Study population

We selected our study cohort from the database of the Consortium on Safe Labor, a study conducted by the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health. In brief, this was a retrospective, cross-sectional study involving deliveries from 2002 through 2008 from 12 clinical centers and 19 hospitals representing 9 American College of Obstetricians and Gynecologists (ACOG) districts. The population was then standardized by assigning a weight to each subject using ACOG district, maternal race/ethnicity, parity, and plurality based on 2004 national data. Institutional review board approval was obtained for this analysis.

The Figure presents the flow diagram for our cohort selection, as originally described in the paper by Gibson et al comparing expectantly managed term pregnancies with electively induced deliveries. Briefly, from the initial data set of all deliveries (n = 233,736), we limited the group to singleton term pregnancies of 37 to 42 weeks’ gestational age in vertex presentation (n = 155,848) in the Consortium on Safe Labor Dataset.

Flow diagram of cohort selection and distribution of subjects by week of pregnancy

The selection of our low-risk term cohort and the distribution of our final cohort into those electively induced or expectantly managed by week of pregnancy.

HIV , human immunodeficiency virus.

Gibson et al. Weekly incidence of hypertensive disorders in low-risk term pregnancies. Am J Obstet Gynecol 2016 .

To limit confounding, we excluded all those with a prior uterine scar or planned (elective) cesarean delivery (n = 136,014). We also excluded those with fetal anomalies, women who had more than 1 pregnancy within the data set (the first was retained), or chronic maternal conditions that may lead to indicated delivery, including diabetes mellitus, chronic hypertension, cardiovascular disease, placental previa, or human immunodeficiency virus–positive status. Finally, we removed the 2 centers with more than 5% missing data for hypertensive outcomes (n = 114,651). All women who developed a hypertensive complication of pregnancy were included in the final cohort of low-risk pregnancies in the expectant management group.

As previously described, we then identified women with elective induction of labors. A predefined variable indicated whether a women had an induction of labor and did not include those receiving only labor augmentation or with less than 2 vaginal examinations. Inductions categorized as elective had no other indications for induction provided with no obstetric, fetal, or maternal conditions complicating the pregnancy.

Study outcomes

Subjects were divided by week of gestational age at delivery ( Figure ). Given the limitations of pregnancy dating and the size of our data set, we chose to evaluate gestational age by week rather than day. Those with an induction of labor coded as elective were the cases of eIOL at each week of gestation. Those not electively induced and who delivered after that week of gestation were considered to be expectantly managed in that week of gestation.

The primary outcome was the frequency of developing a hypertensive complication of pregnancy. The database included codes for gestational hypertension, preeclampsia, and eclampsia. These were combined for a composite outcome of pregnancy-induced hypertensive disorders (PIH). The definitions were based on the contemporaneous ACOG guidelines (published in 2002), and the diagnosis was made based on the clinical judgment of the delivering provider. We assumed the diagnosis of a hypertensive disorder was made within the same week as the gestational age of delivery and that for women who developed hypertensive complications, they did not have that complication prior to the week of their delivery.

Secondary outcomes included mode of delivery and composites for maternal and neonatal comorbidities. Mode of delivery was categorized as a nonoperative vaginal delivery, an operative vaginal delivery (requiring vacuum or forceps assistance), or cesarean delivery. Maternal outcomes included bleeding (blood products, abruption, postpartum hemorrhage, uterine rupture, hysterectomy), maternal intensive care unit admission or death, infections (intrapartum fevers, chorioamnionitis, endomyometritis, wound separation), lacerations (third- or fourth-degree perineal, sulcal, or cervical), and shoulder dystocia.

Neonatal outcomes included a composite of major comorbidities (birth injuries, sepsis, pneumonia, intraventricular hemorrhage, aspiration, hypoxic ischemic encephalopathy, respiratory distress syndrome, seizures, oliguria, myocardial injury, ventilator use, continuous positive airway pressure use, transfusions, or surfactant use), a composite of respiratory morbidities (oxygen use, continuous positive airway pressure use, transient tachypnea of the newborn, or surfactant administration), or perinatal death (intrauterine fetal demise or neonatal demise).

Statistical analysis

The occurrence of the primary outcome was determined for those women expectantly managed in a particular week of gestation until the next week of gestation and compared with those with an eIOL in the same week (for example, those with an eIOL in at 37+0 to 37+6 weeks were compared with women expectantly managed through the 37th week and thus delivered at 38+0 weeks or later). The frequencies of maternal and neonatal outcomes were then compared between those with an eIOL and those expectantly managed who did or did not develop a hypertensive complication of pregnancy. A secondary analysis compared women electively induced each week with those who delivered within the same week (for example, the eIOL at 37+0 to 37+6 weeks were compared with women who delivered also within the same week).

Categorical variables were compared with χ ² or Fisher exact tests, whereas student t tests or an analysis of variance was used for continuous variables, as appropriate. To adjust for possible confounding factors, the primary outcome, a logistic regression was performed, controlling for maternal age, race/ethnicity, parity, body mass index (BMI) at delivery, insurance status, and type of hospital. Centers with more than 5% missing data for an outcome were removed from the analysis on that specific outcome. We performed all analyses using statistical software (SAS, version 9.4; SAS Institute Inc, Cary, NC).

Patients

Our low-risk patient population included 114,651 women. Table 1 presents the maternal demographic characteristics for the cohort, comparing those with an eIOL with those expectantly managed who did or did not develop a hypertensive disorder of pregnancy. Overall, the cohort tended to be nulliparous (49.0%), white (51.4%), and overweight or obese (84.5%), with the majority having private insurance (58.9%) and delivering at a teaching hospital (90.1%). Women who developed a hypertensive complication were significantly younger, more obese, and more likely to be nulliparous when compared with either women with an eIOL or those expectantly managed who did not develop a hypertensive complication.

Study outcomes

The occurrence of hypertensive disorders for those expectantly managed by week of gestation is presented in Tables 2 and 3 . Overall, 5191 women expectantly managed (5.1% of the entire expectantly managed cohort) developed a hypertensive disorder of pregnancy. By week, the frequency of developing any pregnancy-related hypertension in women expectantly managed until at least the next week of gestation ( Table 2 ) was 4.1% at 37 weeks, 3.5% at 38 weeks, 3.2% at 39 weeks, and 2.5% at 40 weeks.

When evaluating the cases expectantly managed but who ultimately delivered within the same week ( Table 3 ), the frequency of hypertensive complications were as follows: 9.2% in the 37th week, 6.4% in the 38th week, 4.7% in the 39th week, and 3.9% in the 40th week. Table 4 presents the specific type of pregnancy-related hypertensive disorder diagnosed for those expectantly managed. Preeclampsia was the most common diagnosis, whereas eclampsia was the most infrequent. This was true for subjects expectantly managed until the next week of gestation and for those expectantly managed but delivering within the same week of gestation.

We next evaluated whether the development of a hypertensive disorder of pregnancy affected the mode of delivery ( Tables 5 and 6 ). For those expectantly managed until the next week of gestation who developed a hypertensive complication, cesarean delivery occurred more often when compared with either the eIOL group or those who were expectantly managed but without development of a hypertensive disorder of pregnancy. eIOL consistently was associated with the lowest observed frequency of cesarean delivery at each week of gestation when compared with either the expectant management group ( P < .001 for each week).

Additionally, the frequency of operative vaginal delivery was also higher for those women expectantly managed who developed PIH at 37, 38, and 39 weeks. Interestingly, eIOL was associated with a decreased risk of operative delivery at 37, 38, and 39 weeks of gestation when compared with the expectantly managed subjects who did not develop a pregnancy-associated hypertensive complication.

In a logistic regression adjusted for confounding variables (maternal age, race/ethnicity, parity, BMI at delivery, insurance status, and type of hospital), the risk of cesarean delivery remained significantly lower at each week of gestation for women with a eIOL when compared with either women who were expectantly managed who developed PIH or when compared with those expectantly managed without the development of PIH. Women who were expectantly managed but without the development of PIH also had a lower adjusted odds of cesarean delivery at each week of gestation when compared with those expectantly managed but who developed PIH. These trends were similar for women expectantly managed but delivering within the same week of pregnancy ( Supplemental Table ).

Table 7 presents the secondary outcome of maternal and neonatal composite morbidities. For only those women who were expectantly managed, the development of PIH was associated with a significantly increased frequency of maternal intensive care unit (ICU) admissions or death at 37 (0.3% vs 0.8%, P < .001), 38 (0.3% vs 0.8%, P < .001), and 39 weeks (0.3% vs 1.2%, P < .001) and maternal infections at all weeks (8.8% vs 12.5%, P < .001 at 37 weeks; 9.3% vs 13.9%, P < .001 at 38 weeks; 11.3% vs 16.5%, P < .001 at 39 weeks; 13.3% vs 23.1%, P < .001 at 40 weeks).

Table 7

Maternal morbidities by week in those with a eIOL, expectantly managed without PIH, or developing PIH after the week of eIOL

Week	Maternal bleeding complications
	Referent		No PIH		PIH		P Value, referent vs no PIH	P Value, referent vs PIH	P Value, no PIH vs PIH
	n	%	n	%	n	%
37	10/197	5.1	5649/78,657	7.2	316/3694	8.6	.140	.040 a	.002 a
38	63/1190	5.3	4331/60,562	7.2	189/2426	7.8	.002 a	< .001 a	.230
39	716/6969	10.3	1927/29,591	6.5	83/1094	7.6	< .001 a	.030 a	.160
40	248/3140	7.9	445/7570	5.9	17/222	7.7	.180	.500 a	.270

Week	Maternal intensive care admission or death
	Referent		no PIH		PIH		P value, referent vs no PIH	P value, referent vs PIH	P value, no PIH vs PIH
	n	%	n	%	n	%
37	1/214	0.5	204/79,702	0.3	22/2945	0.8	.43	1.00	< .001 a
38	1/1283	0.1	154/61,222	0.3	15/1792	0.8	.39	.003 a	< .001 a
39	9/7411	0.1	81/29,601	0.3	9/737	1.2	.02 a	< .001 a	< .001 a
40	5/3706	0.1	23/7336	0.3	1/119	0.8	.07	.170	.320

Week	Maternal infections
	Referent		No PIH		PIH		P value, referent vs no PIH	P value, referent vs PIH	P value, no PIH vs PIH
	n	%	n	%	n	%
37	5/209	2.4	7644/86,931	8.8	488/3920	12.5	< .001 a	< .001 a	< .001 a
38	28/1265	2.2	6267/67,260	9.3	357/2566	13.9	< .001 a	< .001 a	< .001 a
39	144/7404	1.9	3750/33,258	11.3	192/1165	16.5	< .001 a	< .001 a	< .001 a
40	184/3679	5.0	1113/8398	13.3	53/229	23.1	< .001 a	< .001 a	< .001 a

Week	Third- or fourth-degree perineal or sulcal lacerations
	Referent		No PIH		PIH		P value, referent vs no PIH	P value, referent vs PIH	P value, no PIH vs PIH
	n	%	n	%	n	%
37	9/217	4.2	6440/99,598	6.5	332/4213	7.9	.170	.040 a	< .001 a
38	56/1296	4.3	5187/77,181	6.7	225/2767	8.1	< .001 a	< .001 a	.004 a
39	285/7437	3.8	2899/39,024	7.4	103/1269	8.1	< .001 a	< .001 a	.360
40	265/3822	6.9	767/10,181	7.5	12/258	4.7	.230	.160	.080

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Association of cervical precancer with human papillomavirus types other than 16 among HIV co-infected women
2. Neonatal and maternal outcome after blastocyst transfer: a population-based registry study
3. Group B streptococcus alters properties of vaginal epithelial cells in pregnant women
4. Prevalence of anal human papillomavirus infection and anal human papillomavirus–related disorders in women: a systematic review

Stay updated, free articles. Join our Telegram channel

Join