Improvements in maternal and perinatal outcomes have been associated with remarkable paradigm shifts in obstetric practice, including a dramatic increase in cesarean rates and a downward shift in median gestational age at birth. The latter has resulted in an increasing proportion of late-preterm births and a decreasing proportion of births after 40 weeks gestational age.1,2 During the past few years there has been a surge of interest in the outcomes of the late-preterm infant and the possible contribution of the increasing cesarean rates on this cohort. Many studies have reported increased pulmonary complications among infants delivered by cesarean, including respiratory distress syndrome (RDS), transient tachypnea of the newborn and pulmonary hypertension.3–6 The risk seems to be greater if labor has not preceded cesarean delivery or if delivery occurs before 39 weeks of gestation.7–10 Although a number of studies have reported on the association between cesarean delivery and respiratory disease, many have been retrospective in design, applied to a select population, and limited in the scope of included respiratory diseases. Additionally, there is little information on whether changes in obstetric practice have altered the epidemiology of neonatal pulmonary disease over time. The purpose of this study was to analyze the effect of gestational age, delivery mode, and maternal–fetal risk factors on the incidence of respiratory problems among infants born 34 or more weeks of gestation over a 9-year period in a single tertiary care center.
PATIENTS AND METHODS
The study population consisted of all infants 34 or more weeks estimated gestational age delivered at Wilford Hall Medical Center from January 1, 1990, through December 31, 1998. In an effort to analyze an unselected population and minimize bias related with excessive high-risk pregnancies, we excluded newborns who were delivered at our hospital after maternal transport. As military active duty or dependents, these mothers received free medical care throughout their pregnancy. Applying Kotelchuck’s Adequacy of Prenatal Care Utilization Index,11 85% of mothers received “adequate” or “adequate plus” care, 9% were intermediate and 6% were determined to have “inadequate” prenatal care. Two trained nurses and one physician collected data for all births on standardized data sheets from the time of delivery. Maternal antepartum and intrapartum records were reviewed immediately after delivery. Neonatal data were prospectively recorded until discharge from the hospital and entered into a single database. All clinical diagnoses were concurrently monitored and recorded. Infants admitted to the newborn intensive care unit were monitored for details related to cardiopulmonary support, including duration of oxygen and mechanical ventilator support. Data on all infants admitted to the newborn intensive care unit were also cross-referenced to a separate database for that unit.
Antepartum complications coded in the database included maternal diabetes, fetal growth restriction, preeclampsia or gestational hypertension, oligohydramnios, hydramnios, prolonged and premature rupture of membranes, placenta previa, and abruption. Several intrapartum characteristics were also recorded, including fetal heart rate patterns, presence of meconium-stained amniotic fluid, mode of delivery, and presumed chorioamnionitis. The mode of delivery was defined as follows: cephalic spontaneous vaginal delivery; forceps or vacuum assisted operative vaginal delivery; vaginal breech delivery; and cesarean delivery. Cesarean deliveries were further coded as 1) primary cesarean delivery—nonurgent cesarean delivery in a woman with no prior birth by cesarean; 2) repeat cesarean delivery—nonurgent cesarean delivery in a woman with a previous birth also by cesarean; and 3) emergency cesarean delivery—cesarean delivery provided when severe fetal or maternal compromise was suspected, such as due to cord prolapse, suspected abruption, previa with hemorrhage, or markedly nonreassuring fetal heart rate pattern.
Diagnostic criteria for each neonatal pulmonary problem were applied concurrently by a single investigator (B.A.Y.) as follows: 1) meconium aspiration syndrome by presence of meconium-stained amniotic fluid, presence of abnormal signs on physical examination consistent with pulmonary disease, need for supplemental oxygen or ventilator support, a compatible chest radiograph and low suspicion for infection; 2) RDS by presence of abnormal signs on physical examination consistent with pulmonary disease, need for supplemental oxygen or ventilator support, a chest radiograph with diffuse reticular granular pattern and low suspicion for infection; 3) pneumonia by presence of abnormal signs on physical examination consistent with pulmonary disease, abnormal chest radiograph with nonhomogeneous findings, and a high clinical suspicion for infection; 4) transient tachypnea of the newborn by clinical and radiographic features initially identified in the first hours of life, followed by characteristic resolution during the initial 24–48 hours and low suspicion for infection; and 5) pulmonary hypertension by clinical or echocardiographic criteria demonstrating significant right-to-left extrapulmonary shunt or elevated pulmonary artery pressure. Diagnosis of pneumothorax and congenital diaphragmatic hernia were made by characteristic radiographic findings. To provide a more objective measure of the clinical importance of a respiratory problem, we defined respiratory morbidity by the use of any assisted mechanical ventilation or by the need for supplemental oxygen for greater than 24 hours. Because the incidence of respiratory morbidity by estimated gestational age was similar to or less for multiple gestation births (n=283) when compared with singletons, we included all live births 34 or more weeks estimated gestational age in this study.
To estimate the influence of changing care practices over the study period, 3 subgroups were identified: Era 1=1990–1992, Era 2=1993–1995, Era 3=1996–1998. There were distinct epidemiologic reasons for these divisions. During the years 1990–1992 (Era 1), we recorded the lowest rates of late-preterm birth, the highest rates of births at 41 weeks or more, and the proportion of births at 40 weeks of gestation exceeded that of births at 39 weeks. Additionally, during this period amnioinfusion was not used, and the institutional cesarean delivery rate was consistently less than 10%. During the years 1996–1998 (Era 3) we recorded the highest rate of late-preterm births, the lowest rate of births at 41 weeks or more, and the proportion of births at 39 weeks exceeded that of births at 40 weeks. Additionally, during this period amnioinfusion was used in more than 35% of meconium-stained labors, the institutional cesarean delivery rate ranged from 12% to 15%, and antepartum fetal surveillance for women reaching 41 completed weeks of gestation changed from a single nonstress test to a policy of twice-weekly fetal nonstress tests and weekly assessments of amniotic fluid volume with ultrasonography. The “transition” era was 1993–1995 (Era 2). During this time late-preterm birth rates were gradually increasing and births at 41 or more weeks of gestation were gradually decreasing. The cesarean delivery rate ranged from 9–12%, amnioinfusion was just being introduced, and a risk factor–based policy for the prevention of early onset neonatal group B streptococcal sepsis was adopted.12 During the course of this 9-year study, neither antenatal steroids nor tocolysis were routinely provided to any mother threatening to deliver at 34 or more weeks of gestation and cesarean delivery by maternal request was not an offered delivery option.
The primary outcome variable for this study was the diagnosis of respiratory morbidity. Univariable analyses comparing differences between infants with respiratory morbidity and those without were performed by 2-tailed t test, χ2, and Fisher exact test. To evaluate the effect of birth year on outcomes and practices, χ2 with Spearman’s rho correlation was used to compare differences in categorical variables with respect to the three study eras, continuous data were compared by one-way analysis of variance, and ordinal measures were compared using Kruskal- Wallis. Stepwise logistic regression was subsequently used to test for independent risk factors on the occurrence of respiratory morbidity, with Hosmer-Lemeshow goodness-of-fit determination.13 All statistical procedures were performed using SPSS for Windows 14.0 (SPSS Inc., Chicago, IL). Etiologic fraction by gestational age category (34–36 weeks, 37–38 weeks, and 41 weeks or more) were determined by the method of Miettinen as follows: EF=Pi (RRi–1)/ S Pi (RRi–1)+1, where Pi is the prevalence of the ith gestational age category, RRi is the corresponding relative risk of respiratory morbidity in the group relative to birth at 39–40 weeks, and S indicates summation over the i gestational age categories.14 This study was approved by the Wilford Hall United States Air Force Medical Center Institutional Review Committee. Based on the observational nature of the study, informed consent was determined to not be necessary.
During the study period there were 15,997 births. After exclusion of all maternal transfers (n=1,050) and births less than 34 weeks (n=382), there were 14,531 infants 34 or more weeks of gestation included in this study. There were 34 fetal deaths. Fetal death rate increased as gestational age decreased (rate per 1,000 births was 8.9 at 34–36 weeks; 3.7 at 37–38 weeks; 1.4 at 39–40 weeks; and 1.3 at more than 40 weeks; P<.001). There was no difference in fetal death rate between the three time eras. Among the 15 neonatal deaths occurring in this study population before discharge from the neonatal intensive care unit (NICU), 11 were due to severe congenital anomalies or inborn errors, two were attributed to perinatal asphyxia, and two were secondary to severe sepsis. There were three infants whose primary cause of death was pulmonary in nature, and all three had severe lung hypoplasia disorders, including thanatophoric dwarf, limb-body wall syndrome, and severe lung hypoplasia secondary to prolonged oligohydramnios.
Table 1 shows demographic, antepartum, and intrapartum features of the maternal population by delivery era. The most dramatic differences between eras were the twofold increase in the diagnosis of any obstetric complication, the threefold increase in the diagnosis of nonreassuring fetal heart rate patterns, and the transition from limited to near universal use of antenatal ultrasonography, with a marked increase in the proportion of women undergoing first-trimester fetal ultrasonography.
There were 764 respiratory diagnoses made in 705 infants for an overall incidence of 4.9%. Of these, 75% were diagnosed as transient tachypnea of the newborn or pneumonia. Any respiratory diagnosis occurred more frequently as gestation moved away from 39 0/7 to 40 6/7 weeks (P<.001). Respiratory morbidity was diagnosed in 204 infants for an overall incidence of 1.40% (Table 2). The most common diagnoses associated with respiratory morbidity were RDS, meconium aspiration syndrome, or pneumonia. Respiratory morbidity increased in frequency as an infant’s gestation moved away from 39 0/7 to 40 6/7 weeks (Table 2).
The etiologic fractions, crude ratio, and unadjusted relative risk for respiratory morbidity by gestational age group and era of birth are shown in Table 3. In all three periods the highest etiologic fraction was found among the late-preterm infants. There was a noticeable reduction in the etiologic fraction of respiratory morbidity over the 9-year study period for infants born at 37–38 weeks and after 40 weeks of gestation.
There was a significant change in the mode of delivery over the 9 years of this study. Cesarean delivery rates nearly doubled from 8.1% to 14.0%, whereas spontaneous cephalic vaginal deliveries decreased from 73.5% to 68.9%, and any vaginal breech delivery decreased fourfold from 1.9% to 0.5% (P<.001). The greatest proportion of cesarean deliveries were designated as “primary,” but both “primary” and “repeat” cesarean delivery rates increased significantly over time (4.6% compared with 9.0% and 2.5% compared with 4.2%, respectively for Era 1 compared with Era 3; P<.01), whereas the proportion of “emergent” cesarean deliveries actually decreased over time (12.1 compared with 5.7% of all cesarean births in Era 1 compared with Era 3, respectively; P<.01). Compared with cephalic vaginal birth, respiratory morbidity occurred at rates fourfold higher after cesarean delivery and breech vaginal delivery, although statistical significance by mode of delivery was only noted at estimated gestational age of more than 37 weeks (Table 4). Although cesarean delivery rates were increased and cesarean delivery was associated with a greater risk for respiratory morbidity, the diagnosis of respiratory morbidity was similar across all three time eras by each mode of delivery as well as for all live births 34 or more weeks estimated gestational age (Era 1 59 of 4,400=1.3%, Era 2 66 of 4,381=1.5%, Era 3 79 of 5,750=1.4%; P =.780). Rates of respiratory morbidity for each specific respiratory diagnosis were statistically unchanged over time except for pneumothorax, which decreased from 16.9% of all diagnoses in Era 1 to 6.3% in Era 3 (P =.048). The combination of RDS plus transient tachypnea of the newborn significantly increased from 33.9% of all diagnoses in Era 1 to 57.0% in Era 3 (P =.009), whereas the combination of meconium aspiration syndrome and pneumonia significantly decreased from 52.5% of all diagnoses in Era 1 to 30.4% in Era 3 (P =.009).
Over the 9-year study period, there was a significant shift in the estimated gestational age at delivery (Fig. 1). The median age at delivery decreased from 40 weeks to 39 weeks, with a shift in 25–75% from 39–41 weeks to 38–40 weeks. This shift in gestational age at delivery was associated with a 37% increase in births 34–36 weeks and a 39% reduction in births less than 40 weeks (Fig. 2).
Logistic regression modeling demonstrated that five factors were independently related to the occurrence of respiratory morbidity. As shown in Table 5, the most important factor was estimated gestational age at the time of delivery. Using 39–40 weeks of gestation as the reference value, risk markedly increased as gestation decreased below 37 weeks. Low 5-minute Apgar and nonreassuring fetal heart rate patterns also increased risk for respiratory morbidity, with the greatest effect noted for low 5-minute Apgar scores. Finally, male gender and cesarean delivery (presented as any cesarean delivery, as well as by specific type of cesarean delivery) also independently increased the risk for respiratory morbidity. Era of birth, maternal race, maternal age and complicated pregnancy were not independent risk factors.
There are two important findings from this study: first, we confirmed the paradigm shift in obstetric care of increased cesarean delivery rates, increased proportion of late-preterm births, and decrease in median gestational age at delivery over a 9-year period; second, despite the fact that we confirmed the association of increased rates of respiratory morbidity with decreasing gestational age and with cesarean compared with vaginal delivery, we did not observe an overall increase in respiratory morbidity over the 9-year period.
With regard to gestation-dependent respiratory morbidities, the epidemiologic findings of this study are similar to those by Hjalmarson and colleagues,15–17 and only slightly disparate from the more recent studies of Rubatelli and colleagues.18,19 Hjalmarson and colleagues have previously identified an increased risk for significant respiratory disease among term and late-preterm infants related to gestation at birth, low Apgar scores, and cesarean delivery.15–17 Similar to their findings, we noted that the majority of respiratory disease in this population was mild in severity, of limited duration, and did not contribute to mortality. Methodologic differences between our study and those reported by Rubaltelli and colleagues make it difficult to contrast the findings.18,19 They included infants of all gestational ages but focused on births less than 37 weeks, whereas our focus was on late-preterm and term infants. Additionally, the use of supplemental oxygen by diagnosis ranged between 69–90%, and for ventilator support it was between 7% and 73%, whereas we only diagnosed respiratory morbidity among infants who had an objective need for oxygen and/or ventilator support. In both European study series the data were of large populations, but a limited period of 3 months to 12 months did not allow evaluation of possible epidemiologic changes over time. One of the strengths of our study is that all diagnoses were concurrently made by a single investigator. It is possible that subtle shifts in the application of diagnostic criteria may occur over time, even for a single investigator. However, the inclusion of more objective criteria, such as the need for oxygen supplementation or ventilator use for more than 24 hours, should minimize any possible change in subjective application of diagnostic criteria. Additionally, the overall incidence of any specific respiratory morbidity did not change over the 9-year study period, but the pattern shifts that were noted (decreased proportional frequency of meconium aspiration syndrome and pneumonia and increased proportional frequency of RDS and transient tachypnea of the newborn) were consistent with the proportional shift in gestational age at delivery. Finally, consistent approaches in the management of infants with respiratory problems were maintained over this 9-year period, including the use of surfactant replacement for RDS and the availability of high-frequency ventilation and extracorporeal membrane oxygenation.
The increased risk associated with cesarean delivery, particularly elective cesarean birth in the late-preterm and early term infant before the onset of labor, is well described.3–10 This and findings of the present study support the current American College of Obstetricians and Gynecologists recommendations that elective cesarean delivery be delayed until 39 or more completed weeks of gestation (in women with good dating criteria) to minimize the risk of subsequent respiratory disease and NICU transfer.20 Our study confirms the increased risk of serious respiratory morbidity among infants not delivered by a spontaneous vaginal route, but it is limited by the lack of determination of the specific indication for cesarean delivery. Because elective cesarean birth by maternal request was not an accepted indication during these years, other mechanisms were involved in the increased rate of cesarean deliveries. The proportion of cesarean births in our study changed in a fashion consistent with the national trend reported by Menacker et al2 Interestingly, the proportion of “emergency” cesarean deliveries significantly decreased over time. Although the highest rate of cesarean delivery was among the late-preterm infants, cesarean rates increased significantly for each gestational age analyzed. Our cesarean delivery rate nearly doubled over time, but the highest cesarean rate of 15.3% in the last year of this study was well below the United States cesarean rate at that time of 22%, and far less than the 29% section rate reported for the United States in 2004.20 Our cesarean delivery rate during the 9-year study period may be partly attributable to the fact that purely elective cesarean delivery and elective inductions at less than 39 weeks were strongly discouraged at our center.
Etiologic fraction defines the proportion of all cases (in our study, respiratory morbidity) due to a particular risk (here, gestational age at birth) and is a function of both risk and prevalence of the risk factor. We found that the etiologic fraction for respiratory morbidity remained relatively stable for late-preterm infants but decreased markedly in time periods 2 and 3 for infants born at 37–38 weeks and after 40 weeks. Although there was a 37% increase in the number of late-preterm births over the 9-year study period, this was primarily due to births at 35 weeks and 36 weeks (Fig. 2), not the highest risk group at 34 weeks, and the absolute increase in prevalence was rather minimal (5.2% of all study births in 1990–1992 compared with 7.1% in 1996–1998), resulting in little effect on late-preterm etiologic fraction. In contrast, the etiologic fraction reduction for infants born after 40 weeks seems to be primarily due to decreased prevalence, because the proportion of births after 40 weeks decreased from 26.7% of all study births in 1990–1992 to 16.4% in 1996–1998. Because the proportion of births at 37–38 weeks actually increased from 16.4% of all study births in 1990–1992 to 23.0% in 1996–1998, the decreased etiologic fraction among infants born at 37–38 weeks seems primarily due to a reduction in relative risk. Although this decreased risk may be associated with time-related changes in obstetric and/or neonatal care, the reasons are unclear from the data available to us.
There are limitations to this study. First, this is a single-center study representing outcomes in a military dependent population with free access to and adequate use of all maternal and neonatal health care services. Our results may differ from populations without such care. Second, the study patients were all managed at a tertiary care center with 24-hour availability to maternal–fetal and neonatal specialists. Our findings may not be representative of those at a community hospital. Although we attempted to pattern the community hospital population by strict exclusion of all high-risk maternal transfers, differences in availability of subspecialty services (such as anesthesiology, maternal–fetal medicine and neonatology) may affect a variety of decisions made in determining the management of labor, timing of delivery, and subsequent neonatal care. Third, our cesarean delivery rates were considerably lower than the national rates. This may result in an underestimation of respiratory morbidity compared with centers or populations with much higher cesarean delivery rates. Next, over time it is possible that there was a significant change in the patient population we cared for. Although military populations tend to move frequently, this did not alter the general demographic make-up of our study population over time. We did identify differences over time in the rates of several important antenatal and perinatal factors. These include an increased diagnosis of complications such as gestational hypertension or gestational diabetes, as well as in the overall diagnosis of any obstetric problem and an increased recognition of nonreassuring fetal heart rate patterns. Some of these differences may be due to changes in obstetric practice rather than a change in the patient population. One example of this change in practice is the use of fetal ultrasonography, which dramatically changed from a minority of women having any ultrasonography during pregnancy to near universal application of this test, with nearly 50% of the mothers having their first fetal ultrasound scan during the first trimester. Lastly, because this was a retrospective analysis of an observational cohort, the associations that we found between mode of delivery and neonatal respiratory morbidity cannot inform prospective decisions regarding clinical management and planning a route of delivery.
The primary goal of obstetric practice is to produce a healthy neonate in the safest manner possible. The optimal timing and approach for this goal remain elusive.21 Caughey et al22,23 have suggested that maternal and neonatal risk increases as gestation advances beyond 41 weeks. Although a study by Alexander et al24 did not concur with this suggestion, their study group was selective, excluding infants less than 40 weeks as well as for a variety of maternal complications. Our findings in this study, and in a previous report on infants born through meconium-stained amniotic fluid,25 are in agreement with the suggestion by Caughey and others, at least as it relates to clinically important neonatal respiratory complications.
In summary, our study confirms numerous recent reports that late-preterm birth is an important contributor to the incidence of respiratory morbidity.26–28 We also found the same downward shift in gestational age at birth previously reported by Davidoff et al.1 What we have shown that is not seen in isolated studies of only late-preterm infants, is the effect on overall respiratory morbidity of the concurrent decreasing rates of births above 40 weeks of gestation. That is, the associated reduction in “late” term births occurring in conjunction with the downward shift in median gestation at birth is accompanied by a decreased rate of respiratory morbidity that seems to balance the increased rate occurring among the late-preterm group. The implication of this finding is that an overall reduction in respiratory morbidity, and possibly other morbidities, would likely be seen if a reduction in the proportion of “late” term births occurred without an accompanying increase in late-preterm births. Further prospective studies are clearly needed to validate or refute the findings we have presented and to better understand the complex interactions involved in the management of pregnancy, labor and delivery, and gestational age to achieve our goal of optimal maternal and neonatal outcome.
1. Davidoff M, Dias T, Damus K, Russell R, Bettegowda V, Dolan S, et al. Changes in the gestational age distribution among US singleton births: impact on rates of late preterm birth, 1992 to 2002 [published erratum appears in Semin Perinatol 2006;30:313]. Semin Perinatol 2006;30:8–15.
2. Menacker F, Declercq E, Macdorman MF. Cesarean delivery: background, trends, and epidemiology. Semin Perinatol 2006;30:235–41.
3. Kolas T, Saugstad OD, Daltveit AK, Nilsen ST, Oian P. Planned cesarean versus planned vaginal delivery at term: comparison of newborn infant outcomes. Am J Obstet Gynecol 2006;195:1538–43.
4. Levine EM, Ghai V, Barton JJ, Strom CM. Modes of delivery and risk of respiratory disease in newborns. Obstet Gynecol 2001;97:439–42.
5. Morrison JJ, Rennie JM, Milton PJ. Neonatal respiratory morbidity and mode of delivery at term: influence of timing of elective cesarean section. Br J Obstet Gynaecol 1995;102:101–6.
6. Zanardo V, Simbi AK, Franzoi M, Solda G, Salvadori A. Trevisanuto D. Neonatal respiratory morbidity risk and mode of delivery at term: influence of timing of elective cesarean section. Acta Paediatr 2004;93:643–7.
7. Gerten KA, Coonrod DV, Bay RC, Chambliss LR. Cesarean delivery and respiratory distress syndrome: does labor make a difference? Am J Obstet Gynecol 2005;193:1061–4.
8. Roth-Kleiner M, Wagner BP, Bachmann D, Pfenninger J. Respiratory distress syndrome in near-term babies after cesarean section. Swiss Med Wkly 2003;133:283–8.
9. Parilla BV, Dooley SL, Jansen RD, Socol ML. Iatrogenic respiratory distress syndrome following elective cesarean delivery. Obstet Gynecol 1993;81:392–5.
10. Madar J, Richmond S, Hey E. Surfactant-deficient respiratory distress after elective delivery at ‘term.’ Acta Paediatr 1999;88:1244–8.
11. Kotelchuck M. An evaluation of the Kessner Adequacy of Prenatal Care Index and a proposed Adequacy of Prenatal Care Utilization Index. Am J Public Health 1994;84:1414–20.
12. Hankins GV, Chalas E. Group B streptococcal infections in pregnancy: ACOG’s recommendations ACOG Newslett 1993;37:2.
13. Hosmer DW, Hosmer T, Le Cessie S, Lemeshow S. A comparison of goodness-of-fit tests for the logistic regression model. Stat Med 1997;16:965–80.
14. Miettinen OS. Proportion of disease caused or prevented by a given exposure, trait or intervention. Am J Epidemiol 1974;99:325–32.
15. Hjalmarson O. Epidemiology and classification of acute, neonatal respiratory disorders. A prospective study. Acta Paediatr Scand 1981;70:773–83.
16. Krantz M, Wennergren M, Bengtson L, Hjalmarson O, Karlsson K, Sellgren U. Epidemiological analysis of the increased risk of disturbed neonatal respiratory adaptation after caesarean section. Acta Paediatr Scand 1986;75:832–9.
17. Wennergren M, Krantz M, Hjalmarson O, Karlsson K. Low Apgar score as a risk factor for respiratory disturbances in the newborn infant. J Perinat Med 1987;15:153–60.
18. Dani C, Reali MF, Bertini G, Wiechmann L, Spagnolo A, Tangucci M, et al. Risk factors for the development of respiratory distress syndrome and transient tachypnea in newborn infants. Italian Group of Neonatal Pneumology. Eur Respir J 1999;14:155–9.
19. Rubaltelli F, Dani C, Reali M, Bertini G, Wiechmann L, Tangucci M, et al. Acute neonatal respiratory distress in Italy: a one-year prospective study. Italian Group of Neonatal Pneumology. Acta Paediatr 1998;87:1261–8.
20. ACOG educational bulletin. Assessment of fetal lung maturity. Number 230, 1996. Committee on Educational Bulletins of the American College of Obstetricians and Gynecologists. Int J Gynaecol Obstet 1997;56:191–198.
21. Nicholson JM, Kellar LC, Kellar GM. The impact of the interaction between increasing gestational age and obstetrical risk on birth outcomes: evidence of a varying optimal time of delivery. J Perinatol 2006;26:392–402.
22. Caughey AB, Musci TJ. Complications of term pregnancies beyond 37 weeks of gestation. Obstet Gynecol 2004;103:57–62.
23. Caughey AB, Washington AE, Laros RK Jr. Neonatal complications of term pregnancy: rates by gestational age increase in a continuous, not threshold, fashion. Am J Obstet Gynecol 2005;192:185–90.
24. Alexander JM, McIntire DD, Leveno KJ. Forty weeks and beyond: pregnancy outcome by week of gestation. Obstet Gynecol 2000;96:291–4.
25. Yoder BA, Kirsch EA, Barth WH, Gordon MC. Changing obstetric practices associated with decreasing incidence of meconium aspiration syndrome. Obstet Gynecol 2002;99:731–9.
26. Escobar GJ, Clark RH, Greene JD. Short-term outcomes of infants born at 35 to 36 weeks gestation: we need to ask more questions. Semin Perinatol 2006;30:28–33.
27. Gladstone IM, Katz VL. The morbidity of the 34- to 35-week gestation: Should we reexamine the paradigm? Am J Perinatol 2004;21:9–13.
28. 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.