| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Hôpital St-François d’Assise (CHUQ) Research Center, Department of Obstetrics and Gynecology, and Department of Social and Preventive Medicine, Laval University, Québec, Canada.
Address reprint requests to: William D. Fraser, MD, MSc, Hôpital St-François d’Assise (CHUQ) Research Center, 10 rue de l’Espinay, local D1-724, Québec, G1L 3L5, Canada; E-mail: william.fraser{at}ogy.ulaval.ca.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: The database is derived from a multicenter randomized trial of delayed pushing for nulliparous women under continuous infusion epidural. Members of this cohort (n = 1862) were randomly divided into two groups: a "Model Development" and a "Model Validation" group. We used univariate and multivariable techniques to assess associations between anthropometric, sociodemographic, and obstetric variables and difficult delivery.
RESULTS: With the referent defined as the category of lesser risk, the developed model showed that the risk of difficult delivery was increased for women with height less than 160 cm (odds ratio [OR] 2.1, 90% confidence interval [CI] 1.2, 3.4), prepregnancy weight greater than 65 kg (OR 1.6, 90% CI 1.0, 2.6), age greater than or equal to 35 years (OR 3.0, 90% CI 1.1, 8.1), and gestational age greater than or equal to 41 weeks (OR 1.8, 90% CI 1.1, 2.8). Induction of epidural analgesia late in labor (greater than or equal to 6 cm) was associated with a higher risk of difficult delivery than induction between 3 and 5 cm (OR 1.9, 90% CI 1.3, 2.8). An interval of greater than or equal to 360 minutes between epidural induction and full dilatation increased the risk of difficult delivery (OR 3.8, 90% CI 1.5, 9.5). Fetal station above +2 at full dilatation and a posterior fetal position were both strongly associated with difficult delivery (OR 2.7, 90% CI 1.4, 5.0, and OR 11.2, 90% CI 4.9, 25.6, respectively). For the multivariable predictive model, when the sensitivity was 57%, the specificity was 75%, and the positive predictive value was 35%.
CONCLUSION: Our observations concerning maternal characteristics and obstetric variables are consistent with previous observations with the exception of time of induction of the epidural. The predictive model may be useful in defining high-risk populations for subsequent intervention studies designed to assess approaches to reduce difficult delivery.
Epidural anesthesia is an efficacious method of pain relief for women during the second stage of labor.1,2 However, when the epidural block is maintained in the second stage, randomized clinical trials have shown it to be associated with an increase in operative vaginal delivery.3 The effect of epidural on cesarean section is more controversial.4–6 Two recent trials from the same center compared the effect of epidural versus nonepidural analgesia on cesarean section. Although the results of these two studies were divergent, the epidural technique differed between the studies. In the first study, continuous infusion was maintained in the second stage, and cesarean section was increased in the epidural group.7 In the second study, where the infusion was halved or discontinued if there was a delay in second-stage progress,8 no effect on cesarean section was noted. These findings suggest that, similar to its effect on forceps, cesarean section risk is increased when epidural is maintained in the second stage. Other studies have documented a marked increase in maternal pain after the discontinuation of epidural analgesia.9 This outcome is inconsistent with the goal of maintaining analgesia throughout labor. Both forceps and cesarean section are associated with an increased risk of morbidity for the mother and/or the baby.10,11 For this reason, cesarean section and certain types of operative vaginal delivery have been grouped under the heading of "difficult delivery."12
Several approaches have been proposed as methods of reducing difficult delivery among women with epidural. These include delayed pushing,13–15 routine administration of oxytocin in the second stage,16 and modification of the epidural technique.17 Although these approaches are all promising, further research is required to better define the women who are most likely to benefit from these strategies. In planning further research, it would be important to define specific subgroups of women receiving epidural analgesia for whom the risk of difficult delivery is particularly great. This would help to identify appropriate target populations for the evaluation of interventions designed to reduce difficult deliveries.
The objectives of this study are, for nulliparous women in the second stage of labor with a continuous perfusion of epidural analgesia, 1) to identify variables that are predictive of difficult delivery, and 2) to develop a multivariable model to predict this obstetric complication.
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
For the purpose of this secondary cohort analysis, study participants were divided randomly into two groups: 1) the "Model Development group" in which the strength of potential associations were determined, and a multivariate prediction model was developed; and 2) the "Model Validation group" in which the predictive capacity of the model was evaluated.
Data concerning anthropometric, sociodemographic, obstetric, and anesthesiologic characteristics, as well as measures of sensory and motor block, were collected by the labor room nurse. The obstetrician documented variables related to delivery method and neonatal status. In this study, the principal dependent variable is "difficult delivery," defined as any of the following: second-stage cesarean section, midforceps or midvacuum delivery, low forceps or low-vacuum delivery with rotation of the fetal head greater than 45 degrees, and any operative vaginal delivery preceded by manual rotation of the fetal head by greater than 45 degrees. The definition for midforceps is that of the ACOG.18 For the purposes of analysis, potential risk factors were grouped into the following four independent domains: anesthesiologic, obstetric, anthropometric, and sociodemographic.
Maternal pain intensity was measured using a 10-cm visual analogue scale.19 Degree of motor block was measured using the Bromage scale.20
Dystocia is defined as a rate of dilatation less than 0.5 cm per hour over a period more than 4 hours from greater than 3 cm dilatation as defined by Turcot et al21 and as proposed by the Canadian National Consensus Conference on Aspects of Cesarean Birth.22
The two groups were randomly constituted from the whole population of 1862 women. The first group, with 934 women, served to elaborate the predictive model (Model Development group). The second group, composed of 928 women, was used to validate the model (Model Validation group). Four women (three in the first group and one in the second) were excluded because of missing information concerning the station of the fetal head at the time of application of the instrument, and for whom it was impossible to classify the delivery type. The comparability of the two groups was verified by comparisons of anthropometric and sociodemographic variables and of the mode of delivery. To compare the effect of the individual variables on "difficult delivery," crude relative risks (RR) were calculated along with their 90% confidence intervals (CI).
For the purpose of the multivariable analysis, we retained variables with an RR higher than 2 or less than 0.5, as well as variables for which the 90% CI excluded the value of null effect (RR 1) for at least one category. A multiple logistic regression model was developed using a forward stepwise approach, starting with the variable showing the most significant association in the univariate logistic regression. If, based on the likelihood ratio test, this variable significantly changed the model (P 
.1), it was retained. The variable was also retained if any odds ratio or at least one category of the variable showed a statistically significant association. If the deletion of a variable introduced variations in the ß coefficient of another variable in the model, a collinearity test was performed. No interactions were studied. The specificity, sensitivity, and positive and negative predictive values for difficult delivery were derived from the model, and the receiver operating characteristic curve was traced. With the aid of the specificity, sensitivity, predictive value, and the ß coefficient, we tested the validity of the developed model using the Model Validation group.
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The mean (±standard deviation) for age was 27.6 ± 4.9 years in both groups. Groups were similar with respect to prepregnancy weight, weight gain during pregnancy, gestational age at delivery, educational level, and ethnic distribution (data not shown). The groups were similar with respect to distributions of delivery methods (Table 2). The proportion of women with difficult delivery was 20.0% in the Model Development group and 20.3% in the Model Validation group.
|
|
The univariate analyses indicating RRs and their 90% CIs of difficult delivery are presented in Table 3. Small stature, Asian origin, increasing maternal age, and increasing gestational age were all associated with an increased risk of difficult delivery. A woman with a height of less than 160 cm had a 1.6-fold greater risk of difficult delivery than a woman with a height greater than or equal to 170 cm. Prepregnancy weight of 65 kg or more increased the risk of difficult delivery by nearly 1.5-fold. The risk in women aged greater than or equal to 35 was 2.8 times the risk in women of 20 years or less. There was no evidence in our data for an association between labor induction and the risk of difficult delivery (RR 1.2, 90% CI 0.9, 1.4, data not shown). There was no clear relationship between cervical dilatation at admission and risk of difficult delivery. There was a nonsignificant trend toward an association between oxytocin use in the first stage of labor and difficult delivery (RR 0.8, 90% CI 0.6, 1.0, data not shown). The presence of first-stage dystocia by the study definition did increase the risk of difficult delivery (RR 1.3, 90% CI 1.1, 1.6). Fetal station and position at full dilatation were both strongly associated with risk of difficult delivery. Fetal station above +2 was associated with a two- to three-fold increase in the risk of difficult delivery. Women for whom the fetal position at full dilatation was other than occiput anterior had an increased risk of difficult delivery, ranging from RRs of 1.4 and 1.9 for left occiput anterior (LOA) and right occiput anterior (ROA), respectively, to an RR of 5.2 for the occiput posterior position. Women in the lateral decubitus position at full dilatation had an 1.8-fold increase in the risk of difficult delivery.
|
Table 4 presents the predictive model developed using multivariable logistic regression analysis. In general, the results of the multivariable analysis are consistent with the univariate analysis. With the exception of dystocia, the variables for which a statistically significant (
= 0.1) association was noted in the univariate analysis were retained in the multivariable model.
|
ßx 61 0.451 + 0.310 + 0.849 + 0.192 + 0.651 + 0.978 + 0.435 + 0.488 + 0.345). Such a patient has a predicted probability of a difficult delivery of between 34.6% and 37.1% (positive predictive value in Table 5). The predictive values for both model groups are shown in Table 5 and 6. With the aid of the specificity, sensitivity, predictive value, and the sum for ßx, the results permit prediction of difficult delivery for a nullipara receiving epidural anesthesia during the second stage of labor. The specificity, sensitivity, and predictive positive and negative values are also analyzed for the Model Validation group (Table 6). Comparison of Model Development and Model Validation groups shows a high correspondence between groups. The Model Development group can be accepted as valid. The receiver operating characteristic curve is shown in Figure 1. The area under the receiver operating characteristic curve is 0.77 indicating a moderate predictive value of the model.
|
|
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This study has several strengths. Data are derived from a large patient population of women delivering in 12 centers, and with varying sociodemographic and anthropometric characteristics, which increases the generalizability of study results. Trial participants had uncomplicated pregnancies, a normal fetal status at study enrollment, and an effective epidural analgesia at full dilatation. Results may only be generalized to women with these characteristics. As the study personnel were not aware that the data would be used to develop a predictive model of difficult delivery, there is little reason to suspect bias in the recording of data. Information regarding labor progress and cervical status at the time of induction of epidural analgesia were obtained from routine clinical assessments, which are subject to the inter-and intraevaluator errors that are inherent to these measurements. Because the epidural technique was standardized across centers, the context of care with respect to epidural analgesia is clearly defined.
We have chosen "difficult delivery" as the main dependent variable for this secondary analysis of the trial database. Difficult delivery was the primary outcome for the trial from which the study’s database is derived.13 The choice of the mode of delivery in women in the second stage requiring operative delivery depends in part on the practice pattern of the obstetrician.25 The observed increase in both maternal and neonatal morbidity associated with difficult delivery serves to validate the choice of this variable as an outcome of interest. We believe that the categoric outcome "difficult delivery" represents a useful variable for monitoring interventions and the quality of care in the second stage of labor.
Among the variables showing the strongest association with difficult delivery are abnormal fetal position and high fetal station at full dilatation, advanced maternal age, and height less than 160 cm. In our study, neither oxytocin use in the first stage nor method of labor onset was statistically associated with second-stage difficult delivery. In a previous study of women in spontaneous labor,21 we observed that although in a univariate analysis the use of oxytocin was associated with operative delivery, it was not an independent predictor in a multivariable model that included dystocia.
The effects of the timing of epidural administration on cesarean section and forceps delivery remain controversial. Thorp et al suggested that administration of epidural analgesia at a cervical dilatation of less than 5 cm increases the risk of epidural.6 Chestnut et al, in a small randomized controlled trial, found no evidence for an effect of the timing of epidural on either cesarean section or instrumental vaginal delivery.26 However, the contrast achieved between groups with respect to the mean cervical dilatation at the time of epidural was not great (early epidural 4 cm, late epidural 5 cm). Our data do suggest an association between the timing of epidural analgesia and difficult delivery. However, the lowest risk of difficult delivery was observed in the group where induction of analgesia was between 3 and 5 cm. Increased risks were observed when the epidural was placed very early in labor (less than 3 cm) or late in labor (greater than or equal to 6 cm). The interval between induction of epidural and full dilatation was directly related to the risk of difficult delivery. We have no explanation for the discrepancy between the findings in our study and those of Thorp et al6 and Chestnut et al,26 other than the timing of induction of epidural was determined differently across studies. In our study, the timing of epidural was determined primarily by patient request. In the other two studies, the epidural was the experimental intervention, and timing was controlled by the investigator. It is possible that women who initially wish to avoid epidural but who later change their mind because of a slow labor or higher pain level may be at increased risk of difficult delivery. Similarly, women who request epidural early in labor may be at increased risk of difficult delivery.
Previous studies have suggested that the intensity of the motor block may be a factor in determining the risk of difficult delivery.17 We failed to demonstrate such an association. Given that the concentration of the analgesic agent for the continuous infusion was standardized across centers, there may have been less variation in technique than might normally be found in routine clinical care. We cannot generalize this observation to patients receiving higher concentrations of analgesic agent.
It is reassuring to note that the risk of difficult delivery was not associated with the patient’s educational level contrary to some previous studies carried out in the United States, which have observed an association between operative delivery rates and socioeconomic status.27,28 In the randomly selected subgroup of women in the Model Development group, we observed a trend for early pushing to increase the risk of difficult delivery. However, the CIs around the point estimate of the effect (RR 1.13, 90% CI 0.91, 1.40) did not exclude the null effect. The difference between this result and the previously reported trial results where delayed pushing was found to reduce the risk of difficult delivery (RR 0.79, 95% CI 0.66, 0.95)13 can be explained by sampling variation and reduced precision of the CI in the subgroup of patients.
Birth weight has been shown to influence the risk of cesarean delivery.29 However, as fetal weight estimates are not a routine part of intrapartum care, this variable was not included in our predictive model. Lateral decubitus position at full dilatation was unexpectedly found to be associated with difficult delivery. Although there is no clear explanation for this observation, it is possible that women who were counseled to assume this position had preexisting fetal heart anomalies and were at increased risk of intervention for fetal indications.
Regarding the performance of the predictive model in the validation group, our data suggest that the model could be useful as a screening tool but not as a diagnostic test. Even when the ßx increases to 6.2 (which is a value above the 95th centile for the study population), the predictive value of the positive test was only 58%, meaning 42% of women with these characteristics went on to a nondifficult delivery. In its present state, the model does not appear to have the precision to be used as an aid in decision making regarding the need for operative delivery.
In conclusion, our observations concerning maternal characteristics and obstetric variables are consistent with previous observations, with the exception of the time of induction of epidural. The intensity of the epidural block was not associated with risk of difficult delivery. A cervical dilatation of 3–5 cm at induction of epidural was associated with the lowest risk of difficult delivery. A prolonged interval (greater than or equal to 360 minutes) from epidural induction to full dilatation is associated with a nearly four-fold increase in risk of difficult delivery. Occiput posterior position is associated with a five-fold increase in the risk of difficult delivery. The predictive model developed in this study can serve to define high-risk populations that are appropriate for subsequent evaluation of intervention strategies to prevent difficult delivery in the second stage of labor.
|
Footnotes |
|---|
, B. Steffy, RN, Hutzel Hospital, Detroit, MI; M. Helewa, MD, S. Lucy, MD, S. Erickson, RN, St. Boniface Hospital, Winnipeg, Mb.; N. Okun, MD, A. Guest, MD, A. Stuart, MD, D. Schimeck, RN, University of Alberta Hospital, Edmonton, Ab.; M. Sermer, MD, M. Bailey, RN, Toronto General Hospital, Toronto, Ont.; D. Blouin, MD, Y. Claprood, and D. Beaulieu, RN, Center Hospitalier Universitaire de l’Estrie, Sherbrooke, Qc. Data Management Committee: Michel Boulvain, MD, PhD, Sylvie Bérubé, PhD, and Isabelle Faron. Safety and Efficacy Monitoring Committee: François Meyer, MD, PhD, Aida Bairam, MD, PhD, and Jean-Marie Moutquin, MD, MSc. This study was sponsored by the Medical Research Council of Canada (grant # MT-12423). AstraZeneca R&D Montréal, Montréal, Québec, Canada, provided medication. The data presented are part of the Master’s Thesis of Mireille Cayer.
Received June 21, 2001. Received in revised form November 9, 2001. Accepted November 19, 2001.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Howell CJ. Epidural versus non-epidural analgesia for pain relief in labour (Cochrane Review). The Cochrane Library, Issue 4, 2000.
3. Howell C, Chalmers I. A review of prospectively controlled comparisons of epidural with non-epidural forms of pain relief during labour. Int J Obstet Anest 1992;1: 93–110.
4. Rhonda L, Andrew PH. Anesthesia in obstetrics. Current Opinion Obstet Gynecol 1994;6:408–13.
5. Thorp J, Parisi V, Boylan P, Johnston D. The effect of continuous epidural analgesia on cesarean section for dystocia in nulliparous women. Am J Obstet Gynecol 1989; 161:670–5.[Medline]
6. Thorp J, Hu D, Albin R, McNitt J, Meyer B, Cohen G, et al. The effect of intrapartum epidural analgesia on nulliparous labor: A randomized controlled prospective trial. Am J Obstet Gynecol 1993;169:851–8.[Medline]
7. 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.[Abstract]
8. Sharma SK, Sidawi E, Ramin SM, Lucas MJ, Leveno KJ, Cunningham FG. Cesarean delivery: A randomized trial of epidural versus patient-controlled meperidine analgesia during labor. Anesthesiology 1997;87:487–94.[Medline]
9. Chestnut DH, Vanderwalker GE, Owen CL, Bates JN, Choi WW. The influence of continuous epidural Bupivacaine analgesia on the second stage of labor and method of delivery in nulliparous women. Anesthesiology 1987;66: 774–80.[Medline]
10. Hagadorn-Freathy AS, Yeomans ER, Hankins GDV. Validation of the 1988 ACOG Forceps Classification System. Obstet Gynecol 1991;77:356–60.
11. Robertson P, Laros R, Zhao R. Neonatal and maternal outcome in low pelvic and midpelvic operative deliveries. Am J Obstet Gynecol 1990;162:1436–44.[Medline]
12. Thomson ME. Factors predisposing to difficult labour in primiparas. Montréal, Canada: McGill; 1985.
13. Fraser WD, Marcoux S, Krauss I, Douglas J, Goulet C, Boulvain M. Multicenter, 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.[Medline]
14. Maresh M, Choong K, Beard R. Delayed pushing with lumbar epidural analgesia in labour. Br J Obstet Gynaecol 1983;90:623–7.[Medline]
15. Vause S, Congdon HM, Thornton JG. Immediate and delayed pushing in the second stage of labour for nulliparous women with epidural analgesia: A randomised controlled trial (see comments). Br J Obstet Gynaecol 1998; 105:186–8.[Medline]
16. Saunders NJ, Spiby H, Gilbert L, Fraser RB, Hall JM, Mutton PM, et al. Oxytocin infusion during second stage of labour in primiparous women using epidural analgesia: A randomised double blind placebo controlled trial. Br Med J 1989;299:1423–6.
17. Vertommen J, Vandermeulen E, Van Aken H, Vaes L, Soetens M, Van Steenburge A, et al. The effects of the addition of Sufentanil to 0.125% Bupivicaine on the quality of analgesia during labor and on the incidence of instrumental deliveries. Anesthesiology 1991;74:809–14.[Medline]
18. American College of Obstetricians and Gynecologists. Committee on obstetrics, maternal and fetal medicine: Obstetrics forceps. Report number 59. Washington, DC: American College of Obstetricians and Gynecologists; 1988.
19. Melzack R. The short form McGill pain questionnaire. Pain 1987;30:192–209.
20. Bromage P. An evaluation of Bupivacaine in epidural analgesia for obstetrics. Can Anaest Soc J 1969;16:46–7.
21. Turcot L, Marcoux S, Fraser WD. Multivariate analysis of risk factors for operative delivery in nulliparous women. Canadian Early Amniotomy Study Group. Am J Obstet Gynecol 1997;176:395–402.[Medline]
22. National Consensus Conference on Aspects of Cesarean Birth. Indications for cesarean section: Final statement of the panel of the National Consensus Conference on Aspects of Cesarean Birth. Can Med Assoc J 1986;134: 1348–52.[Medline]
23. Thorp JA, Eckert LO, Ang MS, Johnston DA, Peaceman AM, Parisi VM. Epidural analgesia and cesarean section for dystocia: Risk factor in nulliparas. Am J Perinat 1991; 6:402–10.
24. Morton SC, Williams MS, Keeler EB, Gambone JC, Kahn KL. Effect of epidural analgesia for labor on the cesarean delivery rate. Obstet Gynecol 1994;83:1045–52.[Medline]
25. Guillemette J, Fraser W. Differences between obstetricians in caesarean section rates and the management of labour. Br J Obstet Gynaecol 1992;99:105–8.[Medline]
26. Chestnut D, Vincent R Jr, McGrath J, Choi W, Bates J. Does early administration of epidural analgesia affect obstetric outcome in nulliparous women who are receiving intravenous oxytocin? Anesthesiology 1994;80:1193–200.[Medline]
27. de Regt RH, Minkoff HL, Feldman J, Schwarz RH. Relation of private or clinic care to the cesarean birth rate. N Engl J Med 1986;315:619–24.[Abstract]
28. Peipert JF, Hogan JW, Gifford D, Chase E, Randall R. Strength of indication for cesarean delivery: Comparison of private physician versus resident service labor management. Am J Obstet Gynecol 1999;181:435–9.[Medline]
29. Turner MJ, Rasmussen MJ, Turner JE, Boylan PC, Mac-donald D, Stronge JM. The influence of birth weight on labor in nulliparas. Obstet Gynecol 1990;76:159–63.
This article has been cited by other articles:
|
|
 |
|
  J. Senecal, X. Xiong, W. D. Fraser, and for the PEOPLE (Pushing Early Or Pushing Late with Effect of Fetal Position on Second-Stage Duration and Labor Outcome Obstet. Gynecol., April 1, 2005; 105(4): 763 - 772. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
