| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Division of Maternal–Fetal Medicine, Department of Obstetrics and Gynecology, Dalhousie University, Halifax, Nova Scotia, Canada.
Address reprint requests to: Tina Delaney, Department of Obstetrics and Gynecology, 300 Prince Philip Drive, St. John’s, Newfoundland, Canada A1B 3V6; E-mail: martinadelaney{at}hotmail.com.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: Women with one previous cesarean delivery who had spontaneous labor between January 1992 and January 2000 were compared with those whose labor was induced.
RESULTS: Three thousand seven hundred forty-six patients had a trial of labor (2943 spontaneous, 803 induced). Those induced had more frequent early postpartum hemorrhage (7.3% versus 5.0%; odds ratio [OR] 1.66; 95% confidence interval [CI] 1.18, 2.32), cesarean delivery (37.5% versus 24.2%; OR 1.84; 95% CI 1.51, 2.25), and neonatal intensive care unit (NICU) admission (13.3% versus 9.4%; OR 1.69; 95% CI 1.25, 2.29). There was a trend toward higher uterine rupture rates in those with induced versus spontaneous labor (0.7% versus 0.3%, P = .128) and for patients undergoing dinoprostone (prostaglandin E2) induction versus other methods (1.1% versus 0.6%, P = .62), although neither difference achieved statistical significance.
CONCLUSION: Induced labor is associated with an increased rate of early postpartum hemorrhage, cesarean delivery, and neonatal ICU admission. The higher rate of uterine rupture in those who had labor induced was not statistically significant.
Most women with one previous lower segment cesarean delivery are considered candidates for a trial of labor and should be counseled about and offered this option.1, 2 However, if labor does not begin spontaneously and delivery is indicated, the patient requires counseling regarding labor induction in the setting of a previous cesarean delivery. Lydon-Rochelle et al assessed the risk of uterine rupture in patients with one previous cesarean delivery for those who had spontaneous labor and for those who had labor induced with and without prostaglandins (PGs).3 They found a higher risk of uterine rupture in those whose labor was induced, with the highest risk being in those induced with PGs. This was a population-based, retrospective, cohort analysis. Zelop et al, in a retrospective analysis,4 compared spontaneously laboring patients with those whose labor was induced with dinoprostone (PGE2) gel or oxytocin after a previous cesarean delivery for the rate of uterine rupture. They found an increase in the rate of uterine rupture in induced labor relative to spontaneous labor. The rate of rupture was also increased in those whose labor was induced with oxytocin relative to those laboring spontaneously, but not in those induced with PGE2or those augmented with oxytocin.
The purpose of our study was to evaluate maternal and neonatal mortality and morbidity of patients with one previous cesarean delivery undergoing spontaneous versus induced labor. We also examined these outcomes for those undergoing induced labor with dinoprostone (PGE2) versus oxytocin, artificial rupture of membranes, or intracervical mechanical methods.
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our final cohort consisted of 3746 women who underwent a trial of labor. Women were excluded from our analysis for the following reasons: multiple gestation, placenta previa, previous uterine surgery, herpes simplex at time of delivery, nonvertex presentation, and more than one previous cesarean delivery. The database does not allow the linkage of deliveries, and therefore the type of previous uterine incision is not definitely known. Based on practice patterns in the province, only those patients with prior lower transverse uterine incisions would be offered a trial of labor, spontaneous or induced. The indication for the previous cesarean delivery is also unknown.
The 2943 women who had spontaneous labor were compared with the 803 women who had induced labor. The 179 patients induced with PGE2 were compared with the 624 women with labor induced by other means (oxytocin, 600; intracervical laminara tent or catheter, 22; and artificial membrane rupture only, two). The groups were compared with respect to maternal and neonatal characteristics, morbidities, and mortality. Antenatal demographics compared were gravidity, parity, maternal age, previous vaginal delivery, attendance at prenatal classes, antepartum complications (chronic hypertension, eclampsia, pregnancy-induced hypertension, fetal growth concerns, oligohydramnios), antepartum fetal death, and major fetal anomaly. Intra- and postpartum morbidities examined were intrapartum nonreassuring fetal status; mode of delivery; first, second, third, and fourth degree tears; inverted uterus; retained placenta; manual removal of placenta; ruptured uterus; uterine scar defects; hysterectomy; early and late post-partum hemorrhage; postpartum anemia; transfusion; postpartum infectious complications; and postpartum thromboembolism. Intrapartum nonreassuring fetal status was defined as prolonged tachycardia, bradycardia, or decelerations requiring medical intervention such as vacuum, cesarean delivery, or forceps; scalp pH less than 7.0; or as stated in the chart by the physician. Ruptured uterus was a significant and serious event necessitating emergency surgery. Uterine scar defect was described as a nonserious situation often noted at the time of cesarean delivery as an unexpected finding. Data are not available on the number of scars examined manually after a vaginal delivery in an attempt to identify uterine scar defects and therefore are not included in the analysis. Early postpartum hemorrhage was hemorrhage within the first 24 hours postpartum as noted by the physician or if estimated blood loss was greater than 500 mL. Late postpartum hemorrhage occurred more than 24 hours postpartum, as noted by the attending physician. Post-partum anemia was defined as a hemoglobin level of less than 100 g/L. Postpartum infectious morbidity included endometritis, mastitis, urinary tract infection, wound infection, septicemia, peritonitis, pulmonary infection, or puerperal morbidity (temperature of 38C or more on two occasions at least 4 hours apart in any 48-hour period exclusive of the first 24 hours). Pyrexia of unknown cause was defined as one or more episodes of temperature of 38C or more. Neonatal morbidities examined were meconium staining of amniotic fluid, neonatal intensive care unit (NICU) admission, respiratory distress syndrome, low cord arterial pH values (less than 7.0), low Apgar scores (3 or less at 5 minutes), and clinical conditions related to asphyxia.6
SPSS 9 (SPSS Inc., Chicago, IL) was used for statistical analysis. Categoric data were compared using 
2 or Fisher exact tests as appropriate. Continuous data were analyzed using the Student t test. Logistic regression was used to control for covariates. P values less than .05 were considered significant.
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Data on attendance at prenatal class was available in 84% of cases. The antepartum fetal deaths all occurred before labor, and those induced were done so for that reason. The clinical gestational ages of these fetuses ranged from 24 to 41 weeks. One pregnancy in this group had been complicated by an abruption. There were two anomalies in the antepartum fetal demise group, a chromosomal abnormality and a congenital heart defect. The indications for induction are presented in Table 2. Intrapartum outcomes are summarized in Table 3. A chart review of all of the uterine ruptures revealed that the previous cesarean delivery was a lower segment transverse incision in all cases. The indications for delivery by cesarean are recorded in Table 4. There was no statistical difference in first, second, third, or fourth degree tears; inverted uterus; manual removal of the placenta; or retained placenta. There were three intrapartum fetal deaths. All of these were in the spontaneous labor group. The gestational age based on last menstrual period was available for two of these fetuses: 40 weeks and 45 weeks. The clinical estimate of the gestational age of the other fetus was 37 weeks. There were no congenital anomalies in this group of fetuses. Birth weights were 3661, 3195, and 3680 g. These deaths were not associated with uterine rupture or scar dehiscence. There were no maternal deaths. Postpartum outcomes are summarized in Table 5. There were no cases of pulmonary embolism. Neonatal outcomes are summarized in Table 6.
|
|
|
|
|
|
In univariate analysis the outcomes of nonreassuring fetal status in labor, cesarean delivery, early postpartum hemorrhage, postpartum anemia, and admission to the NICU were significantly increased for patients undergoing induction of labor versus those laboring spontaneously. These outcomes were analyzed in logistic regression models to control for possible cofounders. Covariates that were felt to be clinically important were forced into the model and stepwise variable selection was not employed. Table 7 represents the five regression models. In this analysis early postpartum hemorrhage, cesarean delivery, and admission to the NICU remained statistically increased in those patients with induced labor. The odds ratios (ORs) and 95% confidence intervals (CIs) are presented in Table 7.
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A recent review by Sanchez-Ramos et al7 regarding cervical ripening and labor induction after previous cesarean delivery concluded that the use of PGE2 and oxytocin for these purposes is safe in women who are vaginal birth after cesarean delivery (VBAC) candidates. They reviewed ten studies that compared the rate of uterine scar disruption in patients with a previous cesarean delivery who received PGE2 versus those who entered labor spontaneously. In their combined analysis of these studies, a total of 1682 patients received PGE2 and 11,097 had spontaneous labor. Rates of uterine scar disruption were 1.6% and 1.23%, respectively. The pooled OR for uterine scar disruption was 1.46 (95% CI 0.96, 2.22), not statistically significant. They also compared the incidence of uterine scar disruption in women with a previous cesarean delivery who underwent induction with oxytocin versus spontaneous labor. Seven studies were included in this analysis. Rates of uterine scar disruption were 0.83% and 0.62%, respectively. The pooled OR was 1.43 (95% CI 0.76, 2.69), not statistically significant. These results concur with ours in that the rate of uterine rupture is not statistically significantly increased in those induced with oxytocin or PGE2 versus patients who spontaneously labor. This is a reassuring finding given the fact that they had a larger sample size. Uterine rupture, however, is a rare event. To see a 50% increase in the rate of uterine rupture from their quoted rate of 1.23% with spontaneous labor to 2.46% with PGE2 at an 
of .05 and ß of .2, 1878 patients would be required in each group. Even with the pooled analysis this number was not reached. To see a smaller difference, which also would be clinically significant, would require even more patients.
Zelop et al4 reviewed cases of 2214 women who had spontaneous labor and 560 women who had labor induced with oxytocin or PGE2 gel. This study was not included in Sanchez-Ramos’ review.7 The rate of rupture was significantly increased with induction (2.3% versus 0.7% with spontaneous labor) (P = .001). In a logistic regression model, compared with no oxytocin use, induction with oxytocin was associated with a 4.6-fold increase in risk of uterine rupture (95% CI 1.5, 14.1). Induction with PGE2 gel was associated with an OR of 3.2 (95% CI 0.9, 10.9) for uterine rupture, and augmentation with oxytocin was associated with an OR of 2.3 (95% CI 0.8, 7.0) for uterine rupture. Although these latter two ORs were not statistically significant, the CI around the use of PGE2 gel is close to one. The lack of statistical significance may be related to the small numbers in this group.
Lydon-Rochelle et al3 reported on subsequent uterine rupture rates in patients with one previous lower segment cesarean delivery. There were 6980 patients with repeated cesarean delivery without labor, 10,789 with spontaneous onset of labor, 1960 with induction without PGs, and 366 with induction with PGs. The risk of uterine rupture was statistically significantly increased with spontaneous labor and induced labor with and without PGs. Not surprisingly, the rate of postpartum complications was increased in those who had a uterine rupture. This study did not review neonatal complications related to uterine rupture or labor induction and was not included in Sanchez-Ramos’ review.7
Also not included in Sanchez-Ramos’ review is a retrospective cohort study by Ravasia et al that determined and compared the rates of uterine rupture and vaginal delivery among women with previous cesarean delivery whose labors were induced and those who had spontaneous labor.8 This study included women with more than one previous cesarean delivery. Five hundred seventy-five women were induced and 1544 entered labor spontaneously. One hundred seventy-two patients received PGE2, 129 had intracervical Foley, 26 had amniotomy alone, 214 had oxytocin alone, and 34 had both oxytocin and amniotomy. Rates of rupture were 1.4% in those induced versus 0.45% in those with spontaneous labor (P < .001). When induction with PG was excluded, rupture rates were 0.74% for inductions versus 0.45% for spontaneous labor, which was not statistically significant (P = .38). The rate of vaginal delivery was lower in the groups who required cervical ripening with PG or catheter than in those not requiring ripening or those who entered labor spontaneously. These authors did not report neonatal outcome or other maternal morbidities.
To properly address the risk of uterine rupture after a previous cesarean delivery using different methods of labor induction, a randomized trial with an adequate sample size is necessary. There has been one randomized trial on the use of PGE2 versus intravenous oxytocin for induction of labor in women with previous cesarean delivery.9 Only 42 women were randomized. There was one case of uterine rupture. This occurred in the PGE2 group after oxytocin augmentation. This study did not have adequate sample size to address the outcome of uterine rupture.
In conclusion, our population-based study addressed the risk of uterine rupture with labor induction but also other maternal and neonatal morbidities that need to be considered in this setting. There was an increase in early postpartum hemorrhage, repeat cesarean delivery, and neonatal intensive care admission for those with labor induced. Our study did not show a statistically significant increase in the rate of uterine rupture with induction with oxytocin and other methods or PGE2. The possibility that this reflects a type II error (ie, insufficient statistical power) cannot be excluded. A trend towards a higher rate of rupture was observed in those induced relative to those who labored spontaneously and in those who received PGE2 relative to those with labor induced by oxytocin and the other methods. The former observation is in keeping with the results of Lydon-Rochelle et al,3 Zelop et al,4 and Ravasia et al.8 The American College of Obstetricians and Gynecologists practice bulletin on VBAC in 19991 recommended that the use of oxytocin or PG gel for VBAC requires close patient monitoring. This was a level B recommendation based on limited or inconsistent scientific evidence. We acknowledge that our study, like any retrospective, non-randomized study, is subject to bias. Despite this, there is unlikely to be a randomized controlled trial in this area, and physicians must rely on less vigorous evidence to counsel patients, including retrospective reviews such as this one and the references discussed here.
|
Footnotes |
|---|
doi:10.1016/S0029-7844(03)00355-7
Received September 17, 2002. Received in revised form December 2, 2002. Accepted January 8, 2003.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Maternal–Fetal Medicine Committee, Society of Obstetricians and Gynecologists of Canada. Policy statement: Vaginal birth after previous cesarean birth. SOGC Clin Pract Guidelines 1997;68:1–4.
3. Lydon-Rochelle M, Holt VL, Easterling TR, Martin DP. Risk of uterine rupture during labor among women with a prior cesarean delivery. N Engl J Med 2001;345:3–8.
4. Zelop CM, Shipp TD, Repke JT, Cohen A, Caughey AB, Lieberman E. Uterine rupture during induced or augmented labor in gravid women with one prior cesarean delivery. Am J Obstet Gynecol 1999;181:882–6.[Medline]
5. Fair M, Cyr M, Allen AC, Wen SW, Guyon G, Macdonald RC. Validation study for a record linkage of births and infant deaths in Canada. Statistics Canada catalogue no. 84F0013XIE. Ottawa: Statistics Canada, 1999.
6. American College of Obstetricians and Gynecologists. Fetal and neonatal neurologic injury. ACOG technical bulletin no. 163. Washington: American College of Obstetricians and Gynecologists, 1992.
7. Sanchez-Ramos L, Gaudier FL, Kaunitz AM. Cervical ripening and labor induction after previous cesarean delivery. Clin Obstet Gynecol 2000;43:513–23.[Medline]
8. Ravasia DJ, Wood SL, Pollard JK. Uterine rupture during induced trial of labor among women with previous cesarean delivery. Am J Obstet Gynecol 2000;183:1176–9.[Medline]
9. Taylor AVG, Sellars S, Ah-Moye M, MacKenzie IZ. A prospective random allocation trial to compare vaginal prostaglandin E2 with intravenous oxytocin for labor induction in women previously delivered by cesarean section. J Obstet Gynecol 1993;13:333–6.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
