INTRODUCTION

Premature rupture of membranes (P-PROM) occurring in the early second trimester (<24 weeks gestation) is associated with neonatal morbidity and mortality as well as significant maternal morbidity. Delivery at this time may result in a very low birth weight baby with an overall predicted survival rate ranging between 23 and 54%. Of the survivors, as many as 55% may be significantly neurologically and developmentally impaired on long-term follow-up.^1,2 Conservative management leading to prolonged gestational age frequently provides the neonate with a greater chance of survival; however, this approach does not alleviate the lethal hazard of pulmonary hypoplasia. When expectant management is pursued, the incidence of maternal infection has been reported to be as high as 42%.¹

Despite these discouraging data, a consensus appears to have been formed in the last two decades regarding expectant management.^1,2,3 A review of the English language literature^{2,7,8,9,10,11,12,13,14,15} revealed that studies diagnosed "midtrimester P-PROM" as late as 28–32 weeks of gestation and included only "eligible" patients. Our aim was to assess the neonatal survival and maternal complication rate in P-PROM cases prior to 24 weeks of gestation, at which time the remote chance of neonatal survival may be outweighed by the risk of maternal infection. Because of their religious beliefs, the ultraorthodox Jewish pregnant women delivering at our institution invariably refuse to terminate their pregnancies; using this population to test this management option allowed us to eliminate the selection bias.

Based on our critical review of the literature, we also formulated a patient consult strategy for cases of midtrimester P-PROM.

PATIENTS AND METHODS

This is a retrospective review of a case-cohort series using an in-house developed database. The study population consisted of consecutive patients with premature rupture of membranes occurring between 16 and 24 weeks gestation, who were managed at our institution from 1995 to 2001. Cases complicated by fetal malformation or chorioamnionitis at the time of diagnosis of P-PROM were excluded from the study. Premature rupture of membranes was diagnosed by visualization of amniotic fluid on speculum vaginal examination, positive nitrazine test, and amniotic fluid index (AFI) below the 5^th centile for gestational age, and/or less than 5 cm.⁴ Our center principally serves a strictly religiously observant (Jewish Orthodox) population that will not consider terminating any pregnancy at any gestational age. Expectant management was implemented only after the couple had received a comprehensive detailed consultation from a specialist in perinatology, and in all cases requested to be observed unless maternal life hazard is to be expected. If contractions occurred before 34 weeks of gestation, tocolysis was attempted by administration of indomethacin for 48 hours in order to permit the fetus to benefit from antenatal steroid therapy. The patients received a combined antibiotic treatment of ampicillin and erythromycin from the time of admission until delivery (intravenously during the first week and orally thereafter). At 24 weeks of gestation, betamethasone (two doses of 12 mg i.m. at 24-hour intervals) was administrated and repeated once if delivery was considered imminent between 28 and 34 weeks of gestation. Indication for delivery included: (a) suspected clinical acute chorioamnionitis; (b) non-reassuring fetal profile (4); (c) gestational age of 36 weeks. Unless obstetrically contraindicated, vaginal delivery was attempted. AFI was determined at admission and then twice weekly until delivery. The latency period was defined as the time interval between the rupture of membranes and delivery. Vaginal and urine bacterial cultures were obtained at admission and then weekly until delivery. Placental cultures were obtained at delivery.

Neonatal pulmonary hypoplasia was diagnosed by radiologic and clinical criteria.^5,6,7 The radiologic criteria included a small, bell-shaped chest with well-aerated lung fields and an elevated diaphragm up to the seventh rib. The clinical criteria consisted of a small chest circumference, severe pulmonary insufficiency necessitating high ventilatory settings and complicated by air leak or pulmonary hypertension. As a result of religious restrictions, the families refused postmortem examinations. Maternal and neonatal characteristics were obtained from our perinatal database and a database specifically designed for the patients included in the study.

Statistical analysis was performed by descriptive statistics, Student's t-test, Mann Whitney test, and a logistic regression model built for significant indicators of neonatal mortality and maternal infectious morbidity. Statistical significance was defined as P<0.05.

RESULTS

A total of 28 women presented with P-PROM between 16 and 24 weeks of gestation (mean 22.7+1.0). Two patients declined conservative management, and one was lost to follow-up (3/28, 10.7%). The overall perinatal survival was 8/25 (32%). The perinatal characteristics of the survivors and nonsurvivors are summarized in Table 1. Survivors differed significantly from the nonsurvivors with regard to the amniotic fluid index (AFI) at the time of admission, gestational age at delivery, and birth weight. Although not significantly different, a trend of differences between the groups was observed for the following factors: latency period, the use of antenatal steroids, and the occurrence of neonatal sepsis. In a multivariate logistic regression model (89.5% accuracy) applied to determine which of the variables independently affected mortality, we found that low AFI (significant oligohydramnios below the fifth percentile for gestational age) was the sole significant factor for this group. Pulmonary hypoplasia accounted for three deaths (3/28, 12%). The immediate complications of the group of survivors were as follows: five neonates (5/8, 63%) suffered from varying degrees of respiratory distress necessitating assisted ventilatory support, two (2/8, 25%) were diagnosed with intraventricular hemorrhage (IVH) graded III–IV, two (2/8, 25%) were diagnosed with retinopathy grade II and treated by laser therapy before discharge, one (1/8, 12.5%) had severe necrotizing enterocolitis (NEC) with bowel surgical intervention and five (5/8, 63%) had neonatal sepsis. Longer neonatal intensive care stay was significantly associated with the development of nosocomial infections (p<0.05) and not with associated intrauterine infection. Antenatal and placental bacterial cultures showed no pathological growth.

Table 1 - Perinatal Characteristics of Neonatal Survivors Versus Nonsurvivors from Pregnancies Complicated by Early Midtrimester P-PROM.

Full table

The group of ten (40%) patients in whom the P-PROM occurred before 22 weeks of gestation was also analyzed separately. In this group, maternal characteristics were similar to those of the entire group. The mean gestational age at which P-PROM occurred was 19.00.6 weeks. The mean latency period was 15.54.3 days. There were two (20%) survivors; furthermore, at the time of P-PROM, the AFI for this group was also significantly different for the survivors than for the nonsurvivors (p<0.056). Pulmonary hypoplasia accounted for one death in this group (10%). None of the neonates from this group had significant skeletal deformities.

We encountered no case of maternal mortality. All mothers were exposed to antibiotic treatment during the latent period between the diagnosis of P-PROM and delivery, and 8/25 (32%) had postpartum febrile morbidity related to the pelvic organs and thus required additional antibiotic therapy.

DISCUSSION

Early midtrimester P-PROM is an intriguing issue. An informed, educated management decision should consider both potential neonatal as well as maternal complications. The option of intentional delivery at the time of diagnosis should be weighed against the option of expectant management. Over the last two decades, 1059 pregnancies complicated by early midtrimester P-PROM and managed expectantly were reported (Table 2).^{2,6,7,8,9,10,11,12,13,14,15,16} ). The overall neonatal survival was 51.8% (549/1059). There were only rare occurrences of both pulmonary hypoplasia (11.2%) and skeletal deformities (7.2%). Each of these studies considered different primary outcomes for communicating the results: survival, pulmonary hypoplasia, or overall perinatal outcome. In all studies both neonatal mortality and/or lung hypoplasia were found to correlate positively with the degree of oligohydramnios (defined by ultrasound), earlier gestational age on admission and delivery.^{6,13,14,15,17,18,19}

Table 2 - Summary of Studies on Midtrimester P-PROM in the English language Literature from 1985 to 2002.

Full table

In the early 1990s, there were significant changes in the standards of perinatal care, including use of antenatal steroids in cases of P-PROM, use of antenatal prophylactic antibiotics, technical advances in neonatal respiratory support, and use of surfactant. In the light of these changes, we carried out a "re-analysis" of the literature (Table 3). Because the level of care had improved, we expected to observe a rise in the survival rate; in fact we observed a decrease in survival. However, the frequent use of ultrasound examination in determining gestational age and in defining oligohydramnios has led to stringent study inclusion criteria that may account for this trend. Also the clinical diagnosis of lung hypoplasia was strictly defined and not biased by inclusion of severe respiratory distress syndrome of prematurity, especially in neonates who had not received either antenatal steroids and/or postnatal surfactant. We believe that survival rates of 32% (our study) to 46% and a lung hypoplasia incidence of 10.4–12% would be a closer reflection of present-day standards. The mean gestational age at delivery was reported to be 27–28 weeks (as it was in our series), and had no independent influence on the neonatal outcome. Nevertheless, such an extremely early delivery represents an added burden to the neonate.

Table 3 - Midtrimester P-PROM — Reviewed by Study Period.

Full table

In the even more problematic group of very early midtrimester P-PROM (<24 weeks gestation) ("postamniocentesis drip" not included), only 153 cases were reported in the literature and the outcome was rarely reported separately. According to our study, the pregnancies complicated by P-PROM at less than 22 weeks gestation had a significantly higher neonatal mortality rate (80%), despite the 90% use of antibiotics and 40% use of antenatal steroid therapy. The survival in this group was positively influenced by the amount of amniotic fluid at diagnosis. Similar to our series, a study²⁰ on the neonatal short-term follow-up of the survivors of early P-PROM compared to control infants of matched gestational age also showed an increased risk for chronic lung disease (46% vs 17%, respectively), intracranial hemorrhage (31% vs 6%) and death (32% vs 8%). Generally, in the literature and in our series, the rate of skeletal deformities was low.

By the time the study was performed, there was minimal influence on the maternal outcome. The mean maternal antenatal hospitalization remained stable (20.9–17.8 days in our series) and the cesarean section rate also remained relatively high (28.4–33.7% in our series). The only decrease that was observed was in the postpartum febrile complication rate (47.3–26.9%, and 32% in our series), probably due to the more frequent use of antenatal antibiotics. In none of the series did we find any report of maternal mortality due to infectious morbidity.

During the two-decade survey period, we, along with the rest of the medical community, have witnessed a progressive decrease in the definition of neonatal "limit of viability"; over the same period, the survival rate has stabilized, but there has been a continuous rise in the neonatal morbidity.^21,22,23

Pending a prospective, randomized, multicenter study of management options, we suggest that the long-term maternal hospitalization, antibiotic treatment before and after delivery, high rate of cesarean delivery rate, and the burden of neonatal morbidity should be presented to the pregnant patient facing the decision of electing expectant management when diagnosed with P-PROM before 24 weeks of gestation. Special consideration and different emphasis should be given to patients with pregnancies complicated by P-PROM at less than 22 weeks gestation. In this subgroup, the potential maternal complications may outweigh the chances of neonatal survival, and termination of pregnancy should be considered a balanced alternative to conservative management.

References

1. Mercer B. Management of premature rupture of membranes before 26 weeks' gestation. Obstet Gynecol Clin N Am 1992;19(2):338–351.
2. Morales WJ, Talley T. Premature rupture of membranes at <25 weeks: A management dilemma. Am J Obstet Gynecol 1993;168:503–507. | PubMed |
3. Williams. Obstetrics 21st ed. Section VII; Common Complications of Pregnancy. Preterm Birth, Chapter 27. New York: McGraw-Hill Medical Publishing Division; 2001. p. 689–729.
4. Moore TR, Cayle JE. The amniotic fluid index in normal human pregnancy. Am J Obstet Gynecol 1990;162:1168–1173. | PubMed | ChemPort |
5. Leonidas JC, Bhan I, Beatty EC. Radiographic chest contour and pulmonary air leak in oligohydramnios and related pulmonary hypoplasia (Potter's syndrome). Invest Radiol 1982;17:6–10.
6. Vergani P, Ghidini A, Locatelli A, et al. Risk factors for pulmonary hypoplasia in second trimester premature rupture of membranes. Am J Obstet Gynecol 1994;170:1359–1364. | PubMed |
7. Rotchild A, Ling EW, Puterman ML, Farquharson D. Neonatal outcome after prolonged premature rupture of membranes. Am J Obstet Gynecol 1990;162:46–52. | PubMed |
8. Thibeault DW, Beatty EC, Hall RT, Bowen SK, O'Neill DH. Neonatal pulmonary hypoplasia with premature rupture of fetal membranes and oligohydramnios. J Pediatr 1985;107(2):273–277. | PubMed | ISI | ChemPort |
9. Bedouyn S, Yasin S. Premature rupture of membranes before 28 weeks : Conservative management. Obstet Gynecol 1986;155:471–479.
10. Blott M, Greennough A. Neonatal outcome after prolonged rupture of the membranes starting in the second trimester. Arch Dis Childhood 1988;63:1146–1150.
11. Bengtson JM, Van Marter LJ, Brass VA, Greeene MF, Tuomala RF, Epstein MF. Pregnancy outcome after premature rupture of the membranes at or before 26 weeks gestation. Obstet Gynecol 1989;73(6):921–927. | PubMed |
12. Major CA, Kitzmiller JL. Perinatal survival with expectant management of midtrimester rupture of membranes. Am J Obstet Gynecol 1990;163(3):838–844. | PubMed |
13. Hadi HA, Hodson CA, Strickland D. Premature rupture of membranes between 20–25 weeks gestation: Role of amniotic fluid volume in perinatal outcome. Am J Obstet Gynecol 1994;170(4):1139–1144. | PubMed |
14. Kilbride HW, Yeast J, Thibeault DW. Defining limits of survival: Lethal pulmonary hypoplasia after mid-trimester premature rupture of membranes. Am J Obstet Gynecol 1996;175:675–681. | PubMed |
15. Winn HN, Chen M, Amon E, Leet TL, Shumway JB, Mostello D. Neonatal pulmonary hypoplasia and perinatal mortality in patients with mid-trimester rupture of amniotic membranes – A critical analysis. Am J Obstet Gynecol 2000;182(6):1638–1644. | Article | PubMed | ChemPort |
16. Schuker JL, Mercer BM. Midtrimester premature of the membranes. Semin Perinatol 1996;20(5):389–400.
17. Wenstrom KD. Pulmonary hypoplasia and deformations related to premature rupture of membranes. Obstet Gynecol Clin North Am 1992;19(2):297–408.
18. Hoekstra JH, deBoer R. Very early prolonged premature rupture of the membranes and survival. Eur J Pediatr 1990;149:585–586.
19. Shumway JB, Al-Malt A, Cohlan B, Amini S, Abboud M, Winn HN. Impact of oligohydramnios on maternal and perinatal outcomes of spontaneous premature rupture of membranes at 18–28 weeks. J Matern Fetal Med 1999;8:20–23. | PubMed |
20. Lindner W, Pohland F, Grab D, Flock F. Acute respiratory failure and short term outcome after premature rupture of the membranes and oligohydramnios before 20 weeks gestation. J Pediatr 2002;140(2):177–182.
21. Wood NS, Marlow N, Costeloe K, Gibson AT, Wilkinson AR. Neurologic and developmental disability after extremely preterm birth. New Engl J Med 2000;343(6):378–384. | Article | PubMed | ISI | ChemPort |
22. Campbell DE, Fleishman AR. Limits of viability: Dilemmas, decisions, and decision makers. Am J Perinatol 2001;18(3):117–128. | PubMed |
23. Saigal S. The limits of viability – A review for the EPICure Study Group 2000. Pediatr Res 2001;49(4):451.