Volume 59, Number 9 OBSTETRICAL AND GYNECOLOGICAL SURVEY Copyright 2004 by Lippincott Williams & Wilkins

CME REVIEWARTICLE

26

CHIEF EDITORS NOTE: This article is part of a series of continuing education activities in this Journal through which a total of 36 AMA/PRA category 1 credit hours can be earned in 0. Instructions for how CME credits can be earned appear on the last page of the Table of Contents.

An Evidence-Based Approach to the Evaluation and Treatment of Premature Rupture of Membranes: Part I
Timothy P. Canavan, MD, FACOG,* Hyagriv N. Simhan MD, MSCR, and Steve Caritis, MD
*Fellow, MaternalFetal Medicine, Assistant Professor, and Director of MaternalFetal Medicine, Magee Womens Hospital, University of Pittsburgh, Pittsburgh, Pennsylvania Preterm premature rupture of membranes (PPROM) occurs in 3% of pregnancies and is responsible for one third of all preterm births. PPROM will affect 120,000 women in the United States each year. It is associated with significant maternal, fetal, and neonatal morbidity and mortality resulting from infection, umbilical cord compression, abruptio placentae, and prematurity. The etiology is multifactorial, but the most significant risk factors are previous preterm birth and previous preterm premature rupture of membranes. Accurate diagnosis is extremely important to assure proper treatment. Evaluation is based on patient history and clinical examination. This review presents the available evidence and grades it according to the U.S. Preventative Task Force recommendations. In part I of this review, the definition, pathophysiology, and methods of PPROM diagnosis are presented. In part II, the management, treatment, neonatal outcome, and the maternal and fetal evaluation of women with PPROM in the presence of cerclage and medical complications is reviewed. Target Audience: Obstetricians & Gynecologists, Family Physicians Learning Objectives: After completion of this article, the reader should be able to define the term: preterm premature rupture of membranes, to list the factors associated with premature rupture of membranes, and to outline the tests available for the diagnosis of intra-amniotic infection.

Premature rupture of the fetal membranes (PPROM) is defined as the rupture of the amniotic membranes with release of the amniotic fluid more than 1 hour before the onset of labor. There are 2 general categories: 1. Term PROM (TPROM)PROM after 37 weeks gestation (not discussed in this review). 2. Preterm PROM (PPROM)PROM before 37 weeks gestation.
Reprint requests to: Timothy P. Canavan, MD, FACOG, Magee Womens Hospital, Department of Obstetrics, Gynecology & Reproductive Sciences, 300 Halket Street, Pittsburgh, PA 15213. E-mail: tcanavan@mail.magee.edu The authors have disclosed no significant financial or other relationship with any commercial entity.

PATHOPHYSIOLOGY The etiology of PPROM is multifactorial, and many women will have multiple etiologic and associated factors. Many of these factors are suspected to increase the risk for PPROM as a result of membrane stretch or degradation, local inflammation, or increased susceptibility to ascending infection (1). In the majority of cases, the exact etiology of the PPROM is unknown. The most significant risk factors are a previous pregnancy complicated by PPROM or preterm labor (PTL). For more details, the reader is referred to Tables 1 and 2 and the references listed (15). A recent study by Sadler and associates (6) evaluated the association of laser conization and loop

669

670
TABLE 11 4

Obstetrical and Gynecological Survey

Possible Mechanisms for PPROM Choriodecidual infection Collagen degradation Decreased membrane collagen content Localized membrane defects Membrane stretch (uterine overdistension) Programmed amniotic cell death

women (aRR 3.6; 95% CI, 1.87.5). Laser ablation of the cervix was not associated with PPROM (aRR 1.1; 95% CI, 0.52.4). PREDICTION OF PREMATURE RUPTURE OF THE FETAL MEMBRANES Women with a history of PPROM have a 13.5% risk of recurrent preterm birth from PPROM compared with 4.1% (relative risk 3.3) for women without that risk (2). The risk of recurrent PPROM is affected by the gestational age of PPROM in the previous pregnancy. The risk of recurrent PPROM is 13.5 times greater when the previous episode of PPROM occurred before 28 weeks rather than after 28 weeks (1.8% vs. 0.13%) (2). The National Institute of Child Health & Human Development sponsored MaternalFetal Medicine Units Network (MFMU) prospectively evaluated factors in a current pregnancy that might predict PPROM (see Table 3) (7). Women have an increased risk of PPROM in the presence of a short cervical length, medical complications, working during pregnancy, and in the case of multiparous women, a prior episode of PPROM that caused a preterm birth. Fetal fibronectin (fFN) alone is a risk factor for multiparous women. Women with 2 or 3 risk factors had a dramatic increased risk for PTB from PPROM compared with women without these factors as presented in Table 4. A combination of a short cervix and an elevated fFN had a dramatic effect on a nulliparous womens risk for PPROM, whereas the addition of a prior PTB from PPROM increased a multiparous womens risk by an odds ratio of 25. In this same study, previously implicated risk factors such as black race, vaginal bleeding, and

TABLE 2 Conditions Associated with PPROM Amniocentesis1 Cervical cerclage1 Cervical insufficiency7 Chronic abruptio placentae5 Cigarette smoking1,4 Laser conization6 Loop electrosurgical excision procedure6 Low socioeconomic status1,4 Prior cervical conization1,4 Prior preterm delivery1,2,4 Prior preterm labor1,2,4 Prior PPROM1,2,4 Sexually transmitted infection1,3,4 Uterine / amnion distension1,3,4 Vaginal bleeding in pregnancy1,4 Working during pregnancy1,7

electrosurgical excision procedure (LEEP) with PPROM. They found an adjusted relative risk (aRR) of 2.7 (95% confidence interval [CI], 1.35.6) for laser conization and an aRR of 1.9 (95% CI, 1.03.8) for LEEP. The risk of PPROM increased with the increased depth of the conization. They found that women in the highest tertile of resection depth (greater than 1.6 cm) had a greater than 3-fold increased risk of PPROM compared with untreated
TABLE 3

Predictors of preterm birth from preterm premature rupture of membranes7 Prediction Factor Odds Ratio (95% confidence interval) 37 weeks 2.1 (1.1 4.1) 3.7 (1.8 7.7) 2.0 (1.0 4.0) 3.7 (1.59.0) NS 2.2 (1.27.5) 3.0 (1.5 6.1) 2.5 (1.4 4.5) 1.8 (1.13.0) 2.1 (1.1 4.0) 1.8 (1.13.1) 3.1 (1.8 5.4) 35 weeks NS 9.9 (3.8 25.9) NS 4.2 (1.116.0) NS NS 5.3 (1.518.7) 4.2 (2.0 8.9) NS 9.0 (3.6 22.5) 2.6 (1.25.3) 4.1 (2.0 8.7)

Nullipara Bacterial vaginosis Cervical length 25 mm Low body mass index (19.8 kg/m2) Medical complications Positive fetal fibronectin ( 50 ng/ml) Symptomatic contractions Working during pregnancy Multipara Cervical length 25 mm Low body mass index (19.8 kg/m2) Positive fetal fibronectin Prior preterm birth from PTL Prior PPROM causing PTB

NS not significant; PTL preterm labor. Medical complications included diabetes, hypertension, cardiac disease, and endocrine disorders requiring medication and hemoglobinopathy.

Evaluation and Treatment of Premature Rupture of Membranes Y CME Review Article

TABLE 4 Risk of PTB from PPROM among women with multiple risk factors7 Risk Factor Nulliparous All nulliparous women No risk factors present Positive fFN & short cervix (25 mm) Multiparous All multiparous women No risk factors present Previous PTB from PPROM & fFN Previous PTB from PPROM & short cx All 3 risk factors 37 weeks (%) 3.7 3.0 16.7 5.0 3.2 15.4 23.1 25.0

671

35 weeks (%) 1.6 0.9 11.1 2.3 0.8 15.4 15.4 25.0

fFN fetal fibronectin; cx cervix; PTB preterm birth; PPROM preterm premature rupture of membranes.

cigarette smoking were not statistically associated with PTB from PPROM (7). Nutritional status was not evaluated as a risk factor for PPROM by the MFMU study, but Ferguson and associates evaluated both the effects of nutrition and socioeconomic factors on the risk of PPROM (8). These investigators found no nutritional links to PPROM; however, there was a statistically significant association between PPROM and a maternal hemoglobin less than 11.1 gm/dL (odds ratio [OR], 4.33; 95% CI, 1.3814.17) or a low socioeconomic status with a total family income less than $25,000 Canadian (OR, 3.1; 95% CI, 1.66.0) (8). Early studies evaluating bacterial vaginosis (BV) and PPROM showed some association, but these studies were underpowered and of weak study design. More recent investigations with better study designs have disputed these earlier findings. Gravett and associates (9) evaluated genital pathogens in 534 consecutive women in the second and third trimester of pregnancy and found that BV was associated with PPROM by an OR of 2.4 (95% CI, 1.44.3). The mean gestational age at enrollment was 32.6 weeks and the gestational age at delivery in all patients with BV was 37.8 4.1 weeks. Kurki and associates (10) evaluated BV in early pregnancy and found BV predicted PPROM with a sensitivity (SEN) of 41% to 67%, specificity (SPE) of 79%, negative predictive value (NPV) of 96% to 99% and a positive predictive value (PPV) of 4% to 11%. However, this study included only 9 patients with PPROM. McGregor
TABLE 5 Methods of diagnosis for PPROM (%)1214 Method Ferning Friedman12 Tricomi13 Kovacs14 Sensitivity 88 96 96 89 90

and associates (11) performed a double-blind, placebo-controlled treatment trial of BV in 229 women and found no association of BV to PPROM. Mercer and the MaternalFetal Medicine Network found that BV is not an independent risk factor for PPROM at less than 35 weeks (7). The strength of the evidence would support that BV is not associated with PPROM. Evaluation for asymptomatic BV as a means to provide treatment and prevent PPROM has not been shown to be effective. Evaluation and treatment of symptomatic BV can be done to alleviate a patients symptoms like in nonpregnant patients. DIAGNOSISMETHODS Physical Examination Diagnosis of PPROM is made by a combination of clinical suspicion, patient history, and some simple testing. Patient history has an accuracy of 90% for the diagnosis of PPROM and should not be ignored (12). Numerous tests have been recommended for the evaluation of PPROM, but 2 tests have withstood the test of time: Nitrazine (Bristol-Myers Squibb, Princeton, NJ) paper testing and ferning (also referred to as amniotic fluid crystallization testing) of vaginal pool fluid. Freidman and McElin found that if a combination of patient history, Nitrazine testing, and ferning was used to evaluate a patient for PPROM, the accuracy was 93.1% when 2 or more tests were positive (see Table 5) (12).

Specificity 94 96 97 84 89

PPV 94 98 99 83 88

NPV 87 90 91 90 90

Accuracy 91 96 96 87 89

Nitrazine12 Patient history12

PPV positive predictive value; NPV negative predictive value.

672

Obstetrical and Gynecological Survey

The ferning test should be performed on midvaginal or posterior fornix fluid. Contamination with cervical mucous can cause a false-positive result. The slide should be allowed to air dry for a minimum of 10 minutes. The false-negative rate increases with less drying time or with flame-drying (1315). Ferning is unaffected by meconium at any concentration or by changes in pH. Small amounts of blood do not affect ferning; however, blood and amniotic fluid mixed in equal amounts results in no ferning (15). Amniotic fluid specimens are fern- and Nitrazinepositive up to 2 weeks after amniocentesis (15). The Nitrazine test could be more susceptible to alteration by contamination. The presence of bacterial vaginosis, cervicitis, semen, alkaline urine, blood, soap, and antiseptic solutions could cause a false-positive Nitrazine test (1215). During diagnosis and evaluation of PPROM, digital cervical examination should be avoided. Lewis and associates compared digital cervical examination with sterile speculum examination and found that latency was shortened significantly by cervical examination at any gestational age (2.1 4.0 vs. 11.3 13.4 days, P 0.0001) (16). Alexander and associates evaluated data from the NICHD MFMU randomized, controlled trial on antibiotics after PPROM and found that 1 or 2 cervical examinations in patients with PPROM between 24 and 32 weeks was associated with a shorter latency (3 vs. 5 days, P 0.009) but did not worsen either maternal or neonatal outcome (17). Biochemical Testing Several chemical markers of PPROM have been suggested, but there are few data that thoroughly evaluate their usefulness. These tests can add significant cost and should be used only when the diagnosis of PPROM is highly suspicious but unable to be verified by the simple bedside testing discussed here. Fetal fibronectin (fFN), alpha-fetoprotein (AFP), and diamino-oxydase (DAO) have been used to confirm PPROM when other methods have been equivocal. Fetal Fibronectin, Alpha-fetoprotein and Diamino-oxydase Although there is little literature to collaborate this suggestion, fFN has the potential to be a diagnostic test with high NPV for determining PPROM. The amniotic fluid contains high levels of fFN (approximately 50,000 ng/mL). The presently available fFN assay is considered positive when concentrations ex-

ceed 50 ng/mL. In a patient with a history suspicious for PPROM, a negative fFN test would be inconsistent with PPROM and could be used to eliminate that diagnosis. A positive result would not be helpful in the evaluation. Eriksen and associates (18) compared fFN against pooling, ferning, and Nitrazine testing to determine rupture of membranes (ROM) in term patients and found fFN has a SEN of 98.2% with a SPE of 26.8%, PPV of 87.5% and a NPV of 75%. Gaucherand and associates (19) compared 131 women between 23 and 40 weeks gestation with and without gross evidence of ROM and found that fFN was the best marker for ROM with a SEN of 94%, SPE of 97%, PPV of 97%, and a NPV of 94%. AFP has a SEN of 88%, a SPE of 84%, PPV of 86%, and a NPV of 87%, whereas DAO has a SEN of 83%, a SPE of 95%, a PPV of 95%, and a NPV of 84.5%. Ultrasound Testing Ultrasound can be used to evaluate patients for PPROM. There are few data evaluating ultrasound as a modality to diagnose PPROM. Theoretically, absent or markedly reduced amniotic fluid index (AFI) in a patient with a suspicious history for PPROM could be used as evidence of ROM. The results of ultrasound testing should be used in conjunction with other testing. During amniocentesis, dye (indigo carmine) can be infused into the amniotic fluid. Using a vaginal tampon, ROM can be confirmed in those cases with a high suspicion of PPROM but equivocal testing. The presence of dye on the tampon would confirm conclusively that rupture of membranes has occurred. Initial Assessment of Fetal Status With Premature Rupture of the Fetal Membranes Kundavi and associates (20) compared the distribution of amniotic fluid, by ultrasound, between the upper and lower quadrants in patients with premature ROM at term. They found that if the sum of the volumes of the 2 upper quadrants was greater than the sum of the 2 lower quadrants, there was a statistically significant increased risk of meconium staining (P 0.003), abnormal fetal heart rate patterns (P 0.005), operative delivery (cesarean section [P 0.0008] or forceps delivery [P 0.002]) and neonatal intensive-care unit admission (P 0.003). Vermillion and associates (21) evaluated 225 singleton pregnancies with PPROM between 24 and 32 weeks by ultrasound and found that an AFI less than 5 cm was associated with an increased risk of cho-

Evaluation and Treatment of Premature Rupture of Membranes Y CME Review Article

673

rioamnionitis (OR, 5.5; P 0.001), endometritis (OR, 6.4; P 0.01), early-onset neonatal sepsis (OR, 8.5; P 0.034), and a shorter latency to delivery (5.5 vs. 14.1 days, P 0.02). These findings could be helpful in counseling patients with PPROM and predicting complications. Amniocentesis Infection Subclinical intrauterine infection has been implicated as a major etiologic factor in the pathogenesis and subsequent maternal and neonatal morbidity in PPROM (2226). The rate of positive culture obtained by transabdominal amniocentesis at the time of presentation with PPROM in the absence of labor is 25% to 40% (22,2733). Amniotic fluid infection is intimately linked with in utero fetal inflammation (34). Serious newborn and infant adverse outcomes (such as cerebral palsy and chronic lung disease) are closely tied with this fetal inflammatory process (35 38). Reliance on clinical criteria to diagnose amniotic fluid infection, although common practice, might not be particularly useful. Clinical chorioamnionitis is present on admission in only 1% to 2% of women who present with PPROM and subsequently develops clinically in 3% to 8% of women (30,39). The majority of amniotic fluid infection in the setting of PPROM does not produce the signs and symptoms traditionally used as diagnostic criteria for clinical chorioamnionitis. The diagnostic and therapeutic goals of management of the patient with PPROM focus on a reduction of neonatal morbidity. To that end, presently accepted strategy includes the use of hospitalized bedrest/expectant management, broad-spectrum antibiotics, antenatal corticosteroids, and serial maternal/ fetal surveillance. These therapies are designed to prolong pregnancy and reduce neonatal morbidity. The current understanding of the contribution of infection and inflammation to preterm birth and its morbidity supports the notion that intrauterine infecTABLE 6

tion (clinical or subclinical) is an indication for delivery (and, by extension, a contraindication to expectant management or pregnancy prolongation). Documentation of amniotic fluid infection in the patient who presents with PPROM thus enables us to triage our therapeutic decision-making rationally. It is conceptually important to define subclinical amniotic fluid infection as a positive amniotic fluid culture. A culture result could take as long as 48 hours to return as definitively positive or negative. In an effort to provide more rapid information regarding amniotic fluid infection status, several short-term tests have been evaluated. The tests that are currently available for clinical use are Gram stain, white blood cell count, leukocyte esterase, and glucose concentration. Romero and colleagues initially evaluated leukocyte esterase assay (LEA) and Gram stain (prevalence of positive cultures was 33.9%). When both tests were combined, a significant increase in sensitivity to 50% was observed. This was associated with a drop in specificity to 81.4% (40). In a subsequent study, Romero and coworkers evaluated amniotic fluid glucose, white blood cell count, interleukin-6, and Gram stain. The combination of the clinically available tests, glucose less than 10 mg/dL, WBC 30 cells/mm3, and Gram stain (whereby if any 1 of the 3 tests is positive, the combination test is considered positive), provides a sensitivity of 76.2%, specificity of 60.3%, PPV of 61.0%, and NPV of 80.4% (33). Table 6 compares these tests. Interleukin (IL)-6 appears to be the best biomarker for intraamniotic infection (IAI) but is unavailable in most hospitals. The evidence suggests that in addition to amniotic fluid culture, the best, practical rapid tests to obtain on amniotic fluid are glucose, WBC, and Gram stain.

Lung Maturity There is controversy regarding the optimal gestational age at which expectant management is discontinued and delivery is performed. The optimal inter-

Predicting a positive amniotic fluid culture22,24,27,28,31,33 TEST Sensitivity (%) 24 57 19 57 81 76 Specificity (%) 99 78 87 74 75 60 PPV (%) 91 42 57 67 61 NPV (%) 68 68 74 86 80

Gram stain AF-WBC (30/mm3) LEA Glu (10md/dl) IL-6 Glu, AF-WBC & Gram Stain

AF-WBC amniotic fluid white blood cell count; Glu glucose; IL-6 interleukin-6; LEA leukocyte esterase.

674
TABLE 7

Obstetrical and Gynecological Survey

Predicting neonatal morbidity22,33 TEST Odds Ratio 4.3 (P .05) 3.32 (P .05)

AF-WBC (30 cells/mm3) IL-6 (790 pg/ml)

AF-WBC amniotic fluid white blood cell count; IL-6 interleukin-6.

val for delivery occurs when the risks of prematurity are outweighed by the risks of pregnancy prolongation in the setting of PPROM (infection, abruption, cord accident). As addressed in this article (and part II of this series), lung maturity assessment could be useful to guide delivery planning in the 32- to 34week interval. An amniocentesis performed at 32 weeks is helpful with respect to diagnosis of infection (as noted previously) and confirmation of fetal lung maturity. Both a retrospective cohort study and a randomized trial support the use of amniocentesis among women with PPROM to aid in delivery planning (30,32). Cotton et al. (30) performed a randomized, controlled trial (RCT) comparing the use of amniocentesis for lung maturity testing (lecithin to sphingomyelin ratio and phosphatidylglycerol) and amniotic fluid culture with observation to determine delivery timing in patients with PPROM between 26 and 34 weeks gestation. The amniocentesis was repeated weekly until pulmonary maturity was reached or bacteria were identified in the amniotic fluid. They found that fetal distress, as judged by the fetal monitor tracing, was more frequent in the no amniocentesis group (P 0.05), and the number of days the infant remained in the hospital was significantly less in the amniocentesis group (median, 8.5 days; range, 288 days) than in the no amniocentesis group (median, 22 days; range, 2110 days; P 0.01). This difference in neonatal hospital days appears to be the result of a slower resolution of the multiple problems of prematurity. This would suggest that when amniocentesis is used aggressively to determine delivery, neonatal length of stay and morbidity are decreased. The predictive values of the various lung maturity tests are not changed by the presence of PPROM. The clinical test to be used to determine lung maturity is dependent on provider preference and availability and experience of the providers institution. Likelihood of Success of Amniocentesis and Risk of Complications There are 9 English-language trials in the literature that report the likelihood of success for transabdominal amniocentesis in the setting of PPROM. The

success rates range from 49% to 98% with an average of 72% (686 of 950) (29). These studies were published between 1979 and 1996. There is a clear trend toward increased success of amniocentesis with date of publication of study. Intuitively, as ultrasound technology has developed and training in invasive procedures has improved, the likelihood of success is likely to improve. Seven of these 9 studies were published before 1990. Thus, one would anticipate that the likelihood of success today might be somewhat better than 72%. There is only 1 English-language trial in the literature that specifically addresses the frequency of complications of amniocenteses performed in the setting of PPROM (41). Yeast et al. found no cases (0 of 91) of fetal injury after amniocentesis. They also noted that these 91 women did not have a shorter latency interval than a matched group of women who did not have amniocentesis. Noninvasive Testing to Predict Intraamniotic Infection In the hopes of predicting intraamniotic infection (IAI) noninvasively, nonstress testing, biophysical profile (BPP), and umbilical artery Doppler were studied in patients with PPROM. Vintzileos and associates (42) performed an observational cohort study in 73 singleton pregnancies with PPROM at 25 weeks or later and found a BPP score of less than or equal to 7 (performed within 24 hours of delivery) was associated with an infection rate of 93.7% (infection was defined as clinical amnionitis, possible neonatal sepsis, or neonatal sepsis). In a second study, Vintzileos and associates (43) evaluated 3 groups of 73 singleton pregnancies greater than 25 weeks gestation with PPROM. They compared BPP testing, observation, and amniocentesis and found that a BPP score of 7 or less when used to mandate delivery, decreased the overall infectious outcome (10.9% vs. 30.1%, P 0.005), amnionitis (5.4 vs. 20.5%, P 0.01), and 5-minute Apgar scores less than 7 (0% vs. 12.3%, P 0.005). Vintzileos and associates (44) performed a retrospective review of nonstress testing (NST) in 127 patients with PPROM at greater than 25 weeks and found the sensitivity and specificity of the NST (when done within 24 to 48 hours of delivery) to predict infectious outcome (clinical amnionitis, possible neonatal sepsis, or confirmed neonatal sepsis) was 78.1% and 86.3%, respectively, with a NPV of 65.7% and a PPV of 92.1%. Several recent studies have been unable to reproduce these findings. Carroll

Evaluation and Treatment of Premature Rupture of Membranes Y CME Review Article

675

and associates (45) performed a BPP (by the criteria described by Vintzileos (43) in the studies cited previously) in 89 women with PPROM ranging from 20 to 36 weeks gestation and compared these findings with cordocentesis and amniotic fluid culture results. They found no association of BPP score with fetal or amniotic infection. In a randomized trial by Lewis and associates (46), 135 patients with PPROM between 23 and 34 weeks gestation were randomized to either daily BPP or NST and this was compared with amniotic fluid culture or Gram stain, evidence of clinical amnionitis, or culture evidence of neonatal sepsis. They found that neither the NST (sensitivity 39.1%) nor the BPP (sensitivity 25%) had significant sensitivity to predict IAI. This is consistent with the findings of Ghidini and associates (47) who studied 166 singleton pregnancies with PPROM delivered before 32 weeks gestation and compared them with histopathologic placental examination. They found no association between an abnormal BPP score and severe acute placental inflammation. These findings differ from a small study by Yu cel and associates (48) who compared umbilical artery Doppler and BPP with histopathology of placental inflammation in 24 patients with PPROM between 25.6 and 36.4 weeks gestation. They found an association between abnormal BPP scores and increased arterial systolic/diastolic (S/D) ratios with histologic evidence of placental inflammation. However, Leo and associates (49) did not find an association between umbilical artery Doppler S/D ratios and IAI in their study of 51 patients with PPROM. There is a discrepancy between the earlier studies and the more recent studies in regard to whether BPP or NST can predict IAI. Possible neonatal sepsis is a very nonspecific outcome variable and could have confounded the results. Most newborns tend to be observed for possible neonatal sepsis after delivery when the pregnancy had been complicated by PPROM. Amniotic fluid culture and histopathology represent more specific end points to assess IAI. Umbilical artery Doppler S/D ratios are less clear. SUMMARY Preterm premature rupture of membranes affects over 120,000 pregnancies in the United States annually and is associated with significant maternal morbidity and neonatal morbidity and mortality. Healthcare costs are significantly influenced by the prolonged maternal hospital stay, need for frequent testing, and the resulting neonatal costs as a result of prolonged neonatal intensive-care unit care for the

newborn. Management requires an accurate diagnosis and determination of gestational age. A gestational age approach to therapy is important and should be adjusted for each hospitals neonatal intensive-care unit outcomes. These patients are at high risk for infection and amniocentesis can be used to evaluate early markers for infection and provide a sample of amniotic fluid for culture. Any evidence of infection by amniocentesis should be considered carefully as an indication for delivery. We reviewed and evaluated published reports for quality according to the method outlined by the U.S. Preventative Services Task Force: I. Evidence obtained from at least 1 properly designed randomized, controlled trial. II-1. Evidence obtained from well-designed controlled trials without randomization. II-2. Evidence obtained from well-designed cohort or case-control analytic studies, preferably from more than 1 center or research group. II-3. Evidence obtained from multiple time-series with or without the intervention. Dramatic results in uncontrolled experiments also could be regarded as this type of evidence. III. Opinions of respected authorities based on clinical experience, descriptive studies, or reports of expert committees. Based on the highest level of evidence found in the data, our recommendations are provided and graded according to the following categories: Level ARecommendations are based on good and consistent scientific evidence. Level BRecommendations are based on limited or inconsistent scientific evidence. Level CRecommendations are based primarily on consensus and expert opinion. SUMMARY TABLE OF RECOMMENDATIONS Category A (recommendations are based on good and consistent scientific evidence): 1. Digital cervical examination should not be performed on any patient with PPROM who is not for immediate delivery. 2. Amniocentesis for lung maturity testing and amniotic fluid culture to determine delivery timing decreases neonatal length of stay and morbidity. 3. Amniotic fluid culture is the best predictor of IAI.

676

Obstetrical and Gynecological Survey

membranes through clinical findings and ancillary testing. Am J Obstet Gynecol 2000;183:738745. (Level II-1) Ferguson SE, Smith GN, Salenieks ME, et al. Preterm premature rupture of membranes: nutritional and socioeconomic factors. Obstet Gynecol 2002;100:12501256. (Level II-2) Gravett MG, Nelson HP, DeRouen T, et al. Independent associations of bacterial vaginosis and Chlamydia trachomatis infection with adverse pregnancy outcome. JAMA 1986;256: 18991903. (Level II-2) Kurki T, Sivonen A, Renkonen OV, et al. Bacterial vaginosis in early pregnancy and pregnancy outcome. Obstet Gynecol 1992;80:173177. (Level II-3) McGregor JA, French JI, Richter R, et al. Cervicovaginal microflora and pregnancy outcome: results of a double-blind, placebo-controlled trial of erythromycin treatment. Am J Obstet Gynecol 1990;163:15801591. (Level I) Freidman ML, McElin TW. Diagnosis of ruptured membranes. Am J Obstet Gynecol 1969;104:544550. (Level II-3) Tricomi V, Hall JE, Bittar A, et al. Arborization test for the detection of ruptured fetal membranes. Obstet Gynecol 1966; 27:275279. (Level II-3) Kovacs D. Cystallization test for the diagnosis of ruptured membranes. Am J Obstet Gynecol 1962;83:12571260. (Level II-3) Bennett JL, Cullen JB, Sherer DM, et al. The ferning and Nitrazine tests of amniotic fluid between 12 & 41 weeks gestation. Am J Perinatol 1993;10:101104. (Level II-3) Lewis DF, Major CA, Towers CV, et al Effects of digital vaginal examinations on latency period in preterm premature rupture of membranes. Obstet Gynecol 1992;80:630634. (Level II-1) Alexander JM, Mercer BM, Miodovnik M, et al. The impact of digital cervical examination on expectantly managed preterm rupture of membranes. Am J Obstet Gynecol 2000;183:1003 1007. (Level II-2) Eriksen NL, Parisi VM, Daoust S, et al. Fetal fibronectin: a method for detecting the presence of amniotic fluid. Obstet Gynecol 1992;80:451454. (Level II-1) Gaucherand P, Guibaud S, Awada A, et al. Comparative study of three amniotic fluid markers in premature rupture of membranes: fetal fibronectin, alpha-fetoprotein, diamino-oxydase. Acta Obstet Gynecol Scand 1995;74:118121. (Level II-2) Kundavi KM, Kekre AN, Seshadri L. Amniotic fluid distribution for the predicting perinatal outcome in prelabor rupture of membranes at term. Int J Gynecol Onstet 2003;81:4950. (Level II-2) Vermillion ST, Kooba AM, Soper DE. Amniotic fluid index values after preterm premature rupture of the membranes and subsequent perinatal infection. Am J Obstet Gynecol 2000; 183:271276. (Level II-2) Romero R, Quintero R, Oyarzun E, et al. Intraamniotic infection and the onset of labor in preterm premature rupture of the membranes. Am J Obstet Gynecol 1988;159:661666. (Level II-2) Mercer BM, Miodovnik M, Thurnau GR, et al. Antibiotic therapy for reduction of infant morbidity after preterm premature rupture of the membranes. A randomized controlled trial. National Institute of Child Health and Human Development MaternalFetal Medicine Units Network. JAMA 1997;278:989 995. (Level I) Romero R, Mazor M. Infection and preterm labor. Clin Obstet Gynecol 1988;31:553584. (Level III) McGregor JA, French JI, Lawellin D, et al. Preterm birth and infection: pathogenic possibilities. Am J Reprod Immunol Microbiol 1988;16:123132. (Level III) Goldenberg RL, Hauth JC, Andrews WW. Mechanisms of disease: intrauterine infection and preterm delivery. N Engl J Med 2000;342:15001507. (Level III) Asrat T, Nageotte MP, Garite TJ, et al. Gram stain results from amniocentesis in patients with preterm premature rupture of

Category B (recommendations are based on limited or inconsistent scientific evidence): 1. A combination of patient history, Nitrazine testing, and ferning should be used to evaluate a patient for PROM. A positive finding in any 2 has an accuracy of 93%. 2. Evaluation and treatment of BV is not effective at preventing PPROM. 3. Fetal fibronectin can be used to diagnose PPROM in patients with a suspicious history for PPROM but with equivocal pooling, ferning, and Nitrazine testing. 4. Amniocentesis is recommended for patients with PPROM to obtain amniotic fluid for culture and to evaluate the early markers of infection. 5. Amniotic fluid Gram stain, glucose, and leukocyte esterase can be used as rapid tests to predict IAI. 6. Amniocentesis should be repeated if the patient develops a recurrent episode of preterm labor at less than 28 weeks. 7. Amniocentesis in the setting of PPROM is not associated with fetal injury or a shorter latency. 8. The present evidence does not support using a BPP score less than 7 to predict IAI. 9. Present evidence does not support the use of umbilical artery Doppler systolic/diastolic ratio to determine IAI. Category C (recommendations are based primarily on consensus and expert opinion): 1. Cigarette smoking and vaginal bleeding are associated with an increased risk of PPROM. REFERENCES
1. Mercer BM. Preterm premature rupture of membranes. Obstet Gynecol 2003;101:178193. (Level III) 2. Mercer BM, Goldenberg RL, Moawad RL, et al. The preterm prediction study: effect of gestational age and cause of preterm birth on subsequent obstetrical outcome. Am J Obstet Gynecol 1999;181:12161221. (Level II-2) 3. Parry S, Strauss JF. Premature rupture of membranes. N Engl J Med 1998;38:663670. (Level III) 4. Premature rupture of membranes. ACOG Practice Bulletin no. 1. Washington, DC: American College Obstetricians and Gynecologists; 1998. (Level III) 5. Elliott JP, Gilpin B, Strong TH Jr, et al. Chronic abruptionoligohydramnios sequence. J Reprod Med 1998;43:418422. (Level II-2) 6. Sadler L, Saftlas A, Wang W, et al. Treatment for cervical intraepithelial neoplasia and risk of preterm delivery. JAMA 2004;291:21002106. (Level II-2) 7. Mercer BM, Goldenberg RL, Meis PJ, et al. The preterm prediction study: prediction of preterm premature rupture of

8.

9.

10.

11.

12. 13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24. 25.

26.

27.

Evaluation and Treatment of Premature Rupture of Membranes Y CME Review Article

membranescomparison of maternal and fetal characteristics. Am J Obstet Gynecol 1990;163:887889. (Level II-2) Bashiri A, Horowitz S, Huleihel M, et al. Elevated concentrations of interleukin-6 in intra-amniotic infection with Ureaplasma urealyticum in asymptomatic women during genetic amniocentesis. Acta Obstet Gynecol Scand 1999;78:379 382. (Level II-2) Blackwell SC, Berry SM. Role of amniocentesis for the diagnosis of subclinical intra-amniotic infection in preterm premature rupture of the membranes. Curr Opin Obstet Gynecol 1999;11:541547. (Level III) Cotton DB, Gonik B, Bottoms SF. Conservative versus aggressive management of preterm rupture of membranes. A randomized trial of amniocentesis. Am J Perinatol 1984;1: 322324. (Level I) Cotton DB, Hill LM, Strassner HT, et al. Use of amniocentesis in preterm gestation with ruptured membranes. Obstet Gynecol 1984;63:3843. (Level II-3) Dudley J, Malcolm G, Ellwood D. Amniocentesis in the management of preterm premature rupture of the membranes. Aust N Z J Obstet Gynaecol 1991;31:331336. (Level II-2) Romero R, Yoon BH, Mazor M, et al. A comparative study of the diagnostic performance of amniotic fluid glucose, white blood cell count, interleukin-6, and Gram stain in the detection of microbial invasion in patients with preterm premature rupture of membranes. Am J Obstet Gynecol 1993;169:839851. (Level II-1) Gomez R, Romero R, Ghezzi F, et al. The fetal inflammatory response syndrome. Am J Obstet Gynecol 1998;179:194 202. (Level II-1) Yoon BH, Romero R, Kim KS, et al. A systemic fetal inflammatory response and the development of bronchopulmonary dysplasia. Am J Obstet Gynecol 1999;181:773779. (Level II-1) Yoon BH, Kim CJ, Romero R, et al. Experimentally induced intrauterine infection causes fetal brain white matter lesions in rabbits. Am J Obstet Gynecol 1997;177:797802. (Level II-1) Maymon E, Romero R, Pacora P, et al. Human neutrophil collagenase (matrix metalloproteinase 8) in parturition, premature rupture of the membranes, and intrauterine infection. Am J Obstet Gynecol 2000;183:9499. (Level II-1) Dammann O, Leviton A. Maternal intrauterine infection, cytokines, and brain damage in the preterm newborn. Pediatr Res 1997;42:18. (Level II-2)

677

28.

29.

30.

31. 32. 33.

34. 35.

36. 37.

38.

39. Yoon BH, Jun JK, Park KH, et al. Serum C-reactive protein, white blood cell count, and amniotic fluid white blood cell count in women with preterm premature rupture of membranes. Obstet Gynecol 1996;88:10341040. (Level II-2) 40. Romero R, Emamian M, Wan M, et al. The value of the leukocyte esterase test in diagnosing intra-amniotic infection. Am J Perinatol 1988;5:6469. (Level II-2) 41. Yeast JD, Garite TJ, Dorchester W. The risks of amniocentesis in the management of premature rupture of the membranes. Am J Obstet Gynecol 1984;149:505508. (Level II-2) 42. Vintzileos AM, Campbell WA, Nochimson DJ, et al. The fetal biophysical profile in patients with premature rupture of the membranesan early predictor of fetal infection. Am J Obstet Gynecol 1985;152:510516. (Level II-3) 43. Vintzileos AM, Bors-Koefoed R, Pelegano JF, et al. The use of fetal biophysical profile improves pregnancy outcome in premature rupture of the membranes. Am J Obstet Gynecol 1987;157:236240. (Level II-2) 44. Vintzileos AM, Campbell WA, Nochimson DJ, et al. The use of the nonstress test in patients with premature rupture of the membranes. Am J Obstet Gynecol 1986;155:149153. (Level II-3) 45. Carroll SG, Papaioannou S, Nicolaides KH. Assessment of fetal activity and amniotic fluid volume in the prediction of intrauterine infection in preterm prelabor amniorrhexis. Am J Obstet Gynecol 1995;172:14271435. (Level II-2) 46. Lewis DF, Adair CD, Weeks JW, et al. A randomized clinical trial of daily nonstress testing versus biophysical profile in the management of preterm premature rupture of membranes. Am J Obstet Gynecol 1999;181:14951499. (Level I) 47. Ghidini A, Salafia CM, Kirn V, et al. Biophysical profile in predicting acute ascending infection in preterm rupture of membranes before 32 weeks. Obstet Gynecol 2000;96:201 206. (Level II-3) 48. Yucel N, Yucel O, Yekeler H. The relationship between umbilical artery Doppler findings, fetal biophysical score and placental inflammation in cases of premature rupture of membranes. Acta Obstet Gynecol Scand 1997;76:532535. (Level II-3) 49. Leo MV, Skurnick JH, Ganesh VV, et al. Clinical chorioamnionitis is not predicted by umbilical artery Doppler velocimetry in patients with premature rupture of membranes. Obstet Gynecol 1992;79:916918. (Level II-3)