Chorio-amnionitis in relation to mode of delivery at term

SAMJ VOL. 76 2 SEPT 1989 201
14. Tanner JM. Growch ac Adolescence. 2nd ed. London: Blackwell Scientific
Publications, 1962.
IS. Prader A. Testicular size, assessment and clinical importance. Triangle 1966;
7: 240.
16. Grant DB, Laurence BM, Atherden SM ec at. HCG stimulation test in
children with abnormal sexual development. Arch Dis Child 1976; 51:
596-601.
17. Frasier SO, Gafford F, Horton R. Plasma androgens in childhood and
adolescence.] Clin Endocrinol Mecab 1969; 29: 1404-1408.
18. Toublanc JE, Canlorbe P, Job JC Evaluation of Leydig cell function in
normal prepubertal and pubertal boys.] Sceroid Biochem 1975; 6: 95-99.
19. Canlorbe P, Borniche P, Bader JC, Vassal ], Toublanc J-E, Job J-C La
cryptorchidie. Arch Franc Pediacr 1974; 31: 145-156.
20. Vivanco F, Puonzaley G, Ramos F. 5000 IU HCG simple clinical teSI for
measuring serum testosterone response to IV infusion of HCG. Acca Endocrinol
(Copenh) 1973; 73: 790-793 (quoted in Ref. No. 18).
21. Levin SB, Kogen S], Engel RM er al. The impalpable testes: a rational
approach to managemen!.] Uro11978; 120: 515-520.
22. Walsh PC, Curry N, Mils RC ec al. Plasma androgen response to HCG
stimulation in prepubertal boys with hypospadias and cryptorchidism. ]
Endocrinol Mecab 1976; 42: 52-59.
23. Lipstruty L], Caminos Jorees R, Greenpan CS ec al. Testicular function
after orchipexy for unilaterally undescended testes. N Engl] Med 1976; 295:
15-18.
24. Zondak LH, Zondak T. Observations on the testes in anencephaly with
special reference to Leydig cells. Bioi Neonace 1965; 8: 329-347.
25. Gutherie RO, Smith OW, Graham CB. Testosterone treatment for micropenis
during early childhood.] Pediacr 1973; 83: 247-252.
Chorio-amnionitis in relation to mode of
delivery at term
G. H. MOLLER, D. L. WOODS, A. F. MALAN, C. C. SINCLAIR-SMITH
Summary
The incidence of inflammatory changes on histological examination
in the placenta, membranes and umbilical cord of 50
infants born by spontaneous vaginal delivery were compared
with those of 50 infants born by elective caesarean section
before the onset of labour at term. Inflammation was significantly
more frequent after vaginal delivery (28%) than after
caesarean section (6%). This suggests that intra-uterine
bacterial colonisation is uncommon before the onset of labour
and it is argued that chorio-amnionitis in the vaginally
delivered placentas occurred during labour.
S AIr Med J 1989; 76: 201-202.
An acute inflammatory infiltration of the placenta, membranes
and umbilical cord occurs when bacteria spread from the
cervix and vagina into the uterine cavity. 1 The reported incidence
of such an inflammatory response, often referred to as
chorio-amnionitis, varies widely in different studies and may
depend on several factors, including the duration of membrane
rupture, the gestational age at the onset of labour, and the
amount of obstetric intervention during labour.
In addition to complicating labour, it has been suggested
that chorio-amnionitis is a cause of labour owing to the
stimulation of prostaglandin synthesis in the inflamed membranes.
2 If chorio-amnionitis is either a cause or a complication
of labour, then it should often occur in vaginal deliveries after
Departments of Paediatrics and Child Health and Clinical
Pathology, University of Cape Town
G. H. MOLLER, M.B. CH.B., F.CP (SA), D.CH.
D. L. WOODS, M.D, FRCP
A. F. MALAN, M.D., M.MED. (PAED.), DIP. MID. CO.G. (SA)
C C SINCLAIR-SMITH, M.B. CH.B., M.MED. (PATH.)
Accepled 13 OCI 1988.
the spontaneous onset of labour. In contrast, it should be
uncommon with elective caesarean section.
To test the hypothesis that chorio-amnionitis is uncommon
before the onset of labour, the incidence of acute inflammation
in the placentas of infants delivered by caesarean section was
compared with those of infants delivered by spontaneous
vaginal delivery at term.
Material and methods
The placentas of 100 consecutive infants born at term (37 - 42
weeks) were studied. Fifty infants were born by normal vaginal
delivery after the spontaneous onset of labour. The duration of
membrane rupture was less than 24 hours in all infants born
vaginally and none of the mothers showed clinical signs of
chorio-amnionitis during labour. A further 50 infants were
delivered by elective caesarean section before the onset of
labour. The membranes were intact at operation. In most
cases, the indication for surgery was a previous caesearean
section.
The placentas were examined within 24 hours of delivery. A
segment of umbilical cord, a roll of peripheral membranes and
a block of chorionic plate were sampled from each placenta
and fixed in Bouin's fluid. The tissue blocks were impregnated
with paraffin wax, sectioned and stained with haematoxylin
and eosin.
Light microscopy was used to detect the presence of a
polymorphonuclear leucocyte infiltration of the membranes
(membranitis), placenta (subchorionic intervillositis), umbilical
cord (funicitis) or chorionic vessels (chorionic vasculitis). The
method used to evaluate the significance of the polymorphonuclear
leucocyte infiltrate was that of Blanc. 3 Grade I or
greater was regarded as significant. An inflammatory response
in the membranes or subchorionic intervillous area of the
chorionic plate was regarded as maternal while a response in
the umbilical cord or chorionic vessels was regarded as fetal.
Clinical details were not made available to the person assessing
the histological examinations.
The chi-square test was used in the analysis of the data.

202 SAMT VOL 76 2 SEPT 1989
Results
A maternal inflammatory response (membranitis or subchorionic
intervillositis) was present in 14 vaginal deliveries (28%)
and 3 caesarean sections (6%). This difference was significant
(X 2 = 7,09; P < 0,01). Of the 50 placentas delivered vaginally,
membranitis was present in 12 and intervillositis in 9, with 7
placentas having both membranitis and intervillositis. Three of
the 50 placentas delivered by caesarean section had intervillositis,
of which 2 also had a membranitis. In all cases with a
fetal response, a maternal response was also present.
In contrast, a fetal inflammatory response (funicitis or
chorionic vasculitis) was present in the placentas of 6 vaginal
deliveries (12%) and 3 caesarean sections (6%). This difference
was not significant (X = 2 0,49; P > 0,05). In the vaginal
deliveries, funicitis and chorionic vasculitis were present in 4
and 5 cases respectively. In the caesarean sections, funicitis
was present in all 3 cases, of which 2 also had a chorionic
vasculitis. The infants were all clinically well without signs of
bacterial infection at delivery.
A maternal inflammatory response was therefore significantly
more frequent after vaginal delivery than caesarean section,
but there was no significant difference in the incidence of a
fetal response.
Discussion
During pregnancy the intra-uterine cavity and its contents are
usually sterile and the closed cervix is able to prevent an
ascending spread of bacteria from the vagina. However, in
some cases bacteria are able to penetrate the cervical barrier
and migrate between the layers of the peripheral membranes
to reach the placenta. This evokes an acute inflammatory
response by the mother in the chorion, amnion and placenta.
Later the bacteria may colonise the amniotic fluid and stimulate
an inflammatory response by the fetus in the umbilical and
chorionic vessels. Only occasionally is the fetus infected.
Therefore the newborn infant is usually clinically well after
delivery despite the presence of pus cells and bacteria in the
gastric aspirate at delivery and an inflammatory infIltrate ef
the placenta and membranes on histological examination.
The ability of the cervix to prevent the ascending spread of
bacteria before the onset of labour is supported by this study
where an inflammatory response in the placenta, membranes
and umbilical cord was uncommon in pregnancies terminated
by elective caesearean section before the onset of labour. In
the 3 exceptions, dilatation of the cervix with exposure of the
membranes to the vaginal flora before the onset of contractions
may have allowed colonisation of the uterine cavity.
In contrast, chorio-amnionitis was present in the placentas
of 28% of the vaginal delivery patients confIrming that infection
of the placenta and membranes is commonly associated with
the spontaneous onset of labour and vaginal delivery.
It is thought that bacterial invasion of the amnion, chorion
and decidua stimulates the production of prostaglandins, which
initiates myometrial contractions. 2 • 4 It is suggested therefore
that bacterial colonisation of the placenta and membranes,
often before membrane rupture, is a common cause of the
onset of labour,s·6 especially in preterm deliveries.
The fllldings of this study - that a maternal inflammatory
response is unusual before, but common after, labour and
vaginal delivery - support the hypothesis that bacterial colonisation
of the intra-uterine cavity may play a role in the
initiation or maintenance of labour at term. However, the
infrequent fInding of a fetal response suggests that the chorioamnionitis
is of short duration as it had not progressed to the
stage of amniotic fluid colonisation with inflammation in the
umbilical and chorionic vessels. Therefore the inflammatory
changes noted probably reflect bacterial colonisation of the
placenta and membranes, occurring with dilatation of the
cervix and exposure or rupture of the membranes during the
course of labour.
It is concluded that chorio-amnionitis is uncommon before
the onset of labour at term. In contrast, it is signifIcantly more
frequent after spontaneous vaginal delivery, suggesting that
bacterial colonisation of the placenta and membranes frequently
takes place during labour.
REFERENCES
l. Blanc WA. Amniotic infection syndrome - parhogenesis, morphology and
SIgnificance ill orcumnatal mortality. elm Obscet Gyneco/1959; 2: 705-734.
2. Bejar P, Curbelo V, Davis C, Glue!< L. Premature labour: H. Bacterial
sources of phospholipase. Obstet Gyneco/1981; 57: 479-482.
3. Blanc WA. Pathology of the placenta, membranes, and umbilical cord in
bacterial fungal and viral infections in man. In: Naeye RL, Kissane JM,
Kaufman N, eds. Perinatal Diseases. Baltimore: Williams & Wilkins, 1981:
67-132.
4. Lamont RF, Rose M, Elder MG. Effect of bacterial products on prostaglandin-E
prodltction by amnion cells. Lancet 1985; 2: 1331-1333.
5. Bobin JR, Hayslip CC, Damaro ]D. Anmiotic fluid infection as determined
by transabdominal amniocentesis in patients with intact membranes in
premature labor. Am] Obstet Gyneco/1981; 140: 947-952.
6. Naeye RL, Peters EC. Anmioric fluid infections wirh intact membranes
leading to perinatal dearhs: a prospective study. Pediatrics 1978; 61: 171-177.