Health Articles

 

Pii: s0029-7844(00)00794-8

Elective Induction of Labor as a Risk Factor forCesarean Delivery Among Low-Risk Womenat Term ARTHUR S. MASLOW, DO, MSc, AND AMY L. SWEENY, MPH Objective: To determine the effects of elective induction on
tions, indicated and elective inductions, and induction the risk of cesarean delivery in a cohort of women with
techniques. There is continued criticism of the adequacy low-risk term pregnancies and to evaluate the costs of
of control groups and statistical techniques used to elective induction services within our hospital system.
evaluate reported results.
Methods: Records of 1135 eligible women with low-risk,
The purpose of this study was to determine whether singleton, vertex pregnancies at 38 – 41 weeks' gestation who
deliveries with elective induction were associated with were eligible for vaginal delivery were analyzed retrospec-
tively after elective induction (n ⴝ 263) or spontaneous labor
greater risk of cesarean delivery or higher costs because (n ⴝ 872). Outcome measures included cesarean delivery and
of increased in-hospital resource use compared with direct costs. Variables evaluated were parity, maternal age,
noninduced deliveries in a cohort of women with estimated gestational age, birth weight, prior cesarean deliv-
low-risk term pregnancies, some of whom underwent ery, epidural anesthetic use, and provider category. Analysis
elective induction of labor.
was by univariable and multivariable regression modeling.
Results: Elective induction placed nulliparas at a twofold
higher risk for cesarean delivery (odds ratio 2.4, 95% confi-
dence interval 1.2, 4.9) after adjustment for birth weight,
Materials and Methods maternal age, and gestational age. We found a significantly
All 1810 women who delivered live-born infants and increased risk of cesarean delivery with increased birth
were discharged between June 1, 1997, and January 26, weight for nulliparas (2– 66.7%). Increasing maternal age
1998, from St. Joseph's Medical Center were identified increased the risk of cesarean delivery in all parity groups (P
< .05), but particularly among nulliparas (3–26.3%) (P <
through the hospital discharge data system. We ex- .001). Electively induced labors that ended in vaginal deliv-
cluded women with multiple gestations and women ery cost $273 more and required an average of 4 hours more
with gestations of less than 38 or at least 42 weeks.
in the hospital before delivery than did noninduced vaginal
Women were considered ineligible for vaginal delivery deliveries (P < .001).
if they had primary or secondary International Classifi- Conclusion: Elective induction significantly increased the
cation of Diseases, 9th Revision7 (ICD-9) codes of placenta risk of cesarean delivery for nulliparas, and increased in-
previa or abruption (641.0 – 641.23), vasa previa hospital predelivery time and costs. (Obstet Gynecol 2000;
(663.50 – 663.53), cord prolapse (663.0 – 663.03), or breech 2000 by The American College of Obstetri-
or transverse lie (652.0 – 652.33). Women who had in- cians and Gynecologists.)
duced or cesarean deliveries also were excluded if chartreview revealed active herpes, history of classic orvertical incision, more than two prior cesarean deliver- From 1983 through 1996, induction rates worldwide ies, or abdominal delivery without signs of labor. We ranged between 7.5% and 26%, with trends of increas- also excluded women who had scheduled primary or ing rates in the most recent years.1–6 In our hospital repeat cesarean deliveries if they had refused to deliver system, the rate for all inductions was 23.2% in 1997.
vaginally after a previous cesarean delivery or if no Studies of induction have continued to combine low- attempt was made to deliver vaginally. Women who and high-risk pregnancies, preterm and term induc- had cesarean deliveries and were included in the studyhad no prelabor indications for cesarean delivery and From the Departments of Obstetrics, Maternal-Fetal Medicine, and had abdominal deliveries because of nonreassuring Clinical Outcomes and Quality Improvement, Franciscan Health Sys-tem, Tacoma, Washington. fetal heart rate tracings after labor had begun or because VOL. 95, NO. 6, PART 1, JUNE 2000 of failure to progress after a course of labor was delivery dates and times were abstracted, and an admission-to-delivery time in hours was calculated for Women were considered ineligible for elective induc- each patient. That calculation was compared with the tion if they had primary or secondary ICD-9 codes of hospital system's electronically derived admission-to- pregnancy-induced hypertension (642.30 – 642.74), insu- delivery time.
lin-dependent diabetes mellitus (648.0), gestational di- We used SPSS 7.5.2 (Statistical Package for Social abetes (648.80 – 648.84), fetal growth restriction (656.50, Sciences; SPSS Inc., Chicago IL) for analysis. Categoric 656.51, and 656.63), oligohydramnios (658.00 – 658.03), relationships were analyzed using the ␹2 test. Catego- polyhydramnios (657.0 – 657.03), chorioamnionitis rized variables such as age and birth weight, which had (658.40 – 658.43), prolonged rupture of membranes a direct relationship to risk of cesarean delivery, also (658.20 – 658.23), anemia (648.20 – 648.24), or Rh isoim- were tested for significance using the ␹2 test for trend.8 munization. Two women who had vaginal deliveries Differences in averages were tested using the t test were excluded because of chart-confirmed complex where appropriate (age and birth weight) and the medical conditions (endocarditis and severe psychosis).
Mann-Whitney U test for nonnormally distributed vari- Inductions were identified among all deliveries using ables (parity, gestational age, cost, and inpatient prede- the induction logbook, hospital discharge data (ICD-9 livery and postdelivery lengths of stay).
codes 73.01, 73.4, 96.49, and 73.1), and charge data on Stepwise logistic regression model analysis was used induction agents (dinoprostone, oral or vaginal miso- to create a final model of risk factors for cesarean prostol, and oxytocin). We reviewed the charts of all delivery, with a significance level of P ⬍ .05 for variable eligible women. Labors were classified as inductions entry into the model. After the final model was pro- only when the induction agent or method was admin- duced, a second check of interaction terms of all vari- istered or performed before there were contractions 2–5 ables was done, with a significance level of P ⬍ .01 for minutes apart and before the cervix was dilated less interaction terms. The Hosmer-Lemeshow statistic for than 3 cm. The resulting cohort comprised women with goodness of fit was used to test the model.9 With that term pregnancies who delivered singleton infants in test, a model is judged to fit sufficiently well if the test cephalic presentation, without indication for induction assumptions are met and the P value resulting from the and without prelabor indications for cesarean delivery.
test is not significant (P ⬎ .05).
Parity information was collected from Washington State birth certificate data. Inpatient charts missing that information were reviewed and parity was recordedfrom prenatal records. The hospital's discharge data Of 1135 eligible subjects, 263 (23.2%) underwent elective system provided histories of cesarean delivery (ICD-9 induction of labor. One hundred seventy inductions code 654.21), age, actual direct cost, attending provider (64.6%) were done with up to three doses of oral type (certified nurse-midwife, family practitioner, ob- misoprostol 50 ␮g, 70 (26.6%) were done by oxytocin stetrician), and admission, delivery, and discharge infusion, 20 (7.6%) by artificial rupture of membranes dates and times.
only, and three (1.1%) with only prostaglandin (PG) gel.
Actual direct cost was determined by the hospital's Eighty-two (48.2%) of 170 misoprostol inductions also decision support department using Eclipsys (Eclipsys involved oxytocin infusion, and three of those also Corp., Delray Beach, FL), a cost-accounting relative- included use of PG gel. Three (4.2%) of 70 oxytocin weight method that summarizes direct costs for each inductions also included use of PG gel.
patient based on clinical and ancillary resource con- Women who underwent elective induction tended to sumption. Actual direct costs represent all nursing and be older, were more likely to have had prior cesarean supply costs generated by all contributing departments deliveries, had longer gestations, were more likely to that provide service to the patient; not included are receive epidural anesthetic, were less likely to be treated nondepartmental capital supplies, hospital administra- by midwives, delivered larger infants, and had signifi- tion costs, nursery-related costs, and anesthesiology cantly higher cesarean delivery rates compared with women who did not undergo induction (Table 1).
All records with invalid data on hospital admission- Although the average and median birth weights of to-delivery and delivery-to-discharge times were infants of women who underwent induction did not reviewed to determine true admission, delivery, and seem clinically different from those of infants of women discharge dates and times. A quality audit of who did not undergo induction, the infants of women admission-to-delivery calculations was done using 73 who underwent induction were statistically signifi- randomly selected charts from the study period. Moth- cantly heavier and a significantly larger proportion of ers' admission dates and times to delivery and infants' these infants weighed more than 4000 g.
918 Maslow and Sweeny
Elective Induction and Cesarean Delivery Obstetrics & Gynecology Table 1. Characteristics of Women, by Induction Status and Parity
(n ⫽ 263) (n ⫽ 872) (n ⫽ 103) (n ⫽ 349) (n ⫽ 160) (n ⫽ 523) Gestational age (wk) History of cesarean delivery Certified nurse-midwife Epidural anesthetic use Cesarean delivery Data are presented as mean (standard deviation) or n (%).
* P ⬍ .05.
† P ⬍ .001.
‡ P ⬍ .01.
Initial analysis of cesarean deliveries by parity re- underwent induction and whose infants were heavier vealed an almost three-fold increased risk among nul- than 4500 g. Two of the four nulliparas who did not liparas whose labor was induced compared with nul- undergo induction and whose infants were heavier liparas whose labor was not (relative risk 2.9, 95% than 4500 g had cesarean deliveries. Separate regression confidence interval [CI] 1.6, 5.2) and a two-fold increase models for nulliparas, primiparas, and multiparas were in cesarean deliveries among parous women who un- constructed based on those findings.
derwent induction compared with parous women who In the final multivariable logistic regression model did not (Table 1).
predicting cesarean delivery, induction remained a sig- Those data showed important differences in risks of nificant predictor of it (OR 2.4, 95% CI 1.2, 4.9), after cesarean delivery based on relationships between parity adjustment for the other important predictors of cesar- and birth weight categories and parity and maternal ean delivery among nulliparas only. Excluding all nul- age categories. Birth weight and maternal age appeared liparas with infant birth weights greater than 4000 g to have more pronounced effects on the risk of subse- resulted in a model in which elective induction and quent cesarean delivery for nulliparas than for primip- gestational age were the only significant predictors of aras or multiparas (Tables 2 and 3). Although nulliparas cesarean delivery (P ⬍ .01 and ⬍ .05, respectively; who underwent elective induction and delivered large model not shown). The unadjusted risk estimate for infants were not at statistically significantly greater risk cesarean delivery among nulliparas with infants who for cesarean than nulliparas who did not undergo weighed more than 4000 g was the following: OR 4.2, induction and who delivered large infants, small num- 95% CI 2.3, 7.6 (data not shown). A revised model that bers in the high birth weight categories might have included all variables in the final model (Table 4), plus prevented finding such an association. Cesarean deliv- the epidural anesthetic and provider type variables, did eries were done in the case of both nulliparas who not reduce significantly the association between induc- Table 2. Number of Cesarean Deliveries per 100 Deliveries,
by Parity and Birth Weight Table 3. Number of Cesarean Deliveries per 100 Deliveries,
by Parity and Age (n ⫽ 452) (n ⫽ 395) (n ⫽ 288) (n ⫽ 37) (n ⫽ 1135) (n ⫽ 452) (n ⫽ 395) (n ⫽ 288) (n ⫽ 1135) Significant for nulliparas (P ⬍ .001, ␹2 for trend) and all women as a Significant for nulliparas (P ⬍ .001, ␹2 for trend) and all women as a group (P ⫽ .003).
group (P ⫽ .021).
VOL. 95, NO. 6, PART 1, JUNE 2000 Maslow and Sweeny
Elective Induction and Cesarean Delivery Table 4. Logistic Regression Model for Cesarean Delivery:
nificantly longer for noninduced labors that ended in Nulliparas Only (n ⫽ 452) vaginal delivery than for induced labors that ended in vaginal delivery, the average difference of 0.3 hours (20 minutes) is probably not important clinically or finan- cially (Table 5).
Induction (yes, no) Birth weight category* 0.45 (0.30, 0.67) 0.64 (0.48, 0.85) Despite suspected widespread use of elective induction, (41 vs 38 – 40 wk) we found only ten published studies (MEDLINE search, OR ⫽ odds ratio; CI ⫽ confidence interval; NA ⫽ not applicable; 1970 –1998, using the terms "labor," "labour," "induc- NS ⫽ not significant.
Hosmer-Lemeshow goodness-of-fit statistic with 8 degrees of free- tion," "elective," "techniques," and "management") in- dom, P ⫽ .80. Both linear variables (birth weight category and age) volving more than 100 subjects that addressed elective tested P ⬍ .001 for linear-by-linear association with delivery outcome.
induction and a variety of outcome measures since the * Ordinally categorized as ⬍3000, 3000 –3499, 3500 –3999, 4000 – 4499, and 4500⫹ g. Risk estimate is interpreted as incremental increase in advent of continuous fetal monitoring and controlled risk with each increase in birth weight (eg, risk of cesarean delivery oxytocin infusion.10–19 "Elective" was often defined in among deliveries of infants ⬍3000 g ⫽ 0.45; risk for 3000 –3499 g different ways, and consistency of findings varied ac- infants 0.90; risk for infants weighing 4500⫹ g ⫽ 2.25).
† Ordinally categorized as ⬍20, 20–24, 25–29, 30–34, and 35⫹ y.
cording to whether nulliparas were evaluated as adiscrete group. Studies that examined cesarean deliveryrates without regard to parity for women who under- tion and cesarean delivery (OR 2.3, 95% CI 1.1, 4.9) went elective induction yielded inconclusive results on (Hosmer-Lemeshow P ⫽ .80).
cesarean delivery rates between subjects who had in- There were 27 cesarean deliveries among 683 parous duced labor and those who did not.10–12 women. For that group, the only important variable that Studies that evaluated nulliparas as a discrete group predicted cesarean delivery was prior cesarean delivery consistently found increased cesarean delivery rates (OR 38.1, 95% CI 14.5, 100.3). Excluding nulliparas and associated with elective induction,13,17–19 although these women who had had prior cesarean deliveries (n ⫽ 607) increases were not always statistically significant.13,17 In resulted in no variables that were significant for pro- those studies, the cesarean delivery rate for nulliparas ducing increased risk of cesarean delivery, including who underwent elective induction was between 6.2% elective induction.
and 11.7% higher than the base cesarean delivery rate Elective inductions that ended in vaginal delivery for nulliparas who had spontaneous labors. In the study cost $273 more (P ⬍ .001) and required an average of 4 by Yeast et al,18 197 nulliparas who underwent elective hours more in the hospital before delivery than did induction had a cesarean delivery rate of 16.2%, com- noninduced vaginal deliveries (P ⬍ .001) (Table 5). The pared with 7.9% among 4086 nulliparas who labored additional direct cost for elective inductions translated spontaneously. In that study, labor induction among into $63,882 for those that ended in vaginal delivery nulliparas (indicated and elective) was the most impor- during the 8-month study period. The additional ex- tant predictor of cesarean delivery, after adjustment for pense and in-hospital predelivery time associated with age, birth weight, and membrane status. By compari- elective inductions that ended in cesarean delivery son, risk of cesarean delivery among multiparas who compared with noninduced labors that ended in cesar- underwent elective induction was not significantly ean delivery were not statistically significant. Although higher than for multiparas who labored spontaneously.
the postdelivery in-hospital time was statistically sig- Macer et al17 described a subset of 24 nulliparas with Table 5. Average Costs and Lengths of Hospital Stay
Noninduced labor, Noninduced labor, cesarean delivery cesarean delivery (n ⫽ 234) (n ⫽ 835) (n ⫽ 29) (n ⫽ 37) Admission-to-delivery time (h) Delivery-to-discharge time (h) Data are presented as mean (standard deviation).
* P ⬍ .001 for induced labor compared with noninduced labor by the Mann-Whitney rank-sum test.
† P ⬍ .05 for induced labor compared with noninduced labor by the Mann-Whitney rank-sum test.
920 Maslow and Sweeny
Elective Induction and Cesarean Delivery Obstetrics & Gynecology Bishop scores less than or equal to 5, who proved to be Another weakness of this study was incomplete as- at increased risk for cesarean delivery. Although 12 of sessment of induction status within the spontaneous 24 subjects with low Bishop scores required cesarean labor group. We reviewed most charts, but not for all of deliveries, nulliparas with Bishop scores greater than 5 the subjects in the study. All charts for women who had had a cesarean delivery rate not statistically signifi- cesarean deliveries or electively induced labor were cantly higher than that among nulliparas who did not reviewed, which left 835 women who had vaginal undergo induction. In a population of nulliparas with deliveries but did not undergo induction that would be unfavorable cervices who underwent elective induc- noted by our detection methods. If elective inductions tion, there was a significantly increased risk of cesarean were missed in that group, and the misclassification among those who underwent induction before 42 was large enough, that could explain the increased risk weeks, compared with those who underwent induction of cesarean delivery among women who underwent at more than 42 weeks' gestation.19 elective induction. To estimate the misclassification Another key finding of our study was the $273 error, we studied a random sample of 219 women who additional cost per elective induction. For our hospital, had vaginal deliveries and no induction. Applying our that translated into $63,882 in additional costs for definition of elective induction, we found eight subjects women who underwent elective induction and had misclassified as having had no induction, a misclassifi- successful vaginal deliveries. With our MEDLINE cation rate of 3.65% (95% CI 0.55, 6.35). Charts for a search we were unable to locate other studies that subsample of 70 patients were reviewed by both au- focused on the cost-effectiveness of elective induction.
thors to check interobserver reliability, yielding an Our increased costs were principally the result of in- observed agreement of 98.6% and a ␬ statistic of .793. If creased rates of epidural anesthetic use among subjects we assume a maximum error rate of 6.35%, that trans- who underwent induction and the approximate in- lates to a shift of 55 subjects from the noninduced to the crease of 4 hours of predelivery time in the hospital.
induced vaginal delivery group. Our revised cesarean Many studies reported shorter labor times for induced delivery rates would be 9.2% (induced) and 4.2% (not deliveries compared with spontaneous ones. Our study induced), still statistically significant (P ⬍ .01).
did not measure time from induction of labor to spon- As it is practiced currently in our hospital system, taneous delivery and hence could not produce those elective induction resulted in higher direct costs per patient and increased cesarean delivery rates for nullip- Unmeasured or unvalidated variables related to ce- aras with term pregnancies who underwent elective sarean delivery are other weaknesses of this study. Low induction, regardless of age, birth weight, provider Bishop score was one variable we were unable to record type, or epidural anesthetic use. Thus, elective induc- and that has been associated with increased cesarean tion of labor in nulliparas should be discouraged.
delivery rates.17,20 Significant associations between epi-dural anesthetic use and cesarean delivery–related vari-ables were reported recently in nulliparas who had spontaneous labors.21,23 Data on epidural anesthetic use 1. Fleissig A. Prevalence of procedures in childbirth. BMJ 1993;306: were available to us, but it was impossible to tell from the charge data whether epidural anesthetic was used 2. Elferink-Stinkens PM, Van Hemel OJ, Brand R. Differences in for labor pain or for preparation for cesarean delivery.
obstetrical intervention rates between Dutch hospitals. Eur J Obstet The association between provider type and cesarean Gynecol Reprod Biol 1994;53:165–73.
delivery or labor induction might have involved inac- 3. Lampe LG. Indications for induction of labour. Eur J Obstet Gynecol Reprod Biol 1996;65:71–3.
curate assignments of provider type. Local data on 4. Wilailak S, Saropala N, Chaturachinda K. Elective induction of provider type were compromised by the propensity of labor: Ramathibodi Hospital (Jan-Jun, 1990). J Med Assoc Thai medical record coders to assign cesarean deliveries to 1993;76:44 –7.
obstetricians, even when a midwife had treated the 5. Eggebo TM, Okland I, Braut RM. Labor induction. Quality assur- labor up to the point of decision for cesarean delivery.
ance of a method [in Norwegian]. Tidsskr Nor Laegeforen 1993;133:2686 –90.
Although we were able to adjust for epidural anesthetic 6. Wang D. Cesarean section: An analysis of 16,975 cases [in Chinese].
use and provider type in our modeling techniques, the Chung Hua I Hsueh Tsa Chih (Taipei) 1990;70:10 –2.
complete accuracy of those variables has not been 7. World Health Organization. International Classification of Dis- assessed by chart audit. There are no reliable studies eases, 9th Revision. Geneva: World Health Organization, 1977.
that contend that low Bishop score, epidural anesthetic 8. Fisher LD, Van Belle G. Categorical data: Contingency tables. In: Fisher LD, Van Belle G, eds. Biostatistics: A methodology for the use, or provider type can explain the increased risk of health sciences. New York: Wiley, 1993:246 –303.
cesarean delivery for nulliparas who had elective induc- 9. Kleinbaum DG, Kupper L, Muller K. Testing hypotheses in mul- tiple regression. In: Kleinbaum DG, Kupper L, Muller K, eds.
VOL. 95, NO. 6, PART 1, JUNE 2000 Maslow and Sweeny
Elective Induction and Cesarean Delivery Applied logistic regression analysis and other multivariate meth- labor induction in postterm patients. Observations and outcomes.
ods. New York: Wiley, 1989:124 – 43.
J Reprod Med 1992;37:157– 61.
10. Kato K, Nagata I, Furuya K, Seki K, Makimura N. Programmed 21. Lieberman E, Lang JM, Cohen A, D'Agostino R Jr, Datta S, induction of labor for primiparous women to ensure daytime Frigoletto FD Jr. Association of epidural analgesia with cesarean delivery. Asia Oceania J Obstet Gynaecol 1987;13:405–15.
delivery in nulliparas. Obstet Gynecol 1996;88:993–1000.
11. Cole RA, Howie PW, Macnaughton MC. Elective induction of 22. Turcot L, Marcoux S, Fraser WD. Multivariate analysis of risk labour. A randomised prospective trial. Lancet 1975;7910:767–70.
factors for operative delivery in nulliparous women. Canadian 12. Vierhout ME, Out JJ, Wallenburg HC. Elective induction of labor: Early Amniotomy Study Group. Am J Obstet Gynecol 1997;176: A prospective clinical study. I. Obstetric and neonatal effects. J Perinat Med 1985;13:155– 62.
23. Malone FD, Geary M, Chelmow D, Stronge J, Boylan P, D'Alton 13. Prysak M, Castronova FC. Elective induction versus spontaneous ME. Prolonged labor in nulliparas: Lessons from the active man- labor: A case-control analysis of safety and efficacy. Obstet Gynecol agement of labor. Obstet Gynecol 1996;88:211–5.
14. Combs CA, Singh NB, Khoury JC. Elective induction versus spontaneous labor after sonographic diagnosis of fetal macroso- Address reprint requests to: mia. Obstet Gynecol 1993;81:492– 6.
Arthur S. Maslow, DO, MSc 15. Leaphart WL, Meyer MC, Capeless EL. Labor induction with a Department of Maternal and Fetal Medicine prenatal diagnosis of fetal macrosomia. J Matern Fetal Med 1997;6:99 –102.
Franciscan Health System 16. Gonen O, Rosen DJ, Dolfin Z, Tepper R, Markov S, Fejgin MD.
1717 South J Street, 12th Floor Induction of labor versus expectant management in macrosomia: Tacoma, WA 98401 A randomized study. Obstet Gynecol 1997;89:913–7.
17. Macer JA, Macer CL, Chan LS. Elective induction versus sponta- neous labor: A retrospective study of complications and outcome.
Am J Obstet Gynecol 1992;166:1690 – 6.
18. Yeast JD, Jones A, Poskin M. Induction of labor and the relation- Received August 19, 1999. ship to cesarean delivery: A review of 7001 consecutive inductions.
Received in revised form November 24, 1999. Am J Obstet Gynecol 1999;180:628 –33.
Accepted December 17, 1999. 19. Wigton TR, Wolk BM. Elective and routine induction of labor. A retrospective analysis of 274 cases. J Reprod Med 1994;39:21– 6.
Copyright 2000 by The American College of Obstetricians and 20. Shaw KJ, Medearis AL, Horenstein J, Walla CA, Paul RH. Selective Gynecologists. Published by Elsevier Science Inc.
922 Maslow and Sweeny
Elective Induction and Cesarean Delivery Obstetrics & Gynecology

Source: http://www.nurturinglife.com.au/wordpress/wp-content/uploads/2013/02/Elective_Induction_of_Labor_as_a_Risk_Factor_for.26.pdf