Phcenta (1989), IO, 297-307

Villous Oedema of the Placenta:

A Clinicopathological Study

SUSAN SHEN-SCHWARZ”,“, EDUARDO

RUCHELLI” & DAVID BROWNb

Departments of Pathology’ and Pediatrics”, Magee-Wornens Hospital and

University of Pittsburgh Schooi of Medicine, Pittsburgh, Pennsylvanra,
U.S.A.

‘ To whom all correspondence should be addressed: Department of

Pathology, Children’s Memorial Hospital, 2300 Children’s Plaza,
Chicago, Illinois 60614, U.S.A.

Paper accepted m.12.1988.

INTRODUCTION

Villous oedema, the accumulation of fluid in the stroma of the chorionic villi, is a poorly under-
stood entity. To recognize it histopathologically is often difficult since interstitial fluid is a
normal constituent of the mesenchymal stroma in an immature villus. In contrast to oedema in
other organ systems, the pathophysiology of villous oedema in the placenta remains to be elu-
cidated (Fox, 1986).
The functional importance of villous oedema in relation to fetomaternal exchange is open to
speculation since such disturbances have not been documented by haemodynamic studies. A
histomorphometric study of villous oedema in term placentae (Alvarez, Sala and Benedetti,
1972) showed reduction of the intervillous space and a decreased number of chorionic villi per
unit area; the authors suggested that villous oedema might affect maternal intervillous blood
flow and fetal oxygen supply, and therefore could be a cause of the higher perinatal mortality
rate in these pregnancies.
Naeye et al (1983) reported that in placentae examined for pregnancy complications, villous
oedema was more frequent and more severe before 32 weeks gestation. He found that this
lesion strongly correlated with antenatal hypoxia in premature infants, and had a high predict-
ive value for neonatal deaths. From twin placenta studies, he contended that the factors
causing villous oedema were fetal in origin. He postulated that the oedema could compress and
obstruct villous capillaries, and that fetal hypoxia could occur as a result of reduced feto-
placental blood flow (Naeye et al, 1983; Naeye, r987b).
To date, there are no other published studies on the clinical effects of this condition. The
purpose of our study is to determine the incidence of villous oedema in the general obstetric
population, and to analyse the relationship of villous oedema to pregnancy complications, fetal
outcome, chorioamnionitis, and other placental pathology.

MATERIALS AND METHODS

In a prospective study to obtain clinicopathological data for placentae in our patient popula-

0143~4004/89/030297 + I I $05.00/o Ci:)1989 Baillikre Tindall Ltd

298 Placenta (1989), Vol. 10

tion, we examined consecutive placentae from singleton pregnancies of > 19 weeks gestation
for 3 months in 1984. Except for patient identification and date of delivery, no clinical informa-
tion was provided at the time of gross and microscopic examination. The placentae were exam-
ined while fresh and findings on umbilical cord, amnion, chorion and chorionic disc were
recorded on a standard data sheet. A 2 cm strip of chorionic disc, including insertion of the
umbilical cord, was sectioned across the greatest diameter of the disc and placed in IO per cent
buffered formalin overnight. Two sections, each of which included the centre of a placental
lobule, chorionic plate and decidual floor, were taken for histology. One section was taken close
to the site of umbilical cord insertion and the other section was taken midway between cord
insertion and the margin of the placenta. Gross abnormalities of the placental disc were separ-
ately submitted for microscopic examination.
Histologic findings of all placentae were recorded on a standard data sheet. Clinical informa-
tion which included maternal age, gravidity, parity, pregnancy, labour and delivery complica-
tions, gestational age, baby’s sex, birth weight, Apgar scores at I and 5 min, oxygen and
ventilation requirements, and immediate neonatal outcome was subsequently obtained from
delivery room and medical records. Values for clinical and pathological variables were entered
into a data base management program.
Chorioamnionitis was graded microscopically into mild, moderate and severe (Naeye et al,
1983). Criteria for the diagnosis of villous oedema were villous swelling and empty spaces
(microcystic change) in the villous stroma (Naeye et al, 1983). Reactive Hofbauer cells which
had vacuolated cytoplasm and stellate cell borders were often found within the empty spaces. A
second pathologist (ER) reviewed and classified all the placentae with villous oedema. The
extent of the oedema was determined semiquantitatively as the percentage of oedematous villi
present in the two histologic sections for each placenta. The severity of oedema was classified
either as class I, if oedema involved < 50 per cent of the cross-sectional area of the majority of
villi (Figure I), or class 2, if oedema involved > 50 per cent of the cross-sectional area of the
majority of villi (Figure 2). A mean global score was calculated by multiplying the percent of
villi with oedema (extent) by the severity (class). Scores of 20 or less were considered to be mild
oedema, 21 or more were considered to be moderate to severe oedema. A maximum score of
200 was equivalent to all villi having class 2 oedema.
The clinical features and placental findings of cases with villous oedema were analyzed using
the cases without villous oedema as controls. A microcomputer version of Statistical Package
for Social Sciences (SPSS/PC+ V2.0, SPSS Inc., Illinois) was used for data analysis. Chi-
square, Fisher exact test and Student t-test were used to calculate significance, a probability
(P) value of less than 0.05 was considered statistically significant.

RESULTS

Incidence of villous oedema

Villous oedema was present in 259 of 1925 singleton placentae of > 19 weeks of gestation (an
incidence of 13 per cent). It was more prevalent in preterm placentae (84/333 or 25 per cent)
than in term placentae (175/1592 or I I per cent). Moderate to severe villous oedema are more
frequent in preterm (45184 or 54 per cent) than in term placentas (43/175 or 25 per cent).

Relation with pregnancy complication and fetal outcome in term and preterm gesta-
tions
Villous oedema (Table I) was less frequent in term gestations in which the mothers smoked
Shen-Schmarz, Ruchelli, Brown: Vilious oedema 299

Ftgt4re I. Class t severity. The empty spaces occupy < 50 per cent of cross-set tional area of the swollen villi (arrows) in
a preterm placenta. (Haematoxylin-eosin stain. Magnification = x zoo).

Figure 2. Class 2 severity. The empty spaces occupy > 50 per cent of cmss+e&ond area of the swollen villus in a pre-
term placenta. Hofbauer cells (arrows) are seen within the clear spaces. (Haematoxylin-eosin stain. Magnification =
X 400).
 Placenta (I989), Vol. IO

Table I. Clinicopathologic correlation of villous oedema in term and preterm gestations

Term > 37 weeks Preterm < 38 weeks

Total 1592 333
Oedema No ocdema Oedema No oedema
Number of cases ‘75 14’7 84 249

Per cent maternal smoking 10.3 + 20.2 20.2 23.7

Per cent maternal hypertension 2.3 4.0 9.5 9.6
Per cent abruption 1.5 0.9 7.1 4.8
Per cent maternal diabetes I.0 8.3 4.8
Per cent I min AS < 7 i.0 7.2 ‘jJ.0' 26.0
Per cent 5 min AS < 7 2.3 I.1 22.6* 13.3
Per cent fetal death 1.1X 0.07 3.6 3.6
Per cent neonatal death 1.1* 0.07 12.0* 4.8
Per cent chorioamnionitis 20.6 17.2 38.1+ 24.9
Per cent villous dysmaturity 3o.9+ 12.8 30.0’ 16.5
Per cent villous haemorrhage 0.6 0 8.0’ 1.6
Per cent villous fibrosis 38.3+ 25.I 20.2 18.5

AS = Apgar score
* = P < 0.05
+ = P < 0.01

cigarettes (P < 0.0021), but this relationship was not observed in preterm gestations. In term
gestations with villous oedema, intrauterine fetal death and neonatal death were more frequent
(P < 0.0289). Significant correlation with villous oedema was not observed in other maternal
and perinatal complications, such as diabetes, pre-eclampsia/hypertension, abruptio placentae,
placenta previa, prolapsed cord and meconium staining of amniotic fluid.

Relationship with chorioarnnionitis and fetal outcome in preterm gestations

Chorioamnionitis was present in 28 per cent (94/333) of preterm and 18 per cent (280/1592) of
term placentae, with an overall prevalence of 19 per cent (374/1925). Figure 3 shows the incid-
ence of villous oedema and chorioamnionitis at different gestational age groups. The preval-
ence and severity of both villous oedema and chorioamnionitis increased with decreasing
gestational age. However, for gestational ages below 33 weeks, villous oedema remained con-
stant at about 40 per cent, while chorioamnionitis continued to increase from 45 per cent for
28-32 weeks of gestation to 61 per cent for < 28 weeks gestation.
The fact that gestational age was a dependent variable for villous oedema, chorioamnionitis
and low Apgar scores probably accounted for the association of these features with villous
oedema when all placentae below 38 weeks gestation were analysed as a group (Table I). As
shown in Table 2, within a given gestational age range, there was no significant correlation
between villous oedema and chorioamnionitis.
Figure 4 shows the relationship of low Apgar score ( < 7) at I min to gestational age, with the
placentae separated into four groups: those having both villous oedema and chorioanmnionitis,
those having either one of the lesions, and those having neither of the lesions. We found that
the shorter the gestation, the higher was the prevalence of neonates with low Apgar scores at
one minute: 7.3 per cent at term increased to 84 per cent at < 28 week gestation. For’each ges-
tational age groups, chorioamnionitis was not significantly associated with either low Apgar
score at 1 min or at 5 min.
Table 2 is a cross-tabulation of villous oedema with low Apgar scores (<7) at I min and 5
min and severity of chorioamnionitis divided into different gestational age groups. For term as
well as gestations less than 33 weeks, there were no correlations between villous oedema and
Shen-Schmarz, Ruchelli, Brown: Villousoedema 301

Table 2. Relationship between low apgar scores at I and 5 min and chorioamnionitis

Chorioamnionitis
Total ASI<J AS5 ~7 Mi Me Sev

Gestational age = < 28 weeks, N = 31

Absent oedema ‘9 18 16 0 9
Mi oedema 8 6 5 2 0 3
MO oedema 4 3 2 0 0 4
P 0.2795 0.2478 0.1457
Gestational age = 28 to 32 weeks, N = 55
Absent oedema 32 20 II 0 I2
Mi oedema ‘5 9 5 3 I 4
MO oedema 8 3 2 0 0 4
P 0.4334 0.8786 0. I I93
Gestational age = 33 to 37 weeks, N = 246
Absent ocdema ‘97 28 6 I2 IO 17
Mi oedema 36 II 3 5 I 3
MO oedema ‘3 5 2 0 0 2
P 0.0087” 0.0545 0.4955
(iestational age = 38-43 weeks, N = 1591
Absent oedema 1416 I02 14 140 73 3’
Mi oedema 13” IO 3 18 6 6
MO oedema 45 4 4 2 0
P 0.8972 0.3097 0.3493

Mi = mild, MO = moderate, Sev = severe

ASI = Apgar score at I min, AS5 = Apgar score at 5 min, P = probability corrected for degree of free-
dom.

low Apgar scores at I and 5 min. All preterm neonatal deaths were < 29 weeks gestation, and
for this gestational age, there was no significant association between neonatal death and villous
oedema. For preterm gestations from 33 to 37 weeks, villous oedema was significantly asso-
ciated with low Apgar score at I min (P = 0.0087), and showed marginal association with low
Apgar scores at 3 min (P = 0.0545). This association was significant only when chorioamnion-
itis was absent, as shown in Table 3.

Relation with other placental pathology

Villous oedema (Table I) was associated with villous dysmaturity (Figure 5) in both term (P <
0.0001) and preterm placentae (P < 0.0083), with villous haemorrhage (Figure 6) in preterm
placentae (P < 0.0086), and with villous fibrosis (Figure 7) in term placentae (P < o.ooo5).

Table 3 Cross tabulation of villous oedema with low apgar score at I min in rela-
tion to chorioamnionitis at 33-37 weeks gestation

Chorioamnionitis (ASI < 7/N)

Absent Mi MO Sev

Total = 246 196 ‘7 II 22

Absent oedema 22/158 2/12 I/IO 31’7
Mi oedema 9127 115 o/r 113
MO oedema 4/” 010 O/O I/2
P 0.0133 0.6765~ 0.9o90’ o.5249*

ASI < 7/N = Cases with Apgar score at I min <T/Total

Mi = mild, MO = moderate, Sev = severe
P = probability for two degrees of freedom
* = Fisher’s exact text, one tail
302 Placenta (1989), Vol. 10
Incidence of villous oedemo and
chorioomnionitis ot different gestotionol ages

Vil lous oedemo

Grode I and Grode II

60 - ?? Vil lous oedemo

Grode II

??
Chorloomnionitis with
no oedemo

Chorioommonitls
50 - with oedemo

40 -

:
5 30-
&
a

20 -

IO -

Term > 38 GA = 33-37 GA = 20-32 GA= 5 27

Figure 3. Incidence of villous oedema and chorioamnionitis at different gestational ages

We found no significant association between villous oedema and other placental lesions such as
infarction, villitis of unknown etiology, haemorrhagic endovasculitis and thrombosis of fetal
vessels.
Table 4 compares the placental weights of preterm and term placentae in cases with and
without oedema. No significant difference was observed between the two groups in mean pla-
cental weights, birth weights and ratio between placental and birth weights.

DISCUSSION

We found that in a general obstetric hospital with perinatal referral services, the incidence of
villous oedema in nearly zooo thousand consecutive placentae not selected for pregnancy com-
plications was II per cent for term and 25 per cent for preterm gestation. In term placentae,
villous oedema was not only less frequent but also milder in extent and severity. Villous
Shen-.%hwarz, Ruchelli, Brown: Villous oedema 303

Frequency of low Apgor scores ot I minute

of different gestational ages in relotion to
villous oedema and chorioomnionitis

Oedemo and
Chorloomnionltis

Oedemo
No Chorkoomnionitis

No Oedemo
Chorloomnlonltis
0

r No Oedemo
No Chorioamnionitis

Number of cases 36 139244 1173 II 38 39 1513 12 II I?

Term 2 38 GA = 33-37 GA = 28-32 GA=< 27

Figure 4. Frequency of low Apgar scores (<7) at I min at different gestational ages and their relationship with villous
oedema and chorioamnionitis.

oedema correlated with low Apgar scores only for gestational range of 33-37 weeks and only
when chorioamnionitis was absent. Chorioamnionitis neither correlated with villous oedema
nor with low Apgar scores.
Our results are similar to those in a previous study of 164 placentae examined for pregnancy

Table 4. Placental weight

Placenta = P Baby = B
GA (weeks) (9) (9) Ratio = P/B

Terms
Oedema 40 + I.1 526 + 104 3487 + 471 0.15 + 0.03
No oedema 40 + I.2 5’2 + 9.5 3483 + 457 0.15 + 0.02
Preterm
Oedema 32 + 4.3 395 + I39 at83 + 953 0.20 + 0.06
No oedema 34 + 3.7 425 + ‘27 2472 + 874 o. 19 + 0.08

GA = Gestational age in weeks

 Placenta (1989), Vol. 10

Figures. Villous haemorrhage (arrows) in a preterm placenta with moderate to severe villous ocdema. (Haematoxylin-
eosin stain. Magnification = x IOO).

Figure 6. Villous dysmaturity and oedema in a placenta at 41 weeks gestation. (Haematoxylin-eosin stain. Magnifica-
tion = x loo).
Shen-Schmarz, Ruchelli, Brown: Villous oedema 305

Figure 7. Villous fibrosis (arrows) present focally in a term placenta. (Haematoxylin-eosin stain. Magnification =
X 100).

complications (Naeye et al, 1983), however, there are a few significant differences. While
Naeye et al (1983) concluded that villous oedema positively correlated with antepartum fetal
hypoxia, as judged by low Apgar scores, low umbilical arterial blood pH, the need for resuscita-
tion at birth and subsequent assisted ventilation, and the development of hyaline membrane
disease, we did not find such relationship in preterm gestations below 32 weeks (Table 2).
Naeye et al (1983) showed that the villous oedema score (severity) correlated with the severity,
but not the stage, of stage chorioamnionitis; we found no association between the two (Tables 2
and 3). We believe that these differences are due to the fact that since the placentas were not
selected for abnormalities in the pregnancy, delivery and neonate; our study had a wider per-
spective of the normal, as well as a lower incidence and a lesser severity of clinical complica-
tions and pathologic lesions (Table I).
An association between ‘heavier-than-expected’ placentae and antenatal hypoxia, especially
for gestational age <35 weeks was reported from the Collaborative Perinatal Study of the
National Institute of Neurological Communicative Disorders and Stroke (Naeye, r987a). The
author claimed that the relationship between the two was due to placental villous oedema
(Naeye, r987a,b). In our study, we found no significant difference in placental weights and
ratio of placental weight to birth weight between cases with oedema and those without oedema
in both term and preterm placentae (Table 3).
As pregnancy progresses, structural modification of the chorionic villi is particularly evident
within the villous stroma. Five types of villi may be distinguished by their stromal architecture
(Castellucci and Kaufmann, 1982; Kaufmann et al, 1987). During the first 7 weeks of gestation,
the placenta is composed only of mesenchymal villi. Subsequently these develop into the im-
mature intermediate villi which represent the starting points for all the later villous types: stem
villi, mature intermediate villi and terminal villi. The immature intermediate villi are the
306 Placenta (1989), Vol. IO

prevalent villi of the immature placenta and are characterized by a reticular framework of con-
nective tissue cells surrounding globular spaces (stromal channels) containing numerous Hof-
bauer cells. They are the sites of proliferation and branching of the villous tree and are
therefore important for placental growth and maturation (Castellucci and Kaufmann, 1982).
We propose that villous oedema is a condition primarily of the immature intermediate villus
because of the following observations.
(i) Morphologically, villous oedema closely resembles enlarged stromal channels of the im-
mature intermediate villi.
(ii) Villous oedema is significantly associated with prematurity as well as villous dysmatur-
ity (relative immaturity), both of which are known to have a predominant population of im-
mature intermediate villi (Kaufmann et al, 1987).
(iii) In term placentae, immature intermediate villi are present in small numbers and may
account for the low incidence of villous oedema in term gestation (Kaufmann et al, 1987).
(iv) Villous oedema and chorioamnionitis have different pathogenetic mechanisms. We
found that while the prevalence of chorioamnionitis increased steadily from 18 per cent at term
to 61 per cent for gestational age < 28 weeks, the prevalence of villous oedema increased from
I I per cent at term to 20 per cent at 33-37 weeks and then remained constant at 40 per cent for
gestations < 33 weeks. This phenomenon may be explained by the fact that growth of the vil-
lous tree and villous stroma (which may be represented by the proportion of immature inter-
mediate villous) reaches a plateau at around 34-36 weeks gestation and decrease thereafter
until term (Tesdale, 1980).
Cigarette smoking during pregnancy has widespread effects on the maternal-placental-fetal
unit and is associated with a variety of placental changes (Rush et al, 1986); we observed villous
oedema was less frequent in these cases at term. Placental villous haemorrhage was associated
with villous oedema in preterm placentae, and villous fibrosis was associated with villous
oedema in term placentae. Both conditions are vascular lesions of the chorionic villi caused by
reduced fetoplacental blood flow or fetal vascular thrombosis (Fox, 1978). These observations
suggest that fetoplacental and/or maternal factors are also involved in the pathogenesis of
villous oedema.
The stromal channels in the immature intermediate villus facilitate the phagocytic and
immunologic functions of the Hofbauer cells and may have a role in regulating fluid balance in
the placenta and the fetus (Castellucci and Kaufmann, 1982). Villous oedema appears to be an
expansion or distension of these stroma channels. To what extent this change becomes abnor-
mal and interferes with villous function remains a mystery, and would require more research
into the pathophysiology of the chorionic villi.

SUMMARY

Villous oedema was observed in 259 placentae among 1925 consecutive singleton pregnancies
of > 19 weeks gestation. It was present in I I per cent of term placentae in which significant
associations with fetal and neonatal death (P < 0.03), and absence of maternal cigarette smok-
ing (P < 0.002) were found. In preterm placentae, the oedema was usually more severe, and its
prevalence increased from 20 per cent for 33-37 weeks to 40 per cent for < 33 weeks, Our ana-
lysis showed that for a given gestational age, villous oedema was not significantly related to
chorioamnionistis, Apgar scores of < 7 at I and 5 min, or neonatal death, an exception was for
33-37 weeks gestation, in the absence of chorioamnionitis, villous oedema was associated with
low I min Apgar score.
Shen-Schwarz, Ruchellt, Brown: Villousoedema 307

Immature intermediate villi are present in premature placentae as a normal developmental

stage and in dysmature placentae as a result of villous maldevelopment. Since villous oedema
closely resembles the ‘stromal channels’ in this villous type and shows significant association
with prematurity and villous dysmaturity, we postulate that villous oedema is a lesion primar-
ily of the immature intermediate villi. Both fetal and maternal factors are involved in its patho-
genesis.

ACKNOWLEDGEMENT

We thank Dr Bernard Klionsky for his encouragement and support of this study.

REFERENCES

Alvarez, H., Sala, M. A. & Benedetti, W. L. (1972) Intervillous Space Reduction in the Edematous Placenta. AmY-
icanJournaI of Obstetricsand Gynccologv, 112,819-820.
Csstellucci, M. & Kaufmann, P. (1982) A Three-dimensional Study of the Normal Human Placental Villous Core:
II. Stromal Architecture. Placenta, 3,2+286.
Fox, H. (1978) Pathology of the placenta. London: W. B. Saunders.
Fox, H. (1986) Pathology of the Placenta. Clinics in Obstetricsand Gynecology, 13,501~-517.
Kaufmanu, P., Luckhardt, M., Schweikbart, G. & Cnntle, S. J. (1987) Cr oss Sectional Features and Three-
dimensional Structure of Human Placental ViIli. Placenta, 8,235-247.
Nneye, R. L. (1987a) Do Placental Weights have Clinical Significance? Human Pathology, 18,387-391.
Nseye, R. L. (19876) Functionally Important Disorders of the Placenta, Umbilical Cord and Fetal Membranes.
Human Pathology, 18,680-691.
Naeye, R. L., Maisels, M. J., Loretu, R. P. & Botti, J. J. (1983) The Clinical Significance of Placental Villous
Oedema. Pediatrics, 71,588-594.
Rush, D., Kristal, A., Blanc, W., Navarro, C., Chat&an, P., Brown, M. C., Rosso, P., Winick, M., Brnsel, J.,
Naeye, R. & Susser, M. (1986) The Effects of Maternal Cigarette Smoking on Placental Morphology, Histo-
morphology, and Biochemistry. AmericanJournal of Perinatology, 3,263-272.
Tesdale, F. (1980) Gestational changes in the functional structure of the human placenta in relation to fetal growth: a
morphometric study. AmericanJournal of Obstetrics and Gynecology, 137, r&-568.