in millimeters per second. 3) Pulse wave velocity (PWV), expressed in meters per second, and calculated by dividing the electronically given distance between the two sites of measurement, by the time lag. 4) (a) Pulse duration, measured between the onset of two consecutive pulses, expressed in milliseconds; (b) relative crest time ( Tcrestre,),i.e., the time from the onset of the pulse-wave to its peak, expressed in percent of the pulse duration. Nine values of PWV in the SGA group and three in the control group were missing. Recordings of fetal bloodjlow velocity had been performed in 56 fetuses in the SGA group. To measure the blood flow velocity, a combination of a ~-MHZ pulsed Doppler ultrasound instrument (Alfred@, Vingmed, Hot-ten, Norway) and a ~-MHZ real-time linear array scanner (ADR, Tempe, AZ) was used, as described previously by Eik-Nes et al. (I 984). The Doppler transducer was attached to the real-time transducer at a fixed angle of 45”, thereby enabling correction of the measured blood flow velocity for the angle. The fetal aorta was visualised in longitudinal section with the real-time transducer kept parallel to the vessel, the sample volume positioned to cover the lumen of the vessel, the angle of insonation being 45”. The umbilical cord was localised in the amniotic fluid, and the sample volume positioned appropriately. The Doppler shift signals were analysed on-line
STALE ef al.
413
Fig. 1. The fetal aortic diameter pulse wave is the correct and accepted expression (see also Fig. 2). Ddias,= enddiastolit diameter; AD = pulse amplitude; MIV = maximum incremental velocity; LDV = late decremental velocity; T,,,, = crest time. by a spectrum analyser (Daisy@, Vingmed, Hot-ten, Norway). The maximum blood flow velocity in the aorta in the umbilical artery, and the mean blood flow velocity in the aorta were estimated automatically online. The pulsatility index (PI), as defined by Gosling et al. (I 97 l), was used to character& the velocity waveform in the aorta in the umbilical artery. In addition, the velocity waveform in the aorta and in the umbilical artery was described by blood flow classes (BFC) as defined by Lam-in et al. (1987a): BFC 0 (normal PI), BFC I (PI > mean + 2 SO), BFC II (absence of end-diastolic flow) and BFC III (complete absence of diastolic flow or presence of reversed flow). Blood flow velocity measurements performed more than one week before or after the corresponding diameter pulse wave measurement were excluded, at which information of aortic blood flow classes remained in 50 SGA fetuses, and of umbilical artery blood flow classes in 45 SGA fetuses. Values of aortic mean blood flow velocity, of aortic PI, and of umbilical artery PI were available for analysis in 48, 49 and 45 SGA fetuses, respectively. All the women in both groups gave their informed consent before any measurements were made. The use of noninvasive ultrasonic recordings of pulsatile diameter changes and Doppler measurements of blood flow velocity in the fetal aorta has been approved by the Ethics Committee, University of Lund. Statistics The paired t test was used to compare pulse wave characteristics in the SGA group with those in the control group. Linear correlation analysis was used to test for correlation between pulse wave characteristics (AD,, , MIV and LDV) and blood flow velocity variables (mean velocity and PI). Fisher’s exact probability test was used to evaluate fetal outcome and the distribution of pathological blood flow velocity wave-
474
Ultrasound in Medicine and Biology
forms in the SGA group, when the AD,, value was either above or below that of the corresponding matched control. To illustrate the relationship between the pulsewave characteristics in the SGA group and those in the control group, the results were aggregated for twoweek periods and presented as means + 1 SD. Fig. 2 shows the number of fetuses for which results were obtained in each such period.
Volume 17, Number 5, 1991
b-
I f “,
d p
5.
Pp 1
4 .o 3. 0 2. I.
RESULTS
0
.,.,.,.,.,.,.,.,.,
24
Mean (+ 1 SD) fetal heart rate (FHR) calculated from the mean pulse duration was 138 & 10 beats/ min in the SGA group, and 140 + 8 beats/min in the control group. The difference was not significant. There was no significant difference in PWV between the two groups. The mean PWV was 2.34 + 0.6 m/s in the SGA group and 2.46 + 0.6 m/s in the control group. Ddhst was significantly lower (P < 0.02) in the SGA group than in the control group (Fig. 3). However, the weight-related Ddiastof the SGA group exceeded that of the control group. AD,,, was significantly lower (P < 0.0001) in the SGA group than in the control group (Fig. 4). MIV (Fig. 5) and LDV (Fig. 6) were also significantly lower in the SGA group than in the control group (P < 0.000 1). Tcrestre,was significantly shorter in the SGA group than in the control group (P < 0.000 1). In the SGA group, the average aortic mean blood flow velocity was 25.2 + 5.6 cm/s, the mean aortic PI 2.53 4 0.58 and the mean umbilical artery PI 1.65 +- 0.69. Blood flow velocity waveforms in the descendn
26
26
30
32
34
36
38
40
42
gestationalweeks Fig. 3. Aggregated mean end-diastolic diameter for twoweek periods. Open circles: AGA; filled circles: SGA; vertical bars: 1 SD.
ing aorta were pathological in 26 of the 50 fetuses (5 1%) and, of the 26, 20 were BFC II, and six were BFC III. Increased PI (> mean + 2 SD) in the umbilical artery was seen in 30 of the 45 fetuses for which this datum was available (66.7%) and, of the 30, 13 were BFC II and four BFC III. A significant positive correlation (Y = 0.47; P < 0.001) was found between AD,, and aortic mean blood flow velocity (Fig. 7). No other significant correlations were found between AD,, , MIV or LDV, and aortic mean blood flow velocity, aortic PI or umbilical artery PI. Nineteen SGA fetuses differed from the majority in that they had a higher AD,, than their matched controls in six cases also associated with a higher MIV and/or a higher LDV. Three SGA fetuses had a higher MIV only, or a higher MIV in combination with a higher LDV. Table 1 shows the distribution of ODFD, of fetal distress at birth, and of pathological
3530. 25.
5 1
25
27
29
31
33
35
37
39
41
gestational weeks Fig. 2. Number of matched pairs in two-week periods. Open bar: diameter pulse wave measurement; cross-hatching bar: blood flow velocity measurement (SGA only).
04 ., 24
., 26
., 26
., 30
, 32
34
.,
., 36
., 36
. 40
‘
42
gestationalweeks Fig. 4. Aggregated mean relative pulse amplitude for twoweek periods. Open circles: AGA; filled circles; SGA; vertical bars: 1 SD.
Aortic pulses and blood velocity 0 H.
475
STALE et al.
Mean velocity
cm/s 45. .
40.
.
.
25.
35.
.
l
.
30.
. .
,"r..' 0. . l .P .F9 . l * .
. mm*08
25. 20. 15.
s
IO. 5.
04
.,
24
.,
., 28
26
, 30
, 32
*
., 34
,.,.,.( 36
0
38
40
42
gestationalweeks
.,
., 0
5
.,
., IO
15
.
20
AL& AGA
18/41 6119
5141 3119
0 0
5128 8/17
18/33 8117
12128 5117
Significance of difference
NS
NS
NS
NS
NS
NS = nonsignificant; Ao: fetal descending aorta; Umb. a.: umbilical artery.
hypoxia, the hypertensive response being related to increased peripheral resistance (Cohn et al. 1974). In several studies on human blood flow, peripheral resistance has been found to be increased in pregnancies complicated by IUGR (Griffin et al. 1984; Lingman et al. 1986). Increase of peripheral resistance in the placental vascular bed, in the vascular bed of the lower extremities and in the visceral vascular bed have all been suggested to promote the redistribution of blood to the fetal heart and to the cerebral vascular system; i.e., the so-called brain-sparing effect (Wladimiroff et al. 1986). A postulated increase in peripheral resistance distal to the site of the diameter pulse wave measurements might have been the cause of the increase of the diastolic blood pressure suggested among the present SGA fetuses. The net effect of the subsequent, increase in Ddiast would probably be a reduced vessel compliance, as reflected in a loss of AD,, during systole (Imura et al. 1986). This interpretation derives support from the significantly lower A& found in the present SGA group, as compared with the control group. In the present study, the mean aortic PI of the SGA fetuses exceeded the mean aortic PI found in normal pregnancies in the third trimester by Lingman and Mar%1 (1986b). As both aortic PI and umbilical artery PI are considered to increase with increasing peripheral resistance; e.g., in pregnancies complicated by IUGR (Laurin et al. 1987a; Gudmundsson and Ma&Q 1988) we now tested for correlation between A&, and PI in both vessels, but found no such correlations. Ruissen (1990) found a very weak and negative correlation between AD, and umbilical artery PI, a larger aortic systolic distension accompanying a lower PI in AGA fetuses between 23 and 33 gestational weeks. On the other hand, a significantly positive correlation was found between A&, and aortic mean blood flow velocity in the present study (Fig. 7) although, as the significance of this correlation is based on a small number of values, it must be inter-
preted with caution. Nevertheless, the correlation is noteworthy in view of the significantly lower aortic mean blood flow velocity among SGA fetuses previously reported by Laurin et al. (1987a), and with the findings of significant negative correlations between aortic mean blood flow velocity and fetal outcome (Hackett et al. 1987). As no significant difference in mean FHR was found in the present study between the SGA and the control groups, a decrease in A&r might theoretically explain both the observed decrease in MIV and LDV. was, in most cases, significantly However, as Tcrestre, shorter in the SGA than in the controls, this is unlikely-though a decrease in A&, and a shorter Tcresue,, might per se, have been responsible for the decrease in LDV. In our earlier studies in the nonpregnant cat (Gustafsson et al. 1989), quite high positive correlations were found both between MIV and aortic flow acceleration (a measure of cardiac inotropy), and between MIV and cardiac output (CO)/stroke volume (SV); a relatively high negative correlation was found between MIV and total peripheral resistance. Accordingly, the significantly lower MIV in the present SGA group might reflect reduced contractility, decreased absolute CO/SV, increased peripheral resistance, or any combination of these alternatives, but was probably also an effect of the reduced compliance mentioned above. The higher values of the rising slope (RS) of the blood flow velocity waveform found in the nonhypoxaemic, nonacidaemic SGA fetus, on the other hand, suggest an increase in cardiac contractility (Laurin et al. 1987a). In the range of physiological central venous pressures, CO can be expressed mathematically as the ratio between mean arterial pressure and total peripheral resistance. Accordingly, the lower MIV in the present SGA group might reflect both an increased peripheral resistance and a decreased absolute CO, and due to the unchanged FHR a decreased SV, in the SGA group as compared with the controls. The interpretation of a reduced absolute SV also de-
477
Aortic pulses and blood velocity 0 H. STALE et al.
rives support from the lower AD,, , which is a finding in accord with the quite high positive correlation between AD and SV found in the earlier cat study (Gustafsson et al. 1989). Using combined real-time and M-mode techniques, Vosters et al. (1983) demonstrated that the SGA fetus, as compared with the AGA fetus, has a reduced absolute left ventricular SV, but not in relation to fetal weight, especially late in pregnancy. However, the present interpretation of a decrease in the absolute SV, would imply any increase in blood pressure to be relatively smaller than the increase in peripheral resistance, which seems reasonable in view of the exceptional viscoelastic properties of the fetal aorta (Sindberg Eriksen 1984). No significant correlations were found between MIV and aortic mean blood flow velocity, aortic PI and umbilical artery PI, though the expected tendencies could be discerned. A weak correlation was found between MIV and umbilical artery PI (r = -0.28), showing an increasing PI to be accompanied by a decreasing MIV. No correlations were found between LDV and mean blood flow velocity, aortic PI and umbilical artery PI. This is in accord with our earlier findings in the nonpregnant cat (Gustafsson et al. 1989), where, except for a relatively high positive correlation between CO and LDV, there were no other unequivocal relationships between LDV and invasively recorded central haemodynamic variables. The propagation velocity of the pressure pulse in arterial vessels is dependent on blood pressure, increasing with increasing blood pressure and increasing stiffness of the vessel wall. Contrary to the earlier published data by Gennser et al. ( 1984), no significant difference in PWV between the SGA group and the control group was found in the present study; however, the calculation of PWV is susceptible to error. Thus, owing to the very short distance between the two sites of measurement, manual analog-to-digital conversion performed via the digitiser requires a very high degree of accuracy. Divergent results do not necessarily indicate biological differences but may also reflect the limitations of the method, even if the rather small spread of data found in the present study partly argues against this. Higher frequencies of both pathological blood flow velocity waveforms, ODFD, and fetal distress at birth are to be expected among truly IUGR fetuses. On the assumption that the majority of IUGR fetuses are to be found among SGA fetuses, where the pulse wave characteristics diverge from those of controls in the significant manner mentioned above, the SGA fetuses were divided into those where the AD,, value was above that in controls, and those where it was
below. However, no differences were found between these groups in the frequencies of either pathological velocity waveforms, ODFD or fetal distress at birth. CONCLUSION The present results are in accord with and confirm those of an earlier, but smaller study by our group (Gennser and Stale 1988). The pulsatile diameter changes of the fetal descending aorta found among the present SGA fetuses suggest increased diastolic blood pressure to be a possible pathological component of the complex fetal haemodynamic changes seen in connection with intrauterine growth retardation. Irrespective of pathophysiological mechanisms, AD,, would seem to be related to changes in the development of fetal weight and might be useful in the study of IUGR. To some extent, the diameter pulse waveform probably also reflects changes in cardiac performance, characterised in the SGA fetus by a decrease of the absolute SV as compared with the AGA fetus. However, the diameter pulse waveform does not apparently yield any direct information as to peripheral resistance, for which purpose waveform analysis of Doppler ultrasound blood flow velocity recordings would seem to be the only available method. Acknowledgements-Funding for this project was provided by the General Maternity Hospital Foundation, Swedish Medical Research Council (grant no. 05980) and the Faculty of Medicine, University of Lund. Expert assistance in collecting data was provided by Mr. Anders Sjbstrom.
REFERENCES Barker, D. J. P.; Osmond, C.; Golding, J.; Kuh, D.; Wadsworth, M. E. J. Growth in utero, blood pressure in childhood and adult life, and mortality from cardiovascular disease. Br. Med. J. 298564-567; 1989. Barker, D. J. P.; Bull, A. R.; Osmond, 0.; Simmonds, S. J. Fetal and placental size and risk of hypertension in adult life. Br. Med. J. 301:259-262; 1990. Bel, F. van; Bor, M. van de; Stijnen, T.; Ruys, J. H. Decreased cerebrovascular resistance in small for gestational age infants. Eur. J. Obstet. Gynecol. Reprod. Biol. 23:137-144; 1986. Cartier, M. S.; Davidoff, A.; Wameke, L. A.; Hirsh, M. P.; Bannon, S.; St. John Sutton, M.; Doubilet, P. The normal diameter ofthe fetal aorta and pulmonary artery: echocardiographic evaluation in utero. Am. J. Radiol. 149:1003-1007; 1987. Cartier, M. S.; Doubilet, P. Fetal aortic and pulmonary artery diameters: sonographic measurements in growth retarded fetuses. Am. J. Radiol. 151:991-993; 1988. Clarkson, P. M.; Brandt, P. W. T. Aortic diameters in infants and young children: normative angiographic data. Pediatr. Cardiol. 6:3-6; 1985. Cohn, H. E.; Sacks, E. J.; Heymann, M. A.; Rudolph, A. M. Cardiovascular responses to hypoxia and acidemia in fetal lambs. Am. J. Obstet. Gynecol. 120:8 17-824; 1974. Creasy, R. K.; De Swiet, M.; Kahanpti, K. V.; Young, W. P.; Rudolph, A. M. Pathophysiological changes in the foetal lambwith growth retardation. In: Comline, R. S.; Cross, K. W.; Dawes, G. S.; Nathanielz, P. W., eds. Foetal and neonatal physiology. Sir Joseph Barcroft Centenary Symposium. Cambridge: University Press; 1973:398-402.
478
Ultrasound in Medicine and Biology
Eik-Nes, S. H.; Ma&, K.; Kristoffersen, K. Methodology and basic problems related to blood flow studies in the human fetus. Ultrasound Med. Biol. l&329-337; 1984. Gennser, G.; Vetter, K.; Sindberg Eriksen, P.; Huch, R.; Huch, A. Ultrasonic evidence for altered pulsatile dynamics in the aorta of the growth retarded fetus. In: Proceedings of the 4th World Congress International Society Study of Hypertension in Pregnancy. Amsterdam: 1984: 173 (abstr). Gennser, G.; Rymark, P.; Isberg, P. E. Low birth weight and risk for high blood pressure in adulthood. Br. Med. J. 296: 1498-1500; 1988. Gennser, G.; Stale, H. Aortic pulse waves in growth retarded human fetuses. In: Jones, C. T., ed. Fetal and neonatal development. New York: Perinatology Press; 1988543-547. Gosling, R. G.; Dunbar, G.; King, D. H.; Newman, D. L.; Side, C. D.: Woodcock. J. P.: Fitzaerald. D. E.: Keates. J. S.: MacMillan D. The quantitative analysis occlusiveperipheral arterial disease by a non-intrusive ultrasonic technique. Angiology 2252-55; 197 I. Griffin, D.; Bilardo, K.; Masini, L.; Diazrecasens, J.; Pearce, J. M.; Willson, K.: Campbell, S. Doppler waveforms in the descending thoracic aorta of the human fetus. Br. J. Obstet. Gynaecol. 91:997-1006; 1984. Grunewald, P. Chronic fetal distress and placental insufficiency. Biol. Neonat. 5:2 15-265; 1963. Gudmundsson, S.; Matil, K. Umbilical and uteroplacental flow velocity waveforms in pregnancies with fetal growth retardation. Eur. J. Obstet. Gynecol. Reprod. Biol. 27: 187- 196; 1988. Gustafsson, D.; Stale, H.; Bjorkman, J-A.; Gennser, G. Derivation of haemodynamic information from ultrasonic recordings of aortic diameter changes. Ultrasound Med. Biol. 15: 189- 199; 1989. Hackett, G. A.; Campbell, S.; Gamsu, H.; Cohen-Overbeek, T.; Pierce, J. M. F. Doppler studies in the growth retarded fetus and prediction of neonatal necrotising enterocolitis, haemorrhage, and neonatal morbidity. Br. Med. J. 294: 13- 16; 1987. Imura, T.; Yamamoto, K.; Kanamori, K.; Mikami, T.; Yasuda, H. Non-invasive ultrasonic measurement of the elastic properties of the human abdominal aorta. Cardiovasc. Res. 20:208-2 14; 1986.
of
Volume 17, Number 5, 1991 Laurin, J.; Lingman, G.; Ma&l, K.; Persson, P-H. Fetal blood flow in pregnancies complicated by intrauterine growth retardation. Obstet. Gynecol. 69:895-902; 1987a. Laurin, J.; Persson, P-H.; Polberger, S. Perinatal outcome in growth retarded pregnancies dated by ultrasound. Acta Obstet. Gyneco]. Stand. 66:337-343; 1987b. Lindstrom, K.; Gennser, G.; Sindberg Eriksen, P.; Benthin, M.; Dahl, P. An improved echo-tracker for studies on pulse waves in the fetal aorta. In: Rolfe P., ed. Fetal and neonatal physiological measurements. London: Butterworths; 1987:2 17-226. Lingman, G.; Ma&, K. Fetal central blood circulation in the third trimester of normal pregnancy. A longitudinal study: I. Aortic and umbilical blood flow. Early Hum. Dev. 13: 137- 150; 1986a. Lingman, G.; Ma&l, K. Fetal central blood circulation in the third trimester of normal pregnancy. A longitudinal study. II. Aortic blood velocity waveform Early Hum. Dev. 13: 15 I- 159; 1986b. Lingman, G.; Laurin, J.; Ma&l, K. Circulatory changes in fetuses with imminent asphyxia. Biol. Neonate 45:66; 1986. Persson, P-H.; Weldner, B-M. Intra-uterine weight curves obtained by ultrasound. Acta Obstet. Gynecol. Stand. 65: 169- 173; 1986. Ruissen, C. J. Fetoplacental circulation (measurements techniques and applications). Thesis. Limburg University, Maastricht, The Netherlands; 1990. Schauseil-Zipf, U.; Hamm, W.; Stenzel, B.; Bolte, A.; Gladtke, E. Severe intrauterine growth retardation: obstetrical management and follow-up studies in children born between 1970 and 1985. Eur. J. Obstet. Gynecol. Reprod. Biol. 30:1-9; 1989. Sindberg Eriksen, P. Fetal dynamics and maternal smoking. Thesis. University of Lund, Malmo, Sweden; 1984. Stale, H.; Gennser, G. Aortic diameter pulse waves during fetal development. J. Fet. Mat. Invest. 1:41-45; 1991. Wiggers, C. J.; Wegria, R. Active changes in size and distensibility of the aorta during acute hypertension. Am. J. Physiol. 124:603-611; 1938. Vosters, R. P.; Wladimiroff, J. W.; Stewart, P. A.; Tonge, H. M. Cardiac ventricular geometry and function in the growth-retarded fetus. Early Hum. Dev. 8:209-2 15; 1983. Wladimiroff, J. W.; Tonge, H. M.; Stewart, P. A.; Reuss, A. Severe intrauterine growth retardation; assessment of its origin from fetal arterial flow velocitv waveforms. Eur. J. Obstet. Gvnecol. Reprod. Biol. 22:23-28; -1986.
