American Journal of Obstetrics and Gynecology (2006) 194, 103

www.ajog.org

EDITORS CHOICE

A randomized trial of coached versus uncoached

maternal pushing during the second stage of labor
Steven L. Bloom, MD, Brian M. Casey, MD, Joseph I. Schaffer, MD,
Donald D. McIntire, PhD, Kenneth J. Leveno, MD

Department of Obstetrics and Gynecology, University of Texas Southwestern Medical Center, Dallas, TX

Received for publication February 9, 2005; revised April 29, 2005; accepted June 3, 2005

KEY WORDS Objective: The objective of this study was to compare obstetrical outcomes associated with
Coached pushing coached versus uncoached pushing during the second stage of labor.
Uncoached pushing Study design: Upon reaching the second stage, previously consented nulliparous women with
Second-stage labor uncomplicated labors and without epidural analgesia were randomly assigned to coached
management (n = 163) versus uncoached (n = 157) pushing. Women allocated to coaching received
standardized closed glottis pushing instructions by certified nurse-midwives with proper
ventilation encouraged between contractions. These midwives also attended those women
assigned to no coaching to ensure that any expulsive efforts were involuntary.
Results: The second stage of labor was abbreviated by approximately 13 minutes in coached
women (P = .01). There were no other clinically significant immediate maternal or neonatal
outcomes between the 2 groups.
Conclusion: Although associated with a slightly shorter second stage, coached maternal pushing
confers no other advantages and withholding such coaching is not harmful.
2006 Mosby, Inc. All rights reserved.

One potentially modiable obstetrical practice aect-
ing virtually all American women delivering vaginally is
the practice of actively coaching expulsive eorts during
the second stage of labor. The origins of this practice are
obscure. For example, no mention of coaching maternal
expulsive eorts during the second stage of labor is
made in the rst 9 editions of Williams Obstetrics
textbooks, which encompass the rst half of the 20th
century. By the 10th edition (1950),1 however, Eastman
stated, In most cases, bearing-down eorts are reex
Supported by a grant from the National Institute of Child Health
and Human Development (R01-HD-38663).
Reprints not available from the authors.
and spontaneous in the second stage of labor, but
occasionally the patient does not employ her expulsive
forces to good advantage and coaching is desirable.
This statement acknowledges the normal reexive urge
to bear down to push when the fetal presenting part
distends the pelvic oor. It also suggests that coaching is
not routinely indicated. Indeed, the benets of routine
coached expulsive eorts during the second stage are
being increasingly debated.2,3
We previously reported a randomized trial of coached
versus uncoached maternal pushing to determine the
eects of such coaching on postpartum pelvic oor
structure and function measured 3 months following
delivery.4 In this trial, coached expulsive eorts were

0002-9378/$ - see front matter 2006 Mosby, Inc. All rights reserved.
doi:10.1016/j.ajog.2005.06.022
Bloom et al
Table I Techniques used for coached and uncoached pushing
Coached group
Step 1. Head of bed up 30 degrees
Step 2. Position patient, as she desires, on her
back or either side.
Step 3. Coach patient to pull back on both
knees and tuck her chin while the
provider or partner supports the legs.
Step 4. Coach the patient to take a deep breath
and hold during the peak of a
contraction then bear down and push
for 10 seconds; repeat this as long as
the contraction continues.
Uncoached group
Step 1. Head of bed up 30 degrees
Step 2. Position patient, as she desires, on her
back or either side.
Step 3. The patient should be told simply to
do what comes naturally or whatever
the patient feels the urge to do while
in bed.
associated with evidence of impaired pelvic oor function.
Specically, urodynamic testing revealed decreased blad-
der capacity and decreased rst urge to void in the
coached group. We also asked whether coached expulsive
eorts had an impact on delivery and infant outcomes.
We now report a secondary analysis focused on these
outcomes.
Material and methods
Nulliparous women who presented to Parkland Hospital
from November 6, 2000, to August 31, 2002, in spon-
taneous active labor with uncomplicated pregnancies
between 36 and 41 weeks gestation were asked to
participate in an institutional review boardsanctioned
study of coached versus uncoached pushing during the
second stage of labor. Eligible women were those with a
singleton fetus in cephalic presentation and regular uter-
ine contractions with cervical dilatation of at least 4 cm.
Women with a prior history of urinary incontinence, anal
incontinence, pelvic organ prolapse, any known compli-
cation of pregnancy, or an estimated fetal weight greater
than 4000 g were excluded.
Consent was obtained on admission in labor. The
rst stage of labor was managed according to a written
protocol by certied nurse-midwives under the supervi-
sion of house sta and attending faculty. Women who
consented were randomly assigned to coached or
noncoached pushing when rst identied to be in the
second stage of labor, dened as complete cervical
dilatation. Randomization assignment was masked to
the providers by use of opaque envelopes. Patients who
11
received oxytocin or epidural analgesia as well as women
diagnosed with chorioamnionitis prior to the second
stage were not randomized.
Commencing with onset of the second stage of labor,
women randomly allocated to the coached arm received
standardized instruction on pushing during contractions.
Pushing, using a closed glottis, was coached by certied
nurse-midwives, and proper ventilation was encouraged
between contractions. Women randomly assigned to
uncoached pushing were supervised by certied nurse-
midwives but not given specic instructions on pushing
technique. In essence, they were told to do what comes
naturally. The protocols for coached and uncoached
pushing are shown in Table I.
All nurse-midwives participating in this trial attended
training sessions to ensure compliance with the study
protocol. Forceps delivery was considered only for a
prolonged second stage of labor (2 hours or longer) or
fetal heart rate abnormalities at the discretion of the
attending obstetrician. Episiotomy was at the discretion
of the certied nurse-midwife or attending obstetrician.
Perineal lacerations were classied as 1st degree, 2nd
degree, 3rd degree, or 4th degree according to Hughes.5
The sample size was calculated based on the primary
goal of the study, which was to compare the dierence
in maximal urethral closure pressure between the 2
groups.4 For the purpose of that analysis, a total of 106
women allocated to coached or uncoached pushing were
required to undergo urodynamic testing 3 months post-
partum. To ensure that enough women would actually
present for their postpartum urodynamic evaluation, it
was estimated that about 3 times the number of women
actually required should be recruited to the randomized
phase. The study was terminated when the sample size
for urethral closure pressure had been obtained.
Statistical analyses included Pearson c2 test, Student
t test, and Wilcoxon rank sum test. P values less than
.05 were considered statistically signicant. Analysis
was performed using SAS (version 9.2, SAS Institute,
Cary, NC).
Results
A total of 1534 consecutive women presenting in labor
were screened for participation in the primary study.4 Of
these, 988 consented and 546 women declined to par-
ticipate in the study. At the onset of the second stage of
labor, 325 women were randomly assigned to coached
or uncoached pushing. Complete data were available for
analysis in 320 of these women. The major reasons
women were excluded prior to the randomized phase
included labor epidural use and development of inter-
current labor complications such as need for oxytocin
stimulation or fever. A total of 163 women were
randomly allocated to coached pushing and 157 to the
12 Bloom et al

Table II Demographic characteristics Table IV Selected neonatal outcomes in relation to study

Coached Uncoached P group assignment
Characteristic (n = 163) (n = 157) value Coached Uncoached P
Race .404 Characteristic (n = 163) (n = 157) value
Hispanic 153 (94) 150 (96) 5-min Apgar score
Black 7 (4) 7 (5) 7 or less 1 (1) 0 .326
White 2 (1) 0 3 or less 0 0 d
Other 1 (1) 0 Umbilical artery pH
Age, yr 21.1 G 3.7 21.1 G 3.7 .948 Mean 7.2 G 0.8 7.2 G 0.7 .103
Maternal BMI, kg/m2 29.0 G 3.8 29.0 G 4.0 .875 7.1 or less 7 (4) 5 (4) .601
Prenatal care received 157 (96) 155 (99) .168 7.0 or less 1 (1) 0 .326
Birth weight, g Meconium-stained fluid 36 (22) 20 (13) .028
Mean 3327 G 389 3314 G 361 .770 Delivery room resuscitation
4000 g or greater 6 (4) 4 (3) .560 Bag/mask ventilation 7 (4) 5 (3) .601
All data shown as n (%) or mean G SD. BMI, Body mass index. Intubation 0 0 d
Sepsis workup 7 (4) 12 (8) .205
NICU admission 0 1 (1) .308
Stillbirth or neonatal death 0 0 d
All data shown as n (%) or mean G SD. NICU, Neonatal intensive care
Table III Selected maternal outcomes in relation to study unit.
group assignment
Coached Uncoached
Characteristic (n = 163) (n = 157) P value developed chorioamnionitis after randomization (i.e.,
Length of second stage during the second stage of labor), compared with 3
Minutes, mean 46.3 G 41.5 59.1 G 49.1 .014 (2%) uncoached women (P = .80). No women requested
More than 2 h 14 (9) 17 (11) .498 epidural analgesia or required oxytocin stimulation
More than 3 h 2 (1) 4 (3) .383 following randomization.
Route of delivery Infant outcomes are summarized in Table IV. There
Vaginal 152 (93) 149 (95) .537 were no dierences in any measure of neonatal outcome
Forceps 6 (4) 7 (4) .725 with the exception that the incidence of meconium-
Cesarean stained amnionic uid was nearly twice as high in the
Total 5 (3) 1 (1) .109 coached group. Of the 56 cases of meconium-stained
Failed forceps 2 (1) 1 (1) .584
amnionic uid, 50 (89%) were diagnosed during the rst
Episiotomy 42 (26) 32 (20) .253
stage of labor and prior to pushing during the second
Perineal laceration
None or first degree 105 (45) 110 (51) 0.415 stage. We were unable to time the onset of meconium
Second degree 40 (24) 32 (20) passage in the remaining 6 womendall randomly
Third degree 12 (7) 13 (8) assigned to coachingdbecause they underwent amniot-
Fourth degree 6 (4) 2 (1) omy during the second stage, and it is quite possible that
Shoulder dystocia 0 0 d meconium could well have been present prior to the
All data shown as n (%) or mean G SD. second stage.

uncoached study arm. There were no dierences in
maternal demographic characteristics between the 2
study groups (Table II).
Shown in Table III are selected delivery outcomes in
coached versus uncoached parturients. The mean dura-
tion of the second stage of labor was 46 minutes in
coached women, compared with 59 minutes in those
uncoached (P = .014). However, the proportion of
women with second stages of labor exceeding 2 hours or
3 hours was similar. There were no dierences in route of
delivery. Episiotomy use and the frequency and severity
of perineal lacerations were comparable between the
study groups. Four (3%) women in the coached group
Comment
There were two statistically signicant ndings in this
analysis of women randomly assigned to coached or
uncoached pushing during the second stage of labor.
First, the second stage was abbreviated by approxi-
mately 13 minutes in coached women (P = .01). Second,
the incidence of meconium-stained amnionic uid was
signicantly increased in coached women, although most
of this was diagnosed before the second stage and was
not associated with any adverse infant consequences.
These statistically signicant ndings aside, there were
no dierences in any clinically signicant outcomes,
suggesting that coached expulsive eort is not advanta-
geous and withholding such coaching is not harmful.
Bloom et al
The nding of increased meconium-stained amnionic
uid in women coached to push raised the possibility
that pushing may somehow be linked to meconium
passage. Possible mechanisms could include intensied
umbilical cord compression, leading to hyperperistalsis
via parasympathetic stimulation.6 Alternatively, it is
conceivable that coached maternal eorts to bear
down forced mechanical emptying of the fetal bowel
by transmitted abdominal wall compression. We believe
that neither scenario was involved in our study because
the great majority of cases of meconium were identied
prior to the second stage of labor. That is, meconium
passage antedated maternal pushing.
Interest in management of the second stage of labor
has been heightened by the increased use of labor
epidural analgesia because such pain relief during child-
birth has been linked to prolongation of the second
stage.7-13 Given this nding, actively coaching women to
push during the second stage might be expected to
counter the eects of labor epidural analgesia. At least 7
randomized trials have been reported between 1983 and
2002 on the eects of coaching maternal expulsive
eorts in women with epidural analgesia, and a common
theme from these studies is that coached pushing is not
advantageous.7-13 Like the studies specically focused
on women with epidural analgesia, we found no clini-
cally relevant advantages for routinely coaching women
to push during the second stage. Importantly, we
designed this study to avoid the potential confounding
eects of epidural analgesia on labor. That is, we wanted
to measure the eects of maternal expulsive eects
without the interference of paralyzed pushing because
of epidural analgesia. Lastly, we believe our results have
important clinical relevance because about 40% of
women giving birth in the United States do not receive
epidural analgesia.14,15
We are of the view that routinely coaching women to
push during the second stage has no short-term mater-
nal or neonatal benets. Of note, there is increasing
awareness of the longer-term eects of childbirth on
womens health as reected by pelvic oor dysfunction
later in life. It may be that coaching women to push may
even be injurious to the long-term reproductive health of
women. Indeed, we have previously observed that
coached pushing modies pelvic oor function when
measured using urodynamic studies several months
following childbirth. Thus, the short-term outcome
13
ndings that coached pushing confers neither benet
nor harm might be preempted if it is conrmed that
coaching has deleterious long-term eects.
References
1. Eastman NJ. Williams Obstetrics. 10th ed. New York: Appleton-
Century-Crafts, Inc.; 1950.
2. Roberts JE. The PUSH for evidence: management of the second
stage. J Midwifery Womens Health 2002;47:2-15.
3. Petrou S, Coyle D, Fraser WD. Cost-eectiveness of a delayed
pushing policy for patients with epidural anesthesia. Am J Obstet
Gynecol 2000;182:1158-64.
4. Schaer JI, Bloom SL, Casey BM, McIntire DD, Nihira MA,
Leveno KJ. A randomized trial of the eects of coached vs.
uncoached maternal pushing during the second stage of labor on
postpartum pelvic oor structure and function. Am J Obstet
Gynecol 2005;192:1692-6.
5. Hughes EC. Obstetric gynecologic terminology with section on
neonatology and glossary of congenital anomalies. Philadelphia
(PA): F.A. Davis Co.; 1972.
6. Ramin KD, Leveno KJ, Kelly MA, Carmody TJ. Amniotic uid
meconium: a fetal environmental hazard. Obstet Gynecol 1996;
87:181-4.
7. Fitzpatrick M, Harkin R, McQuillan K, OBrien C, OConnell PR,
OHerlihy C. A randomised clinical trial comparing the eects
of delayed versus immediate pushing with epidural analgesia on
mode of delivery and faecal continence. BJOG 2002;109:1359-65.
8. 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 con-
tinuous epidural analgesia. Am J Obstet Gynecol 2000;182:
1165-72.
9. Hansen SL, Clark SL, Foster JC. Active pushing versus passive
fetal descent in the second stage of labor: A randomized controlled
trial. Obstet Gynecol 2002;99:29-34.
10. Manyonda IT, Shaw DE, Drife JO. The eect of delayed pushing
in the second stage of labor with continuous lumbar epidural
analgesia. Acta Obstet Gynecol Scand 1990;69:291-5.
11. Maresh M, Choong K-H, Beard RW. Delayed pushing with
lumbar epidural analgesia in labour. Br J Obstet Gynaecol
1983;90:623-7.
12. Mayberry LJ, Hammer R, Kelly C, True-Driver B, De A. Use of
delayed pushing with epidural anesthesia: ndings from a ran-
domized, controlled trial. J Perinatol 1999;19:26-30.
13. 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. Br J Obstet
Gynaecol 1998;105:186-8.
14. Hawkins JL, Beaty BR, Gibbs CP. Update on U.S. OB anesthesia
practice. Anesthesiology 1999;91(Suppl):A1060 (Abstract).
15. Eltzschig HK, Lieberman ES, Camann WR. Regional anesthesia
and analgesia for labor and delivery. N Engl J Med 2003;348:
319-32.