- Email: [email protected]
Blood velocity in the fetal vein of Galen and the outcome of high-risk pregnancy
European Journal of Obstetrics & Gynecology and Reproductive Biology 99 (2001) 14±18
Blood velocity in the fetal vein of Galen and the outcome of high-risk pregnancy Mariusz Dubiela, Grzegorz H. Breborowicza, Ricardo Laurinib, Saemundur Gudmundssonc,* a
Department of Perinatology and Gynaecology, University School of Medical Sciences, University Hospital, Poznan, Poland b Division of Developmental Pathology, University Hospital, Lausanne, Switzerland c Department of Obstetrics and Gynaecology, University Hospital MAS, Malmo, Sweden Received 7 July 2000; accepted 26 February 2001
Abstract Background: Pulsation in the ¯ow velocity waveform in the umbilical vein is related to perinatal mortality but the ¯ow velocity waveform in the fetal vein of Galen is normally even and without ¯uctuation. Objectives: To establish whether blood ¯ow velocity pulsations in the vein of Galen in high-risk pregnancies are related to outcome. Study Design: The vein of Galen was located by colour Doppler ultrasound in 102 pregnancies complicated by severe pregnancy-induced hypertension. The blood velocity waveform was recorded by pulsed Doppler within 2 days of delivery and the presence pulsations related to pregnancy outcome, including emergency operative intervention and neonatal distress. Umbilical artery and vein and uterine artery blood ¯ow velocity waveform were also recorded at the same time. The clinicians managing the women were unaware of the venous ¯ow results. Results: Pulsation were present in the vein of Galen in 68 cases and in the umbilical vein in 21. Both were signi®cantly related to adverse outcome. Pulsations in the vein of Galen were seen in all seven perinatal deaths. Conclusions: Since umbilical venous pulsation are a late sign of fetal compromise, and pulsations in the vein of Galen seem to appear earlier, thus being an intermediate sign of fetal compromise that might be of great value for fetal surveillance. # 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Vein of Galen; Doppler ultrasound; Umbilical vein; Pregnancy
1. Introduction Over the past 20 years, Doppler ultrasound technology has developed considerably and is now a routine part of clinical surveillance in high-risk pregnancies. This includes recording of the umbilical and uterine arteries to predict increased placental vascular resistance, which itself is predictive of an adverse outcome of pregnancy [1,2]. Studies on fetal lamb have shown that the fetus adapts to reduced placental perfusion and chronic hypoxia by redistributing its blood ¯ow to vital organs such as the brain, heart and adrenals [3]. Studies on fetal cerebral circulation in humans, especially in the middle cerebral artery, have shown increased diastolic blood velocity to be a characteristic sign of brain sparing the high-risk pregnancies with reduced placental perfusion [4]. *
Corresponding author. Tel.: 46-40-332095; fax: 46-40-962600. E-mail address: [email protected] (S. Gudmundsson).
Examination of the fetal venous circulation has advanced during the 1990s. Blood velocity in the systemic venous circulation is pulsatile, re¯ecting central venous pressure [5]. By contrast, blood velocity in the umbilical vein (U) and portal circulation under normal circumstances are even and without pulsations due to the ®ltering effect of the ductus venosus. The tracing of blood velocity in the systemic veins has revealed increased reversal of ¯ow during atrial contraction and pulsations in the umbilical vein in cases of congestive fetal heart failure and imminent asphyxia [6±8] due to increased central venous pressure. Umbilical venous pulsations are thus a late sign of fetal compromise. Furthermore, experience with fetal post-mortems in cases of asphyxial intra-uterine fetal death has shown that the great vein of Galen (G) is frequently dilated [9]. This dilatation in cases of intra-uterine fetal death caused by chronic hypoxemia could be due to increased cerebral blood ¯ow during hypoxia, or increased central venous pressure. In a pilot study at our institutions and as reported by Laurichesse-Delmas et al. [10], blood velocity in the great
0301-2115/01/$ ± see front matter # 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 1 - 2 1 1 5 ( 0 1 ) 0 0 3 4 0 - 2
M. Dubiel et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 99 (2001) 14±18
vein of Galen, is in the normal state even and without the characteristic pulsations observed in the systemic venous circulation. In complicated pregnancies, however, blood velocity pulsations can be seen, but its clinical implication was unknown. The purpose of this study was to determine whether blood velocity in the vein of Galen might predict adverse perinatal outcome of high-risk pregnancies. The clinical predictive value of blood velocity of vein of Galen was also evaluated in relation to umbilical venous blood ¯ow, which presently is frequently used for fetal surveillance in high-risk pregnancies. 2. Material and methods The internal cerebral veins join in front of the cerebellum in a short vein called the great vein of Galen or the great cerebral vein. The vein of Galen has a cranio-posterior direction in front of the cerebellum towards the tentorium cerebelli, joining the inferior sagittal sinus (ISS) to form the straight sinus (SS). The straight sinus is directed backwards (caudally), joining the superior sagittal sinus (SSS) to form the transverse sinuses. The anatomic relationship between the veins is illustrated in Fig. 1. In the present study, the vein of Galen was located by means of an Acuson 128 XP duplex scanner using a 3.5 and 5 MHz transducer with power and colour Doppler options. During the nineth month, the fetal vein of Galen was located in 102 pregnancies complicated by severe pregnancyinduced hypertension (blood pressure 160/110 or higher;
Fig. 1. Magnetic resonance imaging angiography of human cerebral veins. The angle between the vein of Galen (G) and straight sinus (SS) is clearly seen. The inferior sagittal sinus (ISS) is not seen in this picture, but is illustrated by a dotted line. No flow is detected at the crossing of the vein of Galen and inferior sagittal sinus. Superior sagittal sinus (SSS).
15
proteinuria 3) and referred to our ultrasound laboratories for umbilical and uterine artery blood velocimetry as a part of placental vascular resistance evaluation. Fifty-seven cases had also suspected intra-uterine growth retardation, de®ned as below ®fth percentile for population studied. Fifty-three cases were emergency admissions to a tertiary care unit and were delivered shortly after admission. Each fetus was examined once only, the Galen venous blood ¯ow was recorded in all cases. Median gestational age at examination was 36 weeks (27±41). In all cases the gestational age was based on early ultrasound dating of pregnancy. The median interval between the examination and labour was 1 day (1±2 days). The vein of Galen was located by colour Doppler ultrasound with a transducer in the occipital transverse plane. The angle between the vessel and the transducer was <458. The straight sinus could be located easily. The con¯uence of the vein of Galen and the inferior sagittal sinus showed a characteristic bright colour ¯ow due to disturbed blood ¯ow. By moving the transducer slightly downwards (caudally), the vein of Galen could be visualised below the disturbed blood ¯ow in the vein. The sample volume was placed in the vein and trace recorded without fetal or maternal movements. The women were tilted 15% to the left in order to minimise the risk of vena cava occlusion. Normally fetal blood velocity of vein of Galen is even and without pulsations. Signs of pulsations were noted and de®ned as a reduction in blood velocity by more than 15% of the maximum baseline. Fig. 2 illustrates an even and pulsating blood velocity tracing in the vein of Galen. Umbilical artery and venous blood velocity was recorded from a free-¯oating central part of the cord. The spectrum in the artery was analysed for pulsatility index (PI) according to Gosling et al. [11], and compared with normal reference values [12]. Umbilical venous pulsations were de®ned as a diastolic decrease in blood velocity exceeding 15% of the baseline maximum. Uterine artery blood velocity was also located bilaterally by colour ¯ow mapping in an oblique scan, with the sample volume placed in the artery just cranial to the crossing of the iliac blood vessels. The waveform was analysed for PI and the mean of both uterine artery PI exceeding 1.20 was deemed abnormal [12]. The clinicians managing the pregnancies were informed of the blood velocity waveforms in umbilical and uterine arteries, as well as in umbilical vein, but not of the results of the cerebral venous recording, which was performed at the same time. The study was a joint effort by the obstetrical departments at the university hospitals in Poznan, Poland, and Malmo, Sweden. Perinatal outcome variables were de®ned as interventions during labour, such as operative delivery for fetal distress (ODFD) indicated by abnormal cardiotocography tracing and/or fetal scalp blood pH. The neonatal condition was determined by Apgar score at 1 and 5 min, umbilical artery and venous cord pH admissions to a neonatal intensive care unit (NICU), the need of arti®cial ventilation, and perinatal
16
M. Dubiel et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 99 (2001) 14±18
Fig. 2. Blood velocity traces from the vein of Galen. Upper panel: abnormal blood velocity; lower panel: normal blood velocity.
mortality. Neonatal distress was de®ned as an Apgar score of <7 at 5 min and new-born acidemia as umbilical artery (pH < 7:15) and vein (<7.20). Fisher's exact test and unpaired Student's t-test were used for the statistical analysis, P < 0:05 was regarded as signi®cant. Odds ratios (95% con®dence level) were calculated. 3. Results The relation between clinical characteristics, extreme and average pulsations are shown in Table 1. Gestational age and
birthweight were lower when venous pulsations were present and pulsation in both veins were related to acidemia, neonatal distress and emergency intervention and admissions to a neonatal intensive care unit. Pulsations in the vein of Galen were three times more frequent than in the umbilical vein, but were still highly signi®cant correlated to adverse outcome. All fetuses with pulsations in the umbilical vein had also pulsations in the vein of Galen. The seven perinatal deaths included: six neonatal deaths due to severe growth retardation and prematurity (690 g, 29 weeks; 520 g, 27 weeks; 680 g, 33weeks; 710 g, 29 weeks; 670 g, 28 weeks; 720 g, 29 weeks) and one intra-uterine
Table 1 A comparison of blood velocity of vein of Galen and umbilical vein in predicting outcomea Vein of Galen pulsation
Number Gestational age at examination (weeks) Range Gestational age at delivery (weeks) Range Birthweight (g) SGA (< fifth percentile) Apgar score at <7(5 min) Umbilical arterial pH < 7.15 Umbilical vein pH < 7.20 Ventilation ODFD NICU admission Perinatal mortality a
No
Yes
34 37 2 33±41 38 2 33±42 2548 430 16 1 1 1 3 10 12 0
68 35 4 27±41 35 4 27±42 1947 854 41 20 15 19 34 53 54 7
Umbilical vein pulsation P-value <0.001 <0.001 <0.001 NS <0.001 <0.008 <0.001 <0.0001 <0.0001 <0.0001 <0.03
Mean S.D. or numbers are given; SGA: small for gestational age; NS: non-significant.
No
Yes
81 37 3 29±41 37 3 29±42 2362 687 36 10 9 10 20 43 45 3
21 33 4 27±41 33 4 27±41 1322 607 21 11 7 10 17 20 21 4
P -value <0.001 <0.001 <0.001 <0.0001 <0.0003 <0.02 <0.003 <0.0001 <0.0002 <0.0001 <0.03
M. Dubiel et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 99 (2001) 14±18
17
Table 2 Abnormal blood velocity waveform in the vein of Galen and umbilical veina Ventilation G U
n 68 n 21 a
10 (7.8±45) 13 (3.8±45)
Apgar score at <7(5 min)
NICU
pH < 7.15
pH < 7.20
ODFD
14 (2.1±289) 7.8 (2.3±25)
7.1 (2.7±18) 34 (4.2±0.0)
9.3 (1.4±199) 4.0 (1.2±13)
13 (1.9±269) 6.5 (2.0±20)
8.5 (3.2±22) 18 (2.7±370)
Odds ratios and 95% confidence intervals are given for neonatal outcome.
death (690 g, 30 weeks). All six babies that died also manifested abnormal umbilical artery PI. Abnormal umbilical artery PI (> mean 2S.D.) was found in 47 cases, 14 with nil or reversed end-diastolic ¯ow. In 64 cases, uterine artery PI showed abnormal values, and a bilateral notch was seen in 77 cases (75%). The odds ratio for adverse outcome by the vein of Galen and umbilical vein blood velocity readings are shown in Table 2. The venous pulsation in the vein of Galen showed the highest odds ratio for signs of new-born asphyxia (Table 2), but the umbilical vein for the need for assisted ventilation, admissions to the neonatal intensive care unit and emergency operation. 4. Discussion The ®ndings of the present study indicate that pulsations in the umbilical vein or in the vein of Galen are signi®cantly correlated to adverse outcome of pregnancy, including neonatal distress with low Apgar score and affected umbilical cord pH. This strongly suggests, that pulsations in these veins indicate severe fetal compromise. The more frequent ®nding of blood velocity pulsations in the vein of Galen than in the umbilical vein might suggest that the former appears earlier. Severe fetal compromise could therefore possibly be detected before pulsations in the umbilical vein appear, but these are known to be late signs of fetal compromise associated with new-born morbidity and mortality. Umbilical venous pulsations are quite rare and have previously been correlated to adverse outcome of pregnancy, including high mortality in cases of heart failure [6] and imminent fetal asphyxia [13]. The pathophysiology underlying the pulsations in the umbilical vein are thought to be transmission of increased central venous pressure waves out through the ®ltering sphincter of the ductus venosus into the umbilical cord. The pulsations are detected ®rst in the intraabdominal part of the umbilical vein and later in the cord itself. The Galenic vein system drains virtually the whole white matter area of the cerebral hemispheres. Venous congestion and dilatation are frequently associated with asphyxia [9]. In effect, the congestion and dilatation of the vein of Galen is commonly accompanied by a radial periventricular venous congestion or periventricular venous hemorrhage, representing morphological markers of asphyxia in the brain [9]. Therefore, the morphological ®ndings in asphyxial fetal brains lend further support regarding the pathological char-
acter of abnormalities in the deep cerebral venous blood ¯ow. Normally, blood velocity within the fetal vein of Galen is even and without pulsations [10]. Slight ¯uctuation in blood velocity can be detected when the sample volume is placed near the con¯uence sinus, where the vein of Galen meets the inferior sagittal sinus to form the straight sinus. The absence of pulsations in the vein of Galen was an unexpected ®nding as there is no known anatomical sphincter reported. The chinking of the vein of Galen, as it turns over to a straight sinus in the tentorium cerebelli, might be a hindrance to the transmission of pressure waves from the heart. Disturbed blood ¯ow caused by the blood stream coming from the inferior sagittal sinus might also hinder retrograde transmission of pulsations into the vein of Galen. The appearance of pulsations in the vein of Galen in high-risk pregnancies may be attributable to the transmission of pressure waves from the central venous circulation out into the periphery due to increased central venous pressure in combination with dilated veins due to increased cerebral ¯ow. The absence of a ®ltering sphincter might be the reason why the venous pulsation in the vein of Galen is a more frequent ®nding than in the umbilical vein, where the ductus venosus functions as a ®lter of pressure waves out to the vein. Increased cerebral blood ¯ow due to chronic fetal hypoxemia might also open up the fetal cerebral venous system and thus augment transmission of retrograde pressure waves. Pulsations in the vein of Galen might be mainly caused by retrograde transmission of pressure waves rather than being secondary to increased fetal cerebral blood ¯ow. In severe hypoxemia cerebral oedema might (in theory) also cause transmissions of pressure waves from the cerebral arteries. However, in the terminal case the blood ¯ow in the vein of Galen starts to resemble blood ¯ow in the inferior vena cava. This might suggest, that pulsations of the vein of Galen are mainly transmitted from the central veins. The close correlation between venous pulsations in the vein of Galen and adverse outcome might be of clinical importance for fetal surveillance in high-risk pregnancies. The occurrence of pulsations in the vein of Galen might be an intermediate sign of fetal compromise and delivery should thus be considered when pulsations in vein of Galen appear. Acknowledgements This study was supported by the Medical Faculty, UniÊ ke Wiberg versity of Lund, the Swedish Institute, the A
18
M. Dubiel et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 99 (2001) 14±18
foundation, and research funds at the University Hospital, MAS, Malmo. References [1] Gudmundsson S, Marsal K. Fetal aortic and umbilical artery blood velocity waveforms in prediction of fetal outcome Ð a comparison. Am J Perinatol 1991;8:1±6. [2] Hofstaetter C, Dubiel M, Gudmundsson S, Marsal K. Uterine artery color Doppler assisted velocimetry and perinatal outcome. Acta Obstet Gynecol Scand 1996;75:612±9. [3] Beherman RE, Lees MH, Peterson EN, DeLannoy CW, Seeds AE. Distribution of the circulation in the normal and asphyxiated fetal primate. Am J Obstet Gynecol 1970;108:956±69. [4] Vyas S, Nicolaides KH, Bower S, Campbell S. Middle cerebral artery flow velocity waveforms in fetal hypoxemia. Brit J Obstet Gynaecol 1990;97:797±803. [5] Reuss ML, Rudolph AM, Dae MW. Phasic blood flow patterns in the superior and inferior venae cavae and umbilical vein of fetal sheep. Am J Obstet Gynecol 1983;145:70±8. [6] Gudmundsson S, Huhta JC, Wood DC, Tulzer G, Cohen AW, Weiner S. Venous Doppler in the fetus with non-immune hydrops. Am J Obstet Gynecol 1991;164:33±7.
[7] Tulzer G, Gudmundsson S, Wood DC, Cohen AW, Weiner S, Huhta JC. Doppler in non-immune hydrops fetalis. Ultrasound Obstet Gynecol 1994;4:279±83. [8] Gudmundsson S, Tulzer G, Huhta JC, Marsal K. Venous Doppler in the fetus with absent end-diastolic flow in the umbilical artery. Ultrasound Obstet Gynecol 1996;7:262±7. [9] Laurini RN. Fetal brain pathology and ultrasound. In: Chervenak FA, Kurjak A, Comstock CH,editors. Ultrasound and the fetal brain. Carnforth, England, Parthenon Publishing,1995. p. 27± 42. [10] Laurichesse-Delmas H, Grimaud O, Moscoso G, Ville Y. Color Doppler study of the venous circulation in the fetal brain and hemodynamic study of the cerebral transverse sinus. Ultrasound Obstet Gynecol 1999;13:34±42. [11] Gosling RG, Dunbar G, King DH, Newman DL, Side CD, Woodcock JP, FitzGerald DE, Keates JS, MacMillian D. The quantitative analysis of occlusive peripheral arterial disease by a non-intrusive ultrasound technique. Angiology 1971;22:52±5. [12] Gudmundsson S, Marsal K. Umbilical and uteroplacental blood flow velocity waveforms in normal pregnancy Ð a cross sectional study. Acta Scand Obstet Gynaecol 1988;67:347±54. [13] Gudmundsson S, Tulzer G, Huhta JC, MarsaÂl K. Venous Doppler in the fetus with absent end-diastolic flow in the umbilical artery. Ultrasound Obstet Gynecol 1996;7:262±7.
Blood velocity in the fetal vein of Galen and the outcome of high-risk pregnancy Mariusz Dubiela, Grzegorz H. Breborowicza, Ricardo Laurinib, Saemundur Gudmundssonc,* a
Department of Perinatology and Gynaecology, University School of Medical Sciences, University Hospital, Poznan, Poland b Division of Developmental Pathology, University Hospital, Lausanne, Switzerland c Department of Obstetrics and Gynaecology, University Hospital MAS, Malmo, Sweden Received 7 July 2000; accepted 26 February 2001
Abstract Background: Pulsation in the ¯ow velocity waveform in the umbilical vein is related to perinatal mortality but the ¯ow velocity waveform in the fetal vein of Galen is normally even and without ¯uctuation. Objectives: To establish whether blood ¯ow velocity pulsations in the vein of Galen in high-risk pregnancies are related to outcome. Study Design: The vein of Galen was located by colour Doppler ultrasound in 102 pregnancies complicated by severe pregnancy-induced hypertension. The blood velocity waveform was recorded by pulsed Doppler within 2 days of delivery and the presence pulsations related to pregnancy outcome, including emergency operative intervention and neonatal distress. Umbilical artery and vein and uterine artery blood ¯ow velocity waveform were also recorded at the same time. The clinicians managing the women were unaware of the venous ¯ow results. Results: Pulsation were present in the vein of Galen in 68 cases and in the umbilical vein in 21. Both were signi®cantly related to adverse outcome. Pulsations in the vein of Galen were seen in all seven perinatal deaths. Conclusions: Since umbilical venous pulsation are a late sign of fetal compromise, and pulsations in the vein of Galen seem to appear earlier, thus being an intermediate sign of fetal compromise that might be of great value for fetal surveillance. # 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Vein of Galen; Doppler ultrasound; Umbilical vein; Pregnancy
1. Introduction Over the past 20 years, Doppler ultrasound technology has developed considerably and is now a routine part of clinical surveillance in high-risk pregnancies. This includes recording of the umbilical and uterine arteries to predict increased placental vascular resistance, which itself is predictive of an adverse outcome of pregnancy [1,2]. Studies on fetal lamb have shown that the fetus adapts to reduced placental perfusion and chronic hypoxia by redistributing its blood ¯ow to vital organs such as the brain, heart and adrenals [3]. Studies on fetal cerebral circulation in humans, especially in the middle cerebral artery, have shown increased diastolic blood velocity to be a characteristic sign of brain sparing the high-risk pregnancies with reduced placental perfusion [4]. *
Corresponding author. Tel.: 46-40-332095; fax: 46-40-962600. E-mail address: [email protected] (S. Gudmundsson).
Examination of the fetal venous circulation has advanced during the 1990s. Blood velocity in the systemic venous circulation is pulsatile, re¯ecting central venous pressure [5]. By contrast, blood velocity in the umbilical vein (U) and portal circulation under normal circumstances are even and without pulsations due to the ®ltering effect of the ductus venosus. The tracing of blood velocity in the systemic veins has revealed increased reversal of ¯ow during atrial contraction and pulsations in the umbilical vein in cases of congestive fetal heart failure and imminent asphyxia [6±8] due to increased central venous pressure. Umbilical venous pulsations are thus a late sign of fetal compromise. Furthermore, experience with fetal post-mortems in cases of asphyxial intra-uterine fetal death has shown that the great vein of Galen (G) is frequently dilated [9]. This dilatation in cases of intra-uterine fetal death caused by chronic hypoxemia could be due to increased cerebral blood ¯ow during hypoxia, or increased central venous pressure. In a pilot study at our institutions and as reported by Laurichesse-Delmas et al. [10], blood velocity in the great
0301-2115/01/$ ± see front matter # 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 1 - 2 1 1 5 ( 0 1 ) 0 0 3 4 0 - 2
M. Dubiel et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 99 (2001) 14±18
vein of Galen, is in the normal state even and without the characteristic pulsations observed in the systemic venous circulation. In complicated pregnancies, however, blood velocity pulsations can be seen, but its clinical implication was unknown. The purpose of this study was to determine whether blood velocity in the vein of Galen might predict adverse perinatal outcome of high-risk pregnancies. The clinical predictive value of blood velocity of vein of Galen was also evaluated in relation to umbilical venous blood ¯ow, which presently is frequently used for fetal surveillance in high-risk pregnancies. 2. Material and methods The internal cerebral veins join in front of the cerebellum in a short vein called the great vein of Galen or the great cerebral vein. The vein of Galen has a cranio-posterior direction in front of the cerebellum towards the tentorium cerebelli, joining the inferior sagittal sinus (ISS) to form the straight sinus (SS). The straight sinus is directed backwards (caudally), joining the superior sagittal sinus (SSS) to form the transverse sinuses. The anatomic relationship between the veins is illustrated in Fig. 1. In the present study, the vein of Galen was located by means of an Acuson 128 XP duplex scanner using a 3.5 and 5 MHz transducer with power and colour Doppler options. During the nineth month, the fetal vein of Galen was located in 102 pregnancies complicated by severe pregnancyinduced hypertension (blood pressure 160/110 or higher;
Fig. 1. Magnetic resonance imaging angiography of human cerebral veins. The angle between the vein of Galen (G) and straight sinus (SS) is clearly seen. The inferior sagittal sinus (ISS) is not seen in this picture, but is illustrated by a dotted line. No flow is detected at the crossing of the vein of Galen and inferior sagittal sinus. Superior sagittal sinus (SSS).
15
proteinuria 3) and referred to our ultrasound laboratories for umbilical and uterine artery blood velocimetry as a part of placental vascular resistance evaluation. Fifty-seven cases had also suspected intra-uterine growth retardation, de®ned as below ®fth percentile for population studied. Fifty-three cases were emergency admissions to a tertiary care unit and were delivered shortly after admission. Each fetus was examined once only, the Galen venous blood ¯ow was recorded in all cases. Median gestational age at examination was 36 weeks (27±41). In all cases the gestational age was based on early ultrasound dating of pregnancy. The median interval between the examination and labour was 1 day (1±2 days). The vein of Galen was located by colour Doppler ultrasound with a transducer in the occipital transverse plane. The angle between the vessel and the transducer was <458. The straight sinus could be located easily. The con¯uence of the vein of Galen and the inferior sagittal sinus showed a characteristic bright colour ¯ow due to disturbed blood ¯ow. By moving the transducer slightly downwards (caudally), the vein of Galen could be visualised below the disturbed blood ¯ow in the vein. The sample volume was placed in the vein and trace recorded without fetal or maternal movements. The women were tilted 15% to the left in order to minimise the risk of vena cava occlusion. Normally fetal blood velocity of vein of Galen is even and without pulsations. Signs of pulsations were noted and de®ned as a reduction in blood velocity by more than 15% of the maximum baseline. Fig. 2 illustrates an even and pulsating blood velocity tracing in the vein of Galen. Umbilical artery and venous blood velocity was recorded from a free-¯oating central part of the cord. The spectrum in the artery was analysed for pulsatility index (PI) according to Gosling et al. [11], and compared with normal reference values [12]. Umbilical venous pulsations were de®ned as a diastolic decrease in blood velocity exceeding 15% of the baseline maximum. Uterine artery blood velocity was also located bilaterally by colour ¯ow mapping in an oblique scan, with the sample volume placed in the artery just cranial to the crossing of the iliac blood vessels. The waveform was analysed for PI and the mean of both uterine artery PI exceeding 1.20 was deemed abnormal [12]. The clinicians managing the pregnancies were informed of the blood velocity waveforms in umbilical and uterine arteries, as well as in umbilical vein, but not of the results of the cerebral venous recording, which was performed at the same time. The study was a joint effort by the obstetrical departments at the university hospitals in Poznan, Poland, and Malmo, Sweden. Perinatal outcome variables were de®ned as interventions during labour, such as operative delivery for fetal distress (ODFD) indicated by abnormal cardiotocography tracing and/or fetal scalp blood pH. The neonatal condition was determined by Apgar score at 1 and 5 min, umbilical artery and venous cord pH admissions to a neonatal intensive care unit (NICU), the need of arti®cial ventilation, and perinatal
16
M. Dubiel et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 99 (2001) 14±18
Fig. 2. Blood velocity traces from the vein of Galen. Upper panel: abnormal blood velocity; lower panel: normal blood velocity.
mortality. Neonatal distress was de®ned as an Apgar score of <7 at 5 min and new-born acidemia as umbilical artery (pH < 7:15) and vein (<7.20). Fisher's exact test and unpaired Student's t-test were used for the statistical analysis, P < 0:05 was regarded as signi®cant. Odds ratios (95% con®dence level) were calculated. 3. Results The relation between clinical characteristics, extreme and average pulsations are shown in Table 1. Gestational age and
birthweight were lower when venous pulsations were present and pulsation in both veins were related to acidemia, neonatal distress and emergency intervention and admissions to a neonatal intensive care unit. Pulsations in the vein of Galen were three times more frequent than in the umbilical vein, but were still highly signi®cant correlated to adverse outcome. All fetuses with pulsations in the umbilical vein had also pulsations in the vein of Galen. The seven perinatal deaths included: six neonatal deaths due to severe growth retardation and prematurity (690 g, 29 weeks; 520 g, 27 weeks; 680 g, 33weeks; 710 g, 29 weeks; 670 g, 28 weeks; 720 g, 29 weeks) and one intra-uterine
Table 1 A comparison of blood velocity of vein of Galen and umbilical vein in predicting outcomea Vein of Galen pulsation
Number Gestational age at examination (weeks) Range Gestational age at delivery (weeks) Range Birthweight (g) SGA (< fifth percentile) Apgar score at <7(5 min) Umbilical arterial pH < 7.15 Umbilical vein pH < 7.20 Ventilation ODFD NICU admission Perinatal mortality a
No
Yes
34 37 2 33±41 38 2 33±42 2548 430 16 1 1 1 3 10 12 0
68 35 4 27±41 35 4 27±42 1947 854 41 20 15 19 34 53 54 7
Umbilical vein pulsation P-value <0.001 <0.001 <0.001 NS <0.001 <0.008 <0.001 <0.0001 <0.0001 <0.0001 <0.03
Mean S.D. or numbers are given; SGA: small for gestational age; NS: non-significant.
No
Yes
81 37 3 29±41 37 3 29±42 2362 687 36 10 9 10 20 43 45 3
21 33 4 27±41 33 4 27±41 1322 607 21 11 7 10 17 20 21 4
P -value <0.001 <0.001 <0.001 <0.0001 <0.0003 <0.02 <0.003 <0.0001 <0.0002 <0.0001 <0.03
M. Dubiel et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 99 (2001) 14±18
17
Table 2 Abnormal blood velocity waveform in the vein of Galen and umbilical veina Ventilation G U
n 68 n 21 a
10 (7.8±45) 13 (3.8±45)
Apgar score at <7(5 min)
NICU
pH < 7.15
pH < 7.20
ODFD
14 (2.1±289) 7.8 (2.3±25)
7.1 (2.7±18) 34 (4.2±0.0)
9.3 (1.4±199) 4.0 (1.2±13)
13 (1.9±269) 6.5 (2.0±20)
8.5 (3.2±22) 18 (2.7±370)
Odds ratios and 95% confidence intervals are given for neonatal outcome.
death (690 g, 30 weeks). All six babies that died also manifested abnormal umbilical artery PI. Abnormal umbilical artery PI (> mean 2S.D.) was found in 47 cases, 14 with nil or reversed end-diastolic ¯ow. In 64 cases, uterine artery PI showed abnormal values, and a bilateral notch was seen in 77 cases (75%). The odds ratio for adverse outcome by the vein of Galen and umbilical vein blood velocity readings are shown in Table 2. The venous pulsation in the vein of Galen showed the highest odds ratio for signs of new-born asphyxia (Table 2), but the umbilical vein for the need for assisted ventilation, admissions to the neonatal intensive care unit and emergency operation. 4. Discussion The ®ndings of the present study indicate that pulsations in the umbilical vein or in the vein of Galen are signi®cantly correlated to adverse outcome of pregnancy, including neonatal distress with low Apgar score and affected umbilical cord pH. This strongly suggests, that pulsations in these veins indicate severe fetal compromise. The more frequent ®nding of blood velocity pulsations in the vein of Galen than in the umbilical vein might suggest that the former appears earlier. Severe fetal compromise could therefore possibly be detected before pulsations in the umbilical vein appear, but these are known to be late signs of fetal compromise associated with new-born morbidity and mortality. Umbilical venous pulsations are quite rare and have previously been correlated to adverse outcome of pregnancy, including high mortality in cases of heart failure [6] and imminent fetal asphyxia [13]. The pathophysiology underlying the pulsations in the umbilical vein are thought to be transmission of increased central venous pressure waves out through the ®ltering sphincter of the ductus venosus into the umbilical cord. The pulsations are detected ®rst in the intraabdominal part of the umbilical vein and later in the cord itself. The Galenic vein system drains virtually the whole white matter area of the cerebral hemispheres. Venous congestion and dilatation are frequently associated with asphyxia [9]. In effect, the congestion and dilatation of the vein of Galen is commonly accompanied by a radial periventricular venous congestion or periventricular venous hemorrhage, representing morphological markers of asphyxia in the brain [9]. Therefore, the morphological ®ndings in asphyxial fetal brains lend further support regarding the pathological char-
acter of abnormalities in the deep cerebral venous blood ¯ow. Normally, blood velocity within the fetal vein of Galen is even and without pulsations [10]. Slight ¯uctuation in blood velocity can be detected when the sample volume is placed near the con¯uence sinus, where the vein of Galen meets the inferior sagittal sinus to form the straight sinus. The absence of pulsations in the vein of Galen was an unexpected ®nding as there is no known anatomical sphincter reported. The chinking of the vein of Galen, as it turns over to a straight sinus in the tentorium cerebelli, might be a hindrance to the transmission of pressure waves from the heart. Disturbed blood ¯ow caused by the blood stream coming from the inferior sagittal sinus might also hinder retrograde transmission of pulsations into the vein of Galen. The appearance of pulsations in the vein of Galen in high-risk pregnancies may be attributable to the transmission of pressure waves from the central venous circulation out into the periphery due to increased central venous pressure in combination with dilated veins due to increased cerebral ¯ow. The absence of a ®ltering sphincter might be the reason why the venous pulsation in the vein of Galen is a more frequent ®nding than in the umbilical vein, where the ductus venosus functions as a ®lter of pressure waves out to the vein. Increased cerebral blood ¯ow due to chronic fetal hypoxemia might also open up the fetal cerebral venous system and thus augment transmission of retrograde pressure waves. Pulsations in the vein of Galen might be mainly caused by retrograde transmission of pressure waves rather than being secondary to increased fetal cerebral blood ¯ow. In severe hypoxemia cerebral oedema might (in theory) also cause transmissions of pressure waves from the cerebral arteries. However, in the terminal case the blood ¯ow in the vein of Galen starts to resemble blood ¯ow in the inferior vena cava. This might suggest, that pulsations of the vein of Galen are mainly transmitted from the central veins. The close correlation between venous pulsations in the vein of Galen and adverse outcome might be of clinical importance for fetal surveillance in high-risk pregnancies. The occurrence of pulsations in the vein of Galen might be an intermediate sign of fetal compromise and delivery should thus be considered when pulsations in vein of Galen appear. Acknowledgements This study was supported by the Medical Faculty, UniÊ ke Wiberg versity of Lund, the Swedish Institute, the A
18
M. Dubiel et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 99 (2001) 14±18
foundation, and research funds at the University Hospital, MAS, Malmo. References [1] Gudmundsson S, Marsal K. Fetal aortic and umbilical artery blood velocity waveforms in prediction of fetal outcome Ð a comparison. Am J Perinatol 1991;8:1±6. [2] Hofstaetter C, Dubiel M, Gudmundsson S, Marsal K. Uterine artery color Doppler assisted velocimetry and perinatal outcome. Acta Obstet Gynecol Scand 1996;75:612±9. [3] Beherman RE, Lees MH, Peterson EN, DeLannoy CW, Seeds AE. Distribution of the circulation in the normal and asphyxiated fetal primate. Am J Obstet Gynecol 1970;108:956±69. [4] Vyas S, Nicolaides KH, Bower S, Campbell S. Middle cerebral artery flow velocity waveforms in fetal hypoxemia. Brit J Obstet Gynaecol 1990;97:797±803. [5] Reuss ML, Rudolph AM, Dae MW. Phasic blood flow patterns in the superior and inferior venae cavae and umbilical vein of fetal sheep. Am J Obstet Gynecol 1983;145:70±8. [6] Gudmundsson S, Huhta JC, Wood DC, Tulzer G, Cohen AW, Weiner S. Venous Doppler in the fetus with non-immune hydrops. Am J Obstet Gynecol 1991;164:33±7.
[7] Tulzer G, Gudmundsson S, Wood DC, Cohen AW, Weiner S, Huhta JC. Doppler in non-immune hydrops fetalis. Ultrasound Obstet Gynecol 1994;4:279±83. [8] Gudmundsson S, Tulzer G, Huhta JC, Marsal K. Venous Doppler in the fetus with absent end-diastolic flow in the umbilical artery. Ultrasound Obstet Gynecol 1996;7:262±7. [9] Laurini RN. Fetal brain pathology and ultrasound. In: Chervenak FA, Kurjak A, Comstock CH,editors. Ultrasound and the fetal brain. Carnforth, England, Parthenon Publishing,1995. p. 27± 42. [10] Laurichesse-Delmas H, Grimaud O, Moscoso G, Ville Y. Color Doppler study of the venous circulation in the fetal brain and hemodynamic study of the cerebral transverse sinus. Ultrasound Obstet Gynecol 1999;13:34±42. [11] Gosling RG, Dunbar G, King DH, Newman DL, Side CD, Woodcock JP, FitzGerald DE, Keates JS, MacMillian D. The quantitative analysis of occlusive peripheral arterial disease by a non-intrusive ultrasound technique. Angiology 1971;22:52±5. [12] Gudmundsson S, Marsal K. Umbilical and uteroplacental blood flow velocity waveforms in normal pregnancy Ð a cross sectional study. Acta Scand Obstet Gynaecol 1988;67:347±54. [13] Gudmundsson S, Tulzer G, Huhta JC, MarsaÂl K. Venous Doppler in the fetus with absent end-diastolic flow in the umbilical artery. Ultrasound Obstet Gynecol 1996;7:262±7.