It was noted in 1951 that although pregnancies persisting beyond 300 days occurred less than 5% of the time, they accounted for 30% of perinatal deaths.1 Thus, since its advent, one intent of antenatal fetal surveillance has been the prevention of fetal death among postterm pregnancies. In the 1970s and 1980s, this was commonly defined as patients beyond 42 completed weeks, or 294 days,2 which complicates more than 10% of pregnancies,3 and this remains the definition used by the American College of Obstetricians and Gynecologists (ACOG) today. However, the use of a 42-week threshold was questioned by Bochner et al4 in a 1988 study, which showed a decreased rate of stillborn fetuses and fetal distress during labor in a group of patients who began antenatal testing at 41 weeks, as compared with the control group, which began testing at 42 weeks of gestational age. This has led to testing strategies that begin increasingly earlier for postdates testing. Most recently, the gestational age at which clinical concern should be raised was questioned in a study5 that asserted that concerns regarding the morbidity and mortality of pregnancies at and beyond term should be weighed against the risks of induction of labor. Traditionally, there has been concern that induction of labor will lead to an increased rate of cesarean delivery. However, there are an increasing number of studies5,6 that suggest this concern might be outweighed by risks of other pregnancy complications. Furthermore, the concern regarding the increased rate of cesarean delivery related to induction might be unfounded when considering research that finds that cesarean rates are similar between patients managed with induction versus expectant management.6,7
Given these changes over the past decades, the question remains: At what gestational age does the benefit of induction of labor outweigh that of expectant management? In addition to an increased perinatal mortality rate,4,5,8–10 numerous studies have associated postterm pregnancies with increased rates of meconium and meconium aspiration syndrome,4,11 oligohydramnios,12 macrosomia,4,13,14 fetal birth injury,15 fetal distress in labor,4,10,16 and cesarean delivery.4,14 Most studies15,17,18 that examine gestational age do so by establishing thresholds, such as 41 or 42 weeks, and comparing rates of complications beyond this threshold with those in patients delivered below the threshold. However, studies5,19,20 that have examined the risk of fetal death by week of gestational age show that rates increase in a steadily rising fashion before 42 weeks of gestation. If this complication of pregnancy increases not as a discrete risk beyond some particular gestational age, but instead continuously with increasing gestational age, other complications associated with postterm pregnancies might do the same.
Another theoretic concern with the existing literature regarding perinatal complications of pregnancy is the quality of the pregnancy dating. Because we have improved the dating of pregnancy with the use of ultrasound, we are now better able to identify pregnancies that go beyond 280 days of gestation.21,22 Bennett et al recently showed that up to 10% of pregnancies will be redated by a second-trimester ultrasound and more than 20% by a first-trimester scan (Bennett K, Crane J, O’Shea P, Lacelle J, Hutchens D, Copel J. Combined first and second trimester ultrasound screening is effective in reducing postterm labor induction rates: A randomized controlled trial [abstract]. Am J Obstet Gynecol 2002;187:S68). Therefore, studies that examined complications of pregnancy in populations whose pregnancies were dated primarily by history and physical examination alone are likely to suffer from nondifferential misclassification of gestational age.
In this setting of improved pregnancy dating and a desire to find trends by week of gestation rather than simple dichotomous comparisons, we sought to explore complications of pregnancy beyond 37 weeks among an otherwise low-risk group of patients. Specifically, we were interested in estimating at what gestational age the rates of maternal and fetal complications increase over the prior week of gestation. Further, we were interested in whether these complications continued to increase beyond the initial rise and in what fashion.
MATERIALS AND METHODS
We designed a retrospective, cohort study of all women delivered beyond 37 weeks of gestational age from January 1, 1992, to July 31, 2002, at California Pacific Medical Center in San Francisco. California Pacific Medical Center is a community hospital that performs more than 40% of all deliveries in San Francisco and, other than high-risk transfer patients, nearly all patients who deliver at California Pacific Medical Center receive prenatal care from a California Pacific Medical Center–affiliated provider. Institutional review board approval was obtained from the Committee on Human Research at California Pacific Medical Center. Patients were included in the analysis if they delivered a singleton pregnancy beyond 37 weeks of gestation. Gestational age was determined in relation to the estimated date of confinement, as defined by 280 days from the last menstrual period that was either less than 7 days different from a first-trimester ultrasound or 14 days different from a second-trimester ultrasound. Otherwise, the estimated dates of confinement from the earliest ultrasound were used. The following variables were also exported from the California Pacific Medical Center perinatal database: maternal age, ethnicity, profession and education, length of labor, mode of delivery, parity, prior mode of delivery, anesthesia, birth weight, amniotic fluid characteristics, and labor management. The outcome variables intrauterine fetal death, Apgar scores, admission to the intensive care nursery, chorioamnionitis, and endomyometritis were also included. Characterization of amniotic fluid, endomyometritis, and chorioamnionitis were coded by the attending physicians into the clinical database. Macrosomia was defined as birth weight greater than or equal to 4500 g. After variables were abstracted from the database, all patient identifiers were removed before analysis. This study was approved by the investigational review board at California Pacific Medical Center.
The data were then compiled and analyzed with STATA 7 software (Stata Corp., College Station, TX). Because the primary predictor of interest was gestational age by week, the dependent variables of interest were compared in a bivariate fashion with gestational age from 37 weeks and beyond. For those variables of interest, as well as those that exhibited an increasing bivariate trend before 42 weeks of gestation, a multivariable logistic regression was performed, including possible confounders, and with dummy variables for each week of gestation in the model as independent variables. Cross-product terms to examine interaction between predictor variables were created. Their contribution to the model was tested with the maximum likelihood ratio test, and they were only kept in the model if they were statistically significant; this was designated by a P value less than .05.
During the study period, there were 45,673 women who delivered beyond 37 completed weeks of gestation. These patients were predominantly well-educated, as indicated by the 57% who had completed 4 years of college (Table 1). When complications of pregnancy were examined by gestational age, there was a clear increase in the rates of meconium and macrosomia as early as 38 weeks of gestation (Table 2). In addition to the rates of meconium and macrosomia, intensive care nursery admissions, Apgar scores less than or equal to 6, operative vaginal delivery, chorioamnionitis, and endomyometritis all increased beyond 40 weeks of gestation (Tables 2 and 3). All of the perinatal outcomes that increased beyond 40 weeks continued to increase beyond 41 weeks of gestation. At this point, the rates of intrauterine fetal death and cesarean delivery also began to increase and continued to rise beyond 42 weeks of gestation as well.
Gestational age was examined with multivariable logistic regression, controlling for maternal age, ethnicity, and education, mode of delivery, birth weight, length of labor, and induction as appropriate in the different models. Beyond 40 weeks, it was found to predict an increased risk for moderate or thick meconium, intensive care nursery admission, macrosomia, 5-minute Apgar score less than or equal to 6, chorioamnionitis, and operative vaginal delivery (Table 4) when compared with pregnancies delivered before 40 weeks of gestation. These risks were further increased beyond 41 weeks of gestation, and intrauterine fetal death, endomyometritis, and primary cesarean delivery were also found to be increased among these pregnancies. Interestingly, the risk for intrauterine fetal death was more than 2.5 times greater between 41 and 42 weeks of gestation as compared with before 40 weeks of gestation. In this same comparison, rates of macrosomia and moderate or thick meconium were tripled and doubled, respectively. Cross-product terms were not found to be significant in any of the models and were not used in the final models.
We found that there were a number of complications of pregnancy that rose between 39 and 41 weeks of gestation; that is, before the current threshold of 42 weeks of gestation, which is used to define postterm pregnancy. These complications were all examined in multivariable models, and gestational age beyond 40 and 41 weeks were predictive of increased risk even when controlling for known confounders. Most concerning among these were the rates of intrauterine fetal death and intensive care nursery admissions. The fact that intensive care nursery admissions increase is concerning both for neonatal morbidity and mortality, but also for the use of medical resources and costs. Further research will need to investigate long-term neonatal outcomes to see whether these concerns are well founded.
We also found increases in the rates of meconium and macrosomia beyond 38 weeks in the bivariate comparison and 40 weeks in the multivariable analysis. These findings are markers for other neonatal morbidity and mortality associated with meconium aspiration syndrome23 and birth injury.24 We did not examine the rates of these more severe outcomes, and even in our data set with more than 45,000 patients, it is questionable whether we would have enough power to investigate such findings. Thus, in our analysis, meconium and macrosomia served as risk factors for these more serious neonatal complications.
Pregnancies that are more accurately dated are more likely to exhibit complications of pregnancy sooner in population studies. If one were to examine complications of pregnancy in a cohort of patients who were misdated, the findings would be biased toward risk increases occurring later in pregnancy. This is described by epidemiologists as nondifferential misclassification, and is based on the following. Assume that half of the pregnancies are misdated under and half are misdated over the actual gestational age. Thus, the patients who are misdated earlier than they truly are (ie, their due date is set later than what it should be, so they are always perceived as being earlier than their actual gestational age) will have their complications recorded as occurring earlier in gestation than actually happened. This will lead to an increase in the overall number of complications in earlier weeks of gestation. The patients who are misdated later than they truly are will actually be at earlier gestational ages than stated, which will decrease the overall number of complications occurring in later weeks of gestation. Thus, the difference between later and earlier weeks of gestation will be narrowed by nondifferential misdating.
The improved dating of pregnancies might reveal a number of findings in perinatal epidemiology previously hidden by this nondifferential misclassification that occurs by the use of just the history and physical examination for dating, as compared with ultrasound. As patients’ pregnancies are better dated, it is important to elucidate the risks in these pregnancies in a rigorous fashion. Knowing these risks will improve clinicians’ ability to counsel their patients and enable researchers to explore a variety of predictors and explanations. Finally, both clinicians and researchers will be able to investigate the use of interventions to decrease these risks. Currently, antenatal fetal surveillance is used to help identify those postterm patients at an even higher risk of perinatal complications. With further evidence consistent with what we have found in this study, it might be reasonable to consider such screening at an earlier gestational age.
If antenatal testing is begun at an earlier gestational age, it is reasonable to assume that more patients with need for delivery will be identified. The efficacy and risk profile of the existing and new methods for labor induction will continue to change over time. Thus, we will need to reassess the risks and the benefits from expectant management versus labor induction in these patients with only the most current data. If labor induction methods improve and the risks of increasing gestational age begin earlier than previously suspected, there might be an indication to intervene at an earlier gestational age. Given our data, it might be found that the balance of risks and benefits for intervention in low-risk pregnancies should be earlier than current management. The most recent recommendations by ACOG have defined that threshold to be at 42 weeks of gestation. However, based on these data, we would suggest that this threshold should at least be reconsidered.
Our study is not without limitations. Despite the high quality of the data used, we were unable to examine other complications of pregnancy, such as placental abruption and oligohydramnios because of either underreporting or an interruption in the data-entry process. A retrospective study can be complicated by missing data and inaccurate data. However, because this database was prospectively designed to examine complications of pregnancy, this concern is likely unfounded. Of the variables we used, less than 1% were missing data. We might also be missing other confounding variables. For example, we did not have information on family income. However, we felt that socioeconomic status was well accounted for with the use of education as a proxy.
Another possible limitation is regarding the generalizability of our study population to that of all pregnant women. The patients served by this community hospital are predominantly upper middle class, carry health insurance, and seek prenatal care in the first trimester. Thus, we sought to examine complications of pregnancy at term among these patients who are likely to have excellent pregnancy dating and might be considered to have lower rates of pregnancy complications for socioeconomic reasons. If anything, that the findings were significant among these patients suggests that they might only be more so in other, higher-risk groups. Furthermore, our findings were robust, controlling for maternal age, ethnicity, and education.
Examining the association between gestational age and pregnancy complications is important for estimating both when pregnancies should be screened for complications with antenatal fetal testing, as well as when a delivery plan should be initiated. These risks of increasing gestational age of pregnancy need to be compared with the risks of induction to determine the optimal gestational age for induction of labor. This study examining these risks among a relatively low-risk population is a first approximation of the maternal and fetal risks in pregnancies at term. These findings suggest that antenatal fetal testing should begin sooner than current recommendations of 42 weeks of gestation and that the optimal gestational age to initiate delivery requires further investigation.
1. Clifford SH, Reid DE, Worcester J. Postmaturity. Am J Dis Child 1951;82:232.
2. Hauth JC, Goodman MT, Gilstrap LC 3rd, Gilstrap JE. Post-term pregnancy. Obstet Gynecol 1980;56:467.
3. American College of Obstetricians and Gynecologists. Diagnosis and management of post-term pregnancy. ACOG technical bulletin no. 130. Washington: American College of Obstetricians and Gynecologists, 1989.
4. Bochner CJ, Williams J 3rd, Castro L, Medearis A, Hobel CJ, Wade M. The efficacy of starting postterm antenatal testing at 41 weeks as compared with 42 weeks of gestational age. Am J Obstet Gynecol 1988;159:550–4.
5. Rand L, Robinson J, Economy KE, Norwitz ER. Post-term induction of labor revisited. Obstet Gynecol 2000;96:779–83.
6. Hannah ME, Hannah WJ, Hellmann J, Hewson S, Milner R, Willan A. Induction of labor as compared with serial antenatal monitoring in post-term pregnancy. N Engl J Med 1992;326:1587–92.
7. Dyson DC, Miller PD, Armstrong MA. Management of prolonged pregnancy: Induction of labor versus antepartum fetal testing. Am J Obstet Gynecol 1987;156:928–34.
8. Divon MY, Haglund B, Nisell H, Otterblad PO, Westgren M. Fetal and neonatal mortality in the postterm pregnancy: The impact of gestational age and fetal growth restriction. Am J Obstet Gynecol 1998;178:726–31.
9. Feldman GB. Prospective risk of stillbirth. Obstet Gynecol 1992;79:547–53.
10. Hilder L, Costeloe K, Thilaganathan B. Prolonged pregnancy: Evaluating gestation-specific risks of fetal and infant mortality. Br J Obstet Gynaecol 1998;105:169–73.
11. Usher RH, Boyd ME, McLean FH, Kramer MS. Assessment of fetal risk in postdate pregnancies. Am J Obstet Gynecol 1988;158:259–64.
12. Moore TR, Cayle JE. The amniotic fluid index in normal human pregnancy. Am J Obstet Gynecol 1990;162:1168–73.
13. McLean FH, Boyd ME, Usher RH. Postterm infants: Too big or too small? Am J Obstet Gynecol 1991;164:619–24.
14. Arias F. Predictability of complications associated with prolongation of pregnancy. Obstet Gynecol 1987;70:101–6.
15. Campbell MK, Ostbye T, Irgens LM. Post-term birth: Risk factors and outcomes in a 10-year cohort of Norwegian births. Obstet Gynecol 1997;89:543–8.
16. Boyd ME, Usher RH, McLean FH. Obstetric consequences of postmaturity. Am J Obstet Gynecol 1988;158:334–8.
17. Pollack RN, Hauer-Pollack G, Divon MY. Macrosomia in postdates pregnancies: The accuracy of routine ultrasonographic screening. Am J Obstet Gynecol 1992;167:7–11.
18. Moya F, Grannum P, Pinto K, Bracken M, Kadar N, Hobbins JC. Ultrasound assessment of the postmature pregnancy. Obstet Gynecol 1985;65:319–22.
19. Raymond EG, Cnattingius S, Kiely JL. Effects of maternal age, parity, and smoking on the risk of stillbirth. Br J Obstet Gynaecol 1994;101:301–6.
20. Yudkin PL, Wood L, Redman CW. Risk of unexplained stillbirth at different gestational ages. Lancet 1987;1:1192–4.
21. Taipale P, Hilesmaa V. Predicting delivery date by ultrasound and last menstrual period in early gestation. Obstet Gynecol 2001;97:189–94.
22. Mongelli M, Wilcox M, Gardosi J. Estimating the date of confinement: Ultrasonographic biometry versus certain menstrual dates. Am J Obstet Gynecol 1996;174:278–81.
23. Nathan L, Leveno KJ, Carmody TJ, Kelly MA, Sherman ML. Meconium: A 1990’s perspective on an old obstetric hazard. Obstet Gynecol 1994;83:329–32.
24. Gregory KD, Henry OA, Ramicone E, Chan LS, Platt LD. Maternal and infant complications in high and normal weight infants by method of delivery. Obstet Gynecol 1998;92:507–13.