154 l FEBRUARY JOGC FÈVRIER 2012

OBSTETRICS

Key Words: Periviable preterm premature rupture of membranes, amniotic fluid, perinatal survival

Competing Interests: None declared .

Received on July 21, 2011

Accepted on September 30, 2011

Correlation of Residual Amniotic Fluid and Perinatal Outcomes in Periviable Preterm Premature Rupture of MembranesClaudine Storness-Bliss, BSc,1 Amy Metcalfe, MSc,2 Rebecca Simrose, MD,3 R. Douglas Wilson, MD,3 Stephanie l. Cooper, MD3

1Faculty of Medicine, University of Calgary, Calgary AB2Community Health Sciences, University of Calgary, Calgary AB3Department of Obstetrics and Gynecology, University of Calgary, Calgary AB

J Obstet Gynaecol Can 2012;34(2):154–158

Abstract

Objective: To correlate maternal and fetal outcomes of pregnancies affected by preterm premature rupture of membranes (PPROM) at < 24 weeks’ gestational age with the amount of residual amniotic fluid as determined by sonographic evaluation .

Methods: We searched the local maternal-fetal medicine database for the records of all women with PPROM prior to 24 completed weeks of pregnancy . The quantity of residual amniotic fluid determined by ultrasound was recorded and women were separated into two groups: (A) deepest vertical pocket (DVP) ≥ 1 cm, or (B) DVP < 1 cm (severe oligohydramnios) . Hospital chart review was undertaken to determine latency to delivery, perinatal death, and maternal complications . Data were analyzed using Fisher exact and Wilcoxon-Mann-Whitney U tests .

Results: We identified 31 subjects, of whom nine elected termination of pregnancy (6 in group A, 3 in group B) . Six of 10 subjects in group A had a live delivery without neonatal death, whereas only one of 12 subjects in group B had a live delivery (P = 0 .020) . Additional complications included placental abruption in 63% in group A and 45% in group B, chorioamnionitis in 50% and 70%, respectively, and postpartum endometritis in 0% and 9%, respectively . None of these differences were statistically significant . There were no cases of maternal sepsis or maternal death in either group . Group A was associated with a later GA at delivery (27 .5 weeks vs . 23 weeks, P = 0 .07), with the GA at rupture of the membranes similar for both groups .

Conclusion: These results indicate that a higher level of residual amniotic fluid after periviable PPROM is associated with fetal survival and increased latency to delivery without an increase in maternal complications . This information will be valuable in counselling pregnant women with PPROM < 24 weeks .

Résumé

Objectif : Chercher à établir une corrélation entre les issues maternelles et fœtales de grossesses affectées par la rupture prématurée des membranes préterme (RPMP) à un âge gestationnel < 24 semaines et la quantité de liquide amniotique résiduel déterminée par évaluation échographique .

Méthodes : Nous avons mené des recherches dans la base de données locale de médecine fœto-maternelle afin d’en tirer les dossiers de toutes les femmes ayant présenté une RPMP avant 24 semaines complètes de grossesse . La quantité de liquide amniotique résiduel déterminée par échographie a été consignée et les femmes ont été réparties en deux groupes : (A) poche verticale la plus profonde (PVP) ≥ 1 cm ou (B) PVP < 1 cm (oligohydramnios grave) . Une analyse des dossiers hospitaliers a été menée afin de déterminer le délai avant l’accouchement, la présence d’un décès périnatal et la survenue de complications maternelles . Les données ont été analysées au moyen du test exact de Fisher et du test de Wilcoxon-Mann-Whitney U .

Résultats : Nous avons identifié 31 sujets, neuf desquels ont choisi l’interruption de grossesse (6 du groupe A, 3 du groupe B) . Six sujets du groupe A sur 10 ont donné lieu à une naissance vivante sans décès néonatal, tandis que cela n’a été le cas que chez un sujet du groupe B sur 12 (P = 0,020) . Parmi les complications additionnelles, on trouvait le décollement placentaire chez 63 % des sujets du groupe A et 45 % des sujets du groupe B, la chorioamnionite (50 % et 70 %, respectivement) et l’endométrite postpartum (0 % et 9 %, respectivement) . Aucune de ces différences ne s’est avérée significative sur le plan clinique . Aucun cas de septicémie maternelle ou de décès maternel n’a été constaté au sein des groupes . Le groupe A a été associé à un AG plus avancé au moment de l’accouchement (27,5 semaines, par comp . avec 23 semaines, P = 0,07), l’AG au moment de la rupture des membranes étant similaire dans les deux groupes .

Conclusion : Ces résultats indiquent qu’un niveau accru de liquide amniotique résiduel à la suite d’une RPMP périviable est associé à la survie fœtale et à un délai accru avant l’accouchement, sans augmentation de la fréquence des complications maternelles . Cette information s’avérera utile dans le cadre du counseling des femmes enceintes présentant une RPMP < 24 semaines .

FEBRUARY JOGC FÈVRIER 2012 l 155

Correlation of Residual Amniotic Fluid and Perinatal Outcomes in Periviable Preterm Premature Rupture of Membranes

INTRODUCTION

Preterm premature rupture of membranes occurs in approximately 1% of continuing pregnancies, and,

although associated with perinatal morbidity and mortality, generally results in good maternal and fetal outcomes.1 Periviable PPROM occurs in 0.4% of continuing pregnancies, and is defined as rupture of membranes prior to the age of fetal viability, but after most spontaneous abortions have occurred (14 to 24 gestational weeks).2 Perinatal outcomes in these cases have been reported as poor, mainly due to secondary deformational pulmonary hypoplasia, although neurological complications, infection, and congenital malformations have also been described.2,3 Furthermore, maternal health following periviable PPROM may be compromised due to the risks of chorioamnionitis, sepsis, placental abruption, and complications of immobility.4

Many factors are important when counselling the pregnant patient with periviable PPROM, including gestational age, latency to delivery, and risks of expectant management, given the high rate of associated complications. Data on maternal/neonatal morbidity, fetal/neonatal mortality, and fetal treatment are limited, because studies are generally small and differ in their inclusion and exclusion criteria. Dewan and Morris, in a systematic review of PPROM < 23 weeks between the years 1980 to 1994, reported a 20% perinatal survival rate.5 Although neonatal care has advanced, periviable PPROM has typically been associated with a very poor prognosis, and many patients are counselled to consider termination of pregnancy.

More recently, advances in both antenatal and neonatal care (including antenatal corticosteroid use, antibiotic use, use of postnatal surfactant, and neonatal gentle ventilation strategies) have allowed the definition of fetal viability to be revisited and the outcome for periviable PPROM to be questioned.2–8 Everest et al. reported a 55.7% survival rate after conservative management of periviable PPROM.6

To counsel patients in the setting of periviable PPROM appropriately, it is essential to present the best available options, including the risks and potential complications, using pertinent and centre-based outcome data. Fifty percent of pregnancies complicated by periviable PPROM deliver within seven days, with the majority delivering

within 48 hours.4 Therefore, cases that do not present with imminent labour or signs of chorioamnionitis require those involved to make the difficult decision either to pursue expectant management or to terminate the pregnancy.

One clinical variable which may assist in the counselling of these patients is the residual amniotic fluid volume as determined by ultrasound. Hadi et al., in a prospective cohort of 178 singleton pregnancies complicated by PPROM at 20 to 25 weeks’ gestational age, found that the presence or absence of amniotic fluid was associated with latency to labour.9 However, their study excluded PPROM before 20 weeks’ gestation, included amniotic fluid levels measured throughout pregnancy, and excluded all cases with latency to labour of less than seven days. They found that the frequency of chorioamnionitis was higher (33.8% vs. 21.5%, P = 0.07) and perinatal survival was lower (9.9% vs. 85%, P < 0.01) in patients with oligohydramnios (defined as ultrasonographic measurement of the deepest vertical pocket of amniotic fluid as < 2 cm) than in patients without oligohydramnios. Kilbride et al. compared survivors and non-survivors following periviable PPROM.3

They found that non-survivors, in addition to being less mature than survivors, having a longer duration of PPROM, and lower birth weights, were more likely to have severe oligohydramnios, defined as < 1 cm vertical pocket on ultrasonographic measurement of amniotic fluid. The finding of improved outcomes with higher levels of amniotic fluid has been confirmed in women with PPROM at later gestations as well.10

The primary objective of this study was to correlate both the average latency to delivery at the time of PPROM and perinatal mortality (fetal death or neonatal death within 30 days of life) with the level of residual amniotic fluid, as measured by 2-D ultrasound. Secondary outcomes included the incidence of chorioamnionitis, placental abruption, postpartum endometritis, retained placenta, maternal sepsis, and death in cases of periviable PPROM.

METHODS

We evaluated the outcomes of singleton pregnancies with rupture of membranes prior to 24 weeks of gestation and with a minimal latency to delivery of 48 hours, occurring between January 1, 2002, and January 15, 2011, in a retrospective cohort study. Gestational age was determined by last menstrual period and confirmed when possible by first trimester ultrasound. Rupture of membranes was diagnosed by clinical history in combination with a sterile speculum examination demonstrating vaginal pooling of amniotic

ABBREVIATIONSDVP deepest vertical pocket

PPROM preterm premature rupture of membranes

156 l FEBRUARY JOGC FÈVRIER 2012

OBSTETRICS

fluid and microscopic identification of a ferning pattern. Subjects who presented with chorioamnionitis (fever, abdominal pain, and fetal tachycardia), fetal demise, or active labour (uterine contractions with cervical dilatation) at the time of PPROM were excluded from the study. Multiple gestations, PPROM following amniocentesis or chorionic villus sampling, or any pregnancies with a pre-existing diagnosis of major congenital or chromosomal abnormalities were also excluded.

Patients with PPROM were referred to the Southern Alberta Centre for Maternal-Fetal Medicine for evaluation and counselling. The local maternal–fetal medicine electronic database was searched to retrieve the records of PPROM cases and to review sonographic findings including residual amniotic fluid volume. Cases were separated into two groups:

A residual amniotic fluid demonstrating a deepest vertical pocket ≥ 1 cm, and

B residual amniotic fluid DVP < 1 cm (severe oligohydramnios).

Perinatal and maternal outcomes were determined from manual chart review at the delivery hospital. Bivariate associations were assessed using the Fisher exact test for

categorical variables and the Wilcoxon rank sum test for continuous variables. Because of the small sample size and descriptive nature, P < 0.10 was deemed statistically significant.

This study was approved by the Conjoint Health Research Ethics Board at the University of Calgary.

RESUlTS

A total of 31 pregnancies met the inclusion criteria for evaluation. Following sonographic evaluation and counselling, nine subjects elected for pregnancy termination, six in the DVP < 1 cm group and three in the DVP ≥ 1 cm group (Table 1). For analysis of subjects who followed expectant management, there were 12 patients in the DVP < 1 cm group and 10 in the DVP ≥ 1 cm group.

The primary endpoints of this study are summarized in Table 2. There were statistically significant differences both in latency to delivery and in perinatal mortality between the two groups. The overall survival rate was 31.8% and mean latency to delivery was 43 days. Mean latency to delivery in the DVP < 1 cm group was 32 days, in contrast to 57 days in the DVP ≥ 1 cm group (P = 0.014). Further-more, survival was more than seven-fold increased in the DVP ≥ 1 cm group over the DVP < 1 cm group (60% vs. 8.3 %, P = 0.02).

The mean gestational age at the time of rupture of membranes was 18.5 weeks and at the time of delivery was 25 weeks. The mean gestational age at rupture of membranes and at time of delivery for both groups is shown in the Figure. There was no significant difference in gestational age at rupture of membranes between the DVP < 1 cm and DVP ≥ 1 cm groups (18 and 19 weeks respectively). However, gestational age at time of delivery was, on average, 4.5 weeks later in the DVP ≥ 1 cm group than in the DVP < 1 cm group (27.5 weeks vs. 22.9 weeks, P = 0.07).

Perinatal complications were assessed (Table 3). Seventy percent of patients in the DVP < 1 cm group and 50% of patients in the DVP ≥ 1 cm group experienced chorioamnionitis. Similarly, 9% of patients in the DVP < 1 cm group had postpartum endometritis, but none of the patients in the DVP ≥ 1 cm group did. These differences were not statistically significant. Placental abruption occurred in 45% of the DVP < 1 cm group and in 63% of the DVP ≥ 1 cm group. Rates of retained placenta were similar in the two groups (20% and 22%, respectively). There were no cases of maternal sepsis or death recorded during the study. Overall, there were no statistical differences between the two groups with regard to maternal complications.

Table 1. Subject demographics by group Parameter

DVP < 1 cm

DVP ≥ 1 cm

Number 18 13

Terminations 06 03

Expectant management 12 10

Table 2. Primary endpoints Primary endpoints

Both groups

DVP < 1 cm

DVP ≥ 1 cm

P

Latency, days 18 13 57 0 .014

Perinatal survival, n (%) 7 (31 .8) 1 (8 .3) 6 (60) 0 .02

Table 3. Maternal complications Complication

DVP < 1 cm, %

DVP ≥ 1 cm, %

P

Chorioamnionitis 70 50 0 .63

Endometritis 09 00 N/A

Placental abruption 45 63 0 .65

Retained placenta 20 22 > 0 .99

FEBRUARY JOGC FÈVRIER 2012 l 157

Correlation of Residual Amniotic Fluid and Perinatal Outcomes in Periviable Preterm Premature Rupture of Membranes

0

10

20

30

40

GA at ROM GA at delivery

DVP < 1 cm DVP 1 cm

P = 0.07

Gestational age at rupture of membranes and at delivery

DISCUSSION

The clinical scenario of periviable PPROM is challenging for both patient and clinician. In the absence of chorioamnionitis or active labour, the decision to pursue expectant management must be based on careful consideration of maternal/fetal risks and pregnancy outcomes in the setting of limited published data. Our sample size was limited by numerous factors: periviable PPROM is extremely rare, a number of patients either develop chorioamnionitis or progress to active labour, and strict inclusion criteria were used. In an effort to obtain consistent data, cases of multiple gestation, fetal demise, and congenital or chromosomal anomalies, and cases that progressed to active labour within 48 hours were excluded. Nonetheless, the survival rate in the present cohort was higher than previously described.5 This may reflect heterogeneity between patient groups in prior studies. Similarly, cases of iatrogenic PPROM following amniocentesis or chorionic villus sampling were excluded, because these cases typically have better outcomes.11

The present findings agreed with those of Hadi et al.,9 in that there was a significant difference in latency to delivery and perinatal survival between cases with severe oligohydramnios and those without. In contrast to the study of Hadi et al., our study included only the initial residual amniotic fluid measurements after early PPROM, as opposed to a mean of all amniotic fluid level measurements throughout pregnancy. This is an important difference as patients may benefit from timely information regarding the prognosis of their pregnancy.

As there was no difference in GA at onset of membrane rupture between the two groups in the present study, survival may be attributed to increased latency to labour resulting in later GA at delivery. Importantly, enhanced latency to delivery was achieved without increasing the incidence of maternal complications. Alternatively, as amniotic fluid is important for pulmonary development in mid-pregnancy, improved perinatal survival in the absence of severe oligohydramnios may be a direct consequence of the impact of residual amniotic fluid volume on lung development and subsequent postnatal lung function. Physiologically, factors that may contribute to pulmonary hypoplasia due to severe olighydramnios include fetal thoracic compression, restriction of fetal breathing, and disturbances of pulmonary fluid and flow.12

There are a number of therapies that have been proposed for the management of periviable PPROM, including amnioinfusion and artificial sealing of the membranes.13

However, the efficacy and safety of these treatments

have not been adequately studied, and additional research evaluating these strategies should be undertaken.

In animal models, labour can be induced by infusion of adrenocorticotropic hormone or cortisol.12 This mechanism does not appear to be directly responsible for triggering onset of parturition in humans and higher primates.12 We can hypothesize that in periviable PPROM, oligohydramnios could contribute to severe fetal stress, accompanied by an immense rise in fetal cortisol levels. In such a situation, it is possible that the maternal-fetal response to PPROM could revert to a less advanced mechanism, in evolutionary terms, for the onset of labour. If a higher level of residual amniotic fluid is associated with less thoracic compression and fewer disturbances of pulmonary fluid and flow, it could create a reduction in biological stress, resulting in an increased latency to the onset of labour.

CONClUSION

Choosing the appropriate management of periviable PPROM with respect to the level of amniotic fluid, pulmonary hypoplasia, latency to delivery, and survival is difficult. Our study shows an association between higher levels of residual amniotic fluid after PPROM and overall outcomes, but it is not possible to confirm the etiology of this association. A prospective study, including childhood outcomes, is the next logical step in establishing guidelines for the management of periviable PPROM. Further studies to validate the present findings are warranted. In the interim, routine evaluation of the residual amniotic fluid should be considered when counselling pregnant women with periviable PPROM.

158 l FEBRUARY JOGC FÈVRIER 2012

OBSTETRICS

REFERENCES

1. Mercer BM. Management of premature rupture of membranes before 26 weeks’ gestation. Obstet Gynecol Clin North Am 1992;19:339–51.

2. Waters TP, Mercer BM. The management of preterm premature rupture of the membranes near the limit of fetal viability. Am J Obstet Gynecol 2009;201:230–40.

3. Kilbride HW, Yeast J, Thibeault DW. Defining limits of survival: lethal pulmonary hypoplasia after midtrimester premature rupture of membranes. Am J Obstet Gynecol 1996;175:675–81.

4. Taylor J, Garite TJ. Premature rupture of membranes before fetal viability. Obstet Gynecol 1984;64:615–20.

5. Dewan H, Morris J. A systematic review of pregnancy outcome following preterm rupture of membranes at a previable gestational age. Aust N Z J Obstet Gynaecol 2001;41:389–94.

6. Everest NJ, Jacobs SE, Davis PG, Begg L, Rogerson S. Outcomes following prolonged preterm premature rupture of the membranes. Arch Dis Child Fetal Neonatal Ed 2008;93:F207–F211.

7. Dinsmoor MJ, Bachman R, Haney EI. Outcomes after expectant management of extremely preterm premature rupture of the membranes. Am J Obstet Gynecol 2004;190:183–7.

8. Muris C, Girard B, Creveuil C. Management of premature rupture of membranes before 25 weeks, Eur J Obstet Gynecol Reprod Biol 2007;131:163–8.

9. Hadi HA, Hodson CA, Strickland D. Premature rupture of the membranes between 20 and 25 weeks’ gestation: role of amniotic fluid in perinatal outcome. Am J Obstet Gynecol 1994;170:1139–44.

10. Mercer B. The NICHD-MFMU antibiotic treatment of preterm PROM study: impact of initial amniotic fluid volume on pregnancy outcome. Am J Obstet Gynecol 2006;194:438–45.

11. Chauleur C, Rochigneux S, Seffert P, Chene G, Billiemaz K, Collet F. Neonatal outcomes and four-year follow-up after spontaneous or iatrogenic preterm prelabour rupture of membranes before 24 weeks. Acta Obstet Gynecol Scand 2009;88:801–6.

12. Hacker NF, Moore JG, Gambone JC. Essentials of obstetrics and gynecology. 4th ed. Philadelphia: Elsevier; 2004:63–4.

13. Mercer B. Preterm premature rupture of the membranes. Am J Obstet Gynecol 2003;101:178–93.