ELSEVIER Early Human Development 41 (1995) 69-77

Blood levels of vasoactive intestinal polypeptide in normal and growth retarded fetuses:

relationship with acid-base and haemodynamic status

Giuseppe Rizzo* a, Paolo Montuschi b, Alessandra Capponi”, Carlo Romanini a

“Department of Obstetrics and Gynecology, Vniversitk di Roma “Tor Vergata”, Policlinico ,NUOYO 5 Eugenio, P.le Vmanesimo IO, 00144 Rome, Iia1.v

bDepartment of Pharmacology, Vniversit6 Cattolica S. Gore. Rome, Ital!

Received 28 July 1994; revision received 28 November 1994; accepted 30 November 1996

Abstract

The objectives of this study were (1) to detect vasoactive intestinal polypeptide III fetal blood obtained by cordocentesis (2) to examine possible changes in growth retarded fetuses and to establish relationships between its levels and fetal blood acid-base status as well as feta: haemodynamics as assessed by Doppler ultrasonography. Vasoactive intestinal polypeptide was measured in umbilical vein blood obtained at cordocentesis in 12 growth retarded fetuses and in 13 control fetuses. Umbilical vein pH and PO, values were determined in all the cases. Before the procedure, Doppler indices were calculated from umbilical artery, middle cerebral artery, renal artery, cardiac outflow tracts and inferior vena cava. Simple and multiple step- wise regression analysis were performed to examine the relationships between Doppler in- dices, acid-base status and vasoactive intestinal polypeptide levels. In control fetuses, vasoactive intestinal polypeptide was always detectable in cord blood and its levels did nor change with gestational age. In growth retarded fetuses, vasoactive intestinal polypeptide lev- els were higher and significantly related to umbilical vein PO, levels, Pulsatility Index in um- bilical artery, middle cerebral artery and renal artery, while no relationship was found with umbilical vein pH, cardiac and venous Doppler indices. Stepwise multiple regression demonstrated middle cerebral artery Pulsatility Index to be the best explanatory variable for

* Corresponding author.

0378-3782/95/$09.50 0 1995 Elsevier Science Ireland Ltd. All rights reserved SSDI 0378-3782(94)01610-Z

70 G. Rizzo et al. /Early Human Developmenr 41 (1995) 69-77

vasoactive intestinal polypeptide levels. In conclusion, vasoactive intestinal polypeptide blood levels are increased in growth retarded fetuses and this increase is inversely related to the Dop- pler measured impedance to flow in middle cerebral artery.

Keywords: Vasoactive intestinal polypeptide; IUGR; Cordocentesis; Doppler ultrasono- graphy; Hypoxia

1. Introduction

Doppler ultrasonography has allowed the demonstration of the presence of clear haemodynamic changes in the human fetus during uteroplacental insufficiency. These changes include a reduction of impedance to flow in cerebral vessels associ- ated with an increased impedance to flow at the level of fetal peripheral and umbili- cal vessels. This leads to a preferential blood streaming to the brain resulting in the so called brain sparing effect [7,20].

Although these haemodynamic changes have been extensively studied, there is a scarcity of information on the endocrine mechanisms responsible for these modifica- tions and in addition, these studies are limited to animal models [19].

Since vasoactive intestinal polypeptide (VIP) is a powerful vasodilator [15] and high levels of this polypeptide have been found in the blood of infants who under- went fetal distress in labor [ 111, we speculated that this hormone may also be in- volved in the haemodynamic modifications occurring during uteroplacental insufficiency.

The purpose of this study was to relate the VIP levels in fetal blood obtained by cordocentesis with fetal acid-base status and Doppler indices measured from umbili- cal artery and fetal vessels.

2. Material and methods

2.1. Patients After informed consent was obtained from the mothers, 13 normally growing

fetuses (control group) and 12 growth retarded fetuses (IUGR) scheduled for cor- docentesis were considered for this study. All pregnancies were singleton and ac- curately dated by early ultrasonographic examination, Further criteria for inclusion for IUGR fetuses were (a) free from structural and chromosomal abnormalities and (b) presence of Doppler detectable haemodynamic changes suggestive of the brain sparing phenomenon (i.e. ratio between Pulsatility Index values between umbilical artery and middle cerebral artery > 95th centile of our reference limits for gestation PI).

The indications for cordocentesis in the control group were suspected fetal infec- tion (toxoplasmosis, n = 8; cytomegalovirus, n = 3), need for rapid fetal karyotype for late referral (n = 2). All the included fetuses were found to be free from the pathology for which they were tested. In IUGR fetuses, fetal blood sampling was performed to obtain a rapid fetal karyotype and acid-base status.

2.2. Doppler recordings Recordings were performed immediately before the procedure using commercially

available color and pulsed Doppler ultrasound equipment (Ansaldo Esaote Hitachi 590A, Genoa, Italy) with 3.5- or ~-MHZ convex probes. The Doppler carrier fre- quency ranged from 2.5 to 5 MHz. The high-pass filter was set at 100 Hz.

Velocity waveforms were recorded from umbilical artery as well as from differem fetal extracardiac and intracardiac districts including renal artery, middle cerebra) artery, ascending aorta, pulmonary artery and inferior vena cava. The technique followed for recording velocity waveforms from these vessels has been reported elsewhere [2,17,18].

Doppler images were videorecorded for subsequent analysis. Permanent record* were obtained from the videotape by means of a strip chart recorder. Ten con- secutive heart cycles were selected during periods of fetal rest without breathing movements and the values measured were averaged. The following variables were calculated with the aid of a computer-interfaced digitizer pad (Cardio 800, Kontron. Oxford, UK): (1) Pulsatility Index (PI = [systolic velocity - diastolic velocity]/mean velocity) [9] at the level of umbilical artery and fetal peripheral arterial vessels; (21 aortic and pulmonary valves peak velocities [17]; (3) the percentage of reverse flow during atria1 contraction with respect to total forward flow in inferior vena cava (“,c, reverse flow) [ 181.

2.3. Cordocentesis Umbilical venous blood was obtained directly under ultrasonographic control at

the level of placental cord insertion or on an umbilical free loop without maternai sedation or fetal paralysis. The purity of the blood sample was confirmed by the mean corpuscular cell volume distribution dissimilar to that of the mother at the cell analyzer. The cord blood gas analysis was performed within 5 min of the procedure with an automatic blood gas analyzer (ABL 330, Radiometer. Copenhagen, L>en- mark). An extra 1 ml of fetal blood was obtained and collected in ice-chilled tubes containing heparin (50 IU) and aprotonin (500.000 IU) and left on ice until ccn- trifugation at 4°C. Plasma aliquots were then stored at -30°C unfil assayed

2.4. VIP assessment VIP levels were assessed using a specific radioimmunoassay (RIA) procedure.

Human VIP-free plasma (blanks) was prepared as described by Fahrenkrug and Schaffalitzky de Muckadell [7]. Aliquots of standard solution (concentration range 2.9-1250 pg/ml), samples and blanks were then extracted in absolute ethanol and dried under vacuum as described elsewhere. The dried product was stored at -30°C: and, on the day of use, reconstituted to a l/4 of the original volume with 0.01 M phosphate buffer containing 0.154 M NaCl, 0.025 M ethylene diaminetetraacetic acid (EDTA), 0.01% thimerosal, and 0.5% bovine serum albumin at pH 7.2..

Details of the RIA procedure, using a highly specific polycional antiserum, has been provided elsewhere [6]. The detection limit of the standard curve was 2.9 pgml and the IC50 26 pgml. Maximal binding was 4Y% and non-specific binding was below the background value of the gamma counter. The intra-assay and inter-assa! coefficients of variation were, respectively 6.4 and 9.6’!/,:.

12 G. Rizzo et al. /Early Human Development 41 (1995) 69-77

2.5. Data analysis Since Doppler indices and umbilical venous pH and PO, values change with ges-

tational age, values are also expressed as the number of standard deviations from which they differ from the expected mean for gestation (delta values). For Doppler indices, we used the reference ranges previously constructed by us and for pH and blood gas values, the normal range constructed by Nicolaides et al. [ 131.

Statistical comparison was performed by means of Fisher’s exact test for categoric variables and by Student t test or Wilcoxon test for continuous variables. The rela- tionship between VIP levels, gestational age, each of the Doppler indices analyzed and fetal acid-base status was evaluated with simple and stepwise multiple regression analysis. A P-value ~0.05 was considered significant.

3. Results

There were no differences in maternal age, parity and gestational age at cor- docentesis between control and IUGR fetuses (Table 1). However, IUGR fetuses had lower umbilical artery pH and Paz values, were delivered earlier, had a lower birthweight and poorer perinatal outcome (Table 1).

At the time of cordocentesis, the prevalence of fetal acidemia (a pH value 2 SD. below the mean for gestational age) was 66.6% (8/12) and hypoxemia (a PO, value 2 S.D. below the mean for gestational age) was 91.6% (1 l/12) among IUGR fetuses. All IUGR fetuses had umbilical artery PI values above the 95th centile of our refer- ence limits and seven also had absent end diastolic velocity. Middle cerebral artery PI values were below the 5th centile in 91.6% (1 l/12) of IUGR fetuses, while in 83.3% (10/12) renal artery PI was above the 95th centile. Peak velocities from aorta and pul- monary arteries were below the 5th centile in, respectively 50% (6112) and 75% (8112)

Table 1 Characteristics of the two groups of fetuses considered

Control (n = 13)

Maternal (years) age 30.4 zt 3.2 Parity 0.89 f 1.12 Gestational age at cordocentesis (weeks) 30.61 f 3.66 Abdominal circumference (delta value) 0.07 l 1.44

Head/abdomen circumference ratio (delta value) -0.08 f 1.36 Umbilical vein pH 7.391 f 0.03 Umbilical vein PO, (mmHg) 35.99 sz 6.84 Gestational at delivery (weeks) age 38.23 zt 3.53 Birthweight (g) 3290 zt 540 Intrauterine death n (%) 0 (0) Neonatal death n (%) 0 (0) Perinatal mortality n (%) 0 (0)

Data are shown as mean f SD. or n (%); n.s., not significant.

IUGR (n = 12)

31.9 f 4.2 0.78 f 1.34

29.82 -f 3.205 -4.30 f 1.86

2.66 f 0.52 7.270 zt 0.07 16.27 f 3.44 31.58 f 4.46 1125 zt 430 1 (8.3) 3 (25.0) 4 (33.3)

P-value

n.s. n.s. n.s. ~0.001 SO.001 SO.001 SO.001 10.001 ~0.001 n.s. ns. SO.05

G. Rizzo et al. /Early Human Development 4/ f l99Ci 69-V

of IUGR fetuses. Percentage reverse flow in inferior vena cava was above the 95th centile in (7/12) of IUGR fetuses. None of the control fetuses showed Doppler in- dices or pH and PO, values outside reference limits.

There was no significant relationship between VIP values and gestation (r = 0,01: n.s.). However, VIP values were significantly higher in IUGR fetuses than in control fetuses (control median 6.5 pg/ml, range 3.9-26.4; IUGR median 33.6 pg/mf, range 8.1-82.9; P 5 0.0003) (Fig. 1).

There was no relationship between the severity of growth retardation (i.e. delta abdominal circumference values, r = 0.12; n.s.) and the ratio between head and ab- domen circumference (and VIP values). Similarly, among IUGR fetuses, no differ- ence were found in VIP values between the four perinatal deaths (median 30. range 8.5-82.9) and the eight survivors (median 35.65, range 8.1-72.5).

The relationship between VIP values and the Doppler and blood gas indices con- sidered is reported in Table 2. Linear regression analysis indicated significant rela-

90

80

70

60

50 2 2 40

30

20

10

0

0

l

0

: 0

0

0 8

t

i 0

control fetuses

.

IUGR fetuses

.

Fig. I. Concentration of VIP (pglml) in umbilical vein of control and IUGR fetuses

14 G. Riz:o et al. /Early Human Development 41 (1995) 69-77

Table 2 Relationship of Doppler indices (delta values) and acid-base status (delta values) with VIP levels in umbil- ical vein

r Residual S.D. P-value

Delta middle cerebral artery PI 0.686 15.991 0.0002 Delta umbilical vein PO, 0.625 17.165 0.0008 Delta renal artery PI 0.590 17.753 0.0019 Delta umbilical artery PI 0.539 18.153 0.0054 Delta umbilical artery pH 0.338 20.689 0.0982 Delta inferior vena cava % reverse flow 0.312 22.435 0.1234 Delta aortic peak velocity 0.304 23.561 0.1374 Delta pulmonaric peak velocity 0.298 24.956 0.1654

tionships between VIP levels and delta PO*, delta umbilical artery PI, delta middle cerebral artery PI and renal artery PI (Fig. 2). On the other hand, no significant rela- tionships were found between VIP levels and delta pH and delta peak velocity at outflow tract and % reverse flow in inferior vena cava. Stepwise multiple regression analysis indicated that when delta PI values from the middle cerebral artery were entered first into a model to predict VIP levels, none of the individual variables con- sidered added significantly to the prediction of VIP values. Furthermore, when the other variables were entered first, singly or in combination, the addition of delta PI values from the middle cerebral artery introduced a significant improvement in the prediction of VIP values.

4. Discussion

To the best of our knowledge, this is the first study to analyze fetal blood levels of VIP before delivery. We demonstrated that VIP is detectable in cord blood from 24 weeks onwards and that its level remains stable during the third trimester of preg- nancy. The levels of VIP found in this study are similar to the values found in the umbilical vein of normal newborns at birth [ 11,141.

There is no consensus on the origin of VIP levels in umbilical vein, and maternal, fetal and placental origins have been suggested as potential sources [ 141. Our experi- mental design did not allow us to solve this question. However, the lack of relation- ships between maternal serum and umbilical vein levels in some of our subjects (unpublished observations) as well as the results of previous studies makes the fetal and/or placental origins more likely [ 14,151.

Irrespective of its origin, in a small series of IUGR fetuses, we were able to demonstrate the presence of significantly higher levels of circulating VIP with respect to normally grown fetuses. Furthermore, this increase seems to be related to some of the haemodynamic modifications occurring in these fetuses. Although caution should be used in the interpretation of the results of stepwise multiple regression analysis when a relatively low number of subjects is considered, our results indicate

G. Rho et al. /Early Human Development 41 (1995) 69-77

90

80

i

l

70 0

g :I 5 40 s

30

20

10

0

90

80

1

l

70 l

-6 -5 -4 -3 -2 -1 0 1 2 -5 -4 -3 -2 -I 0 1 2

delta middle cerebral artery PI delta pO2

90

80 0

70 l

60

g 50 & 40 5

30

20

10

0

90 -

80 - 0

70 - a

0 2 4

delta renal artery PI

-2 0 2 4 6 8 IO

delta umbilical artery PI

Fig. 2. Relationship between VIP levels and middle cerebral artery PI. umbilical vem PO,, renal artery

PI and umbilical artery PI. Since these latter variables change with gestation, the data are expressed as the differences in standard deviations from the normal mean of gestation (delta).

that the PI value of the middle cerebral artery is the best parameter to explain the variance of fetal VIP blood levels. This is not surprising since VIP is a powerful vasodilator [ 151 and may thus explain the chronic vasodilation in the cerebral artery present in these fetuses. Indeed, VIP-containing nerve fibers have been found in the brain of the human fetus particularly concentrated in the circle of Willis, thus sug- gesting a role for this peptide in the regulation of the vascular tone of this area [12].

16 C. Rizzo et al. /Early Human Development 41 (1995) 69-77

The concomitant relationship between VIP levels and fetal Paz, umbilical artery PI and renal artery PI may be explained on the basis of the high interrelationships existing between these variables and middle cerebral artery PI, all being expressions of uteroplacental insufficiency. Previous studies by us and others have shown that PI values from umbilical artery, fetal middle cerebral and renal arteries are significantly related to fetal PO, levels [1,5]. The results of the stepwise regression analysis support this thesis.

However, mechanisms of action different from that on cerebral vasculature may not be excluded for VIP on the basis of experimental data indicating a direct vasoac- tive action of this polypeptide on placental stem villous arteries [lo], which are con- sidered directly related to umbilical artery PI. Furthermore, VIP stimulates the secretion of vasopressin, a peptide that may induce selective vasoconstriction in fetal peripheral vessels and in the placenta [ 161.

Of interest is the lack of relationships between VIP levels and umbilical vein pH, peak velocity at the level of outflow tracts and % reverse flow in inferior vena cava. Longitudinal studies on deteriorating IUGR fetuses have shown that fetal acidosis and Doppler anomalies at the level of the fetal heart and of venous circulation are concomitant and present at a late stage of uteroplacental insufficiency following a matter of weeks after the haemodynamic changes in fetal arterial peripheral vessels characteristic of the brain sparing phenomenon [3,4]. The cross-sectional design of this study does not allow us to draw conclusions on the temporal relationships be- tween Doppler indices and VIP levels. However, on the basis of the interrelation found between VIP levels and Doppler indices, we may suggest a role for VIP in the adaptive vascular mechanisms occurring at an early stage of uteroplacental insuffi- ciency, while other factors should be considered to explaiu the haemodynamic changes occurring with the progression of the disease. This thesis is validated by the presence of elevated VIP levels found in infants from uncomplicated pregnancies suf- fering from (mild) distress limited to delivery [l 11.

In conclusion, our data suggest that VIP is detectable in fetal blood by 24 weeks of gestation and its levels are unaffected by gestational age. In IUGR fetuses, VIP levels are increased and this increase is significantly related to the PI values of middle cerebral artery thus suggesting a role for this peptide in the genesis and maintenance of the brain sparing phenomenon.

Acknowledgement

This study was supported by a grant of the Italian National Council of Research (CNR grant No. 94 0004 PF41).

References

[l] Arduini, D., R&o, G., Romanini, C. and Mancuso, S. (1989): Are blood flow velocity waveforms related to umbilical cord acid-base balance in the human fetus? Gynecol. Obstet. Invest., 27, 183-187.

[2] Arduini, D. and R&o, G. (1990): Normal values of Pulsatility Index from fetal vessels: a cross sec- tional study on 1556 healthy fetuses. J. Perinatal Med., 18, 165-172.

G. Rizzo et al. /Early Human Developmenr 41 / l!JYS! 69-77

[3] Arduini, D. and Rizzo, G. (1992): Changes of Pulsatility Index from fetal vessels preceding the onset of late decelerations in growth retarded fetuses. Obstet. Gynecol.. 79, 605-610.

[4] Arduini, D., Rizzo, G. and Romanini, C. (1993): The development of abnormal heart rate patterns after absent end diastolic velocity in umbilical artery: analysis of risk factors. Am. J. Obstet Gynecol., 168, 43-49.

[S] Bilardo, CM., Nicolaides, K.H. and Campbell, S. (1990): Doppler measurements of fetal and uteroplacental circulations: relationship with umbilical venous blood gas measured at COT‘ docentesis. Am. J. Obstet. Gynecol., 162, 115-120.

[6] D’Amato, M., Curro, D., Montuschi, P., Ragazzoni, E. and Lefebvre, R.A. (1992): Release of vaso- active intestinal polypeptide from the rat gastric fundus. Br. J. Pharmacol.. 105, 691-695

]7] Fahrenkrug, J. and Schaffalitzky de Muckadell, D.B. (1977): Radioimmunoassay of vasoactrvc in, testinal polypeptide (VIP) in plasma. J. Lab. Clin. Med., 6, l379- 1388.

[8] Fleischer, A., Schulman, H., Farmakides, G., Bracero, L., Blattner, P. and Randolph, G. ( 198.5): Umbilical artery velocity waveforms and intrauterine growth retardation. Am. J. Obstet. Gynecol 151, 502-505.

[9] Gosling, R.G. and King, D.H. (1975): Ultrasound angiology. In: Arteries and Veins, pp. hi -98 Editors: A.W. Marcus and L. Adamson, Churchill Livingstone, Edinburgh.

(IO] Hansen, V., Maigaard, S., Allen, J. and Forman, A. (1988): Effects of vasoactive intestinal polypep- tide and substance P on human intramyometrial arteries and stem villous arteries in term preg- nancy. Placenta, 9, 501-506.

[I I] Ishitani, N, Ohzeki, T, Hanaki, K., Motozumi, H.. Sunaguchi, M. and Shiraki, K. (1993): Elevated venous concentrations of vasoactive intestinal polypeptide in cord blood of Infants with fetal dis- tress. Biol. Neonate, 63, 70-74.

[I21 Kawamura, K., Sakata, N. and Takebayashi, S. (1991): Neuropeptide Y- and vasoactive intestinal polypeptide-containing nerve fibers in the human cerebral arteries: characteristics of distribution Angiology, 21, 35-43.

[I31 Nicolaides, K.H., Economides, D.L. and Soothill, P. W. (1989): Blood gases, pH and lactate m appropriate- and small-for-gestational-age fetuses. Am. J. Obstet. Gynecol., 161, 996-1001,

1141 Ottesen, B., Ulrichsen, H., Fahrenkrug, J., Larsen, J.J., Wagner, G., Schierup, L. and Sondergaard, F. (1982): Vasoactive intestinal polypeptide and the female genital tract: relationship to reproduc- tive phase and delivery. Am. J. Obstet. Gynecol., 143. 414-420.

[I51 Ottensen, B. (1983): Vasoactive intestinal polypeptide as a neurotrasmitter in female genital tract Am. J. Obstet. Gynecol., 147, 208-224.

[I61 Ottesen, B., Hansen, B., Fahrenkrug, J. and Fuchs, A.R. (1984): Vasoactive intestinal peptide (VIP) stimulates oxytocin and vasopressin release from the neurohypophysis. Endocrinology, l ! 5. 1648-1650.

[I71 Rizzo, G. and Arduini, D. (1991): Fetal cardiac function in intrauterine growth retardation. Am. J. Obstet. Gynecol., 165, 876-882.

[I81 Rizzo. G., Arduini, D. and Romanini, C. (1992): Inferior vena cava llow velocity waveforms in appropriate- and small- for- gestational- age fetuses. Am. J. Obstet. Gynecol., 166, 127l- 1280.

[I91 Rudolph, A.M. (1987): Distribution and regulation of blood flow in the fetal and neonatal lamb Circ. Invest., 80, 1096-l 103.

[20] Wladimiroff, J.W., van den Wijngaard, J.A.G.W., Degani. S., Noordam, M.J. and Tonge, H.M. (1987): Cerebral and umbilical blood flow velocity waveforms in normal and growth retarded pregancies: a comparative study. Obstet. Gynecol.. 69. 705-709.