Many obstetric and neonatal management strategies have been developed during the last 40 years in efforts to improve the outcome of preterm births. These strategies, to name but a few, have included regionalized maternal–neonatal transport systems, development of neonatal intensive care units, and interventions such as attempting to delay delivery using tocolytic drugs or enhancing fetal lung maturation by administration of corticosteroids to the mother. This time period is also important because it was the era during which the disciplines of maternal–fetal medicine and neonatology came into being. Throughout this period of progress in perinatal medicine, most studies have shown that interventions intended to improve the outcome of preterm infants primarily benefit those born before 32–34 weeks of gestation. For example, in the studies by Liggins and Howie1 demonstrating that corticosteroids prevented respiratory distress in preterm infants, the beneficial effect was limited to births before 34 weeks. As a result, the National Institutes of Health Consensus Conference on Corticosteroids (1994) and organizations such as the American Academy of Pediatrics (AAP) and the American College of Obstetricians and Gynecologists (ACOG) have endorsed use of corticosteroids before 34 weeks.2 Indeed, most other obstetric interventions aimed at preterm births are generally practiced only before 34 weeks because benefits have not been documented for later preterm births.
Preterm birth, defined as birth before 37 completed weeks, has increased 33%, from 9.4% of live births in the United States in 1981 to 12.5% in 2004.3,4 According to the National Center for Health Statistics, most of this increase is due to increases in births between 32 and 36 weeks because births before 32 weeks have been relatively stable at about 2% during the last 20 years.4 Late preterm infants, recently defined as those born between 34 and 36 weeks of gestation,5 account for about 75% of all preterm births.3 These late preterm births have come to be recognized as the fastest increasing and largest proportion of singleton preterm births in the United States. As such, late preterm birth is receiving increased attention as to optimal obstetric and neonatal management.5,6
Our purpose was to analyze neonatal mortality and morbidity rates at 34, 35, and 36 weeks compared with births at term over the past 18 years at our hospital to estimate the magnitude of increased risk associated with late preterm births compared with births later in gestation. We were particularly interested in obstetric complications during weeks 34, 35, or 36 because it is these complications that must be modified if late preterm births are to be ameliorated.
MATERIALS AND METHODS
Selected obstetric outcomes for all women who give birth at Parkland Hospital, as well as neonatal outcomes, are entered into a computerized database. Nurses attending each delivery complete an obstetric data sheet, and nurses assess the data for consistency and completeness by reviewing each chart before the charts are stored electronically. Data on infants’ outcomes are abstracted from discharge records. Parkland Hospital is a tax-supported institution serving Dallas County and has a level III neonatal intensive care unit adjacent to the labor and delivery units. The obstetric service is staffed by house officers and faculty members of the Department of Obstetrics and Gynecology at the University of Texas Southwestern Medical School, and the neonatology service is staffed by house officers and faculty members of the Department of Pediatrics.
Between January 1988 and December 2005, a total of 264,730 women and adolescents delivered infants at our hospital (Fig. 1). Of these infants, 240,958 were liveborn singletons without malformations and delivered of women with prenatal care. A deidentified data set was created for analysis of these mother-infant pairs. Approximately 60% of the women were enrolled for prenatal care in the first trimester, and 90% enrolled before the end of the second trimester. Multi-fetal births were excluded from our study because gestational age–related mortality and morbidity in twins differs significantly from that of singletons. This analysis was approved by the Institutional Review Board.
The outcomes we studied included neonatal death within 28 days of birth in undischarged infants. Respiratory distress was defined as use of a ventilator in the first 24 hours after birth. Transient tachypnea of the newborn was defined as tachypnea that resolves spontaneously after 6 hours. Intraventricular hemorrhage, grades 1–4, was defined according to Papile and colleagues,7 and necrotizing enterocolitis was limited to those requiring surgical intervention. Sepsis work-ups were performed at the discretion of the attending neonatologists. Sepsis was confirmed if blood cultures were positive. Phototherapy was used for hyperbilirubinemia based on clinical criteria.
Spontaneous ruptured membranes included those women with preterm premature rupture of membranes, and spontaneous labor included those with no other complications except preterm uterine contractions that resulted in progressive cervical dilatation leading to delivery. Women with pregnancy-related hypertensive disorders such as gestational hypertension or preeclampsia were categorized as having hypertension. Fetal complications included chorioamnionitis, disorders of amniotic fluid volume (eg, hydramnios or oligohydramnios diagnosed with ultrasonography), isoimmunization, and fetal growth restriction. Placental accidents included placental abruption and placenta previa confirmed at delivery. Other complications included medical problems such as diabetes or infection (eg, pneumonitis, acute pyelonephritis).
The obstetric estimate of gestational age that was used to manage the care of the women and adolescents during the intrapartum period was used to assign gestational age for this analysis. These estimates were based on the date of the last menstrual period and the results of obstetric ultrasonography, if any, performed during the pregnancy. The reported time of the last menstrual period was accepted as correct if the fundal height, measured between 18 and 30 weeks of gestation, was correlated with the week of gestation within 2 cm.8 Subjects with discrepancies between the two values underwent obstetric ultrasonography. The validity of the obstetric estimate of gestational age was previously assessed (McIntire and colleagues, 1999).9 Briefly, the correlation coefficient for the estimate of gestational age based on ultrasonography and the obstetric estimate based on the last menstrual period was 0.9.
Late preterm births were defined as births occurring at 34, 35, and 36 weeks according to the criteria developed by the National Institute of Child Health and Human Development Workshop on Optimizing Care and Outcome of Late-Preterm (Near-Term) Infants.5 In this analysis, 34 completed weeks included births at 340/7 weeks through 346/7 weeks and so on through 40 weeks.
Statistical analyses using SAS 9.1 (SAS Institute, Cary, NC) included χ2, Wilcoxon rank sum, and multiple logistic regression. Multiple comparisons were adjusted using the method of Bonferroni. P<.05 was considered significant.
Approximately 3% of births at our hospital during the study period occurred between 24 and 32 weeks of gestation, and 9% (n=21,771) were during the late preterm weeks. These late preterm births account for 76% of all preterm births.
Neonatal death rates were analyzed from 34 to 40 weeks of gestation (Fig. 2) to compare mortality rates for late preterm births to births at term and to derive a reference standard. In this analysis, the neonatal mortality rates at 38 and 39 weeks of gestation reached a nadir of 0.2 per 1,000 live births, and 39 weeks was then chosen as the referent for comparison to weeks 34 through 37. The neonatal death rates at 39 weeks were significantly lower when compared with those at 34, 35, 36, or 37 weeks, with P values ranging from .005 to <.001 for each biweekly comparison. These mortality rates were further analyzed by year of accrual, and comparative rates between late preterm and term were stable during the study period. Put another way, the relationship of late preterm to term deaths did not significantly change over the study period.
Maternal demographic characteristics for 133,022 women delivered of singleton live births at 34, 35, 36, 37, and 39 weeks of gestation are shown in Table 1. Maternal age, parity, and race or ethnicity were all significant covariables for gestational age. Logistic regression analysis was performed with these maternal characteristics, and the results shown in the following tables remained unchanged.
Neonatal morbidity rates decreased progressively and significantly from 34 to 39 weeks of gestation (Table 2). This was true for seven morbidities to include ventilator-treated respiratory distress, transient tachypnea, grades 1 or 2 intraventricular hemorrhage, sepsis work-ups, as well culture-proven sepsis, phototherapy, and intubation in the delivery room. One or more of these morbidities were diagnosed in 14% of infants born at 39 weeks compared with 34%, 24%, and 17% at 34, 35, and 36 weeks of gestation, respectively (P<.001 for each biweekly comparison with 39 weeks).
Shown in Table 3 are admissions to the neonatal intensive care unit, duration of hospitalization, and hospital charges. The hospital charges are there for the infant and do not reflect maternal bills or physician fees. Births at 34, 35, 36, and 37 weeks all had significantly increased rates of admission to neonatal intensive care, length of stay, and hospital bills. Hospital stays of 5 days or longer were assessed because such stays were less likely to be influenced by concurrent hospitalization of the mother according to her route of delivery.
Obstetric complications in relation to weeks of gestation for infants born late preterm compared with 37 weeks and 39 weeks are shown in Table 4. Births at 34, 35, and 36 weeks, as well as those at 37 weeks, had significantly higher rates of all the studied complications compared with births at 39 weeks of gestation. These complications were collated for all late preterm births, and 80% had either idiopathic preterm labor (45%) or preterm premature rupture of membranes (35%). Hypertension, placental accidents, and other fetal complications were implicated in 20% of the deliveries.
Late preterm singleton infants born at our hospital constituted approximately 9% of deliveries and experienced increased neonatal mortality and morbidity when compared with those born at 39 weeks. Indeed, virtually all of the studied measures of adverse neonatal outcome at 34, 35, 36, and even including 37 weeks, were significantly increased compared with births at 39 weeks. Some severe adverse outcomes, such as grade 3 or 4 intraventricular hemorrhage, culture proven sepsis, and necrotizing enterocolitis, were rare. The most common adverse outcomes were respiratory distress, sepsis work-ups, and phototherapy for hyperbilirubinemia. When aggregated to include one or more of any of the studied morbidities, 34% of births at 34 weeks had morbidity. The rate of aggregated morbidity then decreased weekly from 24% at 35 weeks, to 17% at 36 weeks, and to 14% at 37 weeks compared with the reference standard of 14% at 39 weeks. These increased morbidities in the late preterm group inevitably were associated with significantly increased use of intensive care, longer hospitalization, and concomitant increased hospital charges. Indeed, the mean hospital bill for the late preterm group was $3,098, compared with $1,258 for the 39 weeks referent. Although the difference in hospital bills is significant, we must emphasize that the absolute dollar amounts at any given week of gestation represent averages over an 18-year study period and are not adjusted for inflation.
Analysis of the obstetric complications associated with late preterm births suggests that approximately 80% are due to idiopathic preterm labor or preterm premature rupture of membranes. Complications such as pregnancy-related hypertension, placental accidents, fetal disorders, and maternal medical disorders were found in approximately 20% of late preterm births. One interpretation of these results vis-à-vis obstetric complications is that any strategy intended to reduce late preterm births would need to target a broad spectrum of disparate causes, as is the case for preterm births at 33 weeks or earlier.
It might be argued that current management strategies used for births at 33 weeks or less should be applied to late preterm births. For example, premature membrane rupture at 33 weeks or less is typically managed “expectantly,” which includes observation for spontaneous labor and administration of antimicrobials to delay labor and forestall intrauterine infection. In contrast, ruptured membranes without labor at 34 weeks or greater is usually managed by effecting delivery for fear of ascending intrauterine infection. Indeed, 46% of our cohort with premature rupture of membranes between 34 and 36 weeks received labor induction and could have potentially been managed expectantly in an effort to delay delivery. The remainder began labor spontaneously. However, there are few studies addressing expectant management of late preterm births,10 and those that have been reported11,12 have failed to show benefits for such management.
Other management strategies that potentially could be used in late preterm births include use of tocolytics to delay delivery and administration of corticosteroids to promote fetal lung maturation. To our knowledge, only corticosteroids have been specifically studied during 34, 35, and 36 weeks of gestation. Dalziel,13 using the Cochrane Collaboration, performed a meta-analysis of the available data on the efficacy of corticosteroids after 33 weeks of gestation subdivided into several different gestational-age subgroups during the late preterm period. This analysis showed that the effects of corticosteroids have been assessed only in subgroup analysis involving small numbers of infants from two studies with delivery findings as to efficiency. We conclude that there is insufficient evidence to find that corticosteroid therapy is beneficial in late preterm births and that this is largely due to the small number of infants studied and the low rate of respiratory distress in infants born after 33 weeks of gestation. We estimate, given the 1.4% rate of respiratory distress requiring ventilator therapy in our late preterm cohort, that 16,128 women would have to be randomized to demonstrate a 33% reduction in respiratory distress using 80% power. Such a study could conceivably be done in a large clinical trials network encompassing approximately 125,000 deliveries per year. The much less frequent neonatal mortality and some of the other morbidities in late preterm births, however, likely obviate studies using these outcomes because of unachievable sample size.
Our analysis has focused only upon morbidities and mortality occurring immediately after birth. We have not, and could not, measure the longer-range associated morbidities. Indeed, the Institute of Medicine analysis of the consequences of preterm birth in the United States emphasized the large human and economic impacts of prematurity later in the child’s life.14 Throughout the modern perinatal medicine era, which began about 1970, the focus of the health care system vis-à-vis prematurity has been on births before the late preterm period. Our results suggest that the health care focus on prematurity should be expanded to include the late preterm period. Having reached this realization, we must also admit to a sense of futility as to the likelihood of preventing late preterm births because approximately two thirds of our cohort began labor spontaneously, and as is the case for births before 32 weeks, attempts to interrupt preterm labor have not been satisfactory. Given the recent Institute of Medicine report “Preterm Birth: Causes, Consequences, and Prevention,”14 where it is acknowledged that treatment of preterm labor has not prevented preterm birth in the United States, we are of the view that a national strategy aimed at prevention of late preterm births is unlikely to provide discernible benefit without new developments in the prevention and management of preterm labor.
1. Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treatment for prevention of the respiratory distress syndrome in premature infants. Pediatrics 1972;50:515–25.
2. American Academy of Pediatrics and The American College of Obstetricians and Gynecologists. Guidelines for perinatal care. 5th ed. Elk Grove Village (IL): American Academy of Pediatrics; Washington (DC): The American College of Obstetricians and Gynecologists; 2002. p. 170–1.
3. Davidoff MJ, Dias T, Damus K, Russell R, Bettegowda VR, Dolan S, et al. Changes in the gestational age distribution among U.S. singleton births: impact on late preterm birth, 1992 to 2002. Semin Perinatol 2006;30:8–15.
4. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker F, Kirmeyer S, Births: final data for 2004. Natl Vital Stat Rep 2006;55:1–101.
5. Raju TN, Higgins RD, Stark AR, Leveno KJ. Optimizing care and outcome for late-preterm (near-term) infants: a summary of the workshop sponsored by the National Institute of Child Health and Human Development. Pediatrics 2006;118:1207–14.
6. Wang ML, Dorer DJ, Fleming MP, Catlin EA. Clinical outcomes of near-term infants. Pediatrics 2004;114:372–6.
7. Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. Pediatrics 1978;92:529–34.
8. Jimenez JM, Tyson JE, Reisch JS. Clinical measures of gestational age in normal pregnancies. Obstet Gynecol 1983;61:438–43.
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. 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.
11. Naef RW 3rd, Albert JR, Ross EL, Weber BM, Martin RW, Morrison JC. Premature rupture of membranes at 34 to 37 weeks’ gestation: aggressive versus conservative management. Am J Obstet Gynecol 1998;178:126–30.
12. Mercer BM, Crocker LG, Boe NM, Sibai BM. Induction versus expectant management in premature rupture of the membranes with mature amniotic fluid at 32 to 36 weeks: a randomized trial. Am J Obstet Gynecol 1993;169:775–82.
13. Roberts D, Dalziel S. Antenatal corticosteroids for accelerating fetal lung maturation for women at risk of preterm birth. Cochrane Database Syst Rev 2006;(3):CD004454.
14. Behrman RE, Butler AS. Preterm birth: causes, consequences, and prevention. Washington (DC): National Academies Press; 2007.