Temporal increases in operative delivery worldwide1–3 may in part be attributed to changes in maternal characteristics and obstetric practice patterns.4 In particular, operative delivery in the second stage of labor has been associated with significant adverse maternal and perinatal outcomes,5–7 which may be related to the time from full cervical dilatation to delivery.8,9 Challenges to evaluating these outcomes associated with a specific duration of the second stage of labor include establishing a definition of prolonged second stage in contemporary practice6,9,10–13 and assessing studies that have small numbers, combined analyses with both nulliparous and multiparous women,6,12 heterogeneous study groups6 and selected outcomes,14–18 and that report variable use of regional anesthesia.19 Although some studies have attempted to evaluate outcomes with varying lengths of the second stage of labor,8 the influence of progressive duration of the second stage remains unclear. A recent systematic review on this topic9 determined that variable definitions of prolonged second stage of labor, diverse study population characteristics and lack of control of confounding variables increase the difficulty in addressing the effect of the duration of the second stage on obstetric outcomes. In addition, mode of delivery was included variably as an outcome measure in the eligible studies, whereas the systematic review identified effect modification caused by mode of delivery as an important analysis not previously performed.
The present study used data extracted from a population-based database to estimate the magnitude of the effect of progressively increasing durations of the second stage of labor on maternal and perinatal outcomes, in nulliparous women separate from multiparous women, after accounting for clinically relevant confounding factors and method of delivery as an effect modifier.
MATERIALS AND METHODS
We conducted a population-based, cohort analysis using data from the Nova Scotia Atlee Perinatal Database for the years 1988 to 2006. This database collects provincial information on all pregnancy outcomes for residents of Nova Scotia, and was used to evaluate maternal, fetal, and neonatal morbidity associated with increasing durations of the second stage of labor. The low-risk population was defined by the inclusion and exclusion criteria. Pregnancies were included if there was a liveborn singleton at or after 37 weeks of gestational age reaching full cervical dilatation. Pregnancies were excluded if deliveries occurred before onset of labor, if there was a major congenital anomaly, if there was at least one previous cesarean delivery, if there were preexisting (including hypertension and diabetes) or severe pregnancy-related (including hypertension and diabetes requiring therapeutic intervention) medical disorders, or if there was missing outcome data. Information on medical conditions, lifestyle factors, labor and delivery events, and neonatal outcomes was available for analysis. Ongoing data quality assurance programs and validation studies have shown that the information in the database is reliable.20,21
Maternal and infant summary characteristics included maternal age, parity, maternal prepregnancy weight, neighborhood socioeconomic status, smoking in pregnancy, prelabor rupture of membranes, regional anesthesia, augmentation of labor, antibiotic use in labor, method of delivery, gestational age at delivery, and birth weight. Smoking was defined as any maternal smoking in pregnancy. Socioeconomic status was approximated using QAIPPE, a variable developed by Statistics Canada,22 to provide an adjusted annual income based on census data averaged over all households in a postal code. Low socioeconomic status was defined as QAIPPE 1–2, with higher incomes defined as QAIPPE 3–5. Prelabor rupture of membranes was defined as rupture longer than 24 hours before delivery. Regional anesthesia included spinal, epidural, or combined anesthesia. For the majority of the study duration, consideration for induction of labor or increased fetal surveillance in a postterm pregnancy in our center typically started at 41 completed weeks.
The length of the second stage of labor was defined, based on available data in the database, as the duration from full cervical dilatation to delivery. Current United States and Canadian national clinical practice guidelines state that the diagnosis of a prolonged second stage of labor should be considered in nulliparous women when the second stage exceeds 2–3 hours and in multiparous women when the second stage exceeds 1–2 hours,9,10 the upper limit modified by the need for regional anesthesia. In the present study, the length of the second stage of labor was grouped into five categories for nulliparous women (2 hours or less, more than 2 hours to 3 hours or less, more than 3 hours to 4 hours or less, more than 4 hours to 5 hours or less, and more than 5 hours), and into four categories for multiparous women (1 hour or less, more than 1 hour to 2 hours or less, more than 2 hours to 3 hours or less, and more than 3 hours). These groupings were based on the objective of the study, which was to estimate risks associated with increasing duration of the second stage of labor.
Maternal morbidity outcome variables included chorioamnionitis by placental pathology, uterine rupture or dehiscence, maternal obstetric trauma (laceration of the uterine artery, laceration of the bladder, bowel, or ureter, severe extension of the uterine incision, or third-degree or fourth-degree perineal lacerations), need for blood transfusions, hysterectomy, early and late postpartum hemorrhage (physician-diagnosed with an estimated blood loss greater than 500 mL for vaginal delivery or greater than 1,000 mL for cesarean delivery), wound complications (including abdominal or perineal wound infections or dehiscence), puerperal febrile morbidity (more than 38°C on two or more occasions in any 48- hour period, excluding the first 24 hours after delivery), peripartum venous thromboembolism, near-miss maternal mortality (transfer to intensive care) or a composite of any of the above.
Neonatal morbidity outcome variables included 5-minute Apgar score less than 7, birth depression (defined as delay in initiating and maintaining respirations after birth requiring resuscitation by mask or endotracheal tube for at least 3 minutes, a 5-minute Apgar score of 3 or less, or neonatal seizures due to hypoxic-ischemic encephalopathy [Sarnat Score higher than stage 1]),23 neonatal trauma (brachial plexus palsy, facial palsy, fractures, or intracranial hemorrhage), sepsis (positive blood culture, septicemia or systemic infection), and neonatal intensive care unit (NICU) admission with duration of stay longer than 24 hours, or a composite of any of the above. Neonatal traumas included in these analyses were intracranial hemorrhage (defined as tentorial tear, spinal cord hemorrhage, subgaleal hemorrhage, or other intracranial hemorrhage), cephalhematoma, brachial plexus palsy, facial palsy, skull fractures (depressed or linear), and other fractures (clavicle, ribs, humerus, or femur). Neonatal trauma was considered major if one or more of the following were present: depressed skull fracture, intracranial hemorrhage, or brachial plexus palsy, and was considered minor if one or more of the following were present: linear skull fracture, other fractures, facial palsy, or cephalhematoma.
To describe the study population with reference to duration of the second stage of labor consistent with national clinical practice guidelines, summary characteristics in nulliparous women with a second stage duration longer than 3 hours and in multiparous women with a second stage duration longer than 2 hours were compared with all other women in the second stage of labor (regardless of regional anesthesia). To remain consistent with national clinical practice guidelines, to estimate the effect of increasing duration of the second stage of labor, and to provide information on normal duration of the second stage, outcomes in nulliparous women with increasing duration of the second stage longer than 2 hours were compared with a duration of 2 hours or less, whereas outcomes in multiparous women with increasing duration of the second stage longer than 1 hour were compared with a duration of 1 hour or less.
Categorical data were compared using χ2. Multivariate logistic regression analysis was performed in a backward, stepwise fashion (factor retained if it changed the point estimate of the variable representing duration of second stage by 5% or more) to generate adjusted odds ratios for all outcomes by duration of the second stage of labor, accounting for potential confounding variables (maternal age, prelabor rupture of membranes, augmentation of labor, antibiotics in labor, regional anesthesia, gestational age, birth weight and year of birth). Effect modification caused by method of delivery on the outcomes of interest was determined with the most parsimonious model using the difference in –2 log likelihood, χ2. The general estimating equation procedure provided standard errors corrected for nonindependence of multiple pregnancies to the same woman. Odds ratios and 95% confidence intervals were determined. Statistical significance was P<.05. Statistical analyses were performed using the SAS 8.0 programming package for Windows (SAS Institute, Inc., Cary, NC), Egret 2.0.3 for Windows (Cytel Statistical Software & Services, Cambridge, MA), and EpiInfo (Centers for Disease Control and Prevention, Atlanta, GA). This study received approval from the Reproductive Care Program of Nova Scotia Data Access Committee and from the IWK Health Centre Research Ethics Board.
The potential study population of women delivering in Nova Scotia for the 19-year period (1988–2006) was 193,823. Of these, 80.0% met the inclusion criteria of having no previous cesarean delivery and delivering liveborn singletons at or after 37 weeks in cephalic presentation after reaching full cervical dilatation. Exclusion of major congenital anomalies (3.2%), serious preexisting or pregnancy-related medical disorders (3.0%), previous cesarean delivery (9.8%), cesarean deliveries without labor (5.8%), and missing outcome data (2.9%) resulted in the actual study population of 129,517, of whom 52% were nulliparous.
To determine expected duration of the second stage of labor in our population in a group representing normally laboring women, duration of the second stage was first examined in a selected group of nulliparous (n=31,065) and multiparous women (n=49,955) within the actual study population who entered labor spontaneously and progressed to spontaneous vaginal delivery. These median durations were found to be 1.0 hours (range 0–10.2 hours) for nulliparous women and 0.3 hours (range 0–7.8 hours) for multiparous women. Based on these findings, duration of the second stage of labor in our study longer than 3 hours for nulliparous women and longer than 2 hours for multiparous women represented duration of the second stage greater than the 75th percentile for each group.
In the actual study population, 9,314 (14.8%) of nulliparous women were identified as having a second stage of labor duration longer than 3 hours and 2,156 (3.2%) of multiparous women were identified as having a second stage of labor duration of longer than 2 hours. From 1988 to 2006 (Fig. 1), the rate of this longer second stage in nulliparous women increased from 10.2% to 16.6% (χ2 for trend P<.001), and the rate of this longer second stage in multiparous women increased from 2.5% to 3.2% (χ2 for trend P<.001). For nulliparous women having this longer second stage, the proportion having regional anesthesia increased from 89.20% in 1988 to 98.17% in 2006 (P<.001). For multiparous women with this longer second stage, the proportion having regional anesthesia increased from 78.76% in 1988 to 90.70% in 2006 (P<.001).
Table 1 summarizes maternal and perinatal characteristics of the study population with reference to the second stage of labor consistent with national clinical practice guidelines (and consistent with duration of the second stage greater than the 75th percentile for each group in the selected normally laboring population). Each variable in Table 1 was considered a clinically relevant characteristic and a potential confounder (except for method of delivery, which was tested as an effect modifier), and the groups were shown to be significantly different for each variable.
Tables 2–7 summarize the unadjusted and adjusted comparisons of selected maternal and perinatal outcomes separately among nulliparous and multiparous women with increasing duration of the second stage of labor. Smoking, prepregnancy weight, and socioeconomic status were not included as confounding variables in the regression analysis owing to the high percentage of missing values (greater than 10%). The outcome variables of uterine rupture or dehiscence, hysterectomy, venous thromboembolism, and transfer to an intensive care unit were not subjected to regression analysis owing to low occurrence (less than 0.1%). Further evaluation by duration longer than 4 hours for multiparous women was difficult owing to diminishing numbers of women in this group.
Examination of unadjusted analyses of maternal outcomes in nulliparous women (Table 2) demonstrated increased risks for chorioamnionitis, maternal obstetric trauma, postpartum hemorrhage, wound complications, puerperal febrile morbidity and composite maternal morbidity for all durations of the second stage longer than 2 hours. Regression analyses evaluating the effect of length of second stage of labor on maternal outcomes in nulliparous women (Table 2), where method of delivery was not a significant effect modifier, demonstrated progressive additional risks of obstetric trauma and postpartum hemorrhage (45–73%) for second stage of labor duration of 3 hours or longer. Evaluation of puerperal febrile morbidity demonstrated ongoing risks with increasing second-stage duration. There was a decreased risk for blood transfusion for durations longer than 4 hours but shorter than 5 hours. No difference in risk was observed for wound complications at any second stage duration longer than 2 hours.
The joint effects of duration of the second stage of labor and method of delivery (where method of delivery was found to be a significant effect modifier) were examined for the outcomes of chorioamnionitis and composite maternal morbidity (Table 3). There was no increase in risk for chorioamnionitis with spontaneous vaginal delivery regardless of the duration of second stage, and an increased risk for all durations with cesarean delivery. Risk for any maternal morbidity was increased with the joint effect of every duration of the second stage of labor and every method of delivery, with up to an additional 426% risk for nulliparous women delivering greater than 5 hours by operative vaginal delivery.
Unadjusted analyses of perinatal outcomes in nulliparous women (Table 4) demonstrated increased risks for a 5-minute Apgar score less than 7, birth depression, minor and major trauma, NICU admission, sepsis, and composite perinatal morbidity for most or all durations second stage longer than 2 hours. Adjusted analyses evaluating the effect of length of second stage of labor on perinatal outcomes in nulliparous women, where method of delivery was not a significant effect modifier (Table 4), demonstrated an increased risk of low 5-minute Apgar score for all second stage of labor durations, and for minor trauma with duration of the second stage greater than 2 to less than or equal to 3 hours. No differences in risk were observed for major trauma or neonatal sepsis.
The joint effects of duration of the second stage of labor and method of delivery (where method of delivery was found to be a significant effect modifier) were examined for the outcomes of birth depression, admission to NICU, and composite perinatal morbidity (Table 5). Increased risk for birth depression was shown for early durations of the second stage with operative deliveries (likely representing intervention for nonreassuring fetal status) but also for longer durations, regardless of method of delivery. Risk of admission to NICU was increased for longer durations with all methods of delivery. Risk for any perinatal morbidity was increased with the joint effect of every duration of the second stage of labor and every method of delivery.
Examination of unadjusted analyses of maternal outcomes in multiparous women (Table 6) demonstrated progressive increases in risk for maternal obstetric trauma, blood transfusion, postpartum hemorrhage, wound complications, puerperal febrile morbidity and composite maternal morbidity for second stage duration longer than 1 hour. Method of delivery was not found to be an effect modifier for any maternal or perinatal outcome among multiparous women. In multiparous women, regression analyses evaluating the effect of length of second stage of labor on maternal outcomes demonstrated a progressive increase in risk for obstetric trauma, postpartum hemorrhage, wound complications, puerperal febrile morbidity, and composite of maternal morbidity (Table 6), with the highest additional risks (7–144%) occurring longer than 3 hours. No differences in risk were observed for chorioamnionitis or need for blood transfusion for any second stage duration.
Unadjusted analyses of perinatal outcomes in multiparous women (Table 7) demonstrated increased risks of 5-minute Apgar scores less than 7, birth depression, minor neonatal trauma, NICU admission, and composite perinatal morbidity for durations of the second stage longer than 1 hour. In multiparous women, regression analyses evaluating the effect of length of second stage of labor on perinatal outcomes demonstrated an increase in risk for 5-minute Apgar score less than 7, birth depression, minor trauma, NICU admission, and any perinatal morbidity (Table 7), with the highest additional risks (15–57%) occurring longer than 2 hours. No differences in risk were observed for major neonatal trauma or sepsis in multiparous women for second stage duration longer than 2 hours.
Although studies have focused on adverse perinatal outcomes associated with a long duration of the second stage of labor, there are few studies that have considered adverse maternal outcomes or the effect of increasing duration of the second stage. Our large population-based study, designed to characterize the magnitude of effect of progressive increases in duration of the second stage of labor on maternal and perinatal morbidity, demonstrated important observations in maternal and perinatal risks for nulliparous and multiparous women delivering after 36 weeks of gestation, especially when durations of the second stage were longer than 3 hours in nulliparous women and longer than 2 hours in multiparous women.
A previous evaluation of outcomes with increasing duration of labor in a cohort of low- and high-risk nulliparous women at term has demonstrated that rates of chorioamnionitis, severe perineal lacerations, and postpartum hemorrhage increased significantly after controlling for potential confounding variables.8 Women were more likely to have adverse maternal outcomes when they had second stages longer than 3 hours and longer than 4 hours compared with durations shorter than 3 hours and shorter than 4 hours, respectively.8 Significant differences (adjusted analyses) also have been observed in rates of postpartum hemorrhage and chorioamnionitis among women who delivered after less than 2 hours compared with those who delivered after more than 2 hours, with no difference in these maternal outcomes when 2 to 4 hours was compared with more than 4 hours.6 An increase in unintentional hysterotomy extensions has been observed in a low-risk cohort of women at term requiring cesarean delivery, when the effect of duration of a second stage of less than 3 hours was compared with more than 4 hours. Differences in disseminated intravascular coagulation, postpartum fever, need for blood transfusion, or composite morbidity were not demonstrated.14 Our findings of increased risks of maternal morbidity with increased duration of the second stage of labor are consistent with other studies, demonstrating progressive increases in risk for the majority of the adverse outcomes evaluated in our study among nulliparous and multiparous, especially when including the joint effects of method of delivery and duration of the second stage. The risk of any maternal morbidity, which was increased with all durations of the second stage among nulliparous and multiparous women regardless of method of delivery, retained significant increased risk in both groups when postpartum hemorrhage and puerperal febrile morbidity was removed from the composite outcome analysis (data not shown). The need for blood transfusion was reduced significantly for second stage durations of more than 4 hours to 5 hours or less, even though the risk for postpartum hemorrhage was significantly increased for all durations, although it becomes difficult to interpret the clinical significance of this finding given the small number of women in this group and the dramatic reduction in transfusion of blood products in Canada because of infection-related concerns during the years of the study.
The majority of studies addressing the effect of duration of the second stage on perinatal or neonatal outcomes have not observed differences in serious morbidity,6,8,9,12–14,18 including low 5-minute Apgar score, admission to NICU, neonatal seizures, and fetal acid-base status. One study has demonstrated that prolonged second stage is not an independent predictor of neonatal Erb’s palsy,17 and others have not demonstrated a relationship with duration of the second stage and related neonatal outcomes such as shoulder dystocia.6 Our study included Erb’s palsy in the group of major trauma and demonstrated no increased risk for major trauma in nulliparous women or multiparous women. However, minor trauma was increased for both nulliparous women with a second stage of more than 2 hours to 3 hours or less and for all durations of the second stage among multiparous women, regardless of method of delivery. Admission to NICU was shown previously to be increased for infants born to women with a second stage longer than 2 hours,6 but this finding has not been supported by other studies.12,13,18 Our study showed increases in risk of admission to NICU with longer second stage durations and with all methods of delivery. The risk of any perinatal morbidity, which was increased with all durations of the second stage among nulliparous and multiparous women regardless of method of delivery, retained significant increased risk among multiparous, but not nulliparous, women when admission to NICU was removed from the composite outcome analysis (data not shown).
Several studies limited their analysis of outcomes to women with either vaginal12 or cesarean8,14 delivery. In a previous study by our group investigating maternal and neonatal outcomes in a group of low-risk women requiring cesarean delivery in labor, an evaluation of morbidity at full cervical dilatation including the effect of duration of the second stage demonstrated no difference in outcomes when the duration of the second stage was longer than four hours (n=204) compared with a duration of shorter than or equal to 4 hours (n=345),7 although these finding may have been a function of small study numbers. However, a meta-analysis of outcomes with prolonged second stage has identified the importance of accounting for method of delivery as a significant effect modifier in studies evaluating the influence of duration of second stage, given the causal nature of the relationship between prolonged second stage and operative delivery.9 Our study was able to consider modification of the effect of longer second stage duration on maternal and perinatal outcomes by method of delivery, because of the comprehensive collection of delivery data in the Nova Scotia Atlee Perinatal Database.
Although there are methodologic challenges to studying labor progression owing to multiple influencing factors,24 we employed a large data set from a validated electronic perinatal database that allowed separate analyses by parity and controlling for other relevant maternal and perinatal factors, including use of regional anesthesia and year of birth. This capability becomes particularly important in the setting of previous studies limited by lack of confounding variables and, in our study, the numerous statistically significant comparisons in the unadjusted analyses. An important limitation of the present study is that the duration of the second stage of labor was defined as the time from full dilatation to delivery based on the information available in the database, which would also include the time from decision to intervene with operative delivery (forceps/vacuum assisted vaginal delivery or cesarean delivery) to the intervention itself. Although the data in the Nova Scotia Atlee Perinatal Database have been shown, through data reabstraction (with a high level of agreement for most routine variables) and validation studies, to contain reliable information,20,21 retrospective studies are limited to available data and so there may be other variables relevant to the present study that are not routinely captured in the database, such as the intrapartum factors of maternal positioning in labor25 or delayed pushing techniques.19
In this population-based study evaluating maternal and perinatal outcomes associated with prolonged second stage of labor, increasing duration of the second stage of labor was associated with increasing risks for both maternal and perinatal adverse outcomes among both nulliparous and multiparous women. The increased risks for many of the outcomes with longer durations of the second stage of labor were independent of method of delivery. Maternal, fetal and intrapartum characteristics must be considered in the management of the second stage of labor. Most of the factors associated with a longer duration of the second stage of labor cannot be modified, other than to assure effective uterine action with oxytocin augmentation. However, knowledge that there is increasing maternal and perinatal morbidity with prolongation may help guide timely intervention when second stage dystocia is diagnosed.
4. Joseph KS, Young DC, Dodds L, O’Connell CM, Allen VM, Chandra S, et al. Changes in maternal characteristics and obstetric practice and recent increases in primary cesarean delivery. Obstet Gynecol 2003;102:791–800.
5. Murphy DJ, Liebling RE, Verity L, Swingler R, Patel R. Early maternal and neonatal morbidity associated with operative delivery in the second stage of labour: a cohort study. Lancet 2001;358:1203–7.
6. Myles TD, Santolaya J. Maternal and neonatal outcomes in patients with a prolonged second stage of labor. Obstet Gynecol 2003;102:52–8.
7. Allen VM, O’Connell CM, Baskett TF. Maternal and perinatal morbidity of caesarean delivery at full cervical dilatation compared with caesarean delivery in the first stage of labour. BJOG 2005;112:986–90.
8. Cheng YW, Hopkins LM, Caughey AB. How long is too long: does a prolonged second stage of labor in nulliparous women affect maternal and neonatal outcomes? Am J Obstet Gynecol 2004;191:933–8.
9. Altman MR, Lydon-Rochelle MT. Prolonged second stage of labor and risk of adverse maternal and perinatal outcomes: a systematic review. Birth 2006;33:315–22.
10. Dystocia and augmentation of labor. ACOG Practice Bulletin No. 49. American College of Obstetricians and Gynecologists. Obstet Gynecol 2003;102:1445–54.
11. Second stage of labour. Healthy beginnings: guidelines for care during pregnancy and childbirth. SOGC Clinical Practice Guidelines. J Soc Obstet Gynaecol Can 1998;20:52–8.
12. Janni W, Schiessl B, Peschers U, Huber S, Strobl B, Hantschmann P, et al. The prognostic impact of a prolonged second stage of labor on maternal and fetal outcome. Acta Obstet Gynecol Scand 2002;81:214–21.
13. Menticoglou SM, Manning F, Harman C, Morrison I. Perinatal outcome in relation to second-stage duration. Am J Obstet Gynecol 1995;173:906–12.
14. Sung JF, Daniels KI, Brodzinsky L, El-Sayed YY, Caughey AB, Lyell DJ. Cesarean delivery outcomes after a prolonged second stage of labor. Am J Obstet Gynecol 2007;197:306.e1–5.
15. Bofill JA, Rust OA, Perry KG, Roberts WE, Martin RW, Morrison JC. Operative vaginal delivery: a survey of fellows of ACOG. Obstet Gynecol 1996;88:1007–10.
16. Rizk DE, Abadir MN, Thomas LB, Abu-Zidan F. Determinants of the length of episiotomy or spontaneous posterior perineal lacerations during vaginal birth. Int Urogynecol J Pelvic Floor Dysfunct 2005;16:395–400.
17. Weizsaecker K, Deaver JE, Cohen WR. Labour characteristics and neonatal Erb’s palsy. BJOG 2007;114:1003–9.
18. Hagelin A, Leyon J. The effect of labor on the acid-base status of the newborn. Acta Obstet Gynecol Scand 1998;77:841–4.
19. Fraser WD, Marcoux S, Krauss I, Douglas J, Goulet C, Boulvain M. Multi-center, randomized, controlled trial of delayed pushing for nulliparous women in the second stage of labor with continuous epidural analgesia. The PEOPLE (Pushing Early or Pushing Late with Epidural) Study Group. Am J Obstet Gynecol 2000;182:1165–72.
20. Fair M, Cyr M, Allen AC, Wen SW, Guyon G, MacDonald RC. An assessment of the validity of a computer system probabilistic record linkage of birth and infant death records in Canada. The Fetal and Infant Health Study Group. Chronic Dis Can 2000;21:8–13.
22. Wilkins R. Postal code conversion file user’s guide. Catalogue no. 82F0086-XDB. Ottawa, Ontario, Canada: Statistics Canada; 2001.
23. Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress. A clinical and electroencephalographic study. Arch Neurol 1976;33:696–705.
24. Vahratian A, Troendle JF, Siega-Riz AM, Zhang J. Methodological challenges in studying labour progression in contemporary practice. Pediatr Perinat Epidemiol 2006;20:72–8.
25. Ragnar I, Altman D, Tydén T, Olsson SE. Comparison of the maternal experience and durations of labour in two upright delivery positions—a randomized controlled trial. BJOG 2006;113:165–70.