Inferior vena cava flow velocity waveforms in appropriate- and small-for-gestational-age fetuses

Inferior vena cava flow velocity waveforms in appropriate- and small-for-gestational-age fetuses Giuseppe Rizzo, MD: Domenico Arduini, MD: and Carlo Romanini, MDb Rome and Ancona, Italy Reference ranges of inferior vena cava flow velocities were constructed from a cross-sectional study of 118 appropriate-for-gestational-age fetuses of 18 to 40 weeks of gestation. Blood flow velocity waveforms were recorded with color and pulsed Doppler equipment. Peak velocities and time velocity integrals were measured from inferior vena cava during systole, early diastole, and atrial contraction. The systolic-to-diastolic ratios between the peak velocities and time velocity integrals were calculated, and the reverse flow with atrial contraction was quantified as the percentage of forward flow (percentage of reverse flow). Recordings were also obtained from 79 small-for-gestational-age fetuses free of structural and chromosomal abnormalities, divided into three groups according to umbilical artery velocity waveforms: normal pulsatility index values (group 1, n = 26), pulsatility index >95th percentile of our reference limits but presence of end-diastolic velocities (group 2, n = 33), and absence of end-diastolic velocities (group 3, n = 20). Fourteen fetuses of groups 2 and 3 were also studied at weekly intervals until the onset of antepartum late heart rate decelerations. In appropriate-for-gestational-age fetuses no changes were evident in peak velocities and time velocity integrals ratios, whereas the percentage of reverse flow significantly decreased with gestation. No significant differences were found between these values and those obtained in small-for-gestational-age fetuses of group 1. A significant increase of peak velocities and time velocity integrals ratios and of percentage of reverse flow was evidenced in fetuses of both group 2 and 3. The fetuses of groups 2 and 3 with a percentage of reverse flow above the 95% confidence interval showed a poorer perinatal outcome when compared with the fetuses of the same groups but with values inside the normal range. In the 14 fetuses longitudinally followed up until the onset of late heart rate decelerations a significant and progressive increase of peak velocities and time velocity integrals ratios and percentage of reverse flow was evidenced in spite of minimal changes in the pulsatility index from both umbilical artery and different peripheral fetal vessels. This study presents evidence that in small-for-gestational-age fetuses with abnormal Doppler-measured placental resistance the modified flow velocity patterns in the inferior vena cava seem to deteriorate progressively with advancing gestation. (AM J OSSTET GVNECOL 1992;166:1271-80.)

Key words: Venous blood Aow, Doppler ultrasonography, fetal growth retardation Abnormal inferior vena cava Aow velocity waveforms have been described in different fetal pathologic conditions, including anemia,1 nonimmune hydrops." and arrythmias. 3 .• Furthermore, abnormalities of inferior vena cava waveforms, were also found in growth-retarded fetuses, characterized by extreme compromise as expressed by the absence of end diastolic Aow in umbilical artery.' There is no account, however, on the longitudinal changes occurring in inferior vena cava blood Aow pat-

terns with the progression of normal pregnancy and on their modifications in growth-retarded fetuses with different degrees of compromise. This study uses Doppler ultrasonography to establish reference ranges of velocity waveforms in the inferior vena cava of healthy fetuses from 18 to 40 weeks of gestation. Furthermore, it investigates possible modifications in the Aow velocity waveforms of small-forgestational-age (SGA) fetuses and relates the Doppler findings to the severity of fetal compromise.

Material and methods From the Laboratory of Fetal Physiology, Department of Obstetrics and Gynecology, Universita Cattolica S. ClIore," and the Department of Obstetrics and Gynecology, Universita di Ancona.' Supported by a grant of the Italian National Council of Research. Receivedfor publication june 13,1991; revised September 27,1991; accepted September 30,1991. Rpprint requests: D,'. Giuseppe Rizzo, 1st. Cl. Ostetrica e Ginecologica, Universita Cat/olica S. Cuore, Largo A. Gemelli. 8, 00168 Roma,

Italv. 6/j/J4060

Subjects. Reference ranges of inferior vena cava velocity waveforms were constructed from the cross-sectional study of 118 appropriate-for-gestational-age (AGA) fetuses (i.e., birth weight ranging between the 10th and 90th percentile). These fetuses were recruited from the routine antenatal clinic of our institution. All pregnancies were singleton and dated by certain last menstrual period and early second-trimester ultraso1271

1272

Rizzo, Arduini, and Romanini

April 1992

Am

J Obstet Gynecol

Table I. Characteristics of fetuses investigated (mean ± I SD or number and percent) SGA AGA (n

Gestational age at Doppler recordings (wk) Umbilical artery pulsatility index (~ value) Descending aorta pulsatility index Renal artery pulsatility index (~-value) I nternal carotid artery

=

118)

I

Group 2 (n = 33)

I

Group 3 (n = 20)

AGA vs group 1

AGA vs I group 2

29.6 ± 5.1

30.3 ± 2.6

31.0 ± 2.5

28.3 ± 5.2

NS

-0.12 ± 0.99

0.33 ± 0.94

3.04 ± 1.13

4.12 ± 1.62

NS

Ps

0.001

\.94 ± 0.21

1.97 ± 0.26

2.46 ± 0.41

2.64 ± 0.55

NS

Ps

0.001

0.03 ± 0.83

0.21

2.24

1.45

3.19 ± 1.84

NS

Ps

0.001

:!:

0.96

:!:

NS

0.87

0.27 ± 0.75

-2.67 ± 1.44

-3.19 ± 1.46

NS

Ps

0.001

0.12 ± 0.88

0.31 ± 0.92

-2.95 ± 1.21

-3.32 ± 1.33

NS

Ps

0.001

39.7 ± 1.9

38.4 ± 2.5

34.8 ± 3.2

32.1 ± 4.2

NS

Ps

0.001

3654 ± 412

2320 ± 316 7.25 I ± 0.029 0(0)

1635 :!: 521 7.223 ± 0.034 13 (39.4)

1135 ± 363 7.188 ± 0.042 14 (70.0)

p s 0.001

Ps

0.001

NS

Ps

0.001

3 (11.5) 0(0) 0(0)

28 (84.5) 9 (27.2) 1(3.0)

20 (100) 9 (45.0) 4 (20.0)

NS NS NS

Ps Ps

0.001 0.001 NS

-0.06

:!:

(~-value)

Middle cerebral artery pulsatility (~-value) Gestational age at birth (wk) Birth weight (gm) Umbilical artery pH Antepartum late heart rate decelerations (%) Cesarean section (%) 5 min Apgar score <7 Neonatal mortality

Group 1 (n = 26)

Significance*

0(0) 9 (7.6) 0(0) 0(0)

Because pulsatility index values obtained from umbilical artery, renal artery, internal carotid artery, and middle cerebral artery change wtih gestation, data are expressed as number of standard deviations by which they differ from their normal mean for gestation (~-values). NS, Not significant. *Statistical significance was evaluated by unpaired t test and Fisher's exact test.

nography. Ultrasonographic examination at the time of Doppler studies showed all the fetuses to be structurally normal and of appropriate size for gestational age. Infants were born in good condition; their characteristics are reported in Table I. Inferior vena cava velocity waveforms were also studied in 79 SGA fetuses. Entry criteria required (I) singleton gestation accurately dated by early secondtrimester fetal biparietal diameter mesurement and (2) abdominal circumference <5th percentile of our reference limits (fetuses with a gestational age <28 weeks) or ultrasonographically estimated fetal weight (fetuses with a gestational age ~28 weeks) <5th percentile for our population. The diagnosis ofSGA (i.e., birth weight <5th percentile and newborn free of structural abnormalities) was confirmed postnatally in all included subjects. SGA fetuses were divided into three groups according to the results obtained at the Doppler examination of the umbilical artery as follows: (I) group I, normal pulsatility index in the umbilical artery with respect to our reference limits for gestation' (n = 26); (2) group 2, abnormal pulsatility index in the umbilical artery (i.e., >95th percentile) but velocity waveforms characterized by the presence of end-diastolic velocities (n = 33); (3) group 3, abnormal pulsatility index in the umbilical artery and absent end-diastolic velocities (n = 20) (none of these fetuses had reversed flow in the umbilical artery).

The characteristics of these three groups of fetuses are reported in Table I. Although umbilical artery pulsatility index values were provided to the clinicians managing the cases, decisions on delivery were not undertaken on the basis of Doppler data. Some of the SGA fetuses with an abnormal pulsatility index in the umbilical artery were also studied longitudinally at weekly intervals until delivery. To obtain a homogeneous group, we retained the data of the 14 fetuses (10 of group 2 and four of group 3) in whom antepartum late heart rate decelerations developed and who had been studied on at least four occasions I week apart (longitudinal study). In these fetuses heart rate monitoring was performed daily for a minimum of 45 minutes from the time of diagnosis onward. At the onset of late decelerations during Braxton Hicks contractions, an elective cesarean section was performed. All the patients gave their informed consent to participate in the study. Doppler recordings. Commercially available Ansaldo Hitachi Esacord 81 color Doppler equipment (Ansaldo Hitachi, Genoa, Italy) with 3.5 or 5 MHz convex probes was used for all measurements. In addition to the color flow mapping function, the machine was equipped with pulsed Doppler that had a carrier frequency ranging from 2.5 to 5 MHz. After identification of the inferior vena cava in a longitudinal plane on a sagittal view of the fetal trunk, the color flow mapping

Inferior vena cava flow velocity waveforms

Volume 166 !\:umber 4

1273

Significance*

AGA vs group 3

Group I vs group 2

Group I tiS Group 3

Group 2 vs group 3

NS

NS

NS

NS

P s 0.001

P s 0.001

P s 0.001

P s 0.001

P s 0.001

P s 0.001

P s 0.001

NS

P s 0.001

P s 0.001

P s 0.001

P s 0.05

P s 0.001

P s 0.001

Ps 0.001

NS

P s 0.001

P s 0.001

P s 0.001

NS

P s 0.00)

P s 0.001

P S 0.001

P S 0.05

P S 0.001

P S 0.001 P S 0.001 P S 0.002

P S 0.001 P S 0.001 P S 0.001

P S 0.00) P S 0.005 P S 0.01

P S 0.00) P S 0.005

P S 0.001 P S 0.002 P S 0.05

NS NS NS

P S 0.001 P S 0.001 P S 0.001 P S 0.005

NS

function was superimposed. On the basis of previously reported techniques in the first 70 cases (45 AGA and 25 SCA) the pulsed Doppler sample volume was placed both at the inlet of inferior vena cava in the right atrium' and in the portion of inferior vena cava between the entrance of the renal vein and the ductus venosus!· ' A good coefficient of correlation was found between the measurements performed at the two sampling sites (r> 0.84). The latter sampling site was selected for the further examinations because of the significantly lower angle (mean ± I SD 21 ± 7 vs 39 ± 6 degrees, p s 0.00 I) between the Doppler beam and the direction of blood flow, as estimated on the color image. Flow velocity waveforms were recorded from each fetus during periods of fetal rest without breathing movements. All the recordings were performed by one of the authors (C.R.). Permanent records of all the velocity waveforms were obtained from the videotape by means of a strip chart recorder. The printouts were labeled with random numbers. The observer involved in the later analysis was not informed of the sequence of the printouts. Three components were identified in velocity waveforms as described by Reed et a1. 3 : (J) systole (S wave), (2) early diastole (0 wave), and (3) reverse flow during atrial contraction (A wave) (Figs. 1 and 2). Estimates of the peak velocities and time velocity integrals of the three components were made on 10 uniform and consecutive heart cycles with the digitizing tablet of a computer (Cardio 800, Kontron, Oxford). The following indices were calculated from these measurements: (I)

Fig. 1.. Flow velocity waveforms from fetal inferior vena cava in AGA fetuses at 20 (toP) and 39 (bottom) weeks of gestation with pe rcentage of reverse fl ow (boUom channel) of 18.5 in form e r and 4.3 in latter.

ratio between the peak velocity of Sand D waves, (2) ratio between time velocity intervals of Sand D waves, and (3) percentage of reverse flow quantified as the percent of time velocity intervals during A wave with respect to total forward time velocity intervals. Intraobserver variability was determined by examining 19 fetuses for two consecutive days. The maximum difference in the indices considered in the same fetus was 8.9% (mean 5.4 %). In all the fetuses studied blood flow velocity waveforms were also recorded at the level of the umbilical artery, middle cerebral artery, descending aorta, and renal artery with previously described techniques.' the pulsatility index was calculated as a measure of vascular impedance. 6

1274 Rizzo, Arduini, and Romanini

April 1992 Am J Obslel Gynecol

Fig. 2. Flow velocity waveforms from fetal inferior vena cava in SGA fetus of group 3 at 29 weeks of gestation with percentage of reverse flow (top channel) of 38.3.

Data analysis. Regression analysis was used to establish the mean and individual 95% confidence intervals of the relationship of the indices considered and gestation. Data are presented as mean ± 1 SD. For measurements changing with gestation data are also expressed as number of standard deviations by which they differ from their normal mean for gestation (d-values). Differences were evaluated by means of one sample and unpaired t test, Wilcoxon rank-sum test, and Fisher's exact test. The longitudinal changes in Doppler indexes were evaluated by the analysis of variance for repeated measurements. Regression analysis was used to test the relationships between the variables considered. Differences yielding p s 0.05 were considered significant. Results AGA fetuses. These fetuses showed a slight but not significant linear increase with gestation of the peak velocity SID ratio (Fig. 3, r = 0.191, P = 0.07, mean value 1.757, SD 0.334). Similarly, no significant changes with gestation were found for the SID time velocity intervals ratio (Fig. 3, r = 0.091, mean value 2.612, SD 0.997), whereas the percentage of reverse flow significantly and linearly decreased with gestation (Fig. 3, constant 24.671, slope -0.499, SD 2.557, r = 0.775, p::s; 0.001). SGA fetuses, group 1. No significant differences were found in peak velocity SID ratio (mean value = 1.63, SD 0.32, t = 1.737, not significant), SID

time velocity intervals ratio (mean value 2.62, SD 0.88, t = 0.146, not significant) and percentage of reverse flow (mean value 8.26, SD 2.41, d-value 0.09 SD, t = 0.474, not significant) when the values obtained in SGA fetuses with normal umbilical artery pulsatility index values were compared with those of AGA fetuses (Fig. 4). SGA fetuses, group 2. A significant increase in peak velocity SID ratio (mean value \.945, SD 0.45, t = 2.514, P s 0.01), SID time velocity intervals ratio (mean value 3.05, SD 1.51, t = 1.934, P sO.05), and percentage of reverse flow (mean value 11.88, SD 6.04, d-value 1.36 SD; t = 3.125, P s 0.005) was found in this group of SGA fetuses when compared with AGA fetuses (Fig. 5). No relationships were evident between these indices and the pulsatility index values (absolute values and d-values) from umbilical artery, descending aorta, renal artery, and middle cerebral artery. Moreover, when the nine SGA fetuses with percentage of reverse flow values >95% confidence interval were compared with the remaining 24 SGA fetuses, a higher incidence of antepartum late heart rate decelerations (719 vs 6/24, Fisher's exact test p s 0.0 I), a shortertime interval between the recordings and delivery (median 11 days, range 3 to 21 vs 32, range 8 to 54, Wilcoxon test p::s; 0.01), and a lower umbilical artery pH (7.181 ± 0.022 vs 7.228 ± 0.031, p::s; 0.001) were evident. In addition, no differences were found between these two subgroups of fetuses either in birth weight (1442 ± 576 vs 1716 ± 354 gm, not significant) or in

Inferior vena cava flow velocity waveforms

Volume 166 Number 4

1275

3,5 3 0

"'U')

2,5

> CL.

2

U

1,5

>

•

•

• • • • • • •• • --. •• • • • • • . •• • • • • .. ..• , . • • • •

•

• •

t

• ••• •

• ,•

•

•

• • .-!.• • • • • •

,•

• • •• , • •• •• • • • • • • •

• ·•• , • •

0,5 0 18

22

20

24

26

28

30

32

34

36

38

40

gestot1onol oge (weeks) 7

>

.... 0

"'-

> .... U')

u >

6 5 4 3 2

• •

•

•

• • • • • • • • • • • • •• • • • • • • • • • • •• • , • • • •• • • • • • • • • • •• • • • • t •t • •• • •

,• • ,

··,

•

*

• • •• I•

• •

• •

.. • •

*• • • • • •• • • •

0 18

20

22

24

26

28

30

32

34

36

38

40

gestotionol oge (weeks) ~

25

c

00-

c»

20

...c»til

15

...

10

> c»

M

u

>

5 0

18

20

22

24

26

28

30

32

34

36

38

40

gestotionol oge (weeks) Fig. 3. Reference range (mean and linear regressions of individual 95% confidence intervals) with gestation of inferior vena cava (lVC) SI D ratios of peak velocities (PV S I D) and time velocity integrals (STVI I DTVI) and of percentage of reverse flow during atrial contraction.

pulsatility index values obtained from the different arterial districts considered. Moreover, the single case of neonatal death (intraventricular hemorrhage) that occurred in this group had a percentage of reverse Row value >95% confidence interval. SGA fetuses, group 3. Highly significant differences were found in the comparison of SGA fetuses with AGA

fetuses for peak velocities SID ratio (mean value 2.39, SD 1.2 1, t = 4.721, p:5 0.001), SID time velocity intervals ratio (mean value 4.18, SD 2.84; t = 5.294, P :5 0.00 I), and percentage of reverse Row (mean value 21.86, SD 10.53, .:l-value 5.06 SD, t = 14.754, P :50.001) values (Fig. 5). Similarly, with group 2 fetuses no relationships were evident between inferior vena

1276

Rizzo, Arduini, and Romanini

April 1992 Am j Obstet Gynecol

grol4> 1 3,5 3

0

"V'I

2 ,5

> Q.

•• 1

1,5

u

>

•

•

•• •• •• I • ··1 ••• •

2 1 0,5 0 18

20

22

24

26

28

30

32

34

36

38

40

gestational age (weeks)

7

>

~

0

6 5

•• • • ·-;:1;;

•

"- 4

>

~

V'I

u

>

3

I

•

2 0

~

-....

18

20

22

18

20

22

24

26

28

30

24

26

28

30

32

36

34

38

40

38

40

gestational age (weeks)

25

0

20

<1l III L. <1l

15

<1l L.

10

~

5

>

u

>

0 32

34

36

gestational age (weeks)

Fig. 4. Inferior vena cava (/VC) peak velocities (PV) SID and SID time velocity intervals ratios (STV/ I DTV/) and percentage of reverse flow values of 33 SGA fetuses of group I (normal umbilical artery pulsatility index) plotted on reference range.

cava indices and pulsatility index values from arterial beds. In 14 fetuses of this group the percentage of reverse flow was >95th confidence interval, whereas the remaining six fetuses showed values inside the reference limits. When these two subgroups were compared, fetuses with abnormal percentage of reverse flow values showed a shorter interval between the recordings and delivery (median 5 days, range 0 to 11 vs

26, range 4 to 36, Wilcoxon test p :5 0.01) and a higher incidence of antepartum late heart rate decelerations (12114 and 2/6, Fisher's exact test p :5 0.03). Three of the four neonatal deaths occurred in the fetuses with abnormal percentage of reverse flow. No further significant differences were noted between the two subgroups. Longitudinal study. Each SGA fetus longitudinally

Inferior vena cava flow velocity waveforms

Volume 166 Number 4

5 0

4

> Q.

3

u

2

....c.n >

grolC)2

=

5

....0c.n

•

•• • •

• I

>

II.:.:;

Q.

I·· ...

4 3

u 2

>

9

9

6

6

> 7

> 7

....c.n> u

>

5 4 3 2

••• • ••• • • • •••••• •• ••• • •• • •••

....

....06 :;5

ti;4 u3

>

-2 1

0

••• ••

•..1 • • • • • • •

0 16 20 22 24 26 26 30 32 34 36 36 40

16 20 22 24 26 28 30 32 34 36 38 40

gestational age (weeks)

50 ~ 45 0 40 35 Il) (J) 30 I.. Il) 25 > Il) L. 20 M 15 u > 10 5 0

•• • •• ••

gestational age (weeks)

gestational age (weeks)

6

•

16 20 22 24 26 26 30 32 34 36 36 40

16 20 22 24 26 26 30 32 34 36 36 40

....

•• ••

•

0

0

....0

grolC»)

1277

gestational age (weeks)

•

50

-

45 ;:. 40 ~

....

••

•

....

35 ~ 30 Il) > 25 Il) I.. 20 M 15 Il)

~

16 20 22 2426 26 30 32 34 36 36 40

gestational age (weeks)

••

••

10 5 0 16 20 22 24 26 26 30 32 34 36 36

gestational age (weeks)

Fig. 5. Inferior vena cava (/VC) peak velocities (PV) SID and SID time velocity intervals ratios (STVI I DTVI) and percentage of reverse flow values of SGA fetuses of group 2 (increased umbilical artery pulsatility index and end-diastolic velocities present, left side) and group 3 (increased umbilical artery pulsatility index and absent end-diastolic velocities, right side) plotted on reference range.

studied underwent a mean of 4.71 Doppler recordings (range 4 to 7). Gestational age at the first study was 28.3 ± 2.5 and 32.1 ± 2.3 weeks at birth. Mean birth weight was 1379 ± 342 gm and mean umbilical artery pH was 7.218 ± 0.041. Although two of the fetuses of

group 2 developed absent end-diastolic velocities in the umbilical artery, minimal overall changes were evidenced in pulsatility index from the umbilical artery (F = 1.912, P s 0.05) and middle cerebral artery (F = 1.893, P < 0.01). No significant changes were

1278 Rizzo, Arduini, and Romanini

April 1992 Am J Obstet Cyoecol

5

o ..... IJ')

4

> a..

3

u

2

>

o 24

26

28

30

32

34

36

34

36

34

36

gestational age (weeks)

>

~

o

"-

>

~

IJ')

u

>

9 8 7 6 5 4 3 2

o 24

26

28

30

32

gestational age (weeks) ~

.... .... (;)

(\) II)

L..

(\)

> (\) L..

35 30

25 20 15

M

10

>

5

u

o 24

26

28

30

32

gestational age (weeks)

Fig. 6. Individual changes of peak velocities (PV) SI D and SID time velocity intervals ratios (STIlI I DTIII) and percentage of reverse flow values in 14 SGA fetuses studied longitudinally plotted on reference ranges. Values of each individual patient are connected by lines. Open diamonds. Absent end-diastolic velocities in umbilical artery, IVC, inferior vena cava.

found in descending aorta (F = 0.348, not significant;) and renal artery (F = 0.879, not significant) pulsatility index values. When analysis of variance was applied to inferior vena cava indexes a significant increase of peak velocity SID ratio (F = 6.123, P ~ 0 .001), the S time velocity intervals/D time velocity intervals ratio (F = 7.423, P ~ 0.00 I), and the percentage of reverse flow (F = 14.23, P ~ 0.001) was evident with advancing gestation (Fig. 6). No significant relationships were

noted between these changes in inferior vena cava and the concomitant modifications of the pulsatility index from the umbilical artery and the middle cerebral artery.

Comment Flow velocity patterns in the inferior vena cava are inversely related to the modifications of right atrial presure.' In the human fetus the inferior vena cava blood

Volume 166 Number 4

How has a pulsatile pattern similar to that described in animal fetuses or in adults." 8 In these models it has been demonstrated that the first forward wave begins to increase with atrial relaxation, reaches the peak during ventricular systole, and then falls to reach the nadir at the end of ventricular systole. The second forward wave occurs during early diastole, whereas a reverse How is usually present in late diastole with atrial contraction. Simultaneous recordings of the inferior vena cava and descending aorta velocity waveforms have also allowed confirmation of the timing of these events in the human fetus.' In normal pregnancy the peak velocities and the time velocity intervals values obtained during systole in the inferior vena cava exceeded the early diastolic values resulting in ratios > 1. The absence of changes of the systolic-to-diastolic ratios with gestation suggests a constant filling pattern during systole and early diastole. Moreover, a significant decrease of reverse How during atrial contraction was evident with advancing gestation. In the fetal lamb and in adults (animals and humans) the percentage of reverse How is related to the pressure gradient present between the right atrium and the right ventricle at end diastole, which is related both to ventricular compliance and to ventricular end-diastolic pressure." 9,10 The noninvasive nature of the human fetal model does not allow differentiation between these factors, but it can be suggested that both may significantly contribute to the decrease of reverse flow evidenced throughout gestation. Previous studies have demonstrated that the fetal heart is less compliant than the neonatal heart ll and that a progressive maturation of Doppler-measured indices of ventricular diastolic performance occurs, thus suggesting an improvement of ventricular compliance with advancing gestation. 12, I' Furthermore, as a consequence of the reduction of placental resistances with gestation' it may be suggested that the right ventricle afterload falls. This might result in decreased residual volume and ventricular end-diastolic pressure. SGA fetuses are a heterogeneous group that includes fetuses that are small because of racial or genetic factors and those that are growth retarded as result of malnutrition caused by utero placental insufficiency. The good outcome of the SGA fetuses belonging to group 1 and the normality of the pulsatility index in fetal arterial peripheral vessels make the former etiology for these fetuses likely. The absence of evident modifications in inferior vena cava flow patterns suggests normal cardiac hemodynamics. In particular, the finding of values of percentage of reverse flow similar to those of fetuses matched for gestational age but of larger size is consistent with a normal development of the ventricular filling properties in spite of the reduced growth. Both the selection criteria and pregnancy outcome

Inferior vena cava flow velocity waveforms

1279

strongly suggest that the growth defect of the fetuses of groups 2 and 3 resulted from uteroplacental insufficiency. These fetuses showed an increase of peak velocities and time velocity integral systolic-to-diastolic ratios associated with an increased percentage of reverse flow. These modifications are similar to those induced by hypoxia in instrumented lamb fetuses' and confirm recent observations in growth retarded human fetuses characterized by absent end-diastolic velocities in the umbilical artery.' These changes are compatible with an impaired cardiac function that may be secondary to different reasons, including a decreased ventricular compliance,13 a higher end-diastolic ventricular pressure caused by an increased afterload consequent to the increased peripheral and placental resistances,14, 15 or a reduced right myocardial contractility. 16, I' As a consequence of these abnormal How patterns, the return of blood from the placenta to the heart is impaired, thus further reducing the supply of oxygen and nutrients to fetal tissues. This may explain our data on the different outcome of fetuses with similar modifications in arterial vascular districts but different venous flow abnormalities. This hypothesis is supported by the data of Indik et al. 18 who found a different trend in perinatal mortality in fetuses with absent end-diastolic flow in the umbilical artery in the presence of pulsation in the umbilical vein, a situation closely associated with increased value of percentage of reverse flow. It is noteworthy that in our series an increased reverse flow is present also in approximately 27% of the SGA fetuses of group 2 (presence of end-diastolic velocities in the umbilical artery). In spite of the absence of any differences in pulsatility index values from various different arterial vascular beds or in fetal smallness these fetuses had a higher incidence of antepartum late heart rate decelerations, a shorter time interval between the recordings and delivery, and lower umbilical artery pH values. These findings suggest that among growthretarded fetuses with uteroplacental insufficiency the study of inferior vena cava velocity waveforms might allow better identification than that with conventional measurements in arterial peripheral vessels, a subgroup of fetuses with a more severe degree of fetal compromise. The lack of correlation between Doppler-measured arterial resistances and inferior vena cava flow patterns suggests that the degree of cardiac dysfunction necessary to yield abnormal venous patterns in the inferior vena cava is not directly related to arterial vascular resistances but may depend on different factors such as the temporal duration of the hemodynamic alterations. Our longitudinal study allowed us to validate this concept. In spite of the minimal changes in the pulsatility index values in arterial vascular districts there was a progressive deterioration of the inferior vena cava indices considered in the time interval elapsing between

1280 Rizzo, Arduini, and Romanini

the diagnosis to the onset of abnormal fetal heart rate patterns suggestive of abnormal fetal oxygenation. 19 These findings are compatible with the fall of both cardiac output and aortic and pulmonary peak velocities described in the growth-retarded fetuses longitudinally followed. 20 All these changes may be expressions of the same phenomenon, namely, a progressive deterioration of cardiac function. Moreover, the advantages of recording inferior vena cava velocity waveforms must be pointed out. Unlike echocardiographic measurements, the indices of inferior vena cava used are angle independent and do not need calculation of vessel area, thus allowing more reproducible and easier recordings. In conclusion, abnormal flow patterns in the inferior vena cava are present in SGA fetuses as a result of utero placental insufficiency. These patterns are unrelated to the abnormalities present in Doppler-measured vascular resistances in fetal arterial vascular beds. However, the noninvasive nature of the human fetal model does not allow for full clarification of the underlying pathophysiologic condition in these abnormal venous velocity patterns. The presence of increased reverse flow in the inferior vena cava in spite of the presence of end-diastolic flow in the umbilical artey may be a warning sign. In fetuses followed up longitudinally until the onset of spontaneous late heart rate decelerations a progressive deviation from the norm of inferior vena cava flow patterns was evidenced, thus suggesting their potential role in the longitudinal monitoring of growthretarded fetuses. However, an evaluation by means of decision-making techniques (e.g., analysis of receiveroperator characteristic curves) on a larger number of fetuses is required before inferior vena cava flow is used as a c1inica:I test. REFERENCES I. Rightmire DA, Nicolaides KH, Rodeck CH, Campbell S.

Fetal blood velocities in Rh isoimmunization: relationship to gestational age and fetal hematocrit. Obstet Gynecol 1986;68:233-6. 2. Gudmundsson S, Huhta JC, Wood DC, Tulzer G, Cohen AW, Weiner S. Venous Doppler ultrasonography in the fetus with nonimmune hydrops. AM J OilSTET GVNECOL 1991; 164:33-7. 3. Reed KL, Appleton CP, Anderson CF, Shenker L, Sahn DJ. Doppler studies of vena cava flows in human fetusesinsights into normal and abnormal cardiac physiology. Circulation 1990;81 :498-505.

April 1992 Am J Obstet Gynecol

4. Chan FY, Woo SK, Ghosh A, Tang M, Lam C. Prenatal diagnosis of congenital fetal arrhythmias by simultaneous Doppler velocimetry of the fetal abdominal aorta and inferior vena cava. Obstet Gynecol 1990;76:200-5. 5. Arduini D, Rizzo G. Normal values of Pulsatility Index from fetal vessels: a cross sectional study on 1556 healthy fetuses. J Perinat Med 1990; 18: 165-72. 6. Gosling RG, King DH. Ultrasound angiology. In: Marcus AW, Adamson L, eds. Arteries and veins. Edinburgh: Churchill Livingstone, 1975:61-98. 7. 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 GVNECOL 1983;145:70-8. 8. Appleton CP, Hade LK, Popp RL. Superior vena cava and hepatic vein Doppler echocardiography in healthy adults. J Am Coil Cardiol 1987;10:1032-9. 9. Wexler L, Berger DH, Gabe IT, Makin GS, Mills CJ. Velocity of blood flow in normal human venae cavae. Circ Res 1968;23:349-59. 10. Brawley RK, Aldham NH, Vasko SS, Henney PR, Morrow AG. Influence of right atrial pressure on instantaneous vena caval blood flow. Am J Physiol 1966;211 :347-53. 11. Romero T, Covell J, Friedman WF. A comparison of pressure-volume relations of the fetal, newborn and adult heart. Am J Physiol 1972;222: 1285-90. 12. Reed KL, Sahn DJ, Scagnelli S, Anderson CF, Shenker L. Doppler echocardiographic studies of diastolic function in the human fetal heart: changes during gestation. J Am Coli Cardiol 1986;8:391-5. 13. Rizzo G, Arduini D, Romanini C, Mancuso S. Doppler echocardiographic assessment of atrioventricular velocity waveforms in normal and small for gestational age fetuses. Sr J Obstet Gynaecol 1988;95:65-9. 14. AI-Ghazali W, Chita SK, Chapman MG, Allan LD. Evidence of redistribution of cardiac output in asymmetrical growth retardation. Br J Obstet Gynaecol 1989;96:697704. 15. Rizzo G, Arduini D, Romanini C, Mancuso S. Doppler echocardiographic evaluation of time to peak velocity in the aorta and pulmonary artery of small for gestational age fetuses. Sr J Obstet Gynaecol 1990;97:603-7. 16. De Vore GR. Examination of the fetal heart in the fetus with intrauterine growth retardation using M-mode echocardiography. Semin Perinatol 1988; 12:66-79. 17. Rasanen J, Kirkinen P, Jouppila P. Right ventricular dysfunction in human fetal compromise. AM J OBSTET CvNECOL 1989;161:136-40. 18. lndik JH, Reed KL. Umbilical venous pulsations are associated with inferior vena cava flow velocities [Abstract 310]. AMJ OBSTETCVNECOL 1991;164:330. 19. Visser CHA, Sadovsky G, Nicolaides KH. Antepartum heart rate patterns in small-for-gestational-age third trimester fetuses: correlation with blood gas values obtained at cordocentesis. AM J OBSTET CVNECOL 1990; 162:698-703. 20. Rizzo C, Arduini D. Fetal cardiac function in intrauterine growth retardation. AMJ OBSTET GVNECOL 1991; 165:87682.