American Journal of Obstetrics and Gynecology (2004) 190, 129e34

www.elsevier.com/locate/ajog

Risk of cesarean delivery in nulliparous women at greater

than 41 weeks’ gestational age with an unengaged vertex
Karen S. Shin, MD,a Katherine L. Brubaker, MD,a,* Lynn M. Ackerson, PhDb

Department of Obstetrics and Gynecology,a Kaiser Permanente Medical Center, and the Division of Research,b
Kaiser Permanente, Santa Clara and Oakland, Calif

Received May 2, 2003; revised June 25, 2003; accepted July 15, 2003

––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
KEY WORDS Objective: The purpose of this study was to determine whether an unengaged vertex significantly
Cesarean delivery increased the risk of cesarean delivery in nulliparous patients at 41 weeks or greater.
Nulliparity Study design: The medical records from all nulliparous patients greater than 41 weeks’ gestation
Engagement delivered at a single institution were reviewed. Patients undergoing both spontaneous and in-
duced labor were included. Multivariate analyses were used to compare the influence of admis-
sion fetal station versus induction of labor on the risk of cesarean delivery.
Results: Four hundred forty-eight nulliparous women at greater than 41 weeks’ gestation were
delivered at our institution during the study period. Sixty-two percent of these patients underwent
induction of labor. There was a statistically significant increase in cesarean delivery rate com-
pared with station (6% of patients at ÿ1 station, 20% at ÿ2 station, 43% at ÿ3 station, and
77% at ÿ4 station; P = .001). Compared with patients with an engaged vertex, patients with
an unengaged vertex had 12.4 times the risk of cesarean delivery. Most of the cesarean deliveries
were performed for failure to progress. On the basis of multivariate analysis, the odds of cesarean
delivery were better predicted by fetal station than induction of labor.
Conclusion: Nulliparous patients at 41 weeks or greater with an unengaged vertex are 12.4 times
more likely to be delivered by cesarean section than a patient with an engaged vertex.
Ó 2004 Elsevier Inc. All rights reserved.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

The optimal management of prolonged pregnancies a pregnancy to be postdates before 42 completed week,1
has been controversial. Although some do not recognize others have recommended closer surveillance and/or in-
duction at 41 weeks.2,3 A recent American College of
Obstetricians and Gynecologists Practice Pattern indi-
Presented at the First Annual Meeting of the Kaiser Permanente
Medical Center Northern California Department of Obstetrics and Gyn-
cated that either induction or expectant management
ecology Resident Research Symposium, Oakland, Calif, May 31, 2003. will result in good outcomes among postterm patients
To be published in abstract form in The Permanente Journal Summer with unfavorable cervices and without additional compli-
Issue (June/July) in a special section for abstracts from the First Annual cations.4 Although many studies have attempted to quan-
Resident Research Symposium. tify the risk associated with either expectant management
* Reprint requests: Katherine Brubaker, MD, Department of
Obstetrics and Gynecology, 770 E Calaveras Blvd, Milpitas, CA
or induction of the prolonged pregnancy,3,5,6 it is not
95035. clear how the cesarean birth rate is affected by the selec-
E-mail: katherine.brubaker@kp.org. tion of one strategy over another.4

0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/S0002-9378(03)00909-8
130 Shin, Brubaker, and Ackerson

delivery for a nulliparous patient at or greater than 41

Table I Patient characteristics (n = 448)
weeks’ gestational age.
Variable No. (%)
Station of vertex on admission Material and methods
ÿ5 1 (0.2%)
ÿ4 39 (8.7%) This is a retrospective cohort study of cesarean section
ÿ3 148 (33.0%) rates in nulliparous women at or beyond 41 weeks’ ges-
ÿ2 164 (36.6%)
tational age. This study was performed at Kaiser Perma-
ÿ1 70 (15.6%)
0 23 (5.1%)
nente Medical Center in Santa Clara, Calif, which is
+1 2 (0.4%) a community-based teaching hospital with a perinatal re-
+2 1 (0.2%) ferral center. Approximately 3000 deliveries are per-
Spontaneous labor 169 (38%) formed a year. A level III neonatal intensive care unit
Induced labor 279 (62%) and 24-hour inhouse physician coverage with full resident
Induction indication* physician participation are available. The primary cesar-
Postdates 169 (38%) ean section rate at our institution is approximately 15%
Nonreassuring testing/ 66 (15%) (Kaiser Santa Clara internal data). Approval for the study
oligohydramnios was obtained from the Institutional Review Board.
Ruptured membranes 28 (6%) Birth logs from January 1999 to May 2002 at our
Other (including 17 (4%)
institution were examined to identify all nulliparous
preeclampsia, growth
restriction, diabetes,
patients with singleton pregnancies and vertex presenta-
and macrosomia) tion delivered at or greater than 41 weeks’ gestational
Indication unclear 6 (2%) age. Routine dating ultrasound examinations to confirm
Gestational age (wk) 41.4 [0.4] (40.9-43.1) menstrual dates are performed on every patient initiat-
(mean [SD] range) ing prenatal care at our institution at 16 to 18 weeks’ ges-
Maternal age (y) (mean [SD] range) 27.6 [5.7] (15-45) tational age. Ultrasounds are also performed on patients
Maternal weight (pounds) 179 [32] (113-315) transferring care from another institution if no prior dat-
(mean [SD] range) ing ultrasound had been performed. Patients undergoing
* Some patients had more than one indication. both spontaneous labor and induction of labor were in-
cluded as the major point of interest was the outcome of
patients at this gestational age. Labor was induced with
One of the highest risk patient populations for cesar- oxytocin, unless the cervix was unfavorable according to
ean delivery is the nulliparous patient with an ‘‘unproven the Bishop score. Misoprostol was the most common
pelvis.’’ Gabbe states that engagement defined as the de- agent used for cervical ripening. Comprehensive, com-
scent of the biparietal plane of the fetal head to a level puterized records of the prenatal and intrapartum course
below the pelvic inlet demonstrates adequate room for were reviewed to determine station of vertex on admis-
the fetus in the pelvic inlet. Because nulliparous women sion to labor and delivery. Maternal age, maternal
usually undergo engagement of the fetal head by 36 weight and weight gain, infant birth weight, use of anal-
weeks, its absence by this time may demonstrate cepha- gesia, use and type of labor induction or augmentation
lopelvic disproportion.2 This was well described by medications, and presence of complications in labor or
Friedman and Sachtleben7 who reported that high fetal at delivery were also recorded.12-16 Records were reviewed
station was associated with higher cesarean section to determine indication for cesarean section. If fetal intol-
rates. A more recent study by Roshanfekr et al8 found erance to labor was present without failure to progress,
that engagement at the time of active labor was associ- this was considered the primary indication. However,
ated with a lower risk of cesarean delivery. Fourteen per- if both indications were present, failure to progress was
cent of nulliparous patients with an unengaged vertex considered the primary indication. Failure to progress
underwent cesarean section compared with 5% in those included both arrest of dilatation and arrest of descent.
with an engaged vertex. This retrospective cohort study Table I summarizes selected characteristics of the study
included all patients at 37 to 42 weeks and excluded pa- population. Given the observational nature of the study,
tients undergoing induction of labor. patients were not randomly assigned to one treatment
Although previous studies have demonstrated an group or another. Clinical decisions regarding labor man-
increased risk of cesarean delivery in term nulliparous agement were made based on the treating physician’s
women in active labor with an unengaged vertex,9-13 judgment.
our review of the literature finds no study limit- Determination of fetal station is traditionally the
ing the focus to nulliparous patients at or beyond 41 most subjective element of the pelvic examination. At
weeks. The purpose of our study is to determine how our institution, fetal station is measured by the +5 to
well an unengaged vertex predicted the risk of cesarean ÿ5 scoring system, which considers the level of the pre-
Shin, Brubaker, and Ackerson
Table II Comparison of engaged group (stations ÿ1, 0, +1) and unengaged group (stations ÿ3, ÿ4, ÿ5). All patients at ÿ2 station were
excluded from this analysis to better delineate the difference between the unengaged and engaged population and to exclude possible
inaccurate examinations
Unengaged group (n = 188)
Initial dilation (cm) (mean [SD]) 1.3 (1.1)
Initial effacement (%) (mean [SD])* 41 (27)
Spontaneous labor 12%
Induced labor 88%
Vaginal delivery 42%
Cesarean section 50%
* Data missing for two patients.
senting part at the ischial spines to be 0 station and each
unit to represent a centimeter above or below this base-
line. If multiple discrepant examinations were performed
in a short time of admission, the examination of the
most senior physician was recorded.
Although the strict definition of engagement is the
presenting part at a 0 station, the difference of a centime-
ter can be obscure. Therefore, patients at 0 station, G1
cm, (+1, 0, ÿ1 stations) were considered the ‘‘engaged’’
group. Likewise, patients at ÿ4 station, G1 cm (ÿ3, ÿ4,
ÿ5 stations) were considered the ‘‘unengaged’’ group.
Although data from all patients were collected, patients
admitted at ÿ2 station were not included in this part of
the analysis to better delineate the difference between
the unengaged vertex from an engaged vertex. This also
served to exclude examinations that may have recorded
patients at ÿ3 and ÿ1 station as being at ÿ2 station.
Moreover, it was thought the designation of ÿ2 station
may represent a relatively unsure examination.
All analyses were conducted with SAS version 8.2.
All tests were two sided with a type I error rate of
5%. Continuous variables were summarized by using
means and SD. Dichotomous variables were summa-
rized by using frequencies. Comparison between the
engaged and unengaged groups was performed with
two-sample t tests for continuous variables and c2 tests
(or Fisher exact tests if cells were small) for dichotomous
variables.
The study population was limited to nulliparous pa-
tients at or greater than 41 weeks’ gestation. Induction
of labor has already been shown to be a risk factor for ce-
sarean delivery,4 a risk that is highly correlated with ini-
tial dilation.17 Its influence was especially relevant in our
study population in which a majority of patients under-
going induction also had an unengaged vertex (Table
II). Therefore, a multivariate logistic regression model
was generated to evaluate the ability of induction of labor
versus engagement to predict cesarean delivery in this
study population. Two multivariate models were generat-
ed: one excluded induction of labor but included station
of fetal vertex in the calculation of risk of cesarean deliv-
ery, and the other included induction of labor and exclud-
131
Engaged group (n = 95) P-value
3.5 (1.7) !.001
86 (18) !.001
71% .001
30% .001
79% .001
5% .001
ed station. The results of those models are reported as
odds ratio (OR) and 95% CI. The ability of a model to
accurately predict an outcome can be measured by the
R2 statistic. If a model perfectly predicted each data
point, the R2 would be 100%. The C statistic is another
measure of how well the model predicts the correct out-
come. Its value ranges from 0 to 1, with 1 being perfect
predictive ability.
Results
Four hundred forty-eight patients delivered at or greater
than 41 weeks were identified during the study period.
The majority of patients at 41 weeks’ gestation or great-
er underwent induction of labor (62%). Similar to find-
ings from other studies, prolonged pregnancy was the
most common indication for induction of labor at this
gestational age.11 The total cesarean delivery rate for
the 448 patients delivering at or greater than 41 weeks
was 30% compared with the primary cesarean delivery
rate of 15% at our institution (Kaiser internal data).
The primary cesarean rate at our institution represents
all patients without prior cesarean deliveries and there-
fore includes multiparous and nulliparous patients as
well as cesarean deliveries performed for other indica-
tions (such as twins, abnormal presentation). There
was a statistically significant increase in cesarean deliv-
ery rate compared with station (6% of the 70 patients
at ÿ1 station, 20% of the 164 patients at ÿ2 station,
43% of the 148 patients at ÿ3 station, 77% of 39
patients at ÿ4 station; P = .001). None of the patients
at +1 and +2 station underwent cesarean delivery,
whereas 4% of 23 patients at 0 station underwent cesar-
ean delivery. The only patient at ÿ5 station did not
undergo cesarean section. Labor was induced with
misoprostol at 41 weeks 4 days. After 50 hours on the
labor and delivery unit, she underwent vacuum-assisted
vaginal delivery for prolonged second stage of labor.
Delivery was complicated by severe perineal lacerations,
postpartum hemorrhage, and severe anemia (hematocrit
of 16%). The patient was discharged home on the third
postpartum day without a blood transfusion.
132
Table III Odds ratio of cesarean delivery based on bivariate and multivariate analyses
Bivariate: how this variable predicts
Variable cesarean section (OR [95% CI])
Initial dilation
!2 cm 17 (2.2-131)
2-4 cm 3.7 (0.5-29.7)
5-10 cm Reference
Induction
Yes 4.2 (2.2-7.9)
No Reference
Station
ÿ5 to ÿ3 18.0 (7.0-49.3)
ÿ1 to +1 Reference
N 280
R2
C statistic
NA, Not applicable.
* ORs were also adjusted for maternal age, maternal weight, fetal weight, presence of diabetes, use of epidural, and presence of chorioamnionitis.
Compared with patients with an engaged vertex, pa-
tients with an unengaged vertex had 18 times the odds
of cesarean delivery (Table III). Most of the cesarean de-
liveries were performed for arrest of dilation or descent:
78 of 94 cesarean sections in the unengaged group were
performed for arrest of dilation or descent, whereas 12
of 94 were performed for fetal intolerance to labor.
For 2 of the women, other reasons were given, and for
2, the indication for cesarean delivery was not available.
There was no statistically significant increased rate of ce-
sarean section for arrest of dilation or descent versus fe-
tal intolerance to labor (2 of 5 and 3 of 5, respectively) in
the engaged group. Other factors such as maternal
weight gain, birth weight, or use of epidural analgesia
were not shown to be significantly different between
the two groups (data not shown). There was no differ-
ence in fetal outcome or maternal complications.
Table III describes the results of the two multivariate
logistic regression models generated to evaluate the abil-
ity of station versus induction to predict risk of cesarean
delivery. Patients with an unengaged vertex had 12.4
times the odds of cesarean delivery compared with those
with an engaged vertex after adjustment of initial dila-
tion and other factors. On the other hand, patients un-
dergoing induction of labor had 1.7 times the odds of
those undergoing spontaneous labor, although this odds
ratio was not statistically significant, possibly because of
low power. (To detect an OR of 1.7 with 80% power,
a two-sided test at the 5% significance level and the
same ratio of sample sizes in the two groups as described
in this study, we would require a total of 588 patients).
The model using engagement better predicted cesarean
delivery than the model using induction as demonstrated
by the R2 statistic (0.29 vs 0.22, respectively).
As expected, the odds of cesarean delivery in patients
undergoing induction of labor appear to be highly corre-
Shin, Brubaker, and Ackerson
Multivariate 1: station but not Multivariate 2: induction but not
induction (OR [95% CI])* station (OR [95% CI])*
5 (0.4-70.6) 9.1 (0.9-87.7)
2.4 (0.2-34.9) 2.6 (0.3-24.1)
NA 1.7 (0.7-4.1)
12.4 (4.2-36.4) NA
Reference
280
0.29 0.22
0.82 0.78
lated with initial dilation. This may account for why
induction of labor was not significant in the second
multivariate analysis (collinearity problem). For exam-
ple, the majority of patients in the unengaged group were
less than or equal to 1 cm in dilation, whereas the majority
of engaged patients were dilated to greater than 2 cm.
Moreover, the majority of patients in the unengaged
group were induced compared with fewer than 50% of pa-
tients who were engaged. This collinearity problem was
accounted for by reanalyzing the multivariate models af-
ter excluding initial dilation. In this analysis, induction of
labor was statistically significantdincreasing the odds to
4.3 times that of spontaneous labor. However, the model
using engagement still outperformed the model using in-
duction as demonstrated by an R2 of 0.28 versus 0.18, re-
spectively (Table IV).
Comment
This study was undertaken to investigate the perceived in-
creased risk of cesarean delivery among nulliparous pa-
tients with an unengaged vertex at greater than 41
weeks’ gestational age. Clinical experience suggests these
women require lengthier stays on labor and delivery with
the eventual need for cesarean sectiondoften for dys-
tocia. The findings of our study showed that in fact these
women have 12.4 times the risk of cesarean delivery com-
pared with those with an engaged vertex. Although signif-
icantly more patients in the unengaged group underwent
induction of labor, multivariate analyses demonstrated
that fetal station was more predictive of cesarean delivery
than induction of labor. In a study of induction of labor in
postdates pregnancies in both multiparous and nullipa-
rous patients, Alexander et al18 found that factors intrin-
sic to the patient rather than induction of labor were
Shin, Brubaker, and Ackerson
Table IV Multivariate analyses excluding initial dilation
Multivariate 1: station but not
Variable induction (OR [95% CI])*
Induction
Yes NA
No
Station
ÿ5 to ÿ3 18.9 (6.9-51.7)
ÿ1 to +1 Reference
N 282
R2 0.28
C statistic 0.82
NA, Not applicable
* ORs were also adjusted for maternal age, maternal weight, fetal weight, presence of diabetes, use of epidural, and presence of chorioamnionitis.
associated with a higher likelihood of cesarean delivery.
Although labor induction appeared to increase the overall
cesarean delivery rate, correction for confounding factors
such as an undilated cervix, epidural analgesia, more ad-
vanced gestational age, and nulliparity revealed that labor
induction itself did not significantly increase the odds of
cesarean delivery (OR 1.1; 95% CI 0.9-1.2).19 Unlike their
study, however, we limited our analysis to postdates preg-
nancies in nulliparous patients and found that fetal sta-
tion was more predictive than dilation in predicting
successful vaginal delivery.
The results of our study also support traditional ob-
stetrical teaching that station of the fetus acts as an ‘‘in-
ternal pelvimeter.’’ 2 A significant, biologically plausible
trend in cesarean delivery rate correlating with fetal sta-
tion was demonstrated with a rate of 77%, 43%, 20%
and 6% for ÿ4, ÿ3, ÿ2, and ÿ1 station, respectively,
on admissionda finding echoed in previous studies of
station and risk of cesarean delivery.7-11 Future prospec-
tive studies would be helpful in confirming our data.
The weaknesses of this study include the inherent lim-
itations of a retrospective study and a relatively small
study number. However, because of the large difference
in cesarean section rate between the unengaged and en-
gaged group, we believe these results would be repro-
ducible in a study including a much larger population
of patients.
We chose to exclude patients at ÿ2 station from both
of our groups (unengaged and engaged) to better delin-
eate the groups and to reduce misclassification. How-
ever, if the ÿ2 station patients were to be included in
the unengaged group, the overall cesarean delivery rate
in that group would be 36% compared with 5% in the
engaged group. This difference is still statistically signif-
icant (P = .001).
The potential confounding influence of the inclusion
of patients undergoing both spontaneous labor and in-
duction of labor was addressed by the development of
several multivariate models. Multivariate analysis dem-
onstrated that fetal station better predicted the risk of
133
Multivariate 2: induction but not
station (OR [95% CI])*
4.3 (2.1-8.5)
NA
282
0.18
0.74
cesarean section than induction of labor. However, there
may also be an additive effect of both labor induction
and high fetal station in a patient’s risk of cesarean
delivery. Therefore, it may be useful to limit the
analysis in a future prospective study to only induction
of labor.
Nulliparous patients at 41 weeks or greater with an
unengaged vertex have a higher risk for cesarean deliv-
ery. This increased risk was independent of induction,
although induction and initial dilation were again iden-
tified as risk factors for cesarean delivery.
References
1. Cunningham GF, Grant NF, Leveno KJ, Gilstrap LC, Hauth JC.
Williams’ obstetrics. 21st ed. Norwalk (CT): Appleton & Lange;
2001. p. 729-42.
2. Divon MY. Prolonged pregnancy. In: Gabbe SG, Niebyl JR,
Simpson JL, editors. Obstetrics: normal and problem pregnancies.
4th ed. Philadelphia: Churchill Livingstone; 2002.
3. Dyson DC, Miller PD, Armstrong MA. Management of prolonged
pregnancy: induction of labor versus antepartum fetal testing. Am J
Obstet Gynecol 1987;156:928-34.
4. American College of Obstetricians and Gynecologists. Management
of postterm pregnancy. Washington, DC: The College; 1997.
Practice patterns No.: 6.
5. National Institute of Child Health and Human Development
Network of Maternal-Fetal Medicine Units. A clinical trial of
induction of labor versus expectant management in the post-term
pregnancy. Am J Obstet Gynecol 1994;170:716-23.
6. Hannah ME, Hannah WJ, Hellmann J, Hewson S, Milner R,
Willan A. Induction of labor as compared with serial antenatal
monitoring in post-term pregnancy: a randomized controlled trial.
The Canadian Multicenter Post-term Pregnancy Trial Group. N
Engl J Med 1992;326:1587-92.
7. Friedman EA, Sachtleben MR. Station of the fetal presenting part,
VI: arrest of descent in nulliparas. Obstet Gynecol 1976;47:129-36.
8. Roshanfekr D, Blakemore KJ, Lee J, Hueppchen NA, Witter FR.
Station at onset of active labor in nulliparous patients and risk of
cesarean delivery. Obstet Gynecol 1999;93:329-31.
9. Falzone S, Chauhan SP, Mobley JA, Berg TG, Sherline DM, Devoe
LD. Unengaged vertex in nulliparous women in active labor.
J Reprod Med 1998;43:676-80.
134
10. Murphy K, Shah L, Cohen WR. Labor and delivery in nulliparous
women who present with an unengaged fetal head. J Perinatol
1998;18:122-5.
11. Yeast JD, Jones A, Poskin M. Induction of labor and the
relationship to cesearean delivery: A review of 7001 consecutive
inductions. Am J Obstet Gynecol 1999;180:628-33.
12. Witter FR, Caulfield LE, Stoltzfus RJ. Influence of maternal
anthropometric status and birth weight on the risk of cesarean
delivery. Obstet Gynecol 1995;85:947-51.
13. Turner MJ, Rasmussen MJ, Boylan PC, MacDonald D, Stronge
JM. The influence of birth weight on labor in nulliparas. Obstet
Gynecol 1990;76:159-63.
14. Witter FR, Repke JT, Niebyl JR. The effect of maternal age on
primary cesarean rate. Int J Gynaecol Obstet 1988;27:51-5.
Shin, Brubaker, and Ackerson
15. 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.
16. Maslow AS, Sweeny AL. Elective induction of labor as a risk
factor for cesarean delivery among low-risk women at term. Obstet
Gynecol 2000;95:917-22.
17. Seyb ST, Berka RJ, Socol ML, Dooley SL. Risk of cesarean
delivery with elective induction of labor at term in nulliparous
women. Obstet Gynecol 1999;94:600-7.
18. Alexander JM, McIntire DD, Leveno KJ. Prolonged pregnancy:
induction of labor and cesarean births. Obstet Gynecol 2001;97:
911-5.
19. Bishop EH. Pelvic scoring for elective induction. Obstet Gynecol
1964;24:266.