Human Reproduction Update 1997, Vol. 3, No. 5 pp. 517–527
European Society for Human Reproduction and Embryology
Accumulation of interleukin-1β and interleukin-6
in amniotic fluid: a sequela of labour at term and
preterm
Susan M.Cox, M.Linette Casey and Paul C.MacDonald1
The Cecil H. and Ida Green Center for Reproductive Biology Sciences and the Departments of Obstetrics-Gynecology and
Biochemistry, The University of Texas, Southwestern Medical School, Dallas, Texas 75235–9051, USA
TABLE OF CONTENTS
Introduction
Materials and methods
Results
Discussion
Acknowlegements
References
517
519
521
523
526
526
From the finding of micro-organisms or inflammatory
mediators, or both, in amniotic fluid (AF), it has been
proposed that intrauterine infection is one cause of
preterm labour (PTL, intact fetal membranes). This
theory, however, remains unproved, i.e. the accumulation of micro-organisms and inflammatory mediators
in AF after labour is in progress may be the consequence,
not the cause, of labour both at term and preterm. This
study was conducted to evaluate this possibility by a
comparison of the concentrations of interleukin (IL)-1β
and IL-6 in AFs collected before and during PTL (<34
weeks gestation) with those in AFs collected at term
(before labour and from the forebag and upper
compartments of the amniotic sac during labour). The
concentrations of IL-1β and IL-6 in AF were also
analysed as a function of the duration of labour (term or
preterm) before fluid collection. In addition, studies were
conducted to define the source of IL-1β in AF. A total of
666 AFs were evaluated. IL-1β was not detected
(<50 pg/ml) in AFs collected before the onset of labour at
any stage of gestation (n = 320), including 170 fluids
obtained at term. During labour, IL-1β was detected
(>50 pg/ml) in 58 out of 106 (54.7%), 17 out of 64 (26.6%)
and 60 out of 176 (34%) of AF samples obtained during
PTL, term labour (upper compartment) and term
labour (forebag) respectively. AF sampling, as well as
labour and delivery, were completed in <18 h in all term
pregnancies. However, labour (with cervical dilation)
was in progress for >18 h before AF was collected in 39
out of 106 (37%) PTL pregnancies. The incidence of
IL-1β-positive samples among AFs collected before 18 h
of PTL (23 out of 67; 34%) was indistinguishable from
that in AFs collected during labour at term. However, in
AFs collected after >18 h PTL, the incidence of
IL-1β-positive samples was 35 out of 39 (89.7%) The
concentrations of IL-1β (pg/ml; mean ± SEM) in AFs
collected during PTL (2680 ± 730; n = 106) were greater
than those in AFs collected from the upper compartment
and forebag during term labour (436 ± 244, n = 64; and
468 ± 119, n = 176) respectively; this difference, however,
was attributable to very high concentrations of IL-1β in
AFs in which PTL was in progress for >18 h before AF
collection (6021 ± 1832; n = 39). The concentrations of
IL-6 in AF were correlated with those of IL-1β
(P < 0.001). We conclude that IL-1β and IL-6 accummulate in AF in a similar proportion of pregnancies during
the first 18 h of term and preterm labour. Therefore, the
accumulation of these cytokines in AF cannot be taken as
evidence for a role for infection in the pathogenesis of
PTL.
Key words: cytokines/forebag/inflammation/parturition/
preterm labour
Introduction
The spontaneous onset of preterm labour (PTL) in pregnancies with intact fetal membranes accounts for 45–55%
of very low birth weight infants (Cunningham et al., 1997);
and regrettably, there has been no reduction in the
1To whom correspondence should be addressed at: The Green Center for Reproductive Biology Sciences, The University of Texas, Southwestern Medical
School, 5323 Harry Hines Boulevard, Dallas, Texas 75235–9051, USA. Phone: 214–648–3260; Fax: 214–648–8683; E-mail: [email protected]
518
S.M.Cox et al.
incidence of preterm births in the USA or Canada in recent
years notwithstanding the wide-spread use of tocolytic
agents to arrest PTL (Leveno et al., 1990; Canadian Preterm
Labor Investigators Group, 1992). In searching for a remedy
for this common complication of human pregnancy, many
investigators have proposed that intrauterine infection is one
cause of PTL (estimates vary from 10–40% of PTL).
The hypothesis that intrauterine infection may be a cause
of PTL became more captivating as the biomolecular
phenomena of the inflammatory response were defined in
greater detail. Both Romero et al. (1988a) and ourselves
(MacDonald et al., 1988) speculated that the sequence of
pathophysiological events in intrauterine infection-induced
PTL might proceed as follows: micro-organisms, resident in
the vagina or uterine endocervix, after ascending spread via
the cervical canal, could colonize tissues of the lower portion
of the uterine decidua parietalis or fetal membranes or else
burrow through the membranes to infect the amniotic fluid
(AF), or both. Bacterial endotoxin [lipopolysaccharide
(LPS)], or other toxins from these organisms, could act on
decidua or fetal membranes (most likely upon mononuclear
phagocytes in decidua or else leukocytes recruited into the
AF) to promote the formation of interleukin (IL)-1β. This
primary cytokine stimulates the synthesis of uterotonins [e.g.
prostaglandins (PG)] as well as other cytokines (e.g. IL6,
IL-8, tumour necrosis factor-α, and many others) in a variety
of tissues, including amnion and decidua (Romero et al.,
1989a; Casey et al., 1990; Mitchell et al., 1990; Bry and
Hallman, 1991). It was then envisioned that PGs, produced
in these tissues in response to inflammation, are transported,
in some undefined fashion in which inactivation is avoided,
from amnion or decidua to the myometrium to induce
myometrial contractions and PTL. Several findings have
been presented in support of this general hypothesis.
Micro-organisms can be cultured from 10–40% of AFs
collected during PTL (Romero et al., 1988a, 1994; Gomez
et al., 1995). In addition, LPS (Cox et al., 1988; Romero et
al., 1988b) as well as an impressive array of mediators of the
inflammatory response, including IL-1β (Hillier et al., 1993;
Romero et al., 1992; Gomez et al., 1995), are found in a
sizeable proportion of AFs that are collected during PTL.
Other cytokines, e.g. IL-6 (Romero et al., 1990a, 1993;
Greig et al., 1993; Hillier et al., 1993; Opsjon et al., 1993;
Silver et al., 1993) and IL-8 (Romero et al., 1991; Cherouny
et al., 1993; Laham et al., 1993; Puchner et al., 1993) are
present in AF of most normal pregnancies throughout human
gestation; however, the concentration of these inflammatory
mediators is increased, sometimes strikingly, in AF in about
one third of PTL pregnancies (Romero et al., 1988c, 1989b,
1990a, 1991; Greig et al., 1993; Gomez et al., 1995).
Despite these interesting findings, important questions
remain concerning the validity of the theory that intrauterine
infection is a cause of PTL. Most of the evidence cited in
favour of this hypothesis comes from the finding of the
accumulation of micro-organisms or mediators of
inflammation (bacterial endotoxin/LPS, cytokines, and
prostaglandins) in the amniotic fluid (AF) of some
pregnancies during PTL. The inflammatory process leading
to these changes in the constituents of AF, however, may be
the consequence, not the cause, of parturition either at term
or preterm. For example, a temporal relationship has not
been established between the development of infection/
inflammation and the subsequent onset of PTL. AFs, blood,
and tissues used in studies to define the pathogenesis of PTL
are collected after labour is in progress; consequently,
mediators of inflammation [including micro-organisms, bacterial toxins, IL-1β, and other mediators of inflammation
(cytokines, PGs and endothelin-1)] may have entered AF as
a result of an inflammatory process that commences as a
normal consequence of labour (MacDonald and Casey,
1993; Casey et al., 1993).
Specific anatomical rearrangements that affect the
amnionic sac take place during the labour-induced process of
cervical effacement and dilatation at term and preterm. The
lower pole of the amnionic sac becomes the forebag.
Importantly, however, the forebag is formed only after
labour is in progress (or with severe anatomical or functional
incompetence of the uterine cervix). As the cervix dilates,
fragments of decidua parietalis tissue, in highly varying
amounts, are pulled away from the lower uterine segment as
the forebag is formed. These fragments of decidual tissue
remain attached to the outer surface of the forebag and are
exposed in the vagina and thence bathed continuously by
vaginal fluid in which a large variety of micro-organisms and
bacterial toxins, together with cytokines (notably IL-1β), and
PGs, are present in high concentrations in non-pregnant and
pregnant women (Cox et al., 1993). Consequently,
inflammation of the decidual tissue fragments lining the
forebag is obligatory and is readily demonstrable as follows:
(i) there is a very high rate of secretion of IL-1β, IL-6 and
PGs directly from the exposed forebag decidua into the
vagina during labour (Cox et al., 1993); (ii) the
concentrations of pro-IL-1β mRNA in decidual tissue of the
forebag are greater than those in the decidua parietalis taken
from other portions of the same amniotic sac (MacDonald
et al., 1991); (iii) PGs produced by the decidua (i.e. PGF2α
and PGE2) accumulate preferentially in AF of the forebag
during spontaneous labour at term (MacDonald and Casey,
1993); (iv) the concentrations of PGF2α and 13,14-dihydro15-keto-PGF2α (PGFM) in AF of the forebag increase as a
Interleukins-1β and -6 in amniotic fluid
function of the surface area of the forebag exposed in the
vagina, i.e. with cervical dilation during labour, in a highly
significant manner (MacDonald and Casey, 1993, 1996).
Thus, PGs that accumulate in AF at parturition arise in the
forebag tissues after labour has begun; (v) it has been shown
that inflammatory mediators, including micro-organisms
and IL-1β, also are present in AF collected from some
normal term pregnancies (with intact fetal membranes)
during spontaneous labour (Romero et al., 1990b; Tsunoda
et al., 1990; Laham et al., 1993; Opsjon et al., 1993; Gomez
et al., 1995). Therefore, an inflammatory response in the
forebag is a normal and expected sequela of labour.
Unfortunately, however, the finding that inflammatory
mediations also accumulate in AF during labour at term has
been largely discounted by most investigators for
inexplicable reasons. As a result, an explanation(s) other
than intrauterine infection preceding the onset of PTL has
not been explored to define labour-associated inflammation.
A comparison of the concentrations of IL-1β and IL-6 in
AFs from a large number of pregnancies from the same
obstetric population during labour at term and preterm after
comparable times in labour before AF was collected has not
been reported.
In this study, we sought to ascertain whether the
accumulation of mediators of inflammation in AF during
spontaneous labour at term and preterm were similar. If this
were the case, and if these agents accumulate only after
labour begins, considerable doubt would be cast upon the
relevance of the finding of inflammatory mediators in AF as
evidence that infection is a causal factor in the pathogenesis
of PTL. Therefore, this study was conducted to compare the
frequency, time of appearance, and values of IL-1β and IL-6
in AFs collected before and during spontaneous labour at
term and preterm. In the analyses of these data, consideration
also was given to the duration of labour, at term and preterm,
before the AF was collected.
Materials and methods
Amniotic fluids
The experimental protocols and consent forms used in the
conduct of this study were approved by the Institute Review
Board of the University of Texas. AFs (n = 666) were
collected into sterile polystyrene tubes and transported
directly to the laboratory. In some studies, aliquots of the AF
(before centrifugation) were evaluated for cellular content.
Thereafter, the fluid was centrifuged at 600 g to remove cells.
The cell pellets were used for selected studies. The
supernatant fractions were removed, divided into aliquots of
0.5–1.0 ml and stored at –80°C. Transabdominal amniocenteses performed before labour were conducted for
519
diagnostic purposes independent of this study: AFs from
preterm, not in labour, pregnancies with normal fetuses were
used in this investigation [midtrimester (n = 45) and between
24–34 weeks (n = 53)]. AFs also were obtained at 35–42
weeks by transabdominal amniocentesis or transuterine
amniocentesis at Caesarean section before (n = 222) or after
the onset of labour (n = 64; upper AF compartment). None of
the women of this study received oxytocin.
The pregnancies in which AF was obtained by
transuterine amniocentesis during labour at Caesarean
section were selected for study on the basis of established
active, normally-progressing labour, which was defined as
regular, forceful, painful uterine contractions occurring at
intervals of ≥2/10 min together with observed effacement
and progressive dilatation of the cervix. Term pregnancies
in which desultory labour existed were excluded from
study. AF was collected from the upper compartment of
term pregnancies during labour only until 7 cm cervical
dilatation; by this stage of labour progress, most
pregnancies that are to be terminated by Caesarean section
have been delivered or the fetal membranes have ruptured
spontaneously, or the membranes have been ruptured by
amniotomy, or inadequate labour progress has been
encountered. It cannot be stated with absolute certainty that
all pregnancies of the ‘in labour/Caesarean section at term’
group would have continued in labour. It is very likely that
this was the case, however, since forceful contractions (two
or more in 10 min) with progressive cervical effacement
and dilatation were established for each of these
pregnancies.
AF also was collected directly from the forebag (n = 176)
during spontaneous labour at term by direct visualization
(with or without the use of a vaginal speculum) or else by use
of a needle guide (placed into the vagina to the level of the
fetal membranes) by 18 gauge spinal needle aspiration as
previously described (MacDonald and Casey, 1993). AF was
not collected from the forebag until the cervix was dilated
≥3 cm because AF aspiration was performed only in cases in
which amniotomy was to be conducted immediately
thereafter for routine obstetric management. On some
occasions, the fetal membranes ruptured during insertion of
the needle into the forebag; these specimens and those
visibly contaminated by blood or meconium were discarded.
AF was collected from 106 pregnancies during PTL (<34
weeks gestation, fetal membranes intact, spontaneous onset
of PTL) by one of three techniques: transabdominal
amniocentesis, transuterine amniocentesis at the time of
Caesarean section, or by direct needle aspiration of the
amnionic sac per vagina. Delivery occurred within 24 h of
the time of AF sampling. None of the PTL cases were treated
with tocolytic agents. In studies previously conducted at this
520
S.M.Cox et al.
Institution, improved perinatal outcome by use of tocolytic
agents could not be established (Leveno et al., 1986; Cox
et al., 1990). In cases of PTL delivered by Caesarean section,
labour progress was well-advanced, i.e. cervical dilatation
≥4 cm with demonstrable uterine contractions occurring at
intervals of ≤5 min, or labour progress was documented after
AF was collected by transabdominal amniocentesis. In cases
of PTL in which AF was obtained at ≤2.5 cm cervical
dilatation, all pregnancies continued in labour. Five were
delivered later in labour by Caesarean section and six
continued to spontaneous vaginal delivery.
Gestational age was estimated from the best fit of the
menstrual history, obstetric milestones, findings of
sonographic evaluation of the fetus (when available), and
evaluation of newborn maturation. Cervical dilatation was
estimated by digital examination conducted per vagina by
experienced obstetricians. If the estimate of cervical
dilatation were given by a range (e.g. 2–3 cm), an average
value (i.e. 2.5 cm) was used.
minimum amount of IL-1β detectable in AF was 50 pg/ml
and that for IL-6 was 150 pg/ml. In computations of mean
levels for groups of fluids, the values of 25 and 75 pg/ml
were used for IL-1β and IL-6 respectively, for fluids in
which a given cytokine was not detected.
IL-1β in the cellular and supernatant fractions of
amniotic fluid
To evaluate the relationship between the cellular content of
IL-1β and the concentration of IL-1β in the supernatant
fraction of AFs, cells were pelleted by centrifugation at
600 g. The supernatant was removed and the cells were
washed three times in phosphate-buffered saline with
centrifugation. The pellet was sonicated and aliquots of the
lysate were assayed for IL-1β. The supernatant fractions
were also assayed for IL-1β and the two values were
compared. The cellular content was expressed as the amount
of IL-1β in the cells of 1 ml of AF.
Estimation of the duration of labour
Assay of IL-1β and IL-6
IL-1β was quantifiable by use of an enzyme-linked
immunosorbent assay (ELISA) kit, specific for IL-1β
(Cistron Biotechnology, Pine Brook, NJ, USA) over the
range of 5–600 pg IL-1β/well. The diluent for the standard
curve and samples was pseudoamniotic fluid (Schwartz
et al., 1977). Intra-assay and inter-assay coefficients of
variation were 10 and 12% respectively. Parallelism was
maintained for assays conducted with various sample
volumes. In the IL-1β assay, there was no cross-reactivity
with 1000-fold molar excess of IL-6. IL-6 in AF was
quantified by use of an ELISA kit (Intertest-6, Genzyme
Corp, Boston, MA, USA). The limits of the assay ranged
were 15–1000 pg IL-6/assay well. The intra- and interassay
coefficients of variation were 5.3 and 9.1% respectively. The
The duration of labour before the time of AF sampling was
estimated from the time each pregnant woman first
perceived regular, painful contractions and, additionally,
from the known duration of cervical dilatation in the
pregnancies complicated by PTL. In some PTL pregnancies,
labour clearly was in progress for >24 h; but in several of
these cases, a reasonable estimate of the true duration of
labour before the pregnancy came under medical supervision
was not possible. In all cases of pregnancies in spontaneous
labour at term, AF collection and delivery occurred in <18 h
from the commencement of labour. In a sizeable number of
PTL pregnancies, labour with known cervical dilatation was
in progress for >18 h before AF was collected. Therefore,
data analyses also were conducted separately for the group of
AF samples collected after >18 h PTL.
Table I. The incidence of interleukin (IL)-1β-positive amniotic fluids (AFs) and the means and ranges of IL-1β concentration in
AFs as a function of the duration of labour before AF collection
Labour
(hours before
AF collection)
Incidence of IL-1β-positive
AFs during labour
Concentration of IL-1β in AF
Term
Range of values for IL-1β (pg/ml)
Preterm
Term
Preterm
All samples
IL-1β-positive samples
All samples
IL-1β-positive samples
Term
Preterm
<6
24/81
7/23
495 ± 200
1680 ± 652
1115 ± 562
3761 ± 1584
ND–14 980
ND–10 760
6–<12
37/128
12/31
466 ± 157
1550 ± 504
663 ± 313
1620 ± 712
ND–17 025
ND–7950
12–18
16/31
4/13
342 ± 139
639 ± 260
202 ± 104
599 ± 254
ND–4160
ND–1150
>18
0/0
35/39
–
–
6021 ± 1832
6706 ± 2012
–
ND–49 050
ND = not detectable (<50 pg/ml).
The mean ± SEM for all samples of AF includes those in which IL-1β was not detected (<50 pg/ml), and for which a value of 25 pg/ml
was assigned.
Interleukins-1β and -6 in amniotic fluid
521
Statistical analysis
Logarithmic transformation of the data was performed, and
comparisons were made by analysis of variance with
Games–Howell post-hoc analysis and t-test as appropriate.
Results
IL-1β in AF before and during labour at term and
preterm
IL-1β in AFs collected before the onset of labour
IL-1β was assayed in 320 samples of AF obtained by transabdominal or transuterine amniocentesis (at Caesarean
section) at various stages of pregnancy before labour.
IL-1β was not detected (<50 pg/ml) in any of these samples
(Figure 1).
Incidence of IL-1β-positive AFs during term and preterm
labour
In AFs collected during spontaneous labour, the incidences of
IL-1β-positive AFs, i.e. those in which IL-1β was detectable
(>50 pg/ml), were as follows: term upper compartment, 17/64
(26.6%); forebag at term, 60/176 (34%); PTL (irrespective of
the duration of labour), 58/106 (54.7%) (Figure 1, panel A). In
the PTL group, labour (with cervical dilatation ≥3 cm) had
been in progress for >18 h before AF was collected in 39/106
(37%) and for >24 h in 35/106 (33%). In AF samples
collected before 18 h of PTL, IL-1β was detected in 23/67
(34.3%), an incidence similar to that in AFs from term labour
pregnancies. However, in AFs obtained after >18 h of PTL),
35/39 (89.7%) contained >50 pg/ml IL-1β. The greater
incidence of IL-1β-positive AFs during PTL was attributable
to the very high occurrence of IL-1β in AFs collected after
labour and cervical dilatation had been in progress for very
long times (Figure 1, panel A). Therefore, in pregnancies in
labour for <18 h, both at term and preterm, approximately one
third contained IL-1β (>50 pg/ml).
Concentrations of IL-1β in AF during term and preterm
labour
The concentrations of IL-1β (pg/ml; mean ± SEM) in AFs
during spontaneous labour at term were: upper compartment,
436 ± 244 [n = 64; range: none detectable (ND) to 14 980] and
forebag, 468 ± 119 (n = 176; range: ND to 17 025) (Figure 1,
panel B). The mean concentration of IL-1β in AF of PTL
pregnancies (irrespective of duration of labour before AF
collection), was 2680 ± 730 (n = 106; range: ND to 49 050)
(Figure 1, panel B). The concentration of IL-1β in AFs
collected before 18 h of PTL, 735 ± 247 (n = 67; range: ND to
10 760), was only 15% of that in AFs collected after >18 h of
PTL, 6021 ± 1832 (n = 39; range: ND to 49 050; P < 0.0001).
Figure 1. The incidence of interleukin (IL)-1β-positive (>50 pg/ml)
amniotic fluid (AF) (panel A) and the levels of IL-1β (pg/ml; mean ±
SEM) (panel B) in AFs collected before and during labour at term
(upper and forebag compartments) and preterm labour (PTL). The
values for all AFs from PTL and those collected before 18 h of PTL
and after >18 h of PTL are presented separately. The number of AF
samples in each group is shown above each bar. NIL = not in labour;
ND = not detectable.
The amounts of IL-1β in PTL AFs collected before 18 h of
labour had elapsed were not significantly different from those
in AFs collected from the upper or forebag compartment
during labour at term. Except for three samples of AF from
PTL pregnancies in which the concentrations of IL-1β were
49 050, 42 050, and 34 690 pg/ml, the range of values for the
entire PTL group (range: ND to 17 267) was similar to the
range of values in AFs during labour at term. Of these three
atypical PTL samples, one (IL-1β = 42 050) was collected
after >48 h of PTL from a pregnancy with cervical cerclage in
place. In the case of a second sample (IL-1β = 49 050), the
pregnancy was complicated by prolonged (>24 h) labour
before AF collection with an IUD in the uterus. In the case of
the third sample (IL-1β = 34 690), the pregnancy was
complicated by protracted cervical dilatation for several days
preceding AF sampling.
522
S.M.Cox et al.
Table II. Immunoreactive interleukin (IL)-β1 in supernatant and
cellular pellet fractions of amniotic fluids (AFs) obtained by
transuterine amniocentesis at Caesarean section before and
during labour and by needle aspiration of the forebag during
labour
IL-β1 (pg/ml)
Pregnancy
1–60
Supernatant
ND
Cellular
ND
61
ND
2
62
ND
10
63
ND
11
64
ND
20
65
ND
135
66
51
ND
67
81
ND
68
85
ND
69
167
ND
70
181
ND
71
190
ND
72
231
ND
73
270
ND
74
330
ND
75
414
ND
76
133
19
77
274
21
78
187
52
79
206
60
80
68
88
81
194
89
82
142
132
83
288
245
84
332
260
85
209
302
86
1283
305
87
418
316
88
124
406
89
156
487
90
712
496
91
1706
624
92
1739
651
93
2433
802
94
116
1904
95
1029
2490
concentrations of IL-1β in AFs as a function of the duration
of labour at term. This was true in analyses of all AF samples
and in separate analyses of IL-1β-positive AFs. Both the
incidence of IL-1β positivity and the concentrations of IL-1β
in AFs collected after 18 h of PTL, however, were
significantly greater than that in any other group (P < 0.01)
There were no AFs collected after spontaneous labour of
>18 h duration with which the AF data of PTL of >18 h
duration could be compared.
Interleukin-6 in AF before and during labour at term
and preterm
IL-6 was present in AF of most pregnancies at all stages of
gestation examined (14–42 weeks) irrespective of the
absence or presence of labour (Figure 2). The incidence of
IL-6-positive AFs (>150 pg/ml) from pregnancies of 16–34
weeks gestation and at term before labour was 23/41 (56%)
and 77/120 (64.2%) respectively. The concentration of IL-6
(pg/ml; mean ± SEM) in the 16–34 wk samples was
529 ± 109 (n = 41; range: ND to 2756), and at term before
labour, it was 446 ± 50 (n = 120; range ND to 2919). IL-6
was detected in 50/60 (83.3%) of AFs from the upper
compartment during labour at term, in 172/175 (98.3%) of
AFs of the forebag, and in 79/90 (87.8%) of the PTL AFs; of
these AFs, 46/90 (51.1%) were collected before 18 h of PTL
and 33/90 (36.7%) were collected after >18 h of PTL.
(Figure 2). In term pregnancies during spontaneous labour,
the concentration of IL-6 in AF of the forebag was
11 544 ± 1740 (n = 175; range: ND to 149 625) and in AF of
the upper compartment, 8974 ± 4378 (n = 60; range: ND to
224 650). The concentration of IL-6 in PTL pregnancies
(irrespective of duration of labour) was 41 851 ± 8673
(n = 90; range: ND to 425 900). The concentration of IL-6 in
AFs collected before 18 h of PTL, 18 094 ± 6865 (n = 55;
range: ND to 296 400), was significantly less (P < 0.001)
than that in AFs collected after >18 h of PTL, 79 185 ±
17 935 (n = 35; range: ND to 425 900).
Correlation between the concentrations of IL-1β and
IL-6
ND = undetectable. The amount of AF available for developing the cell pellet varied from 1 to 15 ml and therefore, the
minimal amount of immunoreactive IL-1β detectable in the
sonicated cell pellet varied from 2 to 30 pg/cells in 1 ml.
During labour at term and preterm (irrespective of the
duration of labour), there were highly significant correlations
between the concentrations of IL-1β and IL-6 (term:
r = 0.678, P < 0.001; preterm: r = 0.679, P < 0.001).
IL-1β in AF and the duration of labour at term and
preterm
Correlation between the IL-1β content of cellular and
supernatant fractions of AF
Neither the incidence nor the concentration of IL-1β in AF
increased with cervical dilatation during term or preterm
labour (Table I). Moreover, there was no increase in the
In 60 AFs from pregnancies at term in which IL-1β was not
detected in the supernatant, IL-1β was not detected in the
cellular pellet. In 35 other AF samples in which IL-1β was
Interleukins-1β and -6 in amniotic fluid
Figure 2. The levels of interleukin (IL)-6 (pg/ml; mean ± SEM) in
amniotic fluids (AFs) collected before and during labour at term
(upper and forebag compartments) and preterm. The values for AFs
collected after <18 h or >18 h of preterm labour (PTL) are presented
separately. The number of samples in each group is shown above each
bar. NIL = not in labour.
detectable in the supernatant or the cellular pellet, or both,
there was a significant correlation (r = 0.421; P < 0.02)
between the two values (Table II).
Discussion
In assessing the proposition that infection is a cause of PTL
by examinations of AFs (collected during labour), it is
essential to ascertain whether the inflammatory process, if
identified, preceded or followed the onset of labour. The
central question is as follows: Do inflammatory mediators
accumulate in AF only after labour is in progress? If this
were the case, the accumulation of mediators of
inflammation in AF should be similar during term labour and
PTL, provided other factors, such as the duration of labour
before AF collection were similar.
523
In this study, IL-1β was not detected in AFs collected
before the onset of labour at any stage of gestation (n = 320),
including 170 AFs collected at term. Similar findings have
been reported in smaller sample numbers by others (Tsunoda
et al., 1990; Romero et al., 1992; Cox et al., 1993; Opsjon
et al., 1993). IL-1β was present (>50 pg/ml); however, in
54.7 and 32.2% of AFs collected during PTL and
spontaneous labour at term respectively. The higher
incidence of IL-1β-positive AFs during PTL, however, was
accounted for by the very high incidence among fluids
collected after prolonged PTL with cervical dilatation.
Importantly, there was no correlation between the extent of
cervical dilatation and the incidence or concentrations of
IL-1β in AF. Rather, there was an increased incidence of and
increased concentrations of IL-1β in AF of PTL pregnancies
after prolonged exposure of the forebag (>18 h) irrespective
of the extent of cervical dilatation. These findings are
suggestive that invasion of the forebag/AF by microorganisms occurs quickly once the cervix is dilated and that
with very prolonged cervical dilatation, invasion by
micro-organisms is almost assured. Medically-supervised
labour at term rarely is allowed to persist for >18 h, even in
those pregnancies in which this might have been the natural
outcome. Slow or desultory labour and failure of labour to
progress are viewed as significant complications of labour at
term; and in most such cases, pharmacological measures or
surgical intervention is chosen to effect a more expedient
delivery. Commonly, however, the obstetrical management
of PTL is quite different. The hope is held that PTL may
cease; and, there is hesitancy to use pharmacological
augmentation of uterine contractions (e.g. with oxytocin)
because of abnormal fetal lie or presentation and fear of
potential injury to the preterm fetus. Consequently,
prolonged cervical dilatation (with desultory labour) preterm
is common. Therefore, groups of AFs from pregnancies with
PTL usually are not comparable (with respect to the duration
of labour before AF collection) with groups of AFs collected
during term labour. This variable was considered in the
analyses of the data of this study. Labour was in progress for
>18 h before AF was collected in 36% of the PTL
pregnancies of this study. Contrarily, AF sampling, as well as
labour and delivery, were completed in <18 h in all 240 term
pregnancies studied. When the duration of PTL at the time of
AF collection was comparable with that of AFs collected
during spontaneous labour at term, i.e. <18 h, the incidence
of IL-1β-positive AFs (34.3%) was indistinguishable from
that AFs collected during labour at term from the upper
(27.3%) or forebag (34%) compartments. When AF was
obtained after >18 h of PTL, however, IL-1β was almost
always present (89.7%) and the concentration was much
greater than that in AF at term (forebag or upper
524
S.M.Cox et al.
compartment). In AFs collected before 18 h of PTL, the
concentration of IL-1β was not significantly different from
that in AF of the upper or forebag compartment during
labour at term. Of the 106 AFs of PTL of this study evaluated
for IL-1β, concentrations greater than those in the range of
values in AFs collected during spontaneous labour at term
were found in only three cases (2.8%). In each of these three
PTL pregnancies, there were confounding complications
(i.e. labour >24 h in all three pregnancies before AF
sampling, cervical cerclage in one, and an intrauterine device
in decidua in one). In these three AFs, the concentrations of
IL-1β were very high, being two or three times greater than
the highest level in AFs collected during term labour.
Therefore, IL-1β is detected in approximately one third of
AFs collected during the first 18 h of labour, both at term and
preterm, and the mean and range of values for IL-1β in these
AFs are similar.
The tissue (cellular) source of IL-1β that is found in
approximately one third of AFs during labour at term and
preterm (<18 h) is not defined. Amnion tissue, fetal urine,
and fetal lung secretions, however, have been excluded as
significant sources. Although 34 kDa pro-IL-1β is
synthesized in many tissues, including endometrial stromal
and decidual cells, the pro-IL-1β formed in cells other than
leukocytes and trophoblast is not processed to 17 kDa IL-1β
and, therefore, it is not secreted (Kostura et al., 1989; Cerretti
et al., 1992). These characteristics of IL-1β processing and
secretion led us to consider the possibility that IL-1β in AF
most likely arises by secretion from mononuclear
phagocytes or neutrophils, cells capable of converting
pr-IL-1β to 17 kDa IL-1β. The rate of IL-1β secretion from
forebag decidual tissue is great (Cox et al., 1993);
nonetheless, IL-1β is not detected in the forebag AF of
approximately two thirds of pregnancies during labour at
term. Therefore, the transfer of IL-1β from decidua across
the fetal membranes into the AF is probably negligible. In
support of this premise, Kent et al. (1994) demonstrated that
the transfer of radiolabelled IL-1β across the fetal
membranes in vitro is very limited. It is likely, therefore, that
IL-1β in AF arises in situ, i.e. from leukocytes recruited into
the AF (Cox et al., 1993). In support of this proposition, a
high correlation was found in this study between the amount
of IL-1β in the cellular pellet of the AF and the concentration
of IL-1β in the 600 g supernates of AFs collected during
labour. Before the onset of labour, there are very few
monocytes or macrophages in AF, except in the case of
selected fetal anomalies, e.g. open tissue defects, as with
anencephaly (Polgar et al., 1984). Leukocytes normally do
not cross amnion tissue; but in the case of cytokine- or
bacterial toxin-activated leukocytes, the transmembrane
migration of these cells is facilitated (Azzarelli and Lafuze,
1987; Kirchheimer and Remold, 1989; Bakowski and
Tschesche, 1992). The recruitment of leukocytes into the AF
should be expedited by the action of two potent
chemoattractants, i.e. IL-8 and monocyte chemotactic
protein-1 (MCP-1), which are synthesized in amnion in
response to IL-1β and bacterial toxins (Trautman et al.,
1992; S.M.Cox, unpublished data).
The obvious and important question is: why is IL-1β
present together with increased concentrations of other
mediators of inflammation in only approximately one third
of amniotic fluids during labour at term and preterm? Firstly,
it is important to recognize that the incidence of positive
cultures for micro-organisms is the same in AFs collected
during labour at term and preterm (Gomez et al., 1995).
Secondly, bacteria that are uniquely capable of burrowing
through the fetal membranes once the forebag is exposed in
the vagina (e.g. fusobacteria; Altshuler and Hyde, 1985,
1988) are present in the vaginal fluid of some, but not all
women (Hill, 1993; Hill et al., 1984). Fusobacterium is the
single most common micro-organism in AF, being identified
in ~28% of positive cultures of AFs collected during PTL
(intact fetal membranes) (Chaim and Mazor, 1992) but in
only 9.9% of positive cultures of AF obtained after
premature rupture of the fetal membranes (Chaim and
Mazor, 1992) and in the vaginal fluid in only 9% of nonpregnant women (Bartlett and Polk, 1984). The relatively
greater incidence of fusobacteria in AF collected during
labour (intact fetal membranes) is consistent with the unique
capacity of these micro-organisms to burrow through the
intact membranes once labour is in progress and the forebag
is exposed through the dilated cervix. Microbial penetration
of the fetal membranes and infection of the AF is a logical
sequela of term and preterm labour in pregnancies in which
micro-organisms capable of amnion penetration are in the
vagina.
Several possible explanations can be offered for the high
incidence and concentrations of IL-1β in AFs collected after
>18 h of PTL. The human amnion and chorion laeve are
avascular and the vascular supply of the decidual tissue
fragments that remain attached to the forebag as it is torn
away from the uterus during cervical dilatation with labour is
severely compromised. After ≥18 h of forebag exposure with
diminished or absent blood supply to the attached decidual
fragments, viability of these tissues declines and
vulnerability to invasion by bacteria, in general, increases.
Ultimately, the integrity of the fetal membranes also may be
compromised appreciably by small rupture sites, allowing
free entry of micro-organisms. After labour and cervical
dilatation for protracted times, i.e. >18 h, evidence of
intra-amnionic inflammation is almost always present.
Therefore, the finding of IL-1β in AF during PTL is not
Interleukins-1β and -6 in amniotic fluid
indicative of an infection-based cause of PTL; rather, the
appearance of IL-1β in AFs of approximately one third of
pregnancies during labour at term and preterm, is a normal
sequela of labour.
A number of investigators also have cited the finding of
very high concentrations of IL-6 or IL-8 in AF as evidence in
support of an intrauterine infection-based cause of PTL
(Romero et al., 1990a, 1991; Cherouny et al., 1993; Hillier
et al., 1993; Silver et al., 1993). IL-6 and IL-8, unlike IL-1,
however, are present in AF of normal pregnancies
throughout most of gestation (Romero et al., 1990a, 1991,
1993; Cherouny et al., 1993; Greig et al., 1993; Hillier et al.,
1993; Laham et al., 1993; Opsjon et al., 1993; Puchner et al.,
1993; Silver et al., 1993). These cytokines, again unlike
IL-1β, are produced by amnion cells (Mitchell et al., 1991;
Trautman et al., 1992). Moreover, the production of IL-8 by
amnion epithelial cells (Trautman et al., 1992) and LL-6 and
IL-8 by amnion mesenchymal cells (unpublished
observations) is increased strikingly by IL-1β. It is not
surprising, therefore, that the concentrations of IL-6 and IL-8
are increased in AFs in which IL-1β or LPS is present.
Indeed, a very high correlation was found between the
concentrations of IL-1β and IL-6 in AFs during labour at
term and preterm in this study. We conclude that increased
concentrations of IL-6, other cytokines, and related
compounds in AF are to be expected whenever IL-1β is
present in AF.
Finally, it should be emphasized that the infection-based
theory of PTL is anchored to the proposition that the
proximate event in the onset of PTL is the accelerated
production of PGs. The concentrations of PGs in AF are
increased modestly in some but not all pregnancies during
PTL (MacDonald and Casey, 1996). Indeed, several
investigators have expressed surprise in finding that the
concentrations of PGs in AF during PTL are quite low
(Gomez et al., 1995). In previous studies, we found that the
increase in the concentration of PGE2, PGF2α, and the
metabolite of PGF2α (PGFM) in AF during labour at term
and preterm was accounted for primarily, if not exclusively,
by the formation of these prostanoids as a consequence of
labour (MacDonald and Casey, 1993, 1996). In the case of
PTL, the concentrations of PGE2 and PGF2α in AF are not
strikingly elevated (Gomez et al., 1995), being similar to
those in AF of the upper compartment during labour at term
(MacDonald and Casey, 1996). The preterm (<34 weeks)
fetal presenting part does not obstruct the maternal pelvis,
hence there is a single AF compartment during labour in
such pregnancies. These findings are supportive of the
conclusion that forebag tissues are the source of PGs in AF
during PTL, as is the case at term (MacDonald and Casey,
1993, 1996; Cox et al., 1993).
525
A slight parturition-independent increase in the values of
PGs in AF as a function of gestational age is to be expected.
This phenomenon can be attributed to: (i) an increase in the
rate of excretion of fetal urine, a major source of PGs in AF
before the onset of labour (Casey et al. 1983) and (ii) a
decrease in AF volume, which is observed in many
pregnancies near term (Gillibrand, 1969). Even this slight
increase in the concentrations of PGs, however, is not
observed in all pregnancies; and indeed, in some, the
concentrations of PGs in AF decrease somewhat. In a study
of the concentrations of PGs in AFs obtained by multiple
amniocenteses conducted from 37 weeks gestation until the
onset of labour, Romero et al. (1996) reported that, on
average, there was an increase in the concentrations of PGE2
and PGF2α in AF before the onset of labour PGF2α. The
effect of multiple amniocentesis on the concentrations of
PGs in AF is not known. In a study of 28 subjects in whom
AF was collected by amniocentesis twice before the onset of
labour, the concentration of PGE2 increased in AF from 22
subjects, decreased in one, and appeared to be unchanged in
five. In two of these pregnancies, the concentration of PGE2
in AF was greater before labour than during labour. And, in
many of these pregnancies, the concentration of PGE2 was
greater before labour than in other pregnancies during
labour. Similar findings were reported for PGF2α. The mean
increases in PGE2 and PGF2α during the time between the
two amniocenteses during the week(s) before labour were
~0.5 ng/ml. Assuming an AF fluid volume of ~1000 ml and
a half-life of 6–8 h for PGs in AF, the increase in the rate of
entry of PGE2 was <1.2 µg/day. The computed rate of entry
of PGF2α is similar.
Even if multiple amniocenteses do not contribute
appreciably to an increase in AF concentrations of PGs, the
modest increase in PGs in some, but not all, pregnancies
before the onset of labour cannot be interpreted as evidence
for increased formation of PGs as a causal event in the
initiation of parturition. Rather, the slight increase in PGs in
AF in some pregnancies during the last few weeks of
pregnancy is most likely attributable to fetal growth,
increased fetal urine output, and a decrease in AF volume. In
view of the very small amount of PGs in AF before the onset
of labour, the great efficiency for PG degradation in chorion
laeve and, consequently, the multiple obstacles in transfer of
PGs from amniotic fluid or amnion to the myometrium, a
role for PGs produced in the amnion in the initiation of
parturition cannot be envisioned.
For decades, many investigators, including ourselves, held
to the hope that infection could be established as an
important cause of PTL. The findings of this and companion
studies (MacDonald and Casey, 1993, 1996; Casey et al.,
1993; Cox et al., 1993) are not supportive of this theory or of
526
S.M.Cox et al.
the utility of analyses of AF for micro-organisms or
mediators of inflammation in establishing a causal
relationship between intrauterine infection and PTL.
Acknowledgements
This investigation was supported, in part, by USPHS Grant
5-P50-HD-11149. Dr Cox was a recipient of Clinical Investigator
Award K08-HD00853. We thank Jess Smith for skilled technical
assistance, and Kathy Loppnow and Rosemary Bell for expert
editorial assistance.
References
Altshuler, G. and Hyde, S. (1985) Fusobacteria. An important cause of
chorioamnionitis. Arch. Pathol. Lab. Med., 109, 739–743.
Altshuler, G. and Hyde, S. (1988) Clinicopathologic considerations of
Fusobacteria chorioamnionitis. Acta Obstet. Gynecol. Scand., 67,
513–517.
Azzarelli, B. and Lafuze, J. (1987) Amniotic basement membrane: a barrier
to neutrophil invasion. Am. J Obstet. Gynecol., 156, 1130–1136.
Bakowski, B. and Tschesche, H. (1992) Migration of polymorphonuclear
leukocytes through human amnion membrane – a scanning electron
microscopic study. Biol. Chem. Hoppe Seyler, 373, 529–546.
Bartlett, J.G. and Polk, B.F. (1984) Bacterial flora of the vagina: quantitative
study. Rev. Infect. Dis., 6, 567–572.
Bry, K. and Hallman, M. (1991) Synergistic stimulation of amnion cell
prostaglandin E2 synthesis by interleukin-1, tumor necrosis factor and
products from activated human granulocytes. Prostagl. Leuk. Essent.
Fatty Acid, 44, 241–245.
Canadian Preterm Labor Investigators Group (1992) Treatment of preterm
labor with the beta-adrenergic agonist ritodrine. N. Engl. J. Med., 327,
308–312.
Casey, M.L., Cutrer, S. and Mitchell, M.D. (1983) Origin of prostanoids in
human amniotic fluid: the fetal kidney as a source of amniotic fluid
prostanoids. Am. J. Obstet. Gynecol., 147, 547–551.
Casey, M.L., Cox, S.M., Word, R.A. et al. (1990) Cytokines and
infection-induced preterm labor. Reprod. Fertil. Dev., 2, 499–509.
Casey, M.L., Brown, C.E.L., Peters, M. et al. (1993) Endothelin levels in
human amniotic fluid at mid-trimester and at term before and during
spontaneous labor. J. Clin. Endocrinol. Metab., 76, 1647–1650.
Cerretti, D.P., Kozlosky, C.J., Mosley, B. et al. (1992) Molecular cloning of
the interleukin-1 beta converting enzyme. Science, 256, 97–100.
Chaim, W. and Mazor, M. (1992) Intraamniotic infection with fusobacteria.
Arch. Gynecol. Obstet., 251, 1–7.
Cherouny, P.H., Pankuch, G.A., Romero, R. et al. (1993) Neutrophil
attractant/activating peptide-1/interleukin-8: Association with
histologic chorioamnionitis, preterm delivery, and bioactive amniotic
fluid leukoattractants. Am. J. Obstet. Gynecol., 169, 1299–1303.
Cox, S.M., MacDonald, P.C. and Casey, M.L. (1988) Assay of bacterial
endotoxin (lipopolysaccharide) in human amniotic fluid: potential
usefulness in diagnosis and management of preterm labor. Am. J.
Obstet. Gynecol., 159, 99–106.
Cox, S.M., Sherman, M.L. and Leveno, K.J. (1990) Randomized
investigation of magnesium sulfate for prevention of preterm birth. Am.
J. Obstet. Gynecol., 163, 767–772.
Cox, S.M., King, M.R., Casey, M.L. et al. (1993) Interleukins-1β, 1α, and -6
and prostaglandins in vaginal/cervical fluids of pregnant women before
and during labor. J. Clin Endocrinol. Metab., 77, 805–815.
Cunningham, F.G., MacDonald, P.C., Gant, N.F. et al. (1997) Williams
Obstetrics. 20th edn. Appleton and Lange, Stamford, CT, USA.
Gillibrand, P.N. Changes in amniotic fluid volume with advancing
pregnancy. J. Obstet. Gynaecol. Br. Commonw., 76, 527–529.
Gomez, R., Ghezzi, F., Romero, R. et al. (1995) Premature labor and
intra-amniotic infection. Clinical aspects and role of the cytokines in
diagnosis and pathophysiology. Clin. Perinatol., 22, 281
Greig, P.C., Ernest, J.M., Teot, L. et al. (1993) Amniotic fluid interleukin-6
levels correlate with histologic chorioamnionitis and amniotic fluid
cultures in patients in premature labor with intact membranes. Am. J.
Obstet. Gynecol., 169, 1035–1044.
Hill, G.B. (1993) Investigating the source of amniotic fluid isolates of
fusobacteria. Clin. Infect. Dis., 16, S423–S424.
Hill, G.B., Eschenbach, D.A. and Holmes, K.K. (1984) Bacteriology of the
vagina. Scand. J. Urol. Nephrol., 86 (Suppl.), 23–39.
Hillier, S.L., Witkin, S.S., Krohn, M.A. et al. (1993) The relationship of
amniotic fluid cytokines and preterm delivery, amniotic fluid infection,
histologic chorioamnionitis, and chorionamnion infection. Obstet.
Gynecol., 81, 941–948.
Kent, A.S.H., Sullivan, M.H.F. and Elder, M.G. (1994) Transfer of cytokines
through human fetal membranes. J. Reprod. Fertil., 100, 81–84.
Kirchheimer, J.C. and Remold, H.G. (1989) Endogenous receptor-bound
urokinase mediates tissue invasion of human monocytes. J. Immunol.,
143, 2634–2639.
Kostura, M.J., Tocci, M.J., Limjuco, G. et al. (1989) Identification of a
monocyte specific pre-interleukin-1β convertase activity. Proc. Natl.
Acad. Sci. USA, 86, 5227–5231.
Laham, N., Rice, G.E., Bishop, G.J. et al. (1993) Interleukin 8 concentrations
in amniotic fluid and peripheral venous plasma during human
pregnancy and parturition. Acta Endocrinol., 129, 220–224.
Leveno, K.J., Little, B.B. and Cunningham, F.G. (1990) The national impact
of ritodrine hydrochloride for inhibition of preterm labor. Obstet.
Gynecol., 76, 12–15.
Leveno, K.J., Klein, V.R., Guzick, D.S. et al. (1986) Single-centre
randomised trial of ritodrine hydrochloride for preterm labour. Lancet, i,
1293–1296.
MacDonald, P.C. and Casey, M.L. (1993) The accumulation of
prostaglandins (PG) in amniotic fluid is an aftereffect of labor and not
indicative of a role for PGE2 or PGF2α in the initiation of human
parturition. J. Clin. Endocrinol. Metab., 76, 1332–1339.
MacDonald, P.C. and Casey M.L. (1996) Preterm Birth. Sci. Am. Sci. Med.,
3, 42–51.
MacDonald, P.C., Cox, S.M. and Casey, M.L. (1988) Infection-associated
preterm labor as a model for parturition. In Genazzani, A.R., Petraglia,
F., Volpe, A. and Facchinetti, F. (eds), Advances in Gynecological
Endocrinology. Parthenon Publishing, Carnforth, UK, pp. 487–493.
MacDonald, P.C., Koga, S., and Casey, M.L. (1991) Decidual activation in
parturition: examination of amniotic fluid for mediators of the
inflammatory response. Ann. N.Y. Acad. Sci., 622, 315–330.
Mitchell, M.D., Edwin, S. and Romero, R.J. (1990) Prostaglandin
biosynthesis by human decidual cells: effects of inflammatory
mediators. Prostagl. Leuk. Essent. Fatty Acid, 41, 35–38.
Mitchell, M.D., Branch, D.W., Lundin-Schiller, S. et al. (1991)
Immunologic aspects of preterm labor. Semin. Perinatol., 15, 210–224.
Opsjon, S.-L., Wathen, N.C., Tingulstad, S. et al. (1993) Tumor necrosis
factor, interleukin-1, and interleukin-6 in normal human pregnancy. Am.
J. Obstet. Gynecol., 169, 397–404.
Polgar, K., Sipka, S., Abel, G. et al. (1984) Neutral red uptake by amniotic
fluid macrophages in neural tube defects: a rapid test. N. Engl. J. Med.,
310, 1463–1464.
Puchner, T., Egarter, C., Wimmer, C. et al. (1993) Amniotic fluid
interleukin-8 as a marker for intraamniotic infection. Arch. Gynecol.
Obstet., 253, 9–14.
Romero, R., Mazor, M., Wu, .K. et al. (1988a) Infection in the pathogenesis
of preterm labor. Semin. Perinatol., 12, 262–279.
Romero, R., Roslansky, P., Oyarzun, E. et al. (1988b) Labor and infection. II.
Bacterial endotoxin in amniotic fluid and its relationship to the onset of
preterm labor. Am. J. Obstet. Gynecol., 158, 1044–1049.
Romero, R., Wu, .K., Mazor, M. et al. (1988c) Amniotic fluid prostaglandin
E2 in preterm labor. Prostagl. Leuk. Essent. Fatty Acid, 34, 141–145.
Romero, R., Durum, S.K., Dinarello, C.A. et al. (1989a) Interleukin
stimulates prostaglandin biosynthesis by human amnion.
Prostaglandins, 37, 13–22.
Romero, R., Wu, .K., Sirtori, M., et al. (1989b) Amniotic fluid concentrations
of prostaglandin F2α, 13, 14-dihydro-15-keto-prostaglandin F2α (PGFM)
and 11-deoxy-13,14-dihydro-15-keto-11, 16-cyclo-prostaglandin E2
(PGEM-II) in preterm labor. Prostaglandins, 37, 149–161.
Interleukins-1β and -6 in amniotic fluid
Romero, R., Avila, C., Santhanam, U. et al. (1990a) Amniotic fluid
interleukin-6 in preterm labor. Association with infection. J. Clin.
Invest., 85, 1392–1400.
Romero, R., Parvizi, S.T., Oyarzun, E. et al. (1990b) Amniotic fluid
interleukin-1 in spontaneous labor at term. J. Reprod. Med., 35,
235–238.
Romero, R., Ceska, M., Avila, C. et al. (1991) Neutrophil
attractant/activating peptide-1/interleukin-8 in term and preterm
parturition. Am. J. Obstet. Gynecol., 165, 813–820.
Romero, R., Mazor, M., Brandt, F. et al. (1992) Interleukin-1α and
interleukin-1β in preterm and term human parturition. Am. J. Reprod.
Immunol., 27, 117–123.
Romero, R., Yoon, B.H., Mazor, M. et al. (1993) The diagnostic and
prognostic value of amniotic fluid white blood cell count, glucose,
interleukin-6, and Gramstain in patients with preterm labor and intact
membranes. Am. J. Obstet. Gynecol., 169, 805–816.
Romero, R., Munoz, H., Gomez, R. et al. (1994) Does infection cause
premature labor and delivery? Sem. Reprod. Endocrinol., 12, 227–239.
Romero, R., Munoz, H., Gomez, R. et al. (1996) Increase in prostaglandin
527
bioavailability precedes the onset of human parturition. Prostaglandins,
Leukotrienes, and Essential Fatty Acids, 54, 187–191.
Schwartz, A.L., Forster, C.S., Smith, P.A. et al. (1977) Human amnion
metabolism. I. In vitro maintenance. Am. J. Obstet. Gynecol., 127,
470–474.
Silver, R.M., Schwinzer, B. and McGregor, J.A. (1993) Interleukin-6 levels
in amniotic fluid in normal and abnormal pregnancies: preeclampsia,
small-for-gestational-age fetus, and premature labor. Am. J. Obstet.
Gynecol., 169, 1101–1105.
Trautman, M.S., Dudley, D.J., Edwin, S.S. et al. (1992) Amnion cell
biosynthesis of interleukin-8: regulation by inflammatory cytokines. J.
Cell Physiol., 153, 38–43.
Tsunoda, H., Tamatani, T., Oomoto, Y. et al. (1990) Changes in interleukin 1
levels in human amniotic fluid with gestational ages and delivery.
Microbiol. Immunol., 34, 377–385.
Received on February 24, 1997; accepted on September 9, 1997