You are here

Induction of labor before 40 weeks is associated with lower rate of cesarean delivery in women with gestational diabetes mellitus

American Journal of Obstetrics and Gynecology, Volume 214, Issue 3, March 2016, Pages 364.e1 - 364.e8

Background

In women with gestational diabetes mellitus, it is not clear whether routine induction of labor at <40 weeks of gestation is beneficial to mother and newborn infant.

Objective

The purpose of this study was to compare outcomes among women with gestational diabetes mellitus who had induction of labor at either 38 or 39 weeks with those whose pregnancy was managed expectantly.

Study Design

We included all women in Ontario, Canada, with diagnosed gestational diabetes mellitus who had a singleton hospital birth at ≥38 + 0 weeks of gestation between April 2012 and March 2014. Data were obtained from the Better Outcomes Registry & Network Ontario, which is a province-wide registry of all births in Ontario, Canada. Women who underwent induction of labor at 38 + 0 to 38 + 6 weeks of gestation (38-IOL; n = 1188) were compared with those who remained undelivered until 39 + 0 weeks of gestation (38-Expectant; n = 5229). Separately, those women who underwent induction of labor at 39 + 0 to 39 + 6 weeks of gestation (39-IOL; n = 1036) were compared with women who remained undelivered until 40 + 0 weeks of gestation (39-Expectant; n = 2162). Odds ratios and 95% confidence intervals were adjusted for maternal age, parity, insulin treatment, and prepregnancy body mass index.

Results

Of 281,480 women who gave birth during the study period, 14,600 women (5.2%) had gestational diabetes mellitus; of these, 8392 women (57.5%) met all inclusion criteria. Compared with the 38-Expectant group, those women in the 38-IOL group had lower odds for cesarean delivery (adjusted odds ratio, 0.73; 95% confidence interval, 0.52–0.90), higher odds for neonatal intensive care unit admission (adjusted odds ratio, 1.36; 95% confidence interval, 1.09–1.69), and no difference in other maternal-newborn infant outcomes. Compared with the 39-Expectant group, women in the 39-IOL group likewise had lower odds for cesarean delivery (adjusted odds ratio, 0.73; 95% confidence interval, 0.58–0.93) but no difference in neonatal intensive care unit admission (adjusted odds ratio, 0.83; 95% confidence interval, 0.61–1.11).

Conclusion

In women with gestational diabetes mellitus, the routine induction of labor at 38 or 39 weeks is associated with a lower risk of cesarean delivery compared with expectant management but may increase the risk of neonatal intensive care unit admission when done at <39 weeks of gestation.

Key words: gestational diabetes mellitus, induction, expectant management, delivery.

Gestational diabetes mellitus (GDM), which is 1 of the most common medical complications of pregnancy, affects >7% of pregnancies in the United States1 and has been shown to be associated with adverse pregnancy outcome.2 and 3

One of the major controversies in the management of pregnancies that are complicated by GDM relates to the optimal timing of delivery. A policy of a routine induction of labor (IOL) at 38-39 weeks of gestation carries the potential benefit of decreasing macrosomia-related complications that include cesarean delivery, shoulder dystocia, anal sphincter injuries, and birth trauma.4, 5, 6, 7, and 8 Furthermore, in some cases, clinicians may choose to induce labor at 38-39 weeks to eliminate the risk of GDM-related stillbirth, although data regarding the association of GDM with stillbirth are conflicting.9, 10, 11, and 12 However, such a policy of routine IOL might also be associated with certain risks that include increasing the risk of cesarean delivery because of induction failure, fetal distress because of uterine hyperstimulation, or neonatal morbidity that is related to early-term delivery (eg, respiratory morbidity, jaundice) when induction takes place at <39 weeks of gestation.13, 14, 15, 16, 17, and 18

Data regarding the benefits and risks of routine IOL in women with GDM are sparse. This question has been addressed by only a small number of studies that were limited by small sample size, lack of adequate control group, lack of adjustment for potential confounders (eg, type of treatment, body mass index [BMI]) and potential for selection bias.19, 20, 21, and 22 Indeed, in a recent systematic review,23 it was concluded that the substantial heterogeneity of these studies precludes the completion of a quantitative synthesis of the data or drawing definite conclusions about timing of delivery in women with GDM. However, recent evidence that IOL in low-risk pregnancies4 and 24 and in pregnancies with large-for-date fetuses25 does not increase the rate of cesarean delivery and may decrease the rate of shoulder dystocia25 provides indirect support to the rationale underlying such a practice of routine IOL in women with GDM. Furthermore, in a recent large population-based study on the risk of stillbirth and infant death by gestational age in women with GDM, the authors concluded that expectant treatment at 39 and 40 weeks of gestation in women with GDM is associated with significantly greater risk of perinatal death (stillbirth and infant death) compared with delivery at these corresponding weeks.9

Our aim was to compare pregnancy outcomes among women with GDM who had IOL at either 38 or 39 weeks of gestation and those who were treated expectantly.

Methods

Study population

This was a retrospective cohort study of all women with GDM who had a singleton hospital birth at ≥38 + 0 weeks of gestation in Ontario, Canada, between April 2012 and March 2014. Data were obtained from the Better Outcomes Registry & Network (BORN) Ontario (https://www.bornontario.ca/en/about-born/) BORN Ontario is a province-wide registry of all births in Ontario, Canada. Whenever a woman is admitted to a hospital to give birth, data are collected by health care providers and hospital staff from charts, clinical forms, and patient interview then entered into the BORN Information System (either directly or by electronic upload from a hospital’s electronic medical records system). The BORN Information System contains maternal demographics, health behaviors, reproductive history, and clinical information related to pregnancy, labor, and birth, fetal, and neonatal outcomes. An ongoing program of data verifications, quality checks, and formal training sessions for individuals who collect and enter data assures that a high level of data quality is maintained.

To create a “low-risk” GDM cohort, women with any of the following conditions were excluded from the study: gestational age at birth of ≥42 + 0 weeks, women who were not candidates for vaginal birth (eg, nonvertex presentation, placenta previa), previous cesarean delivery, major fetal anomalies, or chronic maternal medical conditions that potentially could influence a decision to induce labor at 38-39 weeks of gestation that include pre-GDM, chronic hypertension, chronic renal disease, cardiac disease (congenital or acquired), pulmonary disease (pulmonary hypertension, cystic fibrosis, pulmonary embolism), autoimmune conditions (systemic lupus erythematosus), or hematologic diseases (hemophilia, sickle cell disease; Figure).

gr1

Figure Description of the study and control groupsSelection of the induction and expectant management groups at 38 and 39 weeks. The asterisk refers to chronic medical conditions that include pregestational diabetes mellitus, chronic hypertension, and renal, cardiac, pulmonary, vascular, and hematologic diseases that could influence a primary decision to induce labor at 38-39 weeks of gestation.CS, cesarean delivery; GDM, gestational diabetes mellitus.Melamed et al. Induction of labor in GDM. Am J Obstet Gynecol 2016.

Some of the previous observational studies on the effects of IOL are limited by the use of a control group of women who experience spontaneous onset of labor. However, in clinical practice, the real choice is not between IOL and spontaneous labor, but between IOL and expectant management; the latter option carries that risk of new onset of pregnancy complications and the potential for labor induction at a later stage of gestation. Thus, to simulate the decision faced by physicians in real-life, women who underwent IOL at 38 + 0 to 38 + 6 weeks of gestation in the current study (38-IOL group) were compared with those who were treated expectantly and remained undelivered until 39 + 0 weeks of gestation (ie, delivered anytime between 39 + 0 and 41 + 6 weeks of gestation; 38-Expectant group). A separate comparison was performed between women who underwent IOL at 39 + 0 to 39 + 6 weeks of gestation (39-IOL group) and those who were treated expectantly and remained undelivered until 40 + 0 weeks (ie, delivered anytime between 40 + 0 and 41 + 6 weeks of gestation; 39-Expectant group; Figure). Only women whose sole documented reason for IOL was GDM or macrosomia were included in the 38-IOL and 39-IOL groups (Figure).

Outcomes

The primary outcome was the rate of cesarean delivery. Secondary outcomes included the rate of instrumental delivery, postpartum hemorrhage, intrapartum fever, anal sphincter injury (defined as a third- or fourth-degree perineal laceration), shoulder dystocia, and neonatal morbidity. Neonatal respiratory morbidity was defined as any of the following events: need for respiratory support in the form of continuous positive airway pressure or mechanical ventilation, a diagnosis of transient tachypnea of the newborn infant, or respiratory distress syndrome. Composite neonatal morbidity was defined as the presence of any of the following events: perinatal death, 5-minute Apgar score <7, admission to the neonatal intensive care unit (NICU), hypoglycemia, jaundice that required phototherapy, or neonatal respiratory morbidity.

Diagnosis of GDM

This study spans 2 time periods with regards to criteria for the diagnosis of GDM in Ontario. Before April 2013, the criteria for diagnosis were according to 2008 Canadian Diabetes Associations guidelines.26 These guidelines recommended screening for GDM with a 50 g glucose challenge test (GCT); when the results were positive (>7.8 mmol/L or 140 mg/dL), a 75g oral glucose tolerance test (OGTT) followed (cutoff values: fasting, ≥5.3 mmol/L or 96 mg/dL; 1 hour ≥10.6 mmol or 191 mg/dL; 2 hours ≥ 8.9 mmol/L or 160 mg/dL). GDM was defined as ≥2 abnormal OGTT values or a GCT result of ≥10.3 mmol/L or 185 mg/dL. The presence of a single abnormal OGTT value was defined as impaired glucose tolerance.

In April 2013 new Canadian Diabetes Associations criteria were published.27 The new guidelines allowed two options for screening/testing for GDM. The “Preferred” option was essentially identical to the Canadian Diabetes Associations 2008 guidelines aside from increasing the diagnostic 50g-GCT value from 10.3 mmol (185 mg/dL) to ≥ 11.1 mmol (200 mg/dL), and the 2-hour 75g-OGTT threshold from 8.9 mmol/L (160 mg/dL) to 9.0 mmol/L (162 mg/dL). The distinction between impaired glucose tolerance and GDM was eliminated in these new guidelines.

Data analysis

Multivariable logistic regression analysis was used to adjust for potential confounding variables that included maternal age (as a continuous variable), nulliparity, need for insulin treatment, prepregnancy BMI (as a continuous variable), and macrosomia (birthweight, >4000 g). This analysis was repeated within a subgroup of nulliparous women, because parity is a major determinant of the success of labor induction.28 In addition, because we did not have data about the Bishop’s cervix score (which is associated with the likelihood of successful induction), we assumed that variation in this score would likely be low among nulliparous women.29

Data were analyzed using the SAS statistical software (version 9.4; SAS Institute Inc, Cary, NC). Significance was set at a 2-sided probability value of <.05. The current study was approved by the Sunnybrook Health Sciences Center Research Ethics Board.

Results

Characteristics of the study population

During the study period, 281,480 women gave birth in hospitals in Ontario, Canada, 14,600 (5.2%) of whom were diagnosed with GDM, of whom 8392 were eligible for the study (Figure). Overall 1188 women who underwent IOL at 38 + 0 to 38 + 6 weeks of gestation for GDM-related reasons (38-IOL group) were compared with 5229 women who remained undelivered until at least 39 + 0 weeks (38-Expectant group). Similarly, 1036 women who underwent IOL at 39 + 0 to 39 + 6 weeks of gestation for GDM-related reasons (39-IOL group) were compared with 2162 women who remained undelivered until at least 40 + 0 weeks of gestation (39-Expectant group).

The characteristics of the induction and expectant groups are presented in Table 1. Women in the induction groups were older, less likely to be nulliparous, more likely to require insulin treatment, and had a higher prepregnancy BMI.

Table 1 Characteristics of the induction and expectant management groups

Characteristic Weeks of gestation
38 39
Induction (n = 1188) Expectant management (n = 5229) P value Induction (n = 1036) Expectant management (n = 2162) P value
Maternal age, ya 32.5 ± 5.1 32.1 ± 5.0 .004 32.5 ± 5.0 31.6 ± 5.1 < .001
Age >35 y, n (%) 329 (27.7) 1293 (24.7) .03 278 (26.8) 492 (22.8) .01
Nulliparity, n (%) 483 (40.66) 2632 (50.33) < .001 494 (47.7) 1210 (56.0) < .001
Insulin treatment, n (%) 906 (76.3) 1641 (31.4) < .001 626 (60.4) 399 (18.5) < .001
Body mass index, kg/m2a 29.1 ± 7.2 27.2 ± 6.9 < .001 28.4 ± 7.2 26.9 ± 6.7 < .001
 >25, n (%) 671 (56.5) 2247 (43.0) < .001 559 (54.0) 924 (42.7) < .001
 >30, n (%) 361 (30.4) 1141 (21.8) < .001 313 (30.2) 445 (20.6) < .001
 >35, n (%) 168 (14.1) 509 (9.7) < .001 149 (14.4) 191 (8.8) < .001
Male newborn infant, n (%) 597 (50.3) 2641 (50.5) .5 542 (52.3) 1085 (50.2) .4

a Data are presented as mean ± standard deviation.

Melamed et al. Induction of labor in GDM. Am J Obstet Gynecol 2016.

Pregnancy complications

Women in the 38-IOL and 39-IOL groups had a lower rate of cesarean delivery compared with women in the 38-Expectant and 39-Expectant groups, respectively (16.7 % vs 21.3% [P < .001] and 19.6% vs 22.9% [P = .04], respectively; Table 2). The rates of instrumental deliveries were similar between the induction and expectant groups.

Table 2 Pregnancy outcomes in the induction and expectant management groups

Outcome Weeks of gestation
38 39
Induction (n = 1188), n (%) Expectant management (n = 5229), n (%) P value Induction (n = 1036), n (%) Expectant management (n = 2162), n (%) P value
Hypertensive complicationsa 293 (5.6) N/A 103 (4.8) N/A
 Gestational hypertension 274 (5.2) N/A 97 (4.5) N/A
 Preeclampsia 19 (0.4) N/A 6 (0.3) N/A
Labor induction 1188 (100.0) 2484 (47.5) < .001 1036 (100.0) 1044 (48.3) < .001
Mode of delivery
 Cesarean delivery 198 (16.7) 1114 (21.3) < .001 203 (19.6) 494 (22.9) .04
 Instrumental 144 (12.1) 612 (11.7) .7 129 (12.5) 273 (12.6) .9
Epidural 928 (78.1) 3816 (73.0) < .001 849 (82.0) 1592 (73.6) < .001
Postpartum hemorrhage 25 (2.1) 111 (2.1) .9 20 (1.9) 45 (2.1) .8
Shoulder dystocia 36 (3.0) 149 (2.8) .7 41 (4.0) 61 (2.8) .09
Anal sphincter injury 27 (2.3) 197 (3.8) .01 40 (3.9) 94 (4.4) .5

a Not applicable to induction groups because the presence of hypertensive complications was an exclusion criterion.

N/A, not applicable.

Melamed et al. Induction of labor in GDM. Am J Obstet Gynecol 2016.

The rate of anal sphincter injuries was lower in women in the 38-IOL group compared with those in the 38-Expectant group (2.3% vs 3.8%; P = .01); similar differences were not observed between women in the 39-IOL and 39-Expectant groups (Table 2).

Women in the 38-Expectant and 39-Expectant groups were at risk of new-onset of hypertensive complications (5.6% and 4.8%, respectively) and eventually underwent labor induction in 47.% and 48.3% of cases, respectively (Table 2).

Neonatal outcomes

Infants of mothers in the 38-IOL and 39-IOL groups were less likely to have a birth weight >4000 g compared with women in the 38-Expectant and 39-Expectant groups, respectively (9.1% vs 12.7% [P < .001] and 12.4% vs 15.5% [P = .02], respectively; Table 3).

Table 3 Neonatal outcomes in the induction and expectant management groups

Outcome Weeks of gestation
38 39
Induction (n = 1188) Expectant management (n = 5229) P value Induction (n = 1036) Expectant management (n = 2162) P value
Gestational age at delivery, wksa 38.4 ± 0.3 39.8 ± 0.7 < .001 39.3 ± 0.3 40.5 ± 0.5 < .001
Birthweight, ga 3362 ± 468 3486 ± 474 < .001 3473 ± 483 3557 ± 461 < .001
 >90th Percentileb 117 (9.9) 521 (10.0) .9 103 (9.9) 214 (9.9) .9
 >4000 gr 108 (9.1) 666 (12.7) < .001 128 (12.4) 335 (15.5) .02
 >4500 gr 17 (1.4) 128 (2.5) .03 23 (2.2) 64 (3.0) .2
Composite morbidityc 355 (29.9) 1550 (29.6) .9 282 (27.2) 673 (31.1) .02
Perinatal mortality 0 (0.0) 8 (0.2) .4 0 (0.0) 5 (0.2) .2
Neonatal intensive care unit admission 165 (13.9) 567 (10.8) .002 103 (9.9) 247 (11.4) .2
Total length of stay at neonatal intensive care unit, da 3.0 ± 3.8 3.2 ± 4.8 .6 3.1 ± 5.0 2. 9 ± 3.3 .7
Respiratory morbidityd 36 (3.0) 188 (3.6) .3 27 (2.6) 89 (4.1) .03
Jaundice requiring phototherapy 59 (5.0) 178 (3.4) .01 39 (3.8) 70 (3.2) .4
Hypoglycemia 74 (6.2) 211 (4.0) < .001 49 (4.7) 77 (3.6) .1

a Data are presented as mean ± standard deviation

b The 90th percentile cut-offs were obtained from published national curves36 and included the following information: 3960 g for the 38-week induction group, 4081 g for the 38-week expectant management group, 4094 g for the 39-week induction group, 4150 g for the 39-week expectant management group

c Composite morbidity was defined as any of the following events: perinatal death, 5-minute Apgar score <7, admission to the neonatal intensive care unit, hypoglycemia, jaundice that required phototherapy, or neonatal respiratory morbidity

d Refers to respiratory distress syndrome, transient tachypnea of the newborn infant, or need for respiratory support.

Melamed et al. Induction of labor in GDM. Am J Obstet Gynecol 2016.

Infants of mothers in the 38-IOL group had a higher rate of admission to NICU (13.9% vs 10.8%; P = .002), jaundice that required phototherapy (5.0% vs 3.4%; P = .01) and neonatal hypoglycemia (6.2% vs 4.0%; P < .001) compared with infants of mothers in the 38-Expectant group (Table 3). In contrast, IOL at 39 weeks of gestation was not associated with similar risks and, in fact, was associated with lower composite neonatal morbidity (27.2% vs 31.1%; P = .02) and respiratory morbidity (2.6% vs 4.1%; P = .03) compared with expectant management (Table 3).

Association of induction with adverse outcome: multivariable analysis

Given the differences in the characteristics of the induction and expectant groups and to minimize the risk of selection bias, we used multivariable logistic regression analysis to adjust for potential confounding variables that included maternal age, parity, insulin treatment, BMI, and macrosomia (Table 4). IOL at 38 or 39 weeks of gestation was associated with a similar reduction in the risk of cesarean delivery compared with expectant management (adjusted odds ratio, 0.73; 95% confidence interval, 0.52–0.90 and adjusted odds ratio, 0.73; 95% confidence interval, 0.58–0.93, respectively; Table 4). IOL at 38 weeks of gestation, but not at 39, was associated with a greater risk for neonatal admission to NICU (adjusted odds ratio, 1.36; 95% confidence interval, 1.09–1.69). Otherwise, IOL at 38 or 39 weeks of gestation was not associated with the risk of instrumental delivery, anal sphincter injury, or composite neonatal morbidity (Table 4).

Table 4 Association of labor induction (vs expectant management) with adverse maternal and neonatal outcome: multivariable analysis

Outcome Induction vs expectant management, adjusted odds ratio (95% confidence interval)
At 38 wk gestation At 39 wk gestation
Cesarean delivery 0.73 (0.52–0.90) 0.73 (0.58–0.93)
Instrumental delivery 1.10 (0.89–1.46) 1.19 (0.89–1.57)
Anal sphincter injurya 0.97 (0.61–1.55) 1.16 (0.72–1.87)
Composite neonatal morbidityb 1.10 (0.93–1.30) 0.84 (0.69–1.03)
Neonatal intensive care admission 1.36 (1.09–1.69) 0.83 (0.61–1.11)

a Analysis was limited to vaginal deliveries

b Composite morbidity was defined as any of the following events: perinatal death, 5-minute Apgar score <7, admission to the neonatal intensive care unit, hypoglycemia, jaundice that required phototherapy, or neonatal respiratory morbidity.

Values reflect the results of multivariable logsitic regression analysis, with adjustment for the following potential confounders: maternal age, nulliparity, insulin treatment, prepregnancy body mass index, and macrosomia (birthweight, >4000 g).

Melamed et al. Induction of labor in GDM. Am J Obstet Gynecol 2016.

The association between IOL and mode of delivery was then recalculated within the subgroup of nulliparous women, according to the rationale described in the Methods section (Table 5). IOL at 39 weeks of gestation compared with expectant management in nulliparous women was associated with a similar decrease in the risk of cesarean delivery as was observed in the overall cohort (adjusted odds ratio, 0.75; 95% confidence interval, 0.57–0.99). However, in contrast to the overall cohort, IOL at 38 weeks of gestation in nulliparous women had no association with the risk of cesarean delivery.

Table 5 Association of labor induction (vs expectant management) with mode of delivery in the subgroup restricted to nulliparous women: multivariable analysis

Outcome Induction vs expectant management, adjusted odds ratio (95% confidence interval)
At 38 wk gestation At 39 wk gestation
Cesarean delivery 0.88 (0.68–1.13) 0.75 (0.57–0.99)
Instrumental delivery 1.16 (0.86–1.56) 1.34 (0.96–1.86)

Values reflect the results of multivariable logsitic regression analysis, with adjustment for the following potential confounders: maternal age, insulin treatment, prepregnancy body mass index, and macrosomia (birthweight, >4000 g).

Melamed et al. Induction of labor in GDM. Am J Obstet Gynecol 2016.

Comment

Principal findings of the study

Our aim was to compare pregnancy outcomes among women with GDM who had IOL at either 38 or 39 weeks of gestation and those who were treated expectantly. Our main findings follow: (1) IOL at 38 or 39 weeks of gestation in women with GDM is associated with a similar decrease in the risk of cesarean delivery compared with expectant management at the corresponding weeks of gestation. (2) Among nulliparous women with GDM, such a decrease in the risk of cesarean delivery is seen only with IOL at 39 weeks of gestation, but not at 38 weeks of gestation. (3) IOL at 38 weeks of gestation is associated with an increased risk of NICU admission. (4) There was no association between IOL and the risk of instrumental deliveries, shoulder dystocia, anal sphincter injuries, and composite neonatal morbidity. (5) IOL at either 38 or 39 weeks of gestation may eliminate the approximate 5% risk of future hypertensive complications in those women who remain undelivered.

Results of the study in the context of other observations

Guideline recommendations on the timing of delivery in pregnancies that are complicated by GDM are inconsistent and often conflicting.30, 31, and 32 The clinician is thus often faced with the dilemma of whether to intervene and expedite delivery when no additional obstetric indications for intervention exist. This decision process involves balancing the potential for complications that are caused by increased interventions with the benefit of avoidance of future adverse outcomes. Thus, such a practice theoretically can decrease the risk of macrosomia and the risk of cesarean delivery because of dystocia.

Alternatively, it might increase the risk of cesarean delivery because of induction failure, fetal distress secondary to uterine hypersimulation, and fetal morbidity when induction takes place at <39 weeks of gestation.13, 14, 15, 16, 17, and 18 This question has been addressed by only a small number of studies that were limited by small sample size, lack of adequate control group, lack of adjustment for potential confounders, and potential for selection bias.19, 20, 21, and 22 Although a randomized-controlled study would be the ideal design to address this question (and such a study is currently underway33), such studies may be underpowered to address uncommon adverse outcomes and may be further limited by protocols for indicated cesarean delivery because of suspected macrosomia. In the current study, we tried to overcome these limitations by using a large diverse population-based cohort with an adequate control group of women who were followed expectantly while we were adjusting for multiple confounding variables.

We have found that IOL at 38 or 39 weeks of gestation is associated with a similar and considerable decrease in the risk of cesarean delivery. This finding may imply that the decrease in the rate of macrosomia that was noted in induction groups in the current study plays an important role in lowering the risk of cesarean delivery and is more significant than any increase in the risk of cesarean delivery secondary to induction failure. The fact that 50% of the women who were treated expectantly at 38 or 39 weeks of gestation ultimately underwent IOL at a more advanced gestational age was another factor that likely affected the difference in cesarean delivery rates between the groups. Our findings are supported by data from cohort studies of low-risk women who underwent elective IOL4, 24, and 34 and from a recent randomized controlled trial of IOL in mainly nondiabetic women with large-for-date fetuses.25

Strengths and limitations

Although the decrease in the rate of cesarean delivery persisted after adjustment for multiple confounding variables, a selection bias because of preferential IOL in women with higher Bishop score cannot be ruled out because information on the preinduction Bishop score was not available. However, although such a bias may play a greater role at 38 weeks of gestation, it is less likely to be a dominant factor in the decision to induce labor at 39 weeks of gestation because many physicians who treat GDM pregnancies routinely consider IOL at 39 weeks of gestation, regardless of the Bishop score.35 Thus, we believe that the association of IOL at 39 weeks of gestation with a lower risk of cesarean delivery is less likely to be affected by such a potential selection bias.

Furthermore, given that the proportion of nulliparous women with a high Bishop score before the onset of labor is low,29 the fact that IOL at 39 weeks of gestation was associated with a lower risk of cesarean delivery in the subgroup of nulliparous women provides additional support to the validity of this assumption.

Given the retrospective design, some of the information, such as the method of induction and adequacy of glycemic control, were not available. Thus, residual confounding cannot be ruled out. In addition, despite its large size, this study is powered insufficiently to assess rare outcomes such as stillbirth, although it is clear that any policy of IOL will prevent all stillbirths (and other complications such as preeclampsia) that are destined to occur at a later gestational age.

A strength of this study is the large sample size that represents the largest cohort on this topic that allowed us to adjust for several significant potential confounding variables.

Conclusion

In summary, we have found that IOL in women with GDM and no other medical complications, at 38 or 39 weeks of gestation, is associated with a decrease in the risk of cesarean delivery. Whether these findings can provide support for a practice of routine IOL for all women with GDM at 38 or 39 weeks of gestation is still premature, given the retrospective design and the limitations described earlier. We do believe, however, that when routine induction is considered, these findings may provide support for IOL at 39 rather than 38 weeks of gestation because of the association of induction at 38 weeks of gestation with increased risk for NICU admission, which is an outcome that, at the population level, may have significant implications in term of costs and neonatal risks. It should be emphasized that the impact of routine IOL on the risk of cesarean delivery also depends on local guidelines and protocols, such as having an upper threshold of weight estimation (eg, 4000 g or 4500 g) for vaginal birth in women with GDM; in such settings, the impact of IOL on the risk of cesarean delivery may be even greater.

References

  • 1 V. Woo. Important differences: Canadian Diabetes Association 2008 clinical practice guidelines and the consensus statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetologia. 2009;52:552-555
  • 2 C.A. Crowther, J.E. Hiller, J.R. Moss, et al. Effect of treatment of gestational diabetes mellitus on pregnancy outcomes. N Engl J Med. 2005;352:2477-2486 Crossref
  • 3 O. Langer, Y. Yogev, O. Most, E.M. Xenakis. Gestational diabetes: the consequences of not treating. Am J Obstet Gynecol. 2005;192:989-997 Crossref
  • 4 Y.W. Cheng, A.J. Kaimal, J.M. Snowden, J.M. Nicholson, A.B. Caughey. Induction of labor compared to expectant management in low-risk women and associated perinatal outcomes. Am J Obstet Gynecol. 2012;207:502.e1-502.e8 Crossref
  • 5 T.S. Nesbitt, W.M. Gilbert, B. Herrchen. Shoulder dystocia and associated risk factors with macrosomic infants born in California. Am J Obstet Gynecol. 1998;179:476-480 Crossref
  • 6 C.D. Naylor, M. Sermer, E. Chen, K. Sykora. Cesarean delivery in relation to birth weight and gestational glucose tolerance: pathophysiology or practice style? Toronto Trihospital Gestational Diabetes Investigators. JAMA. 1996;275:1165-1170 Crossref
  • 7 S.L. Boulet, G.R. Alexander, H.M. Salihu, M. Pass. Macrosomic births in the united states: determinants, outcomes, and proposed grades of risk. Am J Obstet Gynecol. 2003;188:1372-1378 Crossref
  • 8 N. Melamed, O. Gavish, M. Eisner, A. Wiznitzer, N. Wasserberg, Y. Yogev. Third- and fourth-degree perineal tears: incidence and risk factors. J Matern Fetal Neonatal Med. 2013;26:660-664 Crossref
  • 9 M.G. Rosenstein, Y.W. Cheng, J.M. Snowden, J.M. Nicholson, A.E. Doss, A.B. Caughey. The risk of stillbirth and infant death stratified by gestational age in women with gestational diabetes. Am J Obstet Gynecol. 2012;206:309.e1-309.e7 Crossref
  • 10 A. Karmon, A. Levy, G. Holcberg, A. Wiznitzer, M. Mazor, E. Sheiner. Decreased perinatal mortality among women with diet-controlled gestational diabetes mellitus. Int J Gynaecol Obstet. 2009;104:199-202 Crossref
  • 11 A. Lapolla, M.G. Dalfra, M. Bonomo, et al. Gestational diabetes mellitus in Italy: a multicenter study. Eur J Obstet Gynecol Reprod Biol. 2009;145:149-153 Crossref
  • 12 H.E. Fadl, I.K. Ostlund, A.F. Magnuson, U.S. Hanson. Maternal and neonatal outcomes and time trends of gestational diabetes mellitus in Sweden from 1991 to 2003. Diabet Med. 2010;27:436-441 Crossref
  • 13 J.H. Booth, V.B. Kurdyak. Elective induction of labour: a controlled study. Can Med Assoc J. 1970;103:245-248
  • 14 J.A. Macer, C.L. Macer, L.S. Chan. Elective induction versus spontaneous labor: a retrospective study of complications and outcome. Am J Obstet Gynecol. 1992;166:1690-1697 Crossref
  • 15 D.A. Luthy, J.A. Malmgren, R.W. Zingheim. Cesarean delivery after elective induction in nulliparous women: the physician effect. Am J Obstet Gynecol. 2004;191:1511-1515 Crossref
  • 16 M. Prysak, F.C. Castronova. Elective induction versus spontaneous labor: a case-control analysis of safety and efficacy. Obstet Gynecol. 1998;92:47-52
  • 17 S. Robson, B. Pridmore, J. Dodd. Outcomes of induced labour. Aust N Z J Obstet Gynaecol. 1997;37:16-19 Crossref
  • 18 S.T. Seyb, R.J. Berka, M.L. Socol, S.L. Dooley. Risk of cesarean delivery with elective induction of labor at term in nulliparous women. Obstet Gynecol. 1999;94:600-607 Crossref
  • 19 S. Lurie, A. Matzkel, A. Weissman, Z. Gotlibe, A. Friedman. Outcome of pregnancy in class A1 and A2 gestational diabetic patients delivered beyond 40 weeks’ gestation. Am J Perinatol. 1992;9:484-488 Crossref
  • 20 S.L. Kjos, O.A. Henry, M. Montoro, T.A. Buchanan, J.H. Mestman. Insulin-requiring diabetes in pregnancy: a randomized trial of active induction of labor and expectant management. Am J Obstet Gynecol. 1993;169:611-615 Crossref
  • 21 S. Lurie, V. Insler, Z.J. Hagay. Induction of labor at 38 to 39 weeks of gestation reduces the incidence of shoulder dystocia in gestational diabetic patients class A2. Am J Perinatol. 1996;13:293-296 Crossref
  • 22 D.L. Conway, O. Langer. Elective delivery of infants with macrosomia in diabetic women: reduced shoulder dystocia versus increased cesarean deliveries. Am J Obstet Gynecol. 1998;178:922-925 Crossref
  • 23 C.T. Witkop, D. Neale, L.M. Wilson, E.B. Bass, W.K. Nicholson. Active compared with expectant delivery management in women with gestational diabetes: a systematic review. Obstet Gynecol. 2009;113:206-217 Crossref
  • 24 B.G. Darney, J.M. Snowden, Y.W. Cheng, et al. Elective induction of labor at term compared with expectant management: maternal and neonatal outcomes. Obstet Gynecol. 2013;122:761-769 Crossref
  • 25 M. Boulvain, M.V. Senat, F. Perrotin, et al. Induction of labour versus expectant management for large-for-date fetuses: a randomised controlled trial. Lancet. 2015;385:2600-2605 Crossref
  • 26 Canadian Diabetes Association Clinical Practice Guidelines Expert Committee. Canadian Diabetes Association 2008 Clinical Practice Guidelines For The Prevention and Management of Diabetes in Canada. Can J Diabetes. 2008;32(suppl 1):S1-S201
  • 27 Canadian Diabetes Association Clinical Practice Guidelines Expert Committee. Canadian Diabetes Association 2013 Clinical Practice Guidelines for the Prevention and Management of Diabetes in Canada. Can J Diabetes. 2013;37(suppl 1):S1-S212
  • 28 Y. Yogev, L. Hiersch, O. Yariv, Y. Peled, A. Wiznitzer, N. Melamed. Association and risk factors between induction of labor and cesarean section. J Matern Fetal Neonatal Med. 2013;26:1733-1736 Crossref
  • 29 F.P. Vrouenraets, F.J. Roumen, C.J. Dehing, E.S. van den Akker, M.J. Aarts, E.J. Scheve. Bishop score and risk of cesarean delivery after induction of labor in nulliparous women. Obstet Gynecol. 2005;105:690-697 Crossref
  • 30 Committee on Practice Bulletins: Obstetrics. Practice Bulletin No. 137: Gestational diabetes mellitus. Obstet Gynecol. 2013;122:406-416
  • 31 National Institute for Health and Clinical Excellence: Guidance Diabetes in Pregnancy: Management of Diabetes and Its Complications from Preconception to the Postnatal Period. London, 2015. Available at: www.nice.org.uk/guidance/ng3. Accessed January 26, 2016.
  • 32 American Diabetes Association. 12. Management of diabetes in pregnancy. Diabetes Care. 2015;38(suppl):S77-S79
  • 33 G. Maso, S. Alberico, U. Wiesenfeld, et al. GINEXMAL RCT: induction of labour versus expectant management in gestational diabetes pregnancies. BMC Pregnancy Childbirth. 2011;11:31 Crossref
  • 34 S.J. Stock, E. Ferguson, A. Duffy, I. Ford, J. Chalmers, J.E. Norman. Outcomes of elective induction of labour compared with expectant management: population based study. BMJ. 2012;344:e2838 Crossref
  • 35 D.A. Ogunyemi, A. Fong, S. Rad, S. Fong, S.L. Kjos. Attitudes and practices of healthcare providers regarding gestational diabetes: results of a survey conducted at the 2010 meeting of the International Association of Diabetes in Pregnancy Study Group (IADPSG). Diabet Med. 2011;28:976-986 Crossref
  • 36 M.S. Kramer, R.W. Platt, S.W. Wen, et al. A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics. 2001;108:E35 Crossref

Footnotes

a Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Sunnybrook Health Sciences Centre, St. Michael’s Hospital, University of Toronto, Ontario, Canada

b Departments of Medicine and Obstetrics and Gynecology, St. Michael’s Hospital, University of Toronto, Ontario, Canada

c Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, St. Michael’s Hospital, University of Toronto, Ontario, Canada

d Better Outcomes Registry & Network Ontario, Children’s Hospital of Eastern Ontario, Ottawa, Ontario, Canada

e Division of Midwifery, Department of Obstetrics and Gynecology, McMaster University, Hamilton, Ontario, Canada

Corresponding author: Nir Melamed, MD, MSc.

The authors report no conflict of interest.

Cite this article as: Melamed N, Ray JG, Geary M, et al. Induction of labor before 40 weeks is associated with lower rate of cesarean delivery in women with gestational diabetes mellitus. Am J Obstet Gynecol 2016;214:364.e1-8.