Original Research

ajog.org

OBSTETRICS

Descent of the fetal head (station) during the first

stage of labor
Emily F. Hamilton, MD; Gabrielle Simoneau, MSc; Antonio Ciampi, PhD;
Philip Warrick, PhD; Kathleen Collins, RN; Samuel Smith, MD; Thomas J. Garite, MD

BACKGROUND: High station at specific points in the first stage of

labor, such as a floating head on admission, or at 4-cm dilation or when
arrest of dilation occurs, is associated with higher rates of failure to deliver
vaginally. Therefore it could be useful to know if station is within an expected range at a given dilation during first stage. Arrest of descent disorders have been defined thus far on criteria applicable in the second
stage. Statistical modeling is an attractive methodology to characterize the
relationship between station and dilation because the resulting mathematical expressions could be used as a reference for comparison in the
future. In addition, they can be used to produce a finely graded assessment of descent using numerical terms such as percentile rankings. A 2step approach to potentially improving the assessment of station could be
to develop a statistical model that describes the general relationship between station and dilation in the first stage of uncomplicated births and
then determine if such a model would have identified births with complications related to poor labor progress. Given the complex nature of labor
data, especially the imprecision of dilation and station measurement, it is
not immediately evident that such a model is identifiable or what its
precision would be.
OBJECTIVE: We sought to characterize in mathematical terms
the relationship of station to dilation during the first stage of labor

Introduction
High station in the rst stage can be a
harbinger of cesarean or difcult birth.
Several clinical studies have reported
that high station at certain points in the
rst stage of labor, such as a oating head
on admission, or at 4-cm dilation or
when arrest of dilation occurs, is associated with higher rates of failure to
deliver vaginally.1-7 Therefore it could be
useful to know if station is within an
expected range at specic dilations during the rst stage. Although descent
begins during the rst stage, arrest disorders of descent are essentially pass-fail

Cite this article as: Hamilton EF, Simoneau G, Ciampi A,

et al. Descent of the fetal head (station) during the
first stage of labor. Am J Obstet Gynecol 2016;214:
360.e1-6.
0002-9378/$36.00
2016 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ajog.2015.10.005

for nulliparous and multiparous women with spontaneous vaginal births.

STUDY DESIGN: This retrospective cohort study included 28,121
exams from 5555 women with singleton cephalic presentations at
37
weeks gestation with electronic fetal monitoring tracings, who delivered
vaginally without instrumentation and had 5-minute Apgar scores >6 at 2
academic community referral hospitals in 2012 through 2013. Women
with a previous cesarean birth were excluded. We used longitudinal statistical techniques suitable to biological data that were irregularly sampled
with repeated measures over time.
RESULTS: A linear relationship was observed between station and
dilation. For both nulliparous and multiparous women the final model was
a linear regression with random effects for intercept and slope and a firstorder autoregressive correlation structure. The 5th-95th range of station at
any given dilation spanned about 3-4 cm.
CONCLUSION: Our results demonstrate a general trend of increasing
descent of the presenting part as dilation advances during the first stage of
labor in women who delivered vaginally without instrumentation. We
propose that the mathematical expressions describing this relationship
may be valuable in the assessment of first-stage labor progression.
Key words: fetal descent, fetal station, first stage of labor

criteria applicable in the second stage

only.8-11
A 2-step approach to potentially
improving the assessment of descent
could be to develop a statistical model
that describes the general relationship
between station and dilation in the rst
stage of labor in uncomplicated births
and then determine if such a model
could identify births with complications
related to labor progress disorders.
Statistical modeling is an attractive
methodology because the resulting
mathematical expressions may be used
as a reference for comparison in the
future. In addition, they can be used to
produce a nely graded assessment of
descent using numerical terms such as
percentile rankings.
The relationship between dilation and
station in the rst stage has been shown
indirectly by plotting both dilation and
station over time on the same graph as in
the classic Friedman or contemporary

360.e1 American Journal of Obstetrics & Gynecology MARCH 2016

labor curves.12-14 A statistical model

of the relationship between station
and dilation has not been reported
previously.
The objective of this study was
to create a mathematical expression
describing the relationship between
station and dilation during the rst
stage for nulliparous and multiparous
women with spontaneous vaginal
deliveries.

Materials and Methods

In this retrospective cohort study,
deidentied data were extracted from
the departmental electronic perinatal
database for the clinical variables on all
births between Jan. 1, 2012, and Dec.
31, 2013, at 2 acute care, academic
community teaching hospitals and
regional referral centers in the
Baltimore-Washington corridor, MedStar Franklin Square Medical Center
and MedStar Washington Hospital

ajog.org
Center. The inclusion criteria were all
labors with singleton cephalic presentations at
37 weeks gestation with
electronic fetal monitoring tracings,
delivered vaginally without instrumentation and with 5-minute Apgar
scores >6. Women with a previous
cesarean birth were excluded.
All data including cervical dilation,
effacement and station and clinical
data were extracted from the perinatal
electronic medical record, PeriBirth
(PeriGen, Cranbury, NJ.) Dilation and
station were measured in centimeters
where station values could range from
e5 to 5.
First we examined the data using
scatter plots, trajectory plots, and
variograms.
Based on the observation of these
plots and inherent biological variation
in the process of labor, we chose a
modeling approach from the extended
linear mixed model family. We t our
data to several models of increasing
complexity within the extended linear
mixed model family, and chose the nal
model using the Akaike information
criterion.15
Since both fetal station and cervical
dilation are closely related to time, we
also tested the effect of time by adding
minutes elapsed as a covariate to the
model.

OBSTETRICS
We checked the nal model regarding
assumptions of normality.
All statistical analyses were conducted
using R, Version 3.0.2.16
This study was reviewed and approved
by the MedStar Research Institute.

Results
The study sample included data from
5555 labors and 28,121 exams recorded
during the 24-hour period ending with
spontaneous vaginal birth. Characteristics of the study population are summarized in Table 1.
We excluded 210 examinations
because either dilation or station values
were missing, 441 examinations that
were in the second stage, 83 examination
with obvious date and time errors, and
13 examinations with obvious sign errors that were inuential points. For
example, after a labor with progressive
descent in a multiparous patient a station
value was entered as e4 and the baby
delivered vaginally 20 minutes later. The
nal data set included 5535 labors with
27,374 observations. There were 14,320
observations in 2507 nulliparous births,
and 13,054 observations in 3028
multiparous births. Figure 1 shows the
percentage of examinations at each level
of station.
The preliminary analysis with scatter
plots showed a clear relationship

TABLE 1

General characteristics of study population

No.

Nulliparity

2510

45.2

Multiparity

3045

54.8

Diabetes

251

4.5

Hypertension

546

9.8

Induction

2055

37.0

Augmentation

1121

20.2

Epidural anesthesia

3325

59.9

Median
Gestational age
Birthweight
BMI

39.6
3320
30.8

BMI, body mass index; IQR, interquartile range.

Hamilton et al. First-stage fetal descent. Am J Obstet Gynecol 2016.

IQR
38.9e40.3
3037e3610
27.4e35.3

Original Research

between dilation and station. Trajectory

plots of station or dilation over time
suggested a linear relationship. The
trajectories were roughly parallel. That
is, not every labor progressed with the
same rate of descent (slope) or had
exactly the same values of station
relative to dilation (intercept). These
observations of variation among individuals suggested that a random
slope term and random intercept term
could be benecial additions to the
model.
Variogram plots showed strong serial
correlation between dilation and station
in both parity groups indicating that the
model should account for this
correlation.
The intersubject (s2inter ) and intrasubject (s2intra ) variability of station was
4.62 and 2.01 for nulliparous women
and was 3.92 and 1.93 for multiparous
women, strongly supporting the choice
of a mixed effect approach.
The choice of a covariance structure
was based on the results from a
cross-correlation plot of residuals
that demonstrated a rst-order autoregressive correlation pattern between
observations taken on the same woman.
Table 2 shows Akaike information
criterion values for models with
progressively more terms. Akaike information criterion is a statistical measurement that reects both the
goodness of t of the model and the
complexity of the model. Lower values
are preferable.
The coefcient associated with the
time factor did not improve the performance of the model, suggesting that time
does not inuence the relation between
fetal descent and cervical dilation. That is
to say, fetal descent and cervical dilation
both varied in a similar way with respect
to time.
The nal model for nulliparous and
multiparous women was a random
intercept and random slope linear
regression with correlated random effects and a rst-order autoregressive
correlation structure.
The assumption of normally distributed errors was clearly conrmed by
graphical inspection, as suggested in
Pinheiro and Bates.15

MARCH 2016 American Journal of Obstetrics & Gynecology

360.e2

Original Research

ajog.org

OBSTETRICS

The equations are shown below.

For nullipara
Station 0:427dilation  3:295

FIGURE 1

The percentage of examinations at each level of station

For multipara
Station 0:466dilation  3:941
The models are shown graphically in
Figures 2 and 3. Although the model is a
population-average prediction model,
the 5th and 95th percentiles in these
graphs account for the subject-specic
variability in station.
In nullipara, the widest 5th-95th
percentile range was 3.7 cm, and found
at the lowest dilation. This range reached
a minimum of 3.0 cm at a dilation of
6.5 cm and then diverged to 3.2 cm at full
dilation.
In multipara, the widest 5th-95th
percentile range was 3.8 cm also seen at
the lowest dilation. This range reached a
minimum of 3.1 cm at a dilation of 5 cm
and then diverged again to 3.6 cm at full
dilation.

Percentage of examinations at each level of station.

Hamilton et al. First-stage fetal descent. Am J Obstet Gynecol 2016.

dilation and station measurement, it was

not immediately evident that such a
model could be identied or what its
precision would be.

Comment
The creation of a model of station vs
dilation is challenging. There are several
factors that affect the relationship
between station and dilation. There is
natural biological variation in how
descent occurs. Clinical measurements
of dilation and station are inexact and
prone to variation especially with
inexperienced trainees.17,18 Individual
women enter hospital at different points
in labor and have repeated examinations at irregular intervals. There is a
correlation between an individual
womans neighboring exam results that
can be affected by the time between
examinations. Finally there will be inuences that we have not measured at
all. It is possible that even if the stationdilation relationship could be represented mathematically, that the range of
variation would be very large. A large
range would severely limit its potential
clinical utility. For example if the 5th95th percentile range of station at every
dilation was e5 to 5 in normal labor,
then station could not possibly be useful
as a potential indicator of abnormal
labor progression.
Given the complex nature of labor
data, especially the imprecision of

Principal findings
We observed a linear relationship between station and dilation. That is, there
was a general trend of descent of the
fetal presenting part with increasing
dilation during the rst stage of labor.
This is the rst report of an equation
describing the general relationship between station and dilation in the rst
stage of labor, derived from contemporary women who delivered vaginally
without instrumentation. In addition,
the graphs show the range of variation
from the 5th-95th percentile that

spanned roughly 3-4 cm. While a specic woman may have experienced a
trajectory that was different at times,
90% of the examinations fell within this
percentile range.
The relationship between station and
dilation is consistent with the ndings
of Zhang et al12 and Graseck et al13
who reported graphs of dilation and
station vs time. From their graphs it is
possible to extract their station values at
each dilation value and superimpose
their results on our graph of station
vs dilation. Attention must be given
to reconcile measuring station in
thirds (e3 to 3) or in centimeters
(e5 to 5) when comparing these
curves. Figure 4 shows the relationship
between station and dilation for

TABLE 2

Akaike information criterion values with addition of more

complexity to models
Akaike information criterion
in nulliparous women

Akaike information criterion

in multiparous women

Simple linear regression

41,151

37,115

Random intercept

37,901

34,643

Imposed correlation
structure for errors

36,014

33,494

Random slope

35,832

33,212

Time as covariate

35,860

33,239

Hamilton et al. First-stage fetal descent. Am J Obstet Gynecol 2016.

360.e3 American Journal of Obstetrics & Gynecology MARCH 2016

ajog.org

OBSTETRICS

FIGURE 2

The relationship between station and dilation in nulliparous women

Original Research

general relationship between these 2

factors are that they can be used to
quantify a specic patients station relative to dilation.

Research implications

Graphical display of final model showing station vs dilation in nulliparous women. Median (solid line)
with 5th and 95th percentiles (dotted lines).
Hamilton et al. First-stage fetal descent. Am J Obstet Gynecol 2016.

nulliparous women reexpressed from

their publications and superimposed on
our ndings. Note that all station measurements in Figure 4 are expressed in
centimeters in the e5 to 5 system and
the horizontal axis is dilation not time.

Our ndings closely follow the values

published by Zhang et al.12

Clinical implications
The clinical implications of the mathematical expressions describing the

FIGURE 3

The relationship between station and dilation in multiparous women

Graphical display of final model showing station vs dilation in multiparous women. Median (solid line)
with 5th and 95th percentiles (dotted lines).
Hamilton et al. First-stage fetal descent. Am J Obstet Gynecol 2016.

Further research with larger or different

data sets could result in improvements.
Newer ultrasound-based techniques
could reduce measurement errors. The
incorporation of other inuential factors
like body mass index or epidural use that
were unmeasured in this exercise could
potentially improve the precision of the
model. Research examining the outliers
or specic subsets could also produce
insights leading to further improvements. Finally it is possible that a
different relationship could be described
with better clinical discrimination.

Strengths and limitations

Generalizing these results to other centers must be considered with caution for
several reasons. The cesarean rates were
higher than the national average. Maryland and the District of Columbia where
the data originated ranked seventh and
ninth among all US states for cesarean
rates in nulliparous women with
singleton term vertex presentations in
the United States in 2013.19 Other patient and health care differences can
make particular centers different and the
models described here to be less representative for them. For example, centers
with very different cesarean rates or
racial mixes might observe a different
pattern of descent. That said, the close
similarity between observations by
Zhang et al12 for nulliparous women in
spontaneous labor and this model provides reassurance about generalization of
this model. Like the study of Graseck
et al,13 our study included a similar
percentage of women with induction
and augmentation of labor. We also
did not include a small number of
mothers (26) with 5-minute Apgar
scores 6, which can be a complication
of abnormal labor.
It is important to emphasize that
a statistical range of values found in
a normal population does not
necessarily translate to limits dening

MARCH 2016 American Journal of Obstetrics & Gynecology

360.e4

Original Research

OBSTETRICS

FIGURE 4

Comparison of the relationship between station and dilation from 3 studies

Graphical display of final model showing station vs dilation in nulliparous women with values
extracted and superimposed from 2 other studies. Median (solid red line) with 5th and 95th percentiles (dotted red lines). Zhang et al12 (dashed black line) and Graseck et al13 (solid gray line).
Hamilton et al. First-stage fetal descent. Am J Obstet Gynecol 2016.

clinical abnormality. In fact, the denition of percentile ensures that 10% of

the examinations in this normal population will fall beyond the thresholds of
5 and 95. The selection of limits
dening increased risk of a specic
clinical abnormality would require an
examination of women with that clinical abnormality.

Conclusions
Our results demonstrate a general
linear trend of increasing descent
of the presenting part as dilation advances during the rst stage of labor
in women who delivered vaginally
without instrumentation. We propose
that the mathematical expressions
describing this relationship may be
valuable in the assessment of rst-stage
labor progression.
n
Acknowledgments
We wish to thank Dr Omer Ben-Yoseph of
PeriGen (Israel), who gave his time and expertise
for data extraction.

References
1. Segel SY, Carreo CA, Weiner SJ, et al.
Eunice Kennedy Shriver National Institute
of Child Health and Human Development
Maternal-Fetal Medicine Units Network.

Relationship between fetal station and successful vaginal delivery in nulliparous women.
Am J Perinatol 2012;29:723-30.
2. Holt RO, Diehl SJ, Wright JW. Station and
cervical dilation at epidural placement in predicting cesarean risk. Obstet Gynecol 1999;93:
281-4.
3. Shin KS, Brubaker KL, Ackerson LM. Risk of
cesarean delivery in nulliparous women at
greater than 41 weeks gestational age with an
unengaged vertex. Am J Obstet Gynecol
2004;190:129-34.
4. Debby A, Rotmensch S, Girtler O, Sadan O,
Golan A, Glezerman M. Clinical signicance of
the oating fetal head in nulliparous women in
labor. J Reprod Med 2003;48:37-40.
5. 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.
6. Falzone S, Chauhan SP, Mobley JA, Berg TG,
Sherline DM, Devoe LD. Unengaged vertex in
nulliparous women in active labor. A risk factor
for cesarean delivery. J Reprod Med 1998;43:
676-80.
7. Oboro VO, Tabowei TO, Bosah JO. Fetal
station at the time of labor arrest and risk of
cesarean delivery. J Obstet Gynaecol 2005;25:
20-2.
8. Friedman EA, Sachtleben MR. Station of
the fetal presenting part. VI. Arrest of descent
in nulliparas. Obstet Gynecol 1976;47:
129-36.
9. Friedman EA, Sachtleben MR. Station of the
fetal presenting part. V. Protracted descent
patterns. Obstet Gynecol 1970;36:558-67.

360.e5 American Journal of Obstetrics & Gynecology MARCH 2016

ajog.org
10. Spong CY, Berghella V, Wenstrom KD,
Mercer BM, Saade GR. Preventing the rst cesarean delivery: summary of a joint Eunice
Kennedy Shriver National Institute of Child
Health and Human Development, Society for
Maternal-Fetal Medicine, and American College
of Obstetricians and Gynecologists workshop.
Obstet Gynecol 2012;120:1181-93.
11. American College of Obstetricians and Gynecologists, Society for Maternal-Fetal Medicine;
Caughey AB, Cahill AG, Guise JM, Rouse DJ.
Safe prevention of the primary cesarean delivery.
Am J Obstet Gynecol 2014;210:179-93.
12. Zhang J, Troendle JF, Yancey MK. Reassessing the labor curve in nulliparous women.
Am J Obstet Gynecol 2002;187:824-8.
13. Graseck A, Tuuli M, Roehl K, Odibo A,
Macones G, Cahill A. Fetal descent in labor.
Obstet Gynecol 2014;123:521-6.
14. Friedman EA, Sachtleben MR. Station of
the fetal presenting part. I. Pattern of descent.
Am J Obstet Gynecol 1965;93:522-9.
15. Pinheiro JC, Bates DM. Mixed effects
models in S and S-Plus. 2000R Core Team.
New York (NY): Springer-Verlag; 2013.
16. R Foundation for Statistical Computing,
Vienna (Austria). R: a language and environment
for statistical computing. Available at: http://
www.R-project.org/. Accessed November 27,
2015.
17. Dupuis O, Silveira R, Zentner A, et al. Birth
simulator: reliability of transvaginal assessment
of fetal head station as dened by the American
College of Obstetricians and Gynecologists
classication. Am J Obstet Gynecol 2005;192:
868-74.
18. Sherer DM, Abulaa O. Intrapartum
assessment of fetal head engagement:
comparison between transvaginal digital and
transabdominal ultrasound determinations.
Ultrasound Obstet Gynecol 2003;21:430-6.
19. Osterman MJ, Martin JA. Trends in low-risk
cesarean in delivery in the United States, 19902013. Natl Vital Stat Rep 2014;63:1-16.

Author and article information

From the Departments of Obstetrics and Gynecology
(Dr Hamilton) and Epidemiology, Biostatistics, and
Occupational Health (Ms Simoneau and Dr Ciampi),
McGill University, Montreal, Quebec, Canada; PeriGen,
Cranbury, NJ, and Westmount, Quebec, Canada (Drs
Hamilton, Warrick, and Garite); Departments of Obstetrics
and Gynecology at MedStar Washington Hospital Center,
Washington, DC (Ms Collins), and MedStar Franklin
Square Medical Center and MedStar Harbor Hospital,
Baltimore, MD (Dr Smith); University of California Irvine,
Orange, CA (Dr Garite); and Pediatrix Medical Group,
Sunrise, FL (Dr Garite).
Received July 6, 2015; revised Sept. 24, 2015;
accepted Oct. 6, 2015.
This study was supported by PeriGen, Cranbury, NJ.
Drs Hamilton, Warrick, and Garite are employed by
PeriGen. Ms Simoneau, Ms Collins, and Drs Ciampi and
Smith have no conflict of interest to declare.
Corresponding author: Emily F. Hamilton, MD. emily.
hamilton@perigen.com

ajog.org

OBSTETRICS

Original Research

Glossary of selected terms in the context of this study

Autoregressive: refers to the fact that in labor, station measured at 1 examination depends in part upon the most recent values and the time
between examinations
Correlation: refers to the interdependence between factors
Goodness of fit: a measure that describes how well a statistical model matches a set of real observations based on the discrepancies
between actual values and the values produced by the model
Fixed effects: Parameters in a model describing the average behavior of a population
Random effects: Parameters in a model describing the variation around a fixed effect due to heterogeneity of individuals
Statistical model: An equation describing the relationship between station and dilation in the first stage of labor
Mixed model: A statistical equation that accounts for both fixed and random effects
Trajectory plot: A graph showing the path of an individuals measurements over time
Variogram: A specific type of graphical display that helps to visualize how much of the variation in station relative to dilation is related to:
(a) measurement inaccuracy; (b) autoregressive effects; and (c) natural biological variation among subjects

MARCH 2016 American Journal of Obstetrics & Gynecology

360.e6