Risks to both mother and infant increase as pregnancy progresses beyond 40 weeks of gestation.1 A conclusion drawn from the information in the Cochrane database is that routine induction of labor after 41 weeks gestation reduces perinatal mortality,2 and many clinicians advocate such management.3 Induction of labor is one of the most frequent obstetric interventions. It affected one fifth of all deliveries in the United States in 2002,4 and the induction rate has more than doubled during the past 10 years.5 Postterm pregnancy has been reported to be the most important reason for increased induction rates, followed by induction for large for gestational age and elective induction.6 There are contradictory opinions regarding induction of postterm pregnancies; it is still a matter of debate whether to follow the development expectantly with antenatal testing or to induce labor.7 Even although induction of labor will reduce the risk of intrauterine death,8 the procedure itself may be associated with increased cesarean delivery rates.9 Induction of labor should result in an uncomplicated vaginal delivery and a healthy newborn; however, this outcome is dependent on factors such as parity, ripening of the cervix, gestational length, reason for induction, and induction methods.
Epidemiologic studies can quantify the medical consequences of gestational age and induction of labor in a population, and describe possible associations. The aim of this study was to evaluate pregnancy complications and perinatal outcomes by week of gestation, and assess these outcomes in groups where onset of labor was spontaneous or induced.
MATERIALS AND METHODS
The study population was from a geographically well-defined area consisting of the city of Trondheim and eight surrounding municipalities. The area is served solely by St. Olavs Hospital, Trondheim University Hospital for all obstetric and perinatal care. In this nonselected population, 98% of the pregnant women in the area had a routine ultrasound scan at 18 weeks and subsequently delivered at St. Olavs Hospital. Demographic characteristics, obstetric history and data from the ultrasound scans were recorded in a computerized database throughout each pregnancy. After delivery, additional data from delivery and neonatal outcomes were recorded from the hospital records and included in the database. The study was submitted to the Regional Ethics Committee for Research in Medicine of Central Norway which exempted the study, since present regulations in Norway do not require approval for research based on data registries that have been made anonymous.
Women were included in this study if they had a singleton pregnancy, attended an ultrasound scan before 22 weeks of gestation, and delivered after 37 weeks of pregnancy during the period from January 1990 to December 2001. A total of 28,566 women were eligible for the study. Women who had elective cesarean deliveries (n=1,052) were excluded, leaving 27,514 women in the study population. Labor was induced in 2,500 cases. Indications for induction were grouped in the following prioritized order: preeclampsia, postterm pregnancy, fetal indication (fetal distress, intrauterine fetal death, malformations), maternal indication (diabetes mellitus, intrahepatic cholestasis, symphysiolysis), macrosomia, oligohydramnion or polyhydramnion, uterine scar or bad obstetric history, and intrauterine growth restriction (Table 1). The reason for induction of labor was missing in 396 (15.8%) (Table 1). All cases with premature rupture of membranes were defined as spontaneous onset of labor.
The methods for induction of labor were unchanged during the study period. When the cervix was favorable at the time of induction (Bishop score 6 or more), labor was routinely induced with amniotomy. If contractions were not regular, oxytocin was given intravenously 10 IU/L, start-dose 30 mL/h, increasing by 30 mL every 30 minutes to a maximum dose of 180 mL/h (Syntocinon, Novartis Pharmaceuticals, East Hanover, NJ). When the cervix was unfavorable (Bishop score less than 6), the cervix was ripened with prostaglandin E2. The routine method was 0.5 mg dinoprostone intracervically once per day (Minprostin endocervical gel, Pfizer, Inc., New York, NY). If necessary, the same procedure was followed on day 2.
The management protocol for postterm pregnancy at our hospital was unchanged during the study period. Routinely, postterm follow-up was initiated 14 days after the estimated date of delivery; nonstress test, fetal weight, and amniotic fluid estimations were done. If the pregnancy was uncomplicated, the women attended for follow-up every second day until spontaneous delivery occurred or labor was induced at 43 completed weeks. This routine is in accordance with the clinical guidelines recommended by the Norwegian Society of Obstetrics and Gynecology and is practiced at all Norwegian delivery units.10 The guidelines recommend vaginal birth after one cesarean delivery, and elective cesarean delivery after two or more previous cesarean deliveries.10
The denominator in calculations of complication rates was the total number of deliveries during a particular week of gestation, except for intrauterine fetal death, where the denominator was the number of ongoing pregnancies. Neonatal outcome measures were umbilical artery blood gases and pH and Apgar scores. The cutoff values for poor neonatal outcomes were defined as umbilical cord pH less than 7.10 and Apgar score less than 7 at 5 minutes. Umbilical cord pH was available in 92% of the cases. The diagnosis of fracture of clavicula, Erbs Paresis, meconium aspiration, pneumothorax, hypoglycemia, and hip dysplasia were according to the International Classification of Diseases, 9th Revision and International Classification of Diseases, 10th Revision classification systems. Acute caesarean delivery was defined as cesarean delivery performed less than 8 hours after the decision was made to operate. Maternal hemorrhage was defined as blood loss greater than 500 mL at delivery. Severe perineal lacerations were defined as grade 3 or 4 lacerations in which musculus sphincter ani was affected. Length of labor was classified as prolonged first stage of labor (less than 1 cm per 1.5 hours) and prolonged second stage of labor (more than 1 hour).
The Pearson χ2 was used to test for differences in the crude prevalence of perinatal outcomes and pregnancy complications. To assess linear associations across gestational weeks, the χ2 test for trend for proportions was used; for continuous variables we used linear regression. We performed a multivariable logistic regression analysis to control for possible confounding effects of gender, smoking, maternal age, parity, and birth weight. The relative risk of complications was estimated and expressed as odds ratios (OR) with 95% confidence intervals (CI). For statistical analyses, SPSS 12 (SPSS Inc., Chicago, IL) was used except for the analysis of intrauterine fetal death, where STATA 9 software (StataCorp LP, College Station, TX) was used.
The study population of 27,514 women comprised 43% para 0, 38% para 1 and 20% para 2 +. There were 21% self-reported daily smokers, and 11% of the women were aged older than 35 years. The cesarean delivery rate was stable around 12% for the entire study period. The overall proportion of operative vaginal deliveries was constant around 8%, but the proportion of vacuum deliveries increased, and the use of forceps decreased during the study period.
Overall, 2,500 (8.8%) births were induced during the study period. The proportion of induced labor increased throughout the study period from 6.1% to 9.1%. The week-specific induction rates varied from 5.4% at 39 weeks to 31% at 42 weeks. Preeclampsia was the main indication for induction at 37–41 weeks (30–50%), whereas postterm pregnancy represented the most frequent indication for induction at 42 weeks (74%) (Table 1). Overall, 48% of the inductions were performed because of preeclampsia or postterm pregnancy (Table 1).
Table 2 present the crude prevalence of neonatal outcomes according to gestational weeks and according to induced or spontaneous onset of labor. The proportion of newborns with low Apgar scores was consistently higher when labor was induced, as opposed to spontaneous onset; however, the difference was not significant at all gestational weeks (Table 2). At 42 weeks, 4.5% of the babies had an Apgar score below 7 at 5 minutes when labor was induced, whereas 2.3% of the newborns delivered after spontaneous labor had low Apgar score at 5 minutes (OR 2.0, CI 1.2–3.5). Hypoglycemia was more frequent when labor was induced before term (OR 2.5–4.8), but not postterm (OR 1.0–1.7) (Table 2).
A statistically significant negative linear association (P<.01) was found between gestational age and umbilical artery pH. The mean pH values were generally lower if labor was induced (Fig. 1, Table 2). An increased risk of low umbilical cord values when labor was induced was seen mainly in 39 and 40 gestational weeks (Table 2).
The fetal death rate in the total population was 0.2%. The proportion of intrauterine fetal death increased with gestational age (0.25% in gestational week 38, 1.55% in week 42), showing a significant trend with increasing gestational age (P value trend = .01) (Table 3). The prevalence of clavicula fracture, Erbs paresis, meconium aspiration, resuscitation, pneumothorax, and hip dysplasia did not differ materially between induced and spontaneous labors. The prevalence of Erbs paresis and pneumothorax did not vary with gestational age (Table 3). The proportion of meconium aspiration and the need for resuscitation had a U-shaped curve over gestational weeks, with the lowest risk at 39 weeks and the highest risk postterm (Table 3). A positive linear association was found between gestational age and fracture of clavicula and hip dysplasia (P<.001) (Table 3).
Deliveries were more complicated (cesarean delivery, operative vaginal deliveries, and maternal hemorrhage) when labor was induced. However, this was not observed for acute cesarean delivery at gestational week 37 and operative vaginal delivery at week 42 (Table 4). Acute cesarean delivery rates at week 42 were 8.9% for spontaneous and 21.6% for induced labors (OR 2.8, CI 2.2–3.7). Labor was more often prolonged both in the first and the second stage when labor was induced (OR 1.3–4.4), and the proportion varied with gestational age as well (P<.05) (Table 4). The rate of severe perineal injury did not vary by onset of labor or by gestational age.
In general, the complication rate was highest for postterm pregnancies. The risk of various maternal complications was significantly different between gestational weeks, independent of onset of labor (P value difference less than .01, ie, at least two values differ within gestational weeks) (Table 4). Apgar score at 5 minutes less than 7 and pH less than 7.10 were significantly different between gestational weeks when labor started spontaneously (P < .05), but not when labor was induced (Table 2).
Multivariable logistic regression analysis demonstrated that gestational age was a risk factor for all complications. Apgar scores had a U-shaped distribution curve according to week of gestation, with highest risk for low scores at 37 and 42 weeks and lowest at 39 weeks (Table 5). Acute cesarean delivery rates also had a U-shaped distribution curve, whereas the rate of operative vaginal deliveries increased consistently from 37 to 42 weeks (Table 5). Maternal hemorrhage had a U-shaped distribution curve, with the lowest prevalence at 38 weeks (Table 5).
In a multivariable logistic regression analysis, the risk for complications was generally higher when labor was induced (OR 1.3–2.8) (Table 5). Maternal age older than 35 years and nulliparity were statistically significant risk factors for all complications, but no substantial association was observed for offspring gender or smoking. Significant negative associations were, however, observed for fetal weight in relation to Apgar score at 1 minute less than 7, operative vaginal delivery, and maternal hemorrhage.
This study suggests that poor pregnancy outcomes vary with gestational age, and the rate seems to be lowest at 39 weeks and highest postterm. Induction of labor is a prognostic factor for delivery complications at or beyond term, also when adjusted for gestational age.
The cesarean delivery rate in Norway is low compared with many other countries,11 and the cesarean delivery rates did not increase during the study period. This contrasts an observed rising trend in the United Kingdom and United States.3,12 We observed, however, a high cesarean delivery rate among postterm pregnancies. When labor was induced at week 42, the risk for acute cesarean delivery was more than doubled compared with spontaneous onset of labor. This observation contrasts the results reported in a multicenter trial by Hannah et al,13 wherein the cesarean delivery rates were 21.2% when labor was induced and 24.5% when the pregnancy was managed expectantly.
The present study has some limitations. For example, no information was registered about the cervical status before induction; it is possible that knowledge of the cervical status before induction may have revealed subgroups of women with differences in success rates.14 Thus, more detailed registration might reveal that maternal risk factors rather than the induction procedure itself may be responsible for the observed increased frequency of poor outcomes.15
We observed that the proportion of different indications for induction varies with gestational week, and this is reflected in the neonatal outcomes. The proportion of poor neonatal outcome was highest in induced labors at 41 weeks. One possible explanation may be that more than 50% were induced in week 41 due to preeclampsia or fetal indications, and this was also reflected in high proportions of operative vaginal delivery and cesarean delivery. Poor neonatal outcomes may be due to antenatal pregnancy complications rather than induction, whereas differences in neonatal outcomes for spontaneous labor reflect an effect of gestational age per se.
In the study by Alexander et al7 an increasing proportion of prolonged second stage of labor was found from 40 to 42 weeks. We found an increasing proportion of both prolonged first and second stage of labor with increasing gestational age, and when labor was induced. The latter might be due to the induction method itself. Dinoprostone was used in our study, which has been found to be less effective than misoprostol.16 Using different induction protocols may affect the results. For example, by using misoprostol, the proportion of prolonged labors and cesarean delivery due to failure to progress might be reduced.
In agreement with previous studies, we found that induction of labor had a greater association with significantly greater blood loss at delivery, than did spontaneous onset of labor.17 Maternal hemorrhage in postterm pregnancies has been associated with macrosomia rather than the postterm delivery itself.18 In the present study, both factors were risk factors (OR 2.1 for high birth weight and OR 1.3 for postterm delivery).
Apgar score below 7 at 5 minutes in term infants has been associated with an increased risk of neonatal morbidity, infant mortality and neurologic impairment.19 In agreement with other studies, we found that the prevalence of low Apgar scores in spontaneous labors was higher at 37 weeks than 39 weeks.19 However, the prevalence of low Apgar scores at 5 minutes at all gestational age groups was found to be higher than the prevalence reported in a Swedish study.20
We found that the proportion of umbilical cord pH values less than 7.10 increased with gestational age, and this is in accordance with previous studies.20,21 Kitlinski and coworkers20 challenge the concept of one reference interval of cord pH values for all gestational weeks, and argue that mean ± 2 standard deviations of pH values stratified into gestational age intervals should be used as a reference guide, to avoid false-positive diagnosis of acidosis.
Most previous studies have focused either on complications related to induction of labor or on complications related to gestational age. Saunders et al22 examined the relationship between induction and gestational week. They found that the complication rate increased with gestational age, whereas we found a U-shaped distribution curve for most complications.
The rising proportion of inductions is of concern worldwide.5,23 During the study period, the rate of labor induction increased, but the induction rate is still quite low in Scandinavia.23 The increased induction rate is commonly explained by an improved ability—and wish—to plan the timing of delivery by physicians and their patients. The term “elective induction” has been introduced, and it is defined as absence of maternal or fetal indications for labor induction. In a study by Seyb and coworkers,24 an increased risk of cesarean delivery in nulliparous women was found if labor was induced, and the risk was the same for elective inductions and for medical inductions (OR 1.89 and 1.69, respectively). A high proportion of acute cesarean delivery is a clinical problem, because the complication rate for cesarean delivery is generally high (more than 20%).25 Elective induction of labor at any gestational age requires expenditures from the medical system,26 thus induction of labor has economic implications. According to a Canadian study, the cost of delivery was higher after induction of labor than after spontaneous onset of labor ($1,715 compared with $1,474, P<.001).27
Many units have adopted a policy to induce labor at or after 41 weeks of gestation, based on the evidence presented in the Cochrane database. In Norway the routine is to follow postterm pregnancies expectantly with traditional monitoring. However, we observed that 4% of inductions at gestational week 41 were due to postterm pregnancy. The best timing of induction of labor for postterm pregnancy is still a matter of controversy.7 The current Cochrane review2 relies heavily on one large randomized controlled trial,13 a study that may be invalid today due to new methods for induction of labor and ultrasound dating of pregnancies. The increased risk of intrauterine fetal death after gestational week 41 is best demonstrated when ongoing pregnancies are used as denominator.28,29 The risk for intrauterine fetal death in our study was 6 times as great in gestational week 42 as compared with gestational week 38. Although this difference was not statistically significant (P=.06), we observed a significant trend over gestational age categories (P=.01), and this risk factor has to be taken into account when management protocols for postterm pregnancies are evaluated. The increasing proportion of macrosomia in postterm pregnancies is well known,21 and one might expect an increasing proportion of fractures of clavicula and Erbs paresis with increasing gestational week as a consequence of macrosomia. We found no association for Erbs paresis. However, there was a statistically significant increased occurrence of hip dysplasia.
Conclusively, this study shows that delivery complications vary with gestational age and are highest at postterm. We found that induction of labor is a risk factor for complications also after adjustments for gestational age. The study indicates that induction as management for the known risk of the postterm pregnancy may introduce additional complications that may not lead to the desired total improvement of perinatal care. Due to the study design, conclusions must be drawn with caution. There is a need for more randomized controlled trials to evaluate antenatal fetal monitoring compared with induction of labor for postterm pregnancy.
1. Caughey AB, Musci TJ. Complications of term pregnancies beyond 37 weeks of gestation. Obstet Gynecol 2004;103:57–62.
2. Crowley P. Interventions for preventing or improving the outcome of delivery at or beyond term. Cochrane Database Syst Rev 2000:CD000170.
3. Roberts CL, Taylor L, Henderson-Smart D. Trends in births at and beyond term: evidence of a change? Br J Obstet Gynaecol 1999;106:937–42.
4. Mercer BM. Induction of labor in the nulliparous gravida with an unfavorable cervix. Obstet Gynecol 2005;105:688–9.
5. MacDorman MF, Mathews TJ, Martin JA, Malloy MH. Trends and characteristics of induced labour in the United States, 1989-98. Paediatr Perinat Epidemiol 2002;16:263–73.
6. Yawn BP, Wollan P, McKeon K, Field CS. Temporal changes in rates and reasons for medical induction of term labor, 1980–1996. Am J Obstet Gynecol 2001;184:611–9.
7. Alexander JM, McIntire DD, Leveno KJ. Forty weeks and beyond: pregnancy outcomes by week of gestation. Obstet Gynecol 2000;96:291–4.
8. Sue-A-Quan AK, Hannah ME, Cohen MM, Foster GA, Liston RM. Effect of labour induction on rates of stillbirth and cesarean section in post-term pregnancies. CMAJ 1999;160:1145–9.
9. Heffner LJ, Elkin E, Fretts RC. Impact of labor induction, gestational age, and maternal age on cesarean delivery rates. Obstet Gynecol 2003;102:287–93.
10. Lichtenberg S, Bungum L, Lande E. Cesarean section. In: Dalaker, K, Nøstdal W, Nilsen ST, Løvset T, Jerve F, editors. Veileder i fødselshjelp [Guideline booklet in obstetrics]. Oslo (Norway): Den norske lægeforening; 1998. p. 137–40.
11. Espedal BE. Intervention. Birth statistics. Bergen (Norway): Medical Birth Registry of Norway; 2004. p. 23–31.
12. Meikle SF, Steiner CA, Zhang J, Lawrence WL. A national estimate of the elective primary cesarean delivery rate. Obstet Gynecol 2005;105:751–6.
13. 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: a randomized controlled trial. The Canadian Multicenter Post-term Pregnancy Trial Group. N Engl J Med 1992;326:1587–92.
14. Shin KS, Brubaker KL, Ackerson LM. Risk of cesarean delivery in nulliparous women at greater than 41 weeks’ gestational age with an unengaged vertex. Am J Obstet Gynecol 2004;190:129–34.
15. Alexander JM, MCIntire DD, Leveno KJ. Prolonged pregnancy: induction of labor and cesarean births. Obstet Gynecol 2001;97:911–5.
16. Hofmeyr GJ, Gulmezoglu AM. Vaginal misoprostol for cervical ripening and induction of labour. Cochrane Database Syst Rev 2001:CD000941.
17. Phillip H, Fletcher H, Reid M. The impact of induced labour on postpartum blood loss. J Obstet Gynaecol 2004;24:12–5.
18. Olesen AW, Westergaard JG, Olsen J. Perinatal and maternal complications related to postterm delivery: a national register-based study, 1978-1993. Am J Obstet Gynecol 2003;189:222–7.
19. Thorngren-Jerneck K, Herbst A. Low 5-minute Apgar score: a population-based register study of 1 million term births. Obstet Gynecol 2001;98:65–70.
20. Kitlinski ML, Källén K, Marsál K, Olofsson P. Gestational age-dependent reference values for pH in umbilical cord arterial blood at term. Obstet Gynecol 2003;102:338–45.
21. 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.
22. Saunders N, Paterson C. Effect of gestational age on obstetric performance: when is “term” over? Lancet 1991;338:1190–2.
23. Cnattingius R, Höglund B, Kieler H. Emergency cesarean delivery in induction of labor: an evaluation of risk factors. Acta Obstet Gynecol Scand 2005;84:456–62.
24. Seyb ST, Berka RJ, Socol ML, Dooley SL. Risk of cesarean delivery with elective induction of labor at term in nulliparous women. Obstet Gynecol 1999;94:600–7.
25. Häger RM, Daltveit AK, Hofoss D, Nilsen ST, Kolaas T, Øian P, et al. Complications of cesarean deliveries: rates and risk factors. Am J Obstet Gynecol 2004;190:428–34.
26. Kaufman KE, Bailit JL, Grobman W. Elective induction: an analysis of economic and health consequences. Am J Obstet Gynecol 2002;187:858–63.
27. Allen VM, O’Connell, CM, Farrell SA, Baskett TF. Economic implications of method of delivery. Am J Obstet Gynecol 2005;193:192–7.
28. 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.
29. Yudkin PL, Wood L, Redman CW. Risk of unexplained stillbirth at different gestational ages. Lancet 1987;1:1192–4.