CESAREAN DELIVERY
The highest CS rates in our study were observed in nulliparous women with prolonged and post-term pregnancies (I–II, V), which is in line with previous reports (262, 263). In our study I, the rate of CS in induced labor at or beyond 41+5 weeks was sixfold compared to spontaneous labor. In contrast, IOL in multiparous women more often resulted in vaginal delivery (I, IV).
Previously IOL, regardless of the method, has been associated with a three- to eightfold increase in CS rate compared to spontaneous labor at or beyond term (261), but more recent studies have demonstrated a reduction in CS following IOL at term compared to expectant management (27-31). In our study I, the groups were different with a selection bias of more nulliparous and post-term women ending up in the induction group. Additionally, we focused on a narrow time frame between 41+5 and 42+1 gestational weeks.
Thus, our results (I) are inapplicable for further conclusions on timing of IOL with regard to CS rates.
Our results show no difference in the CS rates between FC and misoprostol induced labors (III), as also previously demonstrated (19, 20, 173).
Interestingly, a lower CS rate of 4.5% with no increase in infections has been observed with an expectant management policy of 48 hours after PROM (97).
In our study, the overall rate of CS was lower in IOL following PROM (III) compared to other indications (24% vs. 33%; p<0.001) (III, IV−V). In our study, PROM appears to have a positive impact on the progress of induced labor in women with an unfavorable cervix. It may be contemplated that
perhaps PROM occurred as a result of a biochemical activation process, which led to the process of cervical ripening and rupture of membranes, thus resulting also in more successful labor induction in these women. The rate of CS was not significantly different between inpatients and outpatients (IV), which is in accordance with previous studies (24, 26, 288, 310).
MATERNAL INFECTIOUS MORBIDITY
The median (range) maternal infection rates of intrapartum infections 5(2−6)% and postpartum infections 3(1−4)% in our studies I−V are consistent with the previously shown 7% rate of chorionamnionitis, and 3.5%
rate of postpartum endometritis following labor induction (16). The rate of maternal infections is higher in induced compared to spontaneous labor (0.2−1.5%) (313). An older systematic review has linked FC with an increased risk of infections (174), but in more recent studies FC has not been associated with increased infectious morbidity compared to PGs (16, 18, 19, 165, 173).
The rates of maternal infections were not different between the groups of FC and misoprostol (III), as also previously reported (16). Furthermore, outpatients and inpatients undergoing IOL by FC had similar rates of infectious maternal morbidity (IV), which is in accordance with previous studies (24, 26, 310).
The use of FC in women with ruptured amniotic membranes has raised concerns over infectious morbidity. In our study III, however, the maternal infection rates were the lowest following IOL by FC in cases of term PROM.
On the other hand, prophylactic antibiotics were used for all women in our study III, which may have led to a reduced rate of infections. Another factor related to low infectious morbidity may have been the shorter duration of induced labor following PROM compared to IOL in non-PROM cases (22 h vs. 30 h) (II, III). Similar reassuring results on IOL by FC after PROM at term have been previously reported by a small Swedish pilot study (n=18) and a retrospective cohort study (n=122) (314, 315).
The highest rate of maternal infections was seen in nulliparous women (II).
Unfortunately, the placental histopathological diagnosis of chorionamnionitis was not available in all cases. Duration of labor, duration of ruptured membranes, use of internal fetal monitoring, presence of meconium, number of vaginal examinations, use of prophylactic antibiotics, and GBS screening protocols are factors associated with rates of infectious morbidity (313). In our study, gestational diabetes was associated with maternal infections (II). Similar results were reported by a Danish study (n=2492), in which gestational and pregestational type 2 diabetes were associated with an increased risk of postpartum infections (316). Maternal
(III), as also noted previously (317). Increased risk of chorionamnionitis has been demonstrated in nulliparous women remaining in the latent phase for more than 12 hours (284).
NEONATAL INFECTIOUS MORBIDITY
In our studies I−IV, the median (range) rate of neonatal clinical sepsis following IOL was 1.8(1–3) %, which parallels the results of previous studies (16). No blood culture positive neonatal sepsis cases occurred during our studies I−V, whereas a sepsis incidence of approximately 1 per 1000 live births has been previously reported (318). The rates of neonatal infections are known to vary by geographic region, resource use, and management practices, such as GBS screening or use of prophylactic antibiotics.
The neonatal infection rates were similar in the groups of FC and misoprostol IOL (III), as also reported by a recent review and meta-analysis (16).
Consistent with previous studies, outpatient cervical ripening by FC did not result in increased neonatal infectious morbidity compared to inpatient cervical ripening (IV) (24, 26, 310). Furthermore, the neonatal infection rate was low (1%) in women undergoing IOL by FC following term PROM ≥ 18 hours. Our results add to the limited data on the use of FC in women with term PROM, and suggest that with regards to neonatal infectious morbidity, FC seems a safe and feasible method of IOL.
In nulliparous women, the incidence of neonatal infections was higher in induced compared to spontaneous labor (I). However, the rates of clinical neonatal sepsis were similar between these groups, whereas the rate of suspected neonatal infections was higher in induced labors (I). We speculate that the use of FC, a relatively new method at the time of the study, may have resulted in a lower threshold for starting antibiotics in cases of prolonged induction to delivery intervals, and to more neonates being admitted for observation. Unfortunately, further comparisons between spontaneous and induced labors are limited, since the rates of suspected neonatal infections are not included in most studies, and data on the rates of blood culture negative neonatal infections following spontaneous labor are not available.
The majority of neonates diagnosed with infection in our studies I and II were delivered by CS. Neonatal infection may have occurred in the absence of maternal infection, and maternal infection did not always lead to neonatal infection, although maternal chorionamnionitis may increase the risk of neonatal infection by 1–4% (319). Early epidural analgesia was associated with an increased risk of neonatal infection in our study II, but this may rather be related to prolonged latent phase of induced labor, which is a
OTHER DELIVERY OUTCOMES
Maternal and neonatal outcomes were similar between induced or spontaneous labor in prolonged or post-term pregnancy (I), IOL by FC or misoprostol following PROM at term (III), and between inpatients and outpatients (IV). IOL itself has been associated with increased admissions to neonatal unit (93, 284), but this was not seen in our studies I–V. Moreover, admissions to NICU and to neonatal unit were similar between induced and spontaneous labor, between FC and misoprostol (III), and between outpatients and inpatients (IV), as also shown in previous studies (18, 310).
In accordance with previous studies, most women were satisfied with outpatient IOL, and found contacting the delivery unit feasible (220, 288, 289).
A retrospective cohort study by Mackeen et al. reported shorter induction to delivery interval (12.3 h vs. 22.6 h; p<0.01) in women undergoing IOL by FC compared to misoprostol following PROM at term (314). In our study III, no difference in the induction to delivery interval was found between these two methods. The induction to delivery interval was longer in outpatients than in inpatients (IV), as also previously noted (24). This may be explained by the fact that amniotomy was in outpatients often delayed to the next morning in case of FC expulsion at night. However, the durations of the first and second stages of labor were similar in outpatients and inpatients (IV).