Preterm birth, defined as birth before 37 completed weeks of gestation, has steadily increased in the United States over the past 25 years. For example, the rate has increased from 9.4% in 1981 to 12.7% in 2007.1 Despite the development of numerous obstetrical interventions to reduce the burden of prematurity-related morbidity, including tocolysis, antibiotics to prolong latency for preterm rupture of membranes, and home uterine activity monitoring, these interventions have had no apparent effect on reducing the incidence of preterm birth.2
Criteria for the diagnosis of preterm labor lack precision because the underlying etiology and sequence of events that precede preterm birth are not completely understood. Symptoms such as painful uterine contractions, pelvic pressure, increased vaginal discharge, and low back pain have been associated with preterm birth.3–5 However, these symptoms also can be common in women with normal pregnancies, making the diagnosis of preterm labor even more challenging. These challenges often result in overdiagnosis in up to 40% of women presenting with preterm labor symptoms.6
In a recent publication from Parkland Hospital, it was reported that there was a reduction in the rate of preterm birth from 10.4% to 4.9% between 1988 and 2006, in contrast to the increasing rate of preterm birth in the United States.7 The authors hypothesized that this reduction was the result of a public health care program providing integrated obstetric care to minority pregnant women.
Our goal in this study was to estimate the natural course of pregnancies in women with intact membranes who presented to the labor and delivery triage unit at Parkland Hospital with labor symptoms and who were sent home after found not to be in labor between 24 0/7 and 33 6/7 weeks of gestation. The primary outcome was delivery before 37 weeks of gestation. Secondary outcomes included the interval between presentation for preterm labor symptoms and subsequent delivery, as well as maternal and neonatal outcomes. We also evaluated the association between gestational age at time of presentation and the interval to delivery.
MATERIALS AND METHODS
This was a prospective observational study of women presenting to the labor and delivery triage unit at our hospital between November 16, 2009 and November 29, 2010, with symptoms of preterm labor and who had cervical dilation less than 2 cm. These symptoms included uterine contractions, lower abdominal pain, lower back pain, pelvic pressure, and increased vaginal discharge. Data regarding these women, including gestational age, obstetric history, history of preterm birth, cerclage in current pregnancy, presenting symptoms, cervical examination, discharge diagnosis, and disposition, were entered onto encounter forms in the labor and delivery triage unit by nurse practitioners to identify patients who met study inclusion criteria. Inclusion criteria included a singleton gestation, intact membranes, 24 0/7 to 33 6/7 weeks of gestation, labor symptoms, and cervical dilation less than 2 cm. Exclusion criteria included women with preterm rupture of membranes, bleeding, hypertension, multiple gestations, or fetal death. Outcomes of women in the study cohort were analyzed against a comparable general obstetric population, defined as all other women who delivered a singleton liveborn neonate during the same time interval who did not present with preterm labor symptoms and without a diagnosis of hypertension or placenta previa. Women in this control group were identified through the Parkland Obstetrics Operations database, which contains information on all women delivered at our institution, and which a priori excluded women already within the study cohort.
Research nurses reviewed all mother and newborn charts to ensure completeness and accuracy of the data before electronic storage. Electronic audits were used to assure correct data entry. In addition, external audits were periodically performed to evaluate ascertainment and less than 13% of eligible cases were found to be missing from the study database. Using the electronic medical record, data sheets were then completed for missing cases and incorporated into the study database. The data sheets included the obstetric estimate of gestational age that was used to manage the care of women during the intrapartum period. Briefly, this estimate was based on the date of the last menstrual period and the results of ultrasonography performed during the pregnancy. The reported date of the last menstrual period was accepted to be correct if the fundal height measured between 18 and 30 weeks of gestation corresponded to the predicted gestational age. The validity of this estimate has been described previously.8,9
These data were entered into a computerized database operated by the departmental biostatistician (D.D.M.). Maternal and neonatal outcome data from the existing obstetric operations database were linked to the preterm labor subset database, and patients were deidentified. This database contains selected obstetric and neonatal outcomes for all women delivering at Parkland Hospital. The maternal data set included age, parity, ethnicity, gestational age at delivery, mode of delivery (vaginal, forceps-assisted vaginal, or cesarean), and medical complications. Neonatal outcome data included birth weight, Apgar scores, neonatal intensive care unit admission, neonatal death, and respiratory distress, defined as the need for intubation within the first 24 hours of life.
Parkland Hospital is a tax-supported institution serving Dallas County. Although a tertiary neonatal center, there are few maternal transports for threatened preterm birth. The hospital is staffed by faculty of the Department of Obstetrics and Gynecology of the University of Texas Southwestern Medical School, and the neonatology service is staffed by house officers and faculty members of the Department of Pediatrics.8
Women who presented to Parkland Hospital for preterm labor symptoms between 24 0/7 and 33 6/7 weeks of gestation were initially evaluated by nurse practitioners in the labor and delivery triage unit. Women identified to have intact membranes and cervical dilation 2 cm or more were transferred to the high-risk labor unit for evaluation. Women who had intact membranes and cervical dilation less than 2 cm received continuous external fetal monitoring for 2 hours in triage and then were reevaluated. Women who had more than one contraction in 10 minutes or a change in cervical examination (either dilation or effacement) were transferred to the high-risk labor and delivery unit. If there was one contraction or less in 10 minutes and no cervical change, then women were sent home with a diagnosis of false labor. The diagnosis of preterm labor used was based on criteria originally described by Gonik and Creasy.10
Women who were transferred to the high-risk labor and delivery unit received standardized management. This consisted of intravenous hydration with 5% dextrose in lactated Ringer solution and sedation with intravenous meperidine. Group B streptococcal prophylaxis with 2 g ampicillin was administered intravenously every 6 hours until cervical dilation was determined not to be progressive and the woman could be transferred to the high-risk antepartum unit. For fetal lung maturation, a single course of betamethasone 12 mg intramuscularly every 24 hours for two doses was administered. Repeat courses of corticosteroids were not administered. No tocolytic drugs were used.
The projected sample size was estimated using the following rationale. Currently, approximately 15,000 women are delivered annually at Parkland Hospital. Because we did not know the proportion of women presenting for preterm labor symptoms and who would meet the study criteria, we arbitrarily estimated that 10% of 15,000 women, or 1,500 women, would have preterm labor symptoms and one-third (n=500) would meet our study criteria. Assuming there would be 500 women per year who presented with preterm labor symptoms and a subset of those women would be sent home with a diagnosis of false labor, our null hypothesis was that the women who were sent home would deliver at a preterm rate equal to our overall preterm rate, which was 5.7% of singleton deliveries in 2008. To detect a difference of 4.3% (10% − 5.7%) in preterm delivery less than 37 weeks of gestation, with a power of 80% with two-sided test of significance level of 0.05, we estimated that we would need 270 women who are sent home with a diagnosis of false labor.
A planned 6-month interim analysis was performed to more accurately project sample size and to determine study length. Six months into the study, there were 481 women identified to meet inclusion criteria. Of those, 251 (52%) had delivered at our institution at the time of the interim analysis and 151 (61%) of these were sent home with a diagnosis of false labor. Based on this, we decided to continue the study for a total of 1 year, which would exceed our initial projected sample size and provide sufficient power for data analysis.
Categorical data were reported as frequencies and statistical significance was determined using χ2 analysis. Student t test was used for continuous variables. Because of the interim analysis, the P value for significance of the completed trial is 0.045 following the Lan-DeMets modification of the O'Brien-Flemming boundaries. Statistical analyses were performed using SAS 9. The study was approved by the Institutional Review Board of the University of Texas Southwestern Medical Center. Because this was an observational study of outcomes based on the standard of care at our institution, a waiver of informed consent was granted.
A total of 843 women who met the inclusion criteria were encountered in our hospital's labor and delivery triage unit during the study period (Fig. 1). Of these, 690 (82%) were discharged home after a minimum 2-hour observation period. Demographic characteristics of the study cohort (N=843) analyzed against a comparable general obstetric population delivered during the same time period are shown in Table 1. Women in the study cohort were significantly more often African American and less often Hispanic. Women in the study cohort were also significantly younger than the comparable general obstetric population. More than 95% of women in both cohorts underwent an ultrasound examination during pregnancy.
Obstetric characteristics of women determined not to be in preterm labor and sent home are compared in Table 2 to women determined to be in preterm labor and admitted to the labor and delivery unit for further evaluation. Women determined not to be in preterm labor were earlier in gestation compared with those admitted for preterm labor. Women determined to be in preterm labor and admitted more often reported contractions compared with those sent home for false preterm labor (45% compared with 31%, respectively, P<.001). In contrast, women sent home for false preterm labor were more likely to report lower abdominal discomfort (52% compared with 39%, respectively, P=.002). The primary reason for admission of women with presumed preterm labor was more than one persistent uterine contraction per 10 minutes.
Other symptoms often attributed to preterm labor, including lower back discomfort, pelvic pressure, and increased vaginal discharge, were not significantly different between the two groups. Women admitted for preterm labor more often had cervices dilated 1 cm as well as some degree of cervical effacement.
Obstetrical outcomes for women admitted for preterm labor are compared with those with false preterm labor diagnosed in Table 3. The overall rate of preterm birth less than 37 weeks of gestation was 27% compared with 7%, respectively (P<.001). Similarly, the interval from presentation to delivery was significantly shorter in women admitted. These significant differences persisted after adjusting for maternal age, race, and parity using logistic regression.
Neonatal outcomes in women sent home after a diagnosis of false preterm labor compared with a comparable general obstetric population are shown in Table 4. Women sent home with a diagnosis of false preterm labor did not have an increased rate of birth at less than 34 weeks of gestation (2% compared with 1%, P=.28), whereas the risks of late preterm birth were increased (5% compared with 2%, respectively, P<.001). The diagnosis of respiratory distress was more frequent among neonates delivered of women with false preterm labor, but mortality was not significantly increased. These significant differences persisted after adjusting for maternal age, race, and parity using logistic regression.
In an effort to better understand the preterm birth outcomes in women with false preterm labor diagnosed and who were sent home, we subgrouped women into cervical dilation of 0 compared with 1 cm at the time of discharge (Table 5). Five percent of women with 1-cm dilation of the cervix delivered before 34 weeks of gestation compared with 1% when the cervix was undilated (P<.02). Similarly, the interval to delivery was significantly shorter with cervical dilation of 1 cm; however, 89% of these women remained undelivered for more than 21 days. Table 6 presents birth outcomes for women sent home subgrouped by cervical effacement at the time of discharge. Cervical effacement was not significantly associated with preterm birth; however, the degree of effacement was inversely related to the interval between presentation for preterm labor symptoms and delivery.
For delivery before 37 weeks of gestation, we calculated the odds ratios to be 3.28 (95% confidence interval [CI] 1.48–7.24), 1.93 (95% CI 1.27–2.93), and 0.85 (95% CI 0.76–0.94) for cervical dilation of 1 cm, prior preterm birth, and gestational age at triage visit, respectively. For delivery before 34 weeks of gestation, the odds ratios were 8.62 (95% CI 2.05–36.23), 1.72 (95% CI 0.96–3.08), and 0.68 (95% CI 0.53–0.86) for cervical dilation of 1 cm, prior preterm birth, and gestational age at triage visit, respectively.
In this study, we had the opportunity to observe the natural course of events after a woman's discharge from our labor and delivery triage unit for false preterm labor. First, we found that women diagnosed with false preterm labor and who were subsequently sent home did not have an increased rate of early preterm birth at less than 34 weeks of gestation or neonatal death. However, the rate of late preterm births and related neonatal morbidities in this cohort were significantly increased. Second, the subgroup of women with 1-cm dilation of the cervix at the time of discharge had a two-fold increased risk of preterm birth compared with those with closed cervices. Despite this, the interval from a diagnosis of false preterm labor to delivery exceeded 21 days in the majority (89%) of women with cervical dilation of 1 cm.
Our results imply that there may be an identifiable subgroup of women with preterm labor symptoms sent home with a diagnosis of false labor who are at a higher risk for preterm birth. These risk factors include cervical dilation of 1 cm, number of previous preterm births, and an earlier gestational age at the time of presentation. We estimate, given the 14% rate of preterm birth in our cohort of women with cervical dilation of 1 cm, that 1,400 women would need to be randomized for an intervention to demonstrate a 33% reduction in preterm birth using 80% power. At an accrual rate of 76 women per year at Parkland with cervical dilation of 1 cm and false labor, this would require more than 18 years to complete. This would most likely require a multicenter trial to accomplish this task. In addition, to demonstrate a doubling of the neonatal mortality rate in our cohort, with 0.3% incidence in the comparison group, a study would require 7 years to complete at our institution.
The concept that cervical dilation is a predictor of preterm birth is not a novel concept. Our findings are in concert with the findings that Bouyer et al11 reported in a prospective study of 7,937 women presenting for routine prenatal care. They found that preterm birth was associated with at least one of three signs: dilation of the internal os to 1 cm, cervical length less than 1 cm, and presence of painful uterine contractions. Subsequently, numerous studies have evaluated the signs and symptoms of preterm labor, ultrasonographic measurement of cervical length, and fetal fibronectin for the prediction of preterm labor.5,6,12–17 Whereas signs and symptoms such as uterine contractions, a short cervical length, and a positive fetal fibronectin are associated with preterm birth, the sensitivity and positive predictive values of these indicators are poor.
To date, few studies have focused on women who were sent home with a diagnosis of false labor; most studies evaluating women with similar symptoms have focused on those admitted for threatened preterm labor.18–20 One advantage of our study is that all women were from a single institution and received uniform care without interventions such as tocolytics, ultrasonographic measurement of cervical length, fetal fibronectin, and progesterone therapy; thus, we are able to comment on the natural history of preterm labor symptoms without pertubations potentially attributable to common preterm labor interventions.
We found that women diagnosed with false preterm labor and who were sent home are not at risk for early preterm birth and neonatal mortality, but they may be at an increased risk for late preterm birth. The knowledge that late preterm births are associated with increased neonatal morbidity is a relatively recent concept,21–25 and trials demonstrating the efficacy of interventions to prevent complications related to delivery at these gestational ages are not yet available. In the absence of studies that demonstrate that an intervention is effective for women at risk for delivery in the late preterm period, we conclude that it is reasonable to continue our current practice of discharging women home with a diagnosis of false preterm labor and a cervical dilation of less than 2 cm.
1. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Mathews TJ, Kirmeyer S, et al.. Births: Final data for 2007. Natl Vital Stat Rep 2010;58:2.
2. Creasy RK, Resnik R, Iams JD, Lockwood CJ, Moore TR. Preterm labor and birth. Creasy & Resnik's maternal-fetal medicine: principles and practice. 6th ed. Philadelphia (PA): Saunders; 2009. p. 545–73.
3. Cunningham FG, Leveno KJ, Bloom SL, Hauth JC, Rouse DJ, Spong CY. Preterm birth. Williams obstetrics. 23rd ed. New York (NY): McGraw-Hill Professional; 2010. p. 814.
4. Creasy RK, Resnik R, Iams JD, Lockwood CJ, Moore TR. Preterm labor and birth. Creasy & Resnik's maternal-fetal medicine: Principles and practice. 6th ed. Philadelphia (PA): Saunders; 2009. p. 560.
5. Katz M, Goodyear K, Creasy RK. Early signs and symptoms of preterm labor. Am J Obstet Gynecol 1990;162:1150–3.
6. Iams JD, Johnson FF, Parker M. A prospective evaluation of the signs and symptoms of preterm labor. Obstet Gynecol 1994;84:227–30.
7. Leveno KJ, McIntire DD, Bloom SL, Sibley MR, Anderson RJ. Decreased preterm births in an inner-city public hospital. Obstet Gynecol 2009;113:578–84.
8. Bloom SL, Yost NP, McIntire DD, Leveno KJ. Recurrence of preterm birth in singleton and twin pregnancies. Obstet Gynecol 2001;98:379–85.
9. McIntire DD, Bloom SL, Casey BM, Leveno KJ. Birth weight in relation to morbidity and mortality among newborn infants. N Engl J Med 1999;340:1234–8.
10. Gonik B, Creasy RK. Preterm labor: its diagnosis and management. Am J Obstet Gynecol 1986;154:3–8.
11. Bouyer J, Papiernik E, Dreyfus J, Collin D, Winisdoerffer B, Gueguen S. Maturation signs of the cervix and prediction of preterm birth. Obstet Gynecol 1986;68:209–14.
12. Lockwood CJ, Senyei AE, Dische R, Casal D, Shah KD, Thung SN, et al.. Fetal fibronectin in cervical and vaginal secretions as a predictor of preterm delivery. N Engl J Med 1991;325:669–74.
13. Goldenberg RL, Mercer BM, Meis PJ, Copper RL, Das A, McNellis D. The Preterm Prediction Study: fetal fibronectin testing and spontaneous preterm birth. Obstet Gynecol 1996;87(5Pt1):643–8.
14. Peaceman AM, Andrews WW, Thorp JM, Cliver SP, Lukes A, Iams JD, et al.. Fetal fibronectin as a predictor of preterm birth in patients with symptoms: a multicenter trial. Am J Obstet Gynecol 1997;177:13–8.
15. Iams JD, Goldenberg RL, Mercer BM, Moawad A, Thom E, Meis PJ, et al.. The Preterm Prediction Study: recurrence risk of spontaneous preterm birth. National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network. Am J Obstet Gynecol 1998;178:1035–40.
16. Tsoi E, Fuchs IB, Rane S, Geerts L, Nicolaides KH. Sonographic measurement of cervical length in threatened preterm labor in singleton pregnancies with intact membranes. Ultrasound Obstet Gynecol 2005;25:353–6.
17. Herbst A, Nilsson C. Diagnosis of early preterm labour. BJOG 2006;113(Suppl 3):60–7.
18. How HY, Khoury JC, Sibai BM. Cervical dilatation on presentation for preterm labor and subsequent preterm birth. Am J Perinatol 2009;26:1–6.
19. Papatsonis DNM, Van Geign HP, Bleker OP, Ader HJ, Dekker GA. Maternal admission characteristics as risk factors for preterm birth. Eu J Obstet Gynecol 2004;112:43–8.
20. Parant O, Maillard F, Tsatsaris V, Delattre M, Subtil D, Goffinet F, et al.. Management of threatened preterm delivery in France: a national practice survey (the EVAPRIMA study). BJOG 2008;115:1538–46.
21. McIntire DD, Leveno KJ. Neonatal mortality and morbidity rates in late preterm births compared with births at term. Obstet Gynecol 2008;111:35–41.
22. Hauth JC. Spontaneous preterm labor and premature rupture of membranes at late preterm gestations: to deliver or not to deliver. Semin Perinatol 2006;30:98–102.
23. Khashu M, Narayanan M, Bhargava S, Osiovich H. Perinatal outcomes associated with preterm birth at 33 to 36 weeks' gestation: a population-based cohort study. Pediatrics 2009;123:109–13.
24. Woythaler MA, McCormick MC, Smith VC. Late preterm infants have worse 24-month neurodevelopmental outcomes than term infants. Pediatrics 2011;127:e622–9.
© 2011 by The American College of Obstetricians and Gynecologists.
25. Consortium on Safe Labor, Hibbard JU, Wilkins I, Sun L, Gregory K, Haberman S, et al.. Respiratory morbidity in late preterm births. JAMA 2010;304:419–25.