Fetal growth restriction (FGR) is an important cause of perinatal mortality and of short-term and long-term morbidity.1 Before the widespread use of obstetric ultrasound, FGR was a retrospective diagnosis based upon birth weight below the tenth or fifth percentile for gestation. However, up to half of infants with birth weights between the fifth and tenth percentiles for gestation are healthy small for gestational age (SGA) infants,2 and many placentas from histologic studies of FGR at term were in fact from normal pregnancies.
Ultrasound distinguishes between FGR and SGA based on fetal growth velocity, amniotic fluid volume, placental grading, and Doppler ultrasound of the umbilical arteries.3 In SGA fetuses, as in normal fetuses, umbilical blood flow rises during the third trimester and is accompanied by an increase in end-diastolic velocity in the Doppler waveform.4,5 The placental villous dysfunction in growth-restricted fetuses with abnormally low umbilical artery blood flow, indicated by absent or reverse end-diastolic flow, is characterized by reduced elaboration of gas-exchanging peripheral villi, which increases the risk of chronic fetal hypoxia and acidosis.6,7 Most growth-restricted fetuses have only mild abnormalities of umbilical artery blood flow, typified by low but positive end-diastolic flow velocity, and their perinatal outcome is much more favorable.8 Because these differences might be due to the quality of angiogenesis within placental villi,9 we compared placental vascular development in these two forms of FGR.
We studied the placentas from 18 consecutive singleton pregnancies delivered at the Department of Obstetrics and Gynecology of the University of Turin between June 1995 and January 1997. All of them were from SGA infants that were normal structurally and chromosomally. Small for gestational age was defined as eventual birth weight below the tenth percentile for gestational age according to Italian data.10 Each case was considered to be complicated by uteroplacental insufficiency based on bilateral notched high-impedance uterine artery waveforms.11 As such, each woman had developed preeclampsia12 by delivery, which was by cesarean before the onset of labor.
All Doppler studies were done by one of the authors (TT) using an Aloka SSD 680 machine (Aloka Company, Tokyo, Japan) with a 3.5-MHz probe, color-flow mapping, and a 100-Hz high-pass filter. The proximal uterine artery waveform was identified at the crossover point using color flow.11 The umbilical artery waveform was measured from a free-floating loop of cord during fetal quiescence. The last value measured before delivery (within 48 hours in all cases) was used for this study. Eight cases were identified with either absent or reverse end-diastolic flow velocity, and ten further cases had positive end-diastolic flow, but the pulsatility index was above the 90th percentile for gestation.13 The documentation of abnormal uterine and umbilical artery Doppler flow in each of these SGA cases was the basis of a diagnosis of FGR secondary to uteroplacental insufficiency.3 The control group consisted of the placentas of six pregnancies that were delivered preterm for premature rupture of membranes or preterm labor with intact membranes and nonvertex presentations. Predicted and actual birth weights exceeded the tenth percentile in each, and the amniotic fluid index and umbilical and uterine artery Doppler waveforms were normal in each case. In all cases and controls, gestational age was verified by ultrasound dating before 16 weeks' gestation.
After delivery, the placentas were trimmed of membranes and cords, fixed in neutral buffered 4% formaldehyde solution for 12–24 hours, and weighed. Three vertical full-thickness slices, including the chorionic and basal plates and excluding infarcts, were excised: one from the center, but at least 2 cm from the umbilical cord insertion; one from the periphery; and one from the intermediate part. The slices were embedded in paraffin. Sections (4 μm thick) were immunostained with a monoclonal antibody to alpha–smooth muscle actin to identify stem villi and their stem vessels, according to previously published methods,14 and the slides were counterstained with hematoxylin.
Morphometric analysis of the composition of villous trees was done with an image-analysis system (Zeiss, Jena, Germany) and a high-resolution videocamera. A 4 × 4-mm grid was constructed with 49 intersection points. Five fields from each slide (15 fields per placenta) were selected using a table of random numbers. For each placenta, the total amounts of fibrinoid, inter-villous space, and villous tissue were calculated from the grid using the point-counting method, and the percentages of sectional area occupied by each tissue were derived. In the same way, the percentages of the main components of the villous trees were determined, excluding fibrinoid and intervillous space. Stem villi and immature intermediate villi, both characterized by alpha–smooth muscle actin-positive vessel media, were distinguished from the gas-exchanging peripheral villi, which were negative for alpha–smooth muscle actin-positive vessel media. No distinction was made between mature intermediate and terminal villi, the two subtypes of peripheral gas-exchanging villi.
Using 15 fields per placenta, containing a mean of 800 stem arteries and arterioles, we determined the maximum diameter of each stem arterial vessel profile (lumen and actin-positive media), measured vertically to its longitudinal axis, to calculate a frequency distribution of arterial calibers. For each placenta, the mean vessel diameter was used as the summary statistic for intergroup comparisons, as described previously.14 The extent of branching of arteries and arterioles (branching index) was summarized as the percentage of stem villi containing two or more arterial or arteriolar profiles.
The following qualitative characteristics were studied: 1) the degree of trophoblastic flat-sectioning (so-called trophoblastic knotting), as a measure of villous surface deformation; 2) the prevalence of weblike arranged conglomerates of peripheral villi compared with the prevalence of small isolated peripheral villus cross-sections; and 3) the degree of capillarization of peripheral villi. All measurements were done by one of the authors (AS) who was blinded to the case histories and Doppler results.
The Kruskal-Wallis test was used to assess differences among groups. Differences between paired groups were assessed using the Dunn test. Differences were considered significant at P < .05.
Gestational age at delivery and birth weight were significantly lower in the group with absent or reverse end-diastolic flow (median 28 weeks, range 27–36; median 735 g, range 455–1740) compared with the group with positive end-diastolic flow (median 34 weeks, range 28–36, P = .02; median 1280 g, range 830–1740, P = .02). Placental weight at delivery was significantly lower in both study groups than in the control group, although placental-fetal weight ratios did not differ significantly among the three groups (absent or reverse end-diastolic flow: median 0.25, range 0.15–0.36; positive end-diastolic flow: median 0.20, range 0.15–0.30; control: median 0.25, range 0.22–0.56). There was one perinatal death in the control group, one in the positive end-diastolic group, and four in the absent or reverse end-diastolic flow group.
Table 1 summarizes the morphometric composition of the placentas and Table 2 shows the distribution of villous types in each group. The proportion of placenta occupied by villous tissue was significantly lower in the absent or reverse end-diastolic flow group than in the control group, although no such difference was found between the positive end-diastolic flow and control groups. Mean diameter measurements of stem arteries or arterioles did not differ significantly among the three groups (absent or reverse end-diastolic flow: median 43 μm, range 38–55; positive end-diastolic flow: median 39 μm, range 31–50; control: median 35 μm, range 30–44). There was a progressive trend toward reduced branching of the stem arteries from the control group (median 22%, range 2–38%) through the positive end-diastolic flow group (median 17%, range 11–20%) to the absent or reverse end-diastolic flow group (median 13%, range 4–23%). However, these differences were not significant (P = .06).
Peripheral villi in the placentas with positive end-diastolic flow were characterized by trophoblastic flat-sectioning (trophoblastic knotting), resulting in a web-like arrangement of villi (Figure 1a, b). These peripheral gas-exchanging villi contained numerous capillary cross-sections with multiple branching patterns, indicating a richly branching terminal capillary bed within. By contrast, the villous architecture in the group with absent or reverse end-diastolic flow was characterized by slender, elongated, poorly branched, and poorly capillarized peripheral villi (Figure 1c, d).
Villous vascularization during the first and second trimesters is largely determined by sprouting and branching angiogenesis, which leads to the formation of immature villous trees whose vascular core is characterized by continuously growing networks of fetal capillaries and their supply vessels.15 At the end of the second trimester, there is a switch to nonbranching angiogenesis, which forms long, poorly branched terminal capillary loops. As the third trimester progresses, these loops form the capillary network of the mature intermediate and terminal villi responsible for nutrient and gas exchange between mother and fetus.16 The net effect of these angiogenic phases is an exponential increase in the volume of fetal capillaries within the placenta, representing about one-quarter of the fetoplacental blood volume by term.17 This process coincides with the known increase in end-diastolic flow velocities (and a fall in the pulsatility index) in the umbilical artery, suggesting that these capillaries are an important determinant of fetoplacental vascular impedance in the normal placenta.4
In pregnancies complicated by FGR, with major reductions in umbilical artery blood flow (represented by absent or reverse end-diastolic flow in this study), a pattern of villous maldevelopment has been described in which the peripheral villi contain slender, unbranched, and uncoiled capillary loops.6,7 Our data from placentas with absent or reverse end-diastolic flow confirm these observations, but are of interest because the placentas from pregnancies complicated by FGR with positive end-diastolic flow in the umbilical artery before delivery show a completely different pattern of villous development. We found a netlike arrangement of capillaries, forming multiply branched terminal villi.15 These are the same characteristics observed in placentas derived from pregnancies complicated by preeclampsia at term,15 high altitude,18 or maternal anemia.19 Recently, similar findings were observed in SGA infants delivered at term, who showed abnormally increased end-diastolic velocities in the umbilical artery before delivery.20
Several angiogenic growth factors are expressed in human placental villi.9,21 One of the most intensely studied is vascular endothelial growth factor, the expression of which is stimulated by hypoxia in vitro and is known to mediate branching angiogenesis.22 Oxygen availability appears to be an important regulator of angiogenesis within the tissues,23 and chronic hypoxia in the guinea pig model was shown to increase fetoplacental capillarization24 by branching angiogenesis.25 It remains to be determined whether abnormalities of expression or function of placental growth factors, such as vascular endothelial growth factor, might be responsible for the alterations in villous structure seen in these two categories of growth-restricted fetuses.
Our observations suggest that pregnancies complicated by FGR in which positive end-diastolic flow was documented in the umbilical artery before delivery had an adaptive response in the villous placenta, characterized by enhanced branching angiogenesis. This results in greater numbers of highly branched terminal villi23 and compensates, at least in part, for the underlying impairment of uteroplacental blood flow indicated by abnormal bilateral uterine artery Doppler findings. By contrast, the pattern of villous vascularization in the FGR group complicated by absent or reverse end-diastolic flow suggested that this adaptive process failed to occur,6,7 which might be due to an apparent lack of placental villous up-regulation of vascular endothelial growth factor.
The vascular dysfunction within the villi of pregnancies with FGR is not confined to the peripheral villi.3 A reduction in the density of small-stem arteries was the first histologic abnormality reported in the placentas of pregnancies complicated by FGR with abnormal umbilical artery waveforms.26 In addition to vessel density, our present study addressed the extent of arterial branching within the stem villi. We found a nonsignificant trend toward reduced arterial branching in the placentas of pregnancies complicated by FGR. Neither our study nor that of Giles et al26 used systematic random block sampling to permit stereologic analysis. To our knowledge, Jackson et al27 are the only investigators who systematically sampled placentas of pregnancies complicated by FGR. These investigators found no change in stem artery density but found reduced arterial wall thickness in the stem villi of pregnancies with FGR and abnormal umbilical artery Doppler compared with gestational age-matched controls. Their study group, however, was a mixture of cases with positive and absent end-diastolic flow in the umbilical arteries. Further studies are needed to resolve the role of stem arteries in the etiology of FGR.
Stem villi result from the differentiation of immature intermediate villi,15 and the latter are characterized by richly developed capillary networks. Centrally placed capillaries differentiate to form muscularized stem vessels, whereas those in the periphery form the paravascular network, which regresses in later gestation.15 Hypovascularization of stem villi or reduced peripheral villous branching in placentas from growth-restricted fetuses with absent or reverse end-diastolic flow could result from abnormal differentiation of capillary networks in immature intermediate villi during early gestation, rather than from some obliterative process in the third trimester. If the former concept is true, the placental origins of FGR might begin much earlier in gestation than previously thought.
1. Murphy DJ, Sellers S, MacKenzie IZ, Yudkin PL, Johnson AM. Case-control study of antenatal and intrapartum risk factors for cerebral palsy in very preterm singleton babies. Lancet 1995;346:1449–54.
2. Gabbe SG. Intrauterine growth retardation. In: Gabbe SG, Niebyl JR, Simpson JL, eds. Obstetrics: Normal and problem pregnancies. 2nd ed. Edinburgh, United Kingdom: Churchill Livingstone, 1991: 923–44.
3. Kingdom JCP, Burrell SJ, Kaufmann P. Pathology and clinical implications of abnormal umbilical artery Doppler waveforms. Ultrasound Obstet Gynecol 1997;9:271–86.
4. Hendricks S, Sorensen TK, Wang KY, Bushnell JM, Seguin EM, Zingheim RW. Doppler umbilical artery waveform indices—normal values from fourteen to forty-two weeks. Am J Obstet Gynecol 1989;161:761–5.
5. Burke G, Stuart B, Crowley P, Scanaill SN, Drumm J. Is intrauterine growth retardation with normal umbilical artery blood flow a benign condition? BMJ 1990;300:1044–5.
6. Macara L, Kingdom JCP, Kaufmann P, Kohnen G, Hair J, More IAR, et al. Structural analysis of placental terminal villi from growth-restricted pregnancies with abnormal umbilical artery Doppler waveforms. Placenta 1996;17:37–48.
7. Krebs C, Macara LM, Leiser R, Bowman AW, Greer IA, Kingdom JCP. Intrauterine growth restriction with absent end-diastolic flow velocity in the umbilical artery is associated with maldevelopment of the placental terminal villous tree. Am J Obstet Gynecol 1996;75:1534–42.
8. Karsdorp VHM, van Vugt JMG, van Geijn HP, Kostense PJ, Arduini D, Montenegro N, et al. Clinical significance of absent or reversed end diastolic velocity waveforms in umbilical artery. Lancet 1994;344:1664–8.
9. Ahmed A, Kilby MD. Hypoxia or hyperoxia in placental insufficiency? Lancet 1997;350:826–7.
10. Parazzini F, Cortinovis I, Bortolus R, Fedele L. Standard di peso alla nascita in Italia. Ann Ostet Ginecol Med Perinat 1991;112:203–46.
11. Bower S, Bewley S, Campbell S. Improved prediction of preeclampsia by two-stage screening of uterine arteries using the early diastolic notch and color Doppler imaging. Obstet Gynecol 1993;82:78–83.
12. Davey DA, MacGillivray I. The classification and definition of the hypertensive disorders of pregnancy. Am J Obstet Gynecol 1988;158:892–8.
13. Todros T, Ronco G, Fianchino O, Rosso S, Gabrielli S, Valsecchi L, et al. Accuracy of the umbilical arteries Doppler flow velocity waveforms in detecting adverse perinatal outcomes in a high-risk population. Acta Obstet Gynecol Scand 1996;75:113–9.
14. Macara LM, Kingdom JCP, Kohnen G, Bowman AW, Greer IA, Kaufmann P. Elaboration of stem villous vessels in growth restricted pregnancies with abnormal umbilical artery Doppler waveforms. Br J Obstet Gynaecol 1995;101:807–12.
15. Benirschke K, Kaufmann P. Pathology of the human placenta. 3rd ed. New York: Springer Verlag, 1995.
16. Kaufmann P, Luckhardt M, Leiser R. Three dimensional representation of the fetal vessel system in the human placenta. Trophoblast Res 1988;3:113–37.
17. Luckhardt M, Leiser R, Kingdom J, Malek A, Sager R, Kaisig C, et al. Effect of physiologic perfusion-fixation on the morphometrically evaluated dimensions of the term placental cotyledon. J Soc Gynecol Invest 1996;3:166–71.
18. Jackson MR, Mayhew TM, Haas JD. Morphometric studies on villi in human term placentae and the effects of altitude, ethnic grouping and sex of newborn. Placenta 1987;8:487–95.
19. Kadirov M, Kosanke G, Kingdom J, Kaufmann P. Increased fetoplacental angiogenesis during the first trimester in anaemic women. Lancet (in press).
20. Hitschold T, Muentefering H, Ulrich S, Berle P. Does extremely low fetoplacental impedance as estimated by umbilical artery Doppler velocimetry also indicate fetuses at risk? Ultrasound Obstet Gynecol 1996;8(suppl 1):39.
21. Ahmed A, Li XF, Dunk C, Whittle MJ, Rushton DI, Rollason T. Colocalisation of vascular endothelial growth factor and its Flt-1 receptor in human placenta. Growth Factors 1995;12:235–43.
22. Cao Y, Linden P, Shima D, Browne F, Folkman J. In vivo angiogenic activity and hypoxia induction of heterodimers of placenta growth factor/vascular endothelial growth factor. J Clin Invest 1996;98:507–11.
23. Kingdom J, Kaufmann P. Oxygen and placental villous development: Origins of fetal hypoxia. Placenta 1997;18:613–21.
24. Bacon BJ, Gilbert RD, Kaufmann P, Smith AD, Trevino FT, Longo LD. Placental anatomy and diffusing capacity in guinea pigs following long-term maternal hypoxia. Placenta 1984;5:475–87.
25. Scheffen I, Kaufmann P, Philippens L, Leiser R, Geisen C, Mottaghy K. Alterations of the fetal capillary bed in the guinea pig placenta following long-term hypoxia. In: Piiper J, ed. Oxygen transport to tissue XII. New York: Plenum Press, 1990:779–90.
26. Giles WB, Trudinger BJ, Baird PJ. Fetal umbilical artery flow velocity waveforms and placental resistance: Pathological correlation. Br J Obstet Gynaecol 1985;92:31–8.
© 1999 The American College of Obstetricians and Gynecologists
27. Jackson MR, Walsh AJ, Morrow RJ, Mullen JB, Lye SJ, Ritchie JW. Reduced placental villous tree elaboration in small-for-gestational-age pregnancies: Relationship with umbilical artery Doppler waveforms. Am J Obstet Gynecol 1995;172:518–25.