Volume 200, Issue 5 , Pages e30-e33, May 2009
Indications for delivery and short-term neonatal outcomes in late preterm as compared with term births
Article Outline
Objective
The objective of the study was to evaluate the indications for late preterm birth and compare outcomes by gestational age among late preterm (34-36 weeks) and term (≥ 37 weeks) neonates at our institution.
Study Design
This was a retrospective analysis of delivery indications and short-term neonatal outcomes in women who delivered at the University Hospital between January 1, 2005 and Dec. 31, 2006. Data were analyzed using χ2, Student's t-test, analysis of variance, and post hoc Tukey tests.
Results
One hundred forty-nine late preterm (n = 49 for 34, n = 50 for 35, n = 50 for 36 weeks) and 150 term infants (n = 50 for 37, n = 50 for 38, n = 50 for 39 weeks or longer) were evaluated. Differences among groups (ie, 34 vs 35 vs 36 vs 37, etc) as well as combinations of differences between 2 groups (ie, 34-36 weeks vs ≥ 37 or ≥ 38 weeks) were analyzed. Spontaneous labor and/or rupture of membranes were the most common indications for late preterm delivery (92%). Compared with term, late preterm infants had longer hospital stays (5 days vs 2.4 days; P < .001) and higher rates of neonatal intensive care unit (NICU) admissions (56% vs 4%; P < .001), feeding problems (36% vs 5%; P < .001), hyperbilirubinemia (25% vs 3%; P < .001), and respiratory complications (20% vs 5%; P < .001). Neonatal complications were minimal at 38 weeks or longer.
Conclusion
Rates of neonatal intensive care unit admission, length of stay, and neonatal morbidities are significantly higher in late preterm as compared with term births.
Key words: delivery indications, late preterm birth, neonatal complications
Until recently, there have been few data regarding the neonatal outcomes of late preterm infants. In fact, it was not until December 2007 that the American Academy of Pediatrics reported a definition of the “late preterm” gestation,1 based on the recommendations of the 2005 workshop, Optimizing Care and Outcome of the Near-Term Pregnancy and the Near-Term Newborn Infant, sponsored by the National Institutes of Health.2 According to the 2005 workshop, the reason that infants born 340/7 through 366/7 weeks should be called late preterm is to emphasize that these infants are still preterm and are at risk for immaturity-related complications.
Preterm birth has increased by 20%, from 10.6% of live births in the United States in 1990 to 12.7% in 2005. According to the National Center for Health Statistics, most of this increase is due to late preterm births, which have risen 25% since 1990.3 Late preterm births account for about 72% of all preterm births in the United States and have come to be recognized as the fastest growing proportion of singleton preterm births.3, 4
A recent American College of Obstetricians and Gynecologists (ACOG) Committee Opinion5 states that preterm delivery should occur only when an accepted maternal or fetal indication exists, because of the concerns about neonatal morbidities related to iatrogenic prematurity. In response to the recent ACOG committee opinion, this study should demonstrate that late preterm delivery occurs only when indicated, when there are potential increased maternal and perinatal morbidities from continued pregnancy. Our objectives are to evaluate the indications for late preterm birth and to compare neonatal outcomes by gestational age (GA) among infants born late preterm and term at our institution.
Materials and Methods
Approval was obtained from the University of Cincinnati Institutional Review Board. Cases were identified using Current Procedural Terminology codes to create a list of all mothers delivered between Jan. 1, 2005, and Dec. 31, 2006. Infant data were obtained by accessing the infant records for those women. All patient data came from the University Hospital computer patient database. The complete scanned handwritten record for every patient from admission to discharge was reviewed to ensure that diagnoses were made and confirmed by attending obstetricians and pediatricians. Selection bias was minimized, because women were chosen sequentially from the list of all delivered mothers, starting Jan. 1, 2005, until there were 50 mother-infant pairs in each group. GA ranges studied included 34 (340/7-346/7) weeks, 35 (350/7-356/7) weeks, 36 (360/7-366/7) weeks, 37 (370/7-376/7) weeks, 38 (380/7-386/7) weeks, and 39 or more (390/7-416/7) weeks. There were only 49 qualifying mother-infant pairs in the 34-week group during those 2 years. Multifetal gestations, infants with major anatomic malformations, and fetal demise were excluded.
Obstetrical indications for delivery reviewed were nonreassuring fetal heart tracings, oligohydramnios, intrauterine growth restriction, abruption, praevia, spontaneous rupture of membranes, spontaneous labor, hypertensive disorders (chronic hypertension, gestational hypertension, preeclampsia), cardiac disease, and diabetes (gestational or preexisting). Infant outcome variables analyzed included feeding problems, hyperbilirubinemia, hypoglycemia, respiratory complications (respiratory distress syndrome [RDS], transient tachypnea of the newborn [TTN], apnea, use of continuous positive airway pressure [CPAP]), sepsis, rate of admission to the neonatal intensive care unit (NICU), and days of hospitalization. Infant outcomes were based on clinical diagnoses charted by the pediatric attending physician, with ranges as defined by the Department of Pediatrics.
Data were analyzed using Pearson χ2 for categorical variables and Student's t-test and analysis of variance (ANOVA) for continuous variables. ANOVAs were followed up with post hoc Tukey tests. The decision to stop at 50 patients per group was made by our statistician early in the process of data collection, because this was found to be the approximate number required per group to detect a difference where a difference existed, minimizing the possibility of a type II error.
Results
A total of 149 late preterm and 150 term infants (≥ 37 weeks) were studied. Table 1 summarizes the obstetric indications for deliveries among the subgroups. The most common indications for late preterm delivery were spontaneous preterm labor (83%) and/or preterm premature rupture of membranes (PPROM) (50%). There was often more than 1 indication. The rate of abruptio placentae was higher in those born late preterm (4% vs 0%; P < .02).
TABLE 1. Indications for delivery by GA and number (%)
| Indicationa | Weeks of gestation | |||||
|---|---|---|---|---|---|---|
| 34 n = 49 | 35 n = 50 | 36 n = 50 | 37 n = 50 | 38 n = 50 | ≥ 39 n = 50 | |
| Pregestational diabetes mellitus | 5(10) | 5(10) | 1(2) | 3(6) | 5(10) | 1(2) |
| Hypertensive disorder | 4(8) | 5(10) | 5(10) | 6(12) | 5(10) | 6(12) |
| Cardiac disease | 1(2) | 0 | 2(4) | 1(2) | 0 | 0 |
| Nonreassuring fetal heart tracing | 5(10) | 3(6) | 3(6) | 3(6) | 7(14) | 8(16) |
| Intrauterine growth restriction | 2(4) | 2(4) | 1(2) | 1(2) | 0 | 0 |
| Abruptio placentaeb | 1(2) | 4(8) | 1(2) | 0 | 0 | 0 |
| Spontaneous rupture of membranes | 24(49) | 22(44) | 29(58) | 16(32) | 11(22) | 9(18) |
| Spontaneous labor | 36(73) | 39(78) | 42(84) | 28(56) | 31(62) | 24(48) |
| Spontaneous rupture or labor | 45(92) | 45(90) | 48(96) | 36(72) | 33(66) | 27(54) |
| Spontaneous rupture and labor | 15(30) | 16(32) | 23(46) | 8(16) | 9(18) | 6(12) |
aPatients may have > 1 indication for delivery; |
bP < .02, late preterm vs ≥ 37 weeks. |
Table 2 describes neonatal hospital stay and morbidities by GA at delivery. In general, morbidity was minimal at 37 weeks or longer. There was a statistically significant reduction in both the rate of admission to the NICU (P < .001) and length of hospital stay (P < .001) as GA increased. The reduction in the rate of NICU admissions was stepwise from 34 through 38 or more weeks (P < .001), with a significant decrease demonstrated between 34 and 36 weeks (P < .001). Rate of NICU admission and length of hospital stay remained statistically significant (P < .001) when comparing all late preterm births (n = 149) with infants born at 37 weeks (n = 50) and 38 or more weeks (n = 100). The mean length of hospital stay for all late preterm infants was 5.0 days, compared with 2.4 days for term infants (P < .001).
TABLE 2. Neonatal hospitalization and morbidity by gestational age, No (%)
| Weeks of gestation | ||||||
|---|---|---|---|---|---|---|
| 34 (n = 49) | 35 (n = 50) | 36 (n = 50) | 37 (n = 50) | ≥ 38 (n = 100) | P value | |
| NICU | 33(67) | 15(30) | 6(12) | 3(6) | 1(2) | <.001 |
| Infants with > 5 days in NICU, n | 26 | 4 | 7 | 1 | 1 | <.03 |
| Total days hospitalized (mean ± SD)a | 8 ± 7 | 3 ± 4 | 3 ± 4 | 3 ± 2 | 2 ± 2 | <.001 |
| Feeding problems | 25(51) | 17(34) | 11(22) | 7(14) | 1(2) | <.001 |
| Hyperbilirubinemia | 21(43) | 8(16) | 8(16) | 2(4) | 2(4) | <.001 |
| Hypoglycemia (< 40 mg/dL) | 3(6) | 0 | 3(6) | 0 | 1(2) | ns |
| Sepsis | 4(8) | 2(4) | 0 | 0 | 1(2) | <.02 |
| Respiratory complications | 18(36) | 14(28) | 12(24) | 4(8) | 6(12) | <.03 |
| RDS | 3(6) | 4(8) | 2(4) | 0 | 1(2) | ns |
| CPAP | 9(18) | 4(8) | 3(6) | 2(4) | 2(4) | <.008 |
| Transient tachypnea | 6(12) | 6(12) | 7(14) | 2(4) | 2(4) | <.05 |
| Intubation/ventilation | 0 | 1(2) | 0 | 0 | 1(2) | ns |
a34 weeks different from all other groups. |
When comparing all late preterm infants (n = 149) with all term infants (n = 150), there was a significant increase in rates of feeding problems (36% vs 5%; P < .001), hyperbilirubinemia (25% vs 3%; P < .001), RDS (6% vs 0.7%; P < .01), TTN (13% vs 3%; P < .001), use of CPAP (11% vs 3%; P < .005), and sepsis (4% vs 0%; P < .01). When comparing all late preterm infants with 37-week infants only (n = 50), late preterm infants had significantly higher rates of feeding problems (36% vs 14%; P < .004) and hyperbilirubinemia (25% vs 4%; P < .001), with no significant difference in respiratory complications. When comparing 34-36 week infants, there were significant decreases in feeding problems (51% vs 22%; P < .003), hyperbilirubinemia (43% vs 16%; P < .003), and sepsis (8% vs 0%; P < .04). The first significant decrease in respiratory complications was between 34 and 37 weeks (36% vs 8%; P < .01).
Comment
Principal findings of the study
In this retrospective analysis of infants delivered late preterm, no deliveries occurred for iatrogenic reasons. The most common indications for late preterm delivery were spontaneous preterm labor (83%) and/or PPROM (50%). Separation of preterm labor from PPROM is difficult, because some patients have a combination of the 2 on admission (36% of the late preterm group). Ninety-two percent of late preterm deliveries at our institution had preterm labor or PPROM as 1 of the delivery indications, compared with 80% found by McIntire and Leveno.4
Nine percent of our 149 late preterm deliveries were complicated by a hypertensive disorder, similar to the 13% mean of the late preterm group of McIntire and Leveno.4 Also, similar to their study, intraventricular hemorrhage and necrotizing enterocolitis were rare. One percent of their late preterm deliveries were complicated by abruptio vs 4% of ours.
We experienced no neonatal deaths, but this may have been due to our small numbers compared with both McIntire and Leveno4 and Tomashek et al.6 Wang et al7 found no difference in length of hospital stay for late preterm vs term infants, whereas we found that 34-week infants remained hospitalized significantly longer than any subsequent gestational age. Similar to the results of Wang et al7 is the finding of some form of respiratory complication in 30% of late preterm infants for both studies. Overall, we found that late preterm infants experienced increased morbidity compared with term infants, which supports the findings of previous studies.4, 6, 7, 8
Strengths and weaknesses of the study
One strength of the study is that all infants were delivered and cared for at the same academic institution. Another strength is that the data for all mothers and infants were pooled from 1 database, with access to the complete scanned handwritten record for every mother and newborn.
One limitation of our study is that it was retrospective. Additionally, it had a small number of patients, increasing the possibility of type II error, and we studied only short-term infant outcomes.
Clinical implication of the study
The question arises as to whether women presenting late preterm should be managed expectantly at 34, 35, or 36 weeks, weighing the possible benefit of prolonging a late preterm pregnancy against risks such as chorioamnionitis, eclampsia, abruptio placentae, periventricular leukomalacia, pulmonary edema, and fetal demise. The research addressing expectant management of late preterm birth is scant, and studies that have been done failed to demonstrate a benefit from expectant management of PPROM9 or tocolysis10 in late preterm labor. In fact, the ACOG recommendation for PPROM after 33 completed weeks is delivery.11 In addition, other indications for delivery, such as intrauterine growth restriction and abruptio placentae, are also not amenable to expectant management. Thus, at present there are no reported proven methods to prevent late preterm delivery.
A recent commentary questioned the justification for late preterm delivery, stating that “the risks and benefits of immediate delivery vs postponing it need to be closely assessed. … Obstetricians need to avoid delivery of infants in late preterm pregnancy when it is not medically indicated.”12 At our institution, as reflected in this study, late preterm deliveries are not iatrogenic but are indicated because of maternal, fetal, or placental complications. Also of note is that we experienced no neonatal deaths, and the mean length of stay for our late preterm infant population was only 5 days, with a low rate of respiratory complications. This finding should be evaluated in the context of the potential increased maternal and perinatal morbidities from continued pregnancy in such patients.
Future areas of investigation
Any increase in late preterm births is likely, in large part, the result of an ongoing pathophysiology vs the discretion of providers. However, we do not know whether iatrogenic indications are more common at hospitals with different obstetric providers, as suggested by national data. There is a need for evaluating whether iatrogenic deliveries are more likely at small community hospitals.
There is a definite need to investigate novel methods to prevent late preterm birth secondary to either preterm labor or PPROM. Such studies require a large sample size and should include long-term neonatal outcome.
References
- . “Late preterm” infants: a population at risk. Pediatrics. 2007;120:1390–1401
- . Optimizing care and outcome for late-preterm (near-term) gestations and for late-preterm infants: a summary of the workshop sponsored by the National Institute of Child Health and Human Development. Pediatrics. 2006;118:1207–1214
- Births: final data for 2005. Natl Vital Stat Rep. 2007;56:1–104
- . Neonatal mortality and morbidity rates in late preterm births compared with births at term. Obstet Gynecol. 2008;111:35–41
- . American College of Obstetricians and Gynecologists committee opinion number 404: late-preterm infants. Obstet Gynecol. 2008;111:1029–1032
- . Differences in mortality between late-preterm and term singleton infants in the United States, 1995-2002. J Pediatr. 2007;151:450–456
- . Clinical outcomes of near-term infants. Pediatrics. 2004;114:372–376
- Effect of late-preterm birth and maternal medical conditions on newborn morbidity risk. Pediatrics. 2008;121:e223–e232
- . Premature rupture of membranes at 34 to 37 weeks' gestation: aggressive versus conservative management. Am J Obstet Gynecol. 1998;178:126–130
- Tocolysis in women with preterm labor between 32 0/7 and 34 6/7 weeks of gestation: a randomized controlled pilot study. Am J Obstet Gynecol. 2006;194:976–981
- . ACOG Practice Bulletin Number 80: Premature rupture of membranes. Obstet Gynecol. 2007;109:1007–1020
- . Late-preterm births: challenges and opportunities. Pediatrics. 2008;121:402–403
Reprints not available from the authors.
PII: S0002-9378(08)01077-6
doi:10.1016/j.ajog.2008.09.022
© 2009 Mosby, Inc. All rights reserved.
Volume 200, Issue 5 , Pages e30-e33, May 2009
