Material and Methods
The protocol was instituted on February 1, 1996 at the Maternal-Fetal Medicine service of the University of Alabama at Birmingham Hospital, which has approximately 3000 deliveries annually. Deliveries on this service are performed by resident physicians under the direct 24-hour supervision of one of 11 members (faculty and fellows) of the Maternal-Fetal Medicine Division.
All members of the Maternal-Fetal Medicine Division agreed to manage eligible women according to protocol. Before institution of the protocol, the resident staff received in-service education. Women were eligible for the protocol if they were at or beyond 36 weeks' gestation and met two criteria: 1) spontaneous active-phase labor, defined as a cervical dilatation of at least 4 cm and spontaneous, regular uterine contractions (at least two in 10 minutes); and 2) labor arrest, defined as 1 cm or less of cervical progress in 2 hours. We allowed 1 cm of cervical progress over 2 hours to qualify as arrest because during the period of this study, a diagnosis of labor arrest was required for the administration of oxytocin. Because several physicians shared responsibilities for the same woman, we wanted to minimize the possibility that interexaminer variation would result in prolonged periods of no actual labor progress before the administration of oxytocin. Exclusion criteria included a nonvertex presentation, previous cesarean delivery, multiple gestation, and, at the time of labor arrest, a nonreassuring fetal heart rate tracing, chorioamnionitis, or spontaneous uterine contractions equal to or exceeding 250 Montevideo units.
Upon the diagnosis of active-phase labor arrest, an internal uterine pressure catheter was placed (after amniotomy if necessary). Dilute intravenous oxytocin was initiated at one mμ/minute and increased every 15 minutes over 2 hours to 30 mμ/minute, or until either a uterine contraction pattern of greater than 200 Montevideo units was achieved or labor progressed. Before a cesarean delivery for labor arrest was performed, the protocol mandated at least 4 hours of oxytocin augmentation with a sustained uterine contraction pattern of greater than 200 Montevideo units. However, because it is not always possible to achieve a sustained uterine contraction pattern of greater than 200 Montevideo units, cesarean delivery for labor arrest was permitted after 6 hours of oxytocin augmentation regardless of the uterine contraction pattern. For those patients who, on initial assessment, had greater than 200 but less than 250 Montevideo units, oxytocin was also begun and titrated with an intent to achieve at least 250 Montevideo units.
To ensure complete ascertainment of protocol-eligible patients, oxytocin orders were available only in sequentially numbered envelopes on the labor and delivery suite. On a daily basis, a research nurse determined for which patients oxytocin had been ordered and identified protocol-eligible patients. The charts of eligible patients and their infants were then abstracted for selected demographic and clinical variables. Specific maternal complications assessed included clinically diagnosed chorioamnionitis, postpartum hemorrhage, postpartum endometritis, need for blood transfusion, and abdominal wound infection. Intrapartum chorioamnionitis was diagnosed with a temperature of at least 37.8C and at least one of the following supporting symptoms or signs: uterine tenderness, maternal or fetal tachycardia (greater than 100 and greater than 160 beats per minute, respectively), or purulent amniotic fluid or cervical discharge. The diagnosis of intrapartum chorioamnionitis precluded the diagnosis of post-partum endometritis, which was defined as a temperature of at least 38.0C in the postpartum period and at least one of the following symptoms or signs: uterine tenderness, maternal tachycardia (greater than 100 beats per minute), or purulent cervical discharge. Throughout the study, intrapartum antibiotics were administered to patients with risk factors for early-onset neonatal group B streptococcal disease.4
Neonatal outcomes and complications assessed included 1- and 5-minute Apgar scores, umbilical cord blood gas indices including fetal acidemia (pH less than 7.0), confirmed sepsis (by blood or cerebrospinal fluid culture), need for antibiotics, supplemental oxygen requirement outside of the delivery room, use of mechanical ventilation, phototherapy for hyperbilirubinemia, and a diagnosis of pneumonia, necrotizing enterocolitis, intraventricular hemorrhage, seizures, or death.
The primary outcomes for this investigation were the overall vaginal delivery rate and protocol safety (rates of maternal and fetal or neonatal complications). Vaginal delivery rates and safety were analyzed both by parity and by whether labor had progressed (defined as more than 1 cm of cervical dilatation) or delivery had occurred by both 2 and 4 hours after the initiation of oxytocin augmentation. Vaginal delivery rates were analyzed further for the subset of patients with no labor progress despite 2 hours of uterine activity greater than 200 Montevideo units. For the purposes of analysis, a patient was considered to have had a “2-hour cervical examination” if a cervical examination was documented in the chart from 91 to 179 minutes after the initiation of oxytocin, and a “4-hour cervical examination” if the examination was performed between 211 and 299 minutes after the initiation of oxytocin. Patients who did not have a cervical examination in the 2- or 4-hour examination window were also analyzed, and their outcomes were reported for comparison.
Using Fisher exact test, we compared complication rates among three groups: the group that progressed, the group that did not progress, and the group that was not examined 2 or 4 hours after the initiation of oxytocin augmentation. Data were analyzed using the SAS system 6.12 for personal computers (SAS Institute, Cary, NC). P < .05 was considered significant. We established a target sample size of at least 500 women. A cohort of this size yields high precision (narrow confidence intervals) around observed event rates. For example, the upper limit of the 99% confidence interval for an event with an observed rate of 1% is less than 2%.
From February 1, 1996 through April 6, 1998, 554 protocol-eligible women experienced active-phase labor arrest. Twelve (2%) of these women did not receive oxytocin because of spontaneous progress of labor before initiation of the drug. These women all delivered vaginally, and because they did not receive oxytocin, they were not considered further in the analysis of this protocol. Thus, 542 women were eligible for the protocol and received oxytocin.
Of the women managed by the protocol, 288 (53%) were nulliparous and 254 (47%) were parous. Parous and nulliparous women were similar demographically: The mean age was 23 and 20 years, respectively; mean weight was 81 and 79 kg; 68% and 74% were black; and 27% and 24% were white. The mean gestational age at delivery was 40 weeks for both groups; the mean birth weight was 3427 and 3301 g, respectively; and 89% and 96% received lumbar epidural analgesia for labor.
The vaginal delivery rate for the cohort of protocol-managed women was 92%: 97% for parous women and 88% for nulliparas. The median time from the diagnosis of active-phase labor arrest to the initiation of oxytocin was 15 minutes. Analysis by whether labor had progressed (including delivery) by 2 and 4 hours after the initiation of oxytocin demonstrated that the majority of women with no progress despite 2 or even 4 hours of oxytocin augmentation eventually underwent a vaginal delivery (Tables 1 and 2). Parous women with no progress despite 2 hours of oxytocin augmentation had a vaginal delivery rate of 91%, whereas nulliparas had a vaginal delivery rate of 74%. Even with no labor progress 4 hours after the initiation of oxytocin, 88% of parous women and 56% of nulliparas eventually delivered vaginally.
No patients experienced uterine rupture or underwent hysterectomy. Rates of chorioamnionitis and endometritis for nulliparas were 10% and 5%, respectively, and both were 2% for parous women. Labor progress (and lack thereof) correlated with the risk of maternal infectious complications: Rates of chorioamnionitis and endometritis were increased among patients who had not progressed (or were not examined) by 2 and 4 hours after the initiation of oxytocin (Tables 1 and 2). Four percent of the patients experienced postpartum hemorrhage and 1% received a red blood cell transfusion. The risk of red blood cell transfusion was less consistently (and not significantly) related to labor progress than was the risk of maternal infection (Tables 1 and 2). One patient experienced a postcesarean abdominal wound infection.
In general, fetal and neonatal outcomes were excellent. There were no stillbirths or neonatal deaths. Two infants, one with congenital diaphragmatic hernia and the other with multiple severe congenital malformations, died after protracted courses of disease. With the exception of the latter infant, the lowest 5-minute Apgar score was 6. Two infants had an umbilical artery pH of less than 7.0. This acidemia was respiratory in nature, and both infants were vigorous at birth. One infant, born to a parous woman who had progressed after 2 hours of oxytocin, had a positive blood culture, received antibiotics, and was discharged without apparent sequelae on day 14 of life. Four infants were placed on mechanical ventilation: the two who died, one with Noonan syndrome and an associated cardiac malformation, and one with persistent fetal circulation. This last infant, born to a parous woman who did not have a 2-hour cervical examination, was on the ventilator for 48 hours and was discharged on day 10 of life without apparent sequelae. Six percent of the newborns received antibiotics, and none suffered necrotizing enterocolitis, intraventricular hemorrhage, or seizures. One percent had hyperbilirubinemia requiring phototherapy and 2% required supplemental oxygen (exclusive of mechanical ventilation) outside of the delivery room. In the infants of both parous and nulliparous patients, these complications did not differ significantly among the three groups (progress, no progress, and no examination groups) at either 2 or 4 hours after the initiation of oxytocin.
Because most women progressed after 2 hours of oxytocin, we identified only 52 women with no progress 2 hours after achieving a uterine contraction pattern of greater than 200 Montevideo units. The eventual vaginal delivery rate was 63% for the 40 nulliparas in this group and 58% for the 12 parous women. Although this group had a high rate of maternal infection (chorioamnionitis 27%, endometritis 12%), their other outcomes were essentially equivalent to those of the overall cohort of protocol-managed women, with no severe adverse neonatal outcomes, including sepsis.
Indications for the 42 cesarean deliveries performed included labor arrest (n = 24), nonreassuring fetal status (n = 10), and both labor arrest and nonreassuring fetal status (n = 8). Women who ultimately underwent cesarean for labor arrest either had no cervical progress after the diagnosis of labor arrest or made progress initially but had arrest again later. These women received oxytocin for a median duration of 7.1 hours (range 4.3–13.0). Before cesarean for labor arrest, 88% of the women achieved a uterine contraction pattern of greater than 200 Montevideo units.
We conducted the present investigation to evaluate a protocol that focused on three principal elements: 1) an intent to achieve a sustained uterine contraction pattern of greater than 200 Montevideo units; 2) a more liberal minimum of 4 hours (as opposed to the currently sanctioned 2 hours5) of oxytocin-augmented labor arrest with a sustained uterine contraction pattern of greater than 200 Montevideo units before proceeding to cesarean delivery for active-phase labor arrest; and 3) for patients who could not achieve a sustained uterine contraction pattern of greater than 200 Montevideo units, a minimum of 6 hours of oxytocin augmentation before proceeding to cesarean delivery for active-phase labor arrest. Thus, we were able to address what we perceived as two practical limitations of the current recommendations5 for active-phase labor arrest: 1) that a 2-hour minimum might deny some women the opportunity for a safe vaginal delivery, and 2) that some women are never able to achieve an augmented uterine contraction pattern of greater than 200 Montevideo units.
This protocol was effective, resulting in a high rate of vaginal delivery (92%), and safe, with no severe adverse maternal and fetal or neonatal outcomes. Although the risk of maternal infection was increased for those women who had not progressed in labor by 2 and 4 hours after the initiation of oxytocin, the majority (80%) of these women delivered vaginally. They therefore avoided the attendant surgical and increased infectious risks of cesarean delivery. Other adverse outcomes, both maternal and neonatal, were unrelated to labor progress 2 and 4 hours after the initiation of oxytocin. For example, the only neonate with a positive blood culture was born to a woman who had progressed 2 hours after the initiation of oxytocin. Thus, in terms of decreasing the cesarean delivery rate for those women who had not made progress in labor, the clinical benefit of extending the duration of oxytocin for active-phase labor arrest from 2 to at least 4 hours seemed to more than offset the observed rates of maternal and neonatal morbidity associated with (but potentially unrelated to) continued oxytocin augmentation.
Few studies have focused on the optimal duration of oxytocin administration for active-phase labor arrest. Although a strict definition of labor arrest, in terms of uterine activity and duration of arrest, has been a component of recent institutional programs to lower the cesarean birth rate, in general these programs have used the 2-hour definition of labor arrest.6,7 Thus, although these programs are successful, they provide no insight into the efficacy and safety of policies that require longer oxytocin administration in the face of continued labor arrest. Perhaps the most relevant investigation of this issue was performed by Arulkumaran et al.8 In a study of 319 women of mixed parity, they evaluated the effect of an additional 4-hour period of oxytocin augmentation in women with “unsatisfactory” progress in the active phase of labor 4 hours after the commencement of oxytocin (less than 1 cm/hour of cervical progress over the 4-hour period). Like us, they demonstrated that an additional 4 hours of oxytocin administration was beneficial by allowing 48 of the 92 women (42%) with unsatisfactory labor progress to achieve safe vaginal delivery.
The majority of women (93%) in this cohort received continuous lumbar epidural analgesia, and this probably had an impact on their labor course. Three randomized clinical trials have shown that epidural analgesia prolongs the duration of labor.9–11 In response to this observed effect, it was suggested recently12 that when epidural analgesia is used for pain relief during labor, physicians should consider modifying the guidelines for management of the first stage of labor, as has been done for the second stage.13 The results of our study support that suggestion.
We believe our findings may have important implications for reducing the high cesarean delivery rate in the United States. Although repeat cesarean is the leading indication for all cesareans performed in the United States, more than 50% of all primary cesareans are performed for dystocia. Because dystocia was the original indication for most repeat cesareans, it follows that the majority of cesareans in the United States are ultimately related to the diagnosis of dystocia.14 Moreover, the principal reason that the cesarean delivery rate in the United States is so much higher than in other developed countries is that cesarean for dystocia occurs at a much higher rate in this country. In 1990, 7.1 births per 100 in the United States were cesareans performed for dystocia. In turn, the United States had a high rate of repeat cesarean delivery—8.5 births per 100. Thus, as a result of primary cesareans for dystocia (and the downstream consequence of repeat cesareans), the United States has a much higher cesarean rate than other western countries such as Sweden and Scotland (Table 3).15 Clearly, reducing the number of cesareans performed for dystocia would dramatically lower our high national cesarean delivery rate.
The potential for reducing the number of cesarean deliveries by adopting a more stringent definition of dystocia (specifically, of active-phase labor arrest) is evident by considering the benefits of continued oxytocin augmentation in the face of continued labor arrest. For example, 126 women in this cohort had no labor progress despite 2 hours of oxytocin augmentation. Of these women, 101 achieved vaginal delivery. Had we delivered all of them by cesarean after 2 hours of oxytocin, the cesarean delivery rate for this cohort would have been 26%, as opposed to the 8% rate that was achieved. In the smaller subset of women with no progress 2 hours after achieving a uterine contraction pattern of greater than 200 Montevideo units, continued oxytocin augmentation allowed 32 of 52 women (62%) to achieve vaginal delivery. Even women with no labor progress despite 4 hours of oxytocin augmentation benefited from the protocol: 29 of 43 such women (67%) achieved vaginal delivery and were thereby spared an unnecessary cesarean.
Certain limitations of our data warrant comment. We studied only women in spontaneous labor who had experienced active-phase labor arrest. Therefore, our findings are not directly applicable to women undergoing induction of labor. The data also may not be applicable to women who experience active-phase labor arrest while receiving oxytocin (eg, women who receive oxytocin for protraction disorders or according to the principles of active management of labor). Because we excluded women with previous cesareans and those with chorioamnionitis at the time of active-phase labor arrest, the safety of this protocol for such women and their fetuses requires further investigation. Finally, although we studied more than 500 women, our ability to perform indepth subanalyses (eg, by cervical dilatation at labor arrest) was limited by the relatively small numbers of women in specific subgroups.
The above limitations notwithstanding, this labor-management protocol was safe and effective. We encourage others to investigate this protocol. If, with larger samples, the efficacy and safety of this protocol are confirmed, the minimum recommended duration of oxytocin augmentation for active-phase labor arrest should be increased.
1. National Institutes of Health. Cesarean Birth Task Force Development Report. Obstet Gynecol 1981;57:537–45.
2. American College of Obstetricians and Gynecologists. Dystocia. ACOG technical bulletin no. 137. Washington DC: American College of Obstetricians and Gynecologists, 1989.
3. Clarke SC, Taffel SM. Rates of cesarean and VBAC delivery, United States, 1994. Birth 1996;23:166–8.
4. American College of Obstetricians and Gynecologists. ACOG committee opinion no. 173. Prevention of early-onset group B streptococcal disease in newborns. Washington DC: American College of Obstetricians and Gynecologists, 1996.
5. American College of Obstetricians and Gynecologists. Dystocia and the augmentation of labor. ACOG technical bulletin no. 218. Washington DC: American College of Obstetricians and Gynecologists, 1995.
6. Sanchez-Ramos L, Kaunitz AM, Peterson HB, Martinez-Schnell B, Thompson RJ. Reducing cesarean sections at a teaching hospital. Am J Obstet Gynecol 1990;163:1081–8.
7. Myers SA, Gleicher N. A successful program to lower cesarean-section rates. N Engl J Med 1988;319:1511–6.
8. Arulkumaran S, Koh CH, Ingemarsson I, Ratnam SS. Augmentation of labour—mode of delivery related to cervimetric progress. Aust N Z J Obstet Gynaecol 1987;27:304–8.
9. Sharma SK, Sidawi JE, Ramin SM, Lucas MJ, Leveno KJ, Cunningham FG. Cesarean delivery. A randomized trial of epidural versus patient-controlled meperidine analgesia during labor. Anesthesiology 1997;87:472–6.
10. Thorp JA, Parisi VM, Boylan PC, Johnston DA. The effect of continuous epidural analgesia on cesarean section for dystocia in nulliparous women. Am J Obstet Gynecol 1989;161:670–5.
11. Ramin SM, Gambling DR, Lucas MJ, Sharma SK, Sidawi JE, Leveno KJ. Randomized trial of epidural versus intravenous analgesia during labor. Obstet Gynecol 1995;86:783–9.
12. Alexander JM, Lucas MJ, Ramin SM, McIntire DD, Leveno KJ. The course of labor with and without epidural analgesia. Am J Obstet Gynecol 1998;178:516–20.
13. American College of Obstetricians and Gynecologists. Operative vaginal delivery. ACOG technical bulletin no. 196. Washington DC: American College of Obstetricians and Gynecologists, 1994.
14. Centers for Disease Control and Prevention. Rates of cesarean delivery. MMWR Morb Mortal Wkly Rep 1993;42:287–9.
15. Notzon FC, Cnattingius S, Bergsjø P, Cole S, Taffel S, Irgens L, et al. Cesarean section delivery in the 1980s: International comparison by indication. Am J Obstet Gynecol 1994;170:495–504.