Content uploaded by Judy Slome Cohain
Author content
All content in this area was uploaded by Judy Slome Cohain on Aug 31, 2015
Content may be subject to copyright.
Background
O
ne important aspect of a midwife’s work is to
observe and reflect on what these observations
mean. Most authorised views of anatomy and
physiology evolved from new observations. This is never
so important or as appropriate as when the midwife works
independently, since her setting involves more continuity
of care, offering her a more holistic viewpoint, and since
she has less outside support her outcomes are usually
directly a result of her own management.
In Western society, women have adopted lifestyles that are
far removed from their evolutionary species of primates.
The development of chairs led to the adoption of a sitting
rather than squatting position. Over some considerable
time, sophistication in lifestyles has meant that the
majority of women in the Western world have lost the
habit, and muscle and knee strength, involved in routine
squatting to urinate or defaecate where there are now
widespread use of toilets for these functions. There are
also aspects of protection and preservation which can also
be seen in higher primate behaviour, and which are now
largely hidden in societal norms for childbirth in the
Western world. An example of this is if an unknown
primate approaches a mother gorilla and her newborn
immediately after birth, the gorilla will sense threat and
attack. The instinct of human mothers to protect their
children from strangers may or may not be totally lost, but
it is rare to see a glimpse of it at birth where the majority
of women deliver in hospital and entrust their newborn to
a whole range of strangers previously unknown to them.
The argument for maintaining physiological birth as it
evolved over millions of years is difficult in the present
birth environment and this is probably even truer when it
comes to placental delivery.
Neither active nor expectant management
are physiological
There are currently two main approaches to management
of the third stage within formal Western health care.
The first is referred to as ‘actively’ managing placental
expulsion and the second is referred to as ‘expectant’
management or the ‘non-interventionist, passive, and
physiological approach’ (McDonald 2007). However, in
fact they both actively manage the third stage. More
importantly, neither approach has evidence supporting its
ability to decrease or prevent PPH in low-risk women
(Fahy 2009).
Active management does what its description suggests in
being proactive in assisting the descent of the placenta
through the cervix by pulling on the cord. The protocol
calls for uterotonic within one minute of birth and
controlled cord ‘traction’. Active management has the
serious drawback of higher rates of manual extraction as
well as pain, caused by the uterotonics, which may
interfere with the mother’s ability to focus on the
establishment of breastfeeding. Active management is
associated with a reduction in PPH, but not in the absolute
sense, only when compared to the relatively poorer
outcomes of expectant management (McDonald 2007).
Magann et al (2005) concluded that the length of the third
stage was the critical factor in lowering the PPH rate, not
uterotonics. Any decrease in PPH using active management
may have little or nothing to do with the uterotonic but
rather because timely yanking on the cord gets the placenta
out more quickly than expectant management, albeit with
some cords breaking, sometimes leaving parts inside,
which necessitates manual extraction.
MIDIRS Midwifery Digest 20:3 2010
348
Towards a physiological management of the third stage that prevents postpartum haemorrhage
Towards a physiological management
of the third stage that prevents
postpartum haemorrhage
Judy Slome Cohain
Postpartum haemorrhage (PPH) rates for low-risk women
having vaginal births in the Western world are reported as
5% (500ml), 1% (>1000ml), and 1% of women receiving
blood transfusions as a result of these events. While it
could be argued that these are accepted by-products of
birth, there is a need to question why both active and
expectant management appear to show no reduction in
those rates and whether this is because speed of
placental delivery is the critical factor that has not been
emphasised. Active management causes PPH because of
increased manual removal rates and expectant
management causes increased atony by delaying
placental delivery. Where there is greater
acknowledgement that a shorter third stage results in a
lower PPH rate, this leads to reappraisal of third stage
management. The knowledge that the vast majority of
placentas deliver in five minutes was seemingly forgotten
during the period of putting women under anaesthesia
during delivery. Waiting for signs of separation causes
unnecessary delay in third stage, thereby increasing the
PPH rate. A protocol was devised based on timing the
actions and non actions necessary at three, four and five
minutes from the birth for delivery of the placenta using
a squatting position. This idea was born out of an attempt
to eliminate PPH, based on the logic that vaginal delivery
of a placenta weighing one kilo requires maternal effort,
not passivity. In my practice, 350 consecutive attended
homebirths resulted in a 0.6% PPH rate (500ml ), which
compares favourably to the published PPH rates of other
third stage protocols.
*
MIDIRS Midwifery Digest 20:3 2010
349
Towards a physiological management of the third stage that prevents postpartum haemorrhage
labour & birth
Expectant management entails waiting for signs of
placental separation, delayed cutting of the cord and
w
aiting for the placenta to deliver preferably
spontaneously, and if it does not deliver by 10–15
m
inutes, intervening with squatting, pushing, nipple
stimulation and/or infant sucking (McDonald 2007). Its
shortcoming derives from the incorrect assumption that
maternal effort is not needed to deliver the placenta unless
it doesn’t deliver by 10–15 minutes after birth. What
miracle would explain why the woman pushes hard to get
even a 2 kilo or 2.5 kilo baby out, but a sticky 1 kilo
placenta attached to a slimy bag of membranes is
expected to slide down the uterus through a partially
closed cervix and down the narrow diameter of the vagina
by itself and quickly enough to avoid associated excessive
blood loss? In about 96% of cases, the placenta does
deliver uneventfully using expectant management, albeit
with maternal effort. But 4% of the time expectant
management results in PPH >500ml (Combs et al 1991,
Janssen et al 2002, Hofmeyr et al 2008, Janssen et al
2009). A theory, supported by current research, explaining
why expectant management results in a 4% PPH rate, is
because it delays the placental delivery needed to empty
the uterus enabling the uterus to contract the wound where
the placenta was attached.
Since expectant management often entails intervention,
‘passive’ or ‘non-interventionist’ are misnomers.
‘Physiological’ may be a misnomer as well, since the
resistance by women to squat immediately after birth may
not be physiological, but rather due to the relatively recent
invention of indoor plumbing. Counting off 10–15
minutes and then squatting is neither physiological nor
research-based, since a delay of even five minutes has
been shown to double the PPH rate (Magann et al 2005).
Instinctual third stage management is an oxymoron.
Instinctive birth is not divided into first, second and third
stages with various management protocols for each. One
might hypothesise that instinctive third stage would be
where the placenta followed immediately after the baby,
pulled along by the suction created by the body of the
fetus as it leaves a relatively airless cavity. The only
argument against delivering the placenta in a squatting
position at five minutes is a non-evidence based feeling
that it is not physiological to interfere with delivery of the
placenta until 10–15 minutes if it hasn’t delivered ‘on its
own’. The desire for a physiological third stage may be
admirable, but it is out of context with the way birth
currently takes place in the Western world. How many
women elect to birth physiologically when it involves
walking to the hospital, fetching drinking water from the
river, refusing the use of electricity, or food and drinks
that used electricity or gas to be produced, or going
outside to use an outhouse in the trees behind the
hospital? Most women forsake such a birth when they
leave the privacy of their home to give birth among
strangers, and not only that, take a car to get there. It is
unclear why the desire to return to nature suddenly
surfaces for the first 10–15 minutes after the baby is born.
Women rarely object to squatting to push the placenta out
at 4–5 minutes when it is explained to them that waiting
five minutes doubles their risk of PPH.
Placental separation takes ‘a very few minutes’
The 16
th
edition of Williams Obstetrics (Williams et al
1
980) describes the delivery of the placenta in great detail:
‘…as the baby is born (NB not after, but DURING the
delivery of the infant) immediately the uterus
spontaneously contracts down on its diminishing
c
ontents. Normally, by the time the infant is delivered, the
uterine cavity is nearly obliterated and the organ consists
of an almost solid mass of muscle, the fundus lying below
the umbilicus. This sudden diminution in uterine size
inevitably is accompanied by a decrease in the area of the
placental implantation site. Because of the limited
elasticity of the placenta it is forced to separate. During
cesarean section, this phenomenon can be observed
directly when the placenta is implanted posteriorly…
Placental separation occurs within a very few minutes
after delivery… The membranes usually remain in place
until the separation of the placenta is nearly completed.
They are then peeled off the uterine wall, partly by
further contraction of the myometrium and partly by
traction exerted by the separated placenta as it falls into
the lower uterine segment with the force of gravity and
uterine contraction… When the woman is upright, the
placenta may be expelled by abdominal pressure, but
women in the recumbent position frequently cannot expel
the placenta spontaneously.’
Rates of 50% of placental deliveries within five minutes
(Dombrowski et al 1995) have been reported; however,
this rate was confounded by the routine use of oxytocin
before placental delivery, which can slow or prevent
delivery of the placenta by closing the cervix.
Recent ultrasound studies (Herman 2002, Mo & Rogers
2008) have taken great care to document whether
placentas separate all at once (monophasic) or from the
sides (multiphasic). The multiphasic separation was
further observed to take place either from the right, or the
left, or bipolar. These detailed observations have no
relevance for prevention of PPH. The length of time that
it took for the placenta to separate was not reported in a
single ultrasound study. These studies may only be
reporting part of the story, since they relate only to
placentas delivered in the prone position. No study has
examined when or how the placenta separates when the
woman is upright or squatting.
Seconds count
The median length of the third stage of labour was seven
minutes for women without a PPH and nine minutes for
women with a PPH (Magann et al 2005). Seconds make
large differences in PPH rates. Magann found the risk of
postpartum haemorrhage (>1000ml) to be twice as high
for third stages over 10 minutes compared to those under
10 minutes, four times as high for third stages over 20
minutes as under 20 minutes, and six times as high for
third stages over 30 minutes compared to those under 30
minutes. Each of the 6,588 women in this study received
10u of oxytocin upon delivery of the shoulder and 5.1%
of vaginal deliveries caused the loss of >1000 ml of blood.
Upright position eliminated during the period of
twilight sleep
Although the version of the textbook Williams Obstetrics
(Williams et al 1980) reports on the expedience of upright
position for placental delivery, protocols apparently had to
b
e adjusted for anaesthetised women during vaginal
d
elivery, who were unable to be upright for third stage.
D
espite the fact that twilight sleep and chloroform haven’t
been used for the past 30 years, current third stage
p
rotocols appear not to have readjusted to reflect this
change. One explanation for this may be an underlying
preference for intervention by carers rather than deferring
to woman-initiated actions. Typical Western practice
includes high rates of routines that are done to low risk
women such as prenatal ultrasound, induction,
augmentation, epidurals, vacuum and caesareans which
are promoted as being critical to delivering a healthy baby,
when in fact planned attended homebirth involving very
low rates of those interventions results in equally good
outcomes (Janssen et al 2002, Janssen et al 2009).
Fitting with the distrust of women’s capacity to give birth
unaided, widespread objection to getting the woman into
an expedient position like squatting would be expected,
despite its obvious use of gravity and the diaphragm
muscles to facilitate separation and delivery of the
placenta in a timely fashion. Both Magann et al (2005)
and Dombrowski et al (1995), together observing over
52,000 births, found that the women whose placentas
were delivered within five minutes had the lowest PPH
rates. The researchers admit that the ‘vast majority of
women had epidural anesthesia’ which would have made
it challenging to get them into a squatting position.
However, their conclusion was not to assist the woman to
expedite delivery of the placenta herself, but rather to
decrease the protocol for manual removal by the carer
from 30 minutes postpartum to 18 minutes (Magann et al
2005) or to 15 minutes (Dombrowski et al 1995). In a
population which does not use epidurals, taking five
minutes as a reputable baseline, I invited women to
expedite placental delivery themselves, and created
the protocol of 3, 4, 5, 10 minute third stage protocol
(Cohain 2010).
Redefining what is acceptable blood loss
If a contractor who builds houses had a reputation that 5%
leaked when it rained (PPH), and 1% of the houses needed
a costly overhaul (blood transfusion) every time it was
used, he would not have much business. Neither active nor
expectant management has ‘shown a significant,
consistent reduction in the PPH rates reported in
industrialized countries in recent times’ (McDonald
2007). What this means is that both usually result in PPH
rates of 5% or more and blood transfusion rates of
1%–2%. I question the need for this excessive blood
loss in one of every 20 births and blood loss adequate
enough to justify blood transfusion in about one of every
100 births.
Exact blood loss is hard to measure but PPH is not
Wide ranges of PPH rates can be explained by inaccurate
assessments of blood loss (McDonald 2004). If a woman
delivers in a squatting position, using a bowl to catch all
blood at birth and subsequently measuring it with a
measuring cup is probably the most accurate way to
m
easure blood loss. There is still the problem of how
m
uch is amniotic fluid or urine, and so overestimation of
the amount of blood loss within the total is likely but
reduces the risk of underestimation. Without a bowl, it
may be difficult to distinguish exactly between 200 and
300ml or 600 and 700ml. However, defining the blood
loss as over 2 cups, or 500ml, is relatively easy, since
blood overflows the standard 60 x 90 disposable underpad
and flows on to the bed, chair, floor or bathtub, and clots
appear on the pad and floor. Alternatively, the blood clots
inside the uterus and if one or more 10cm diameter clots
emerge upon massage of the uterus, this is considered to
be a PPH. This is based on calculations that the volume of
one 10cm round clot = (4/3)π(5)³ = 524ml. Measuring the
diameter of a round clot is a skill which practitioners are
familiar with from measuring cervical dilation. If the clot
is 15cm in diameter, the volume of blood loss is 1766ml.
Protocol
The protocol for the 3, 4, 5, 10 minute third stage
management is to use a digital watch or Programmed
Talking Timer that announces at three minutes, ‘cut cord
if you like’, at four minutes, ‘up to squatting and push’,
five minutes. ‘push harder if placenta not out yet’, and ten
minutes, ‘check bleeding’. Use of the bowl is optional.
There will have already been a discussion with the woman
about adopting the squatting position for delivery of the
placenta. The exact time of the birth is noted on a digital
watch and said quietly out loud to help keep track of the
time. Immediate continuous skin-to-skin contact with the
baby is initiated for the first 3½ minutes postpartum. The
following is optional but recommended. At three minutes:
check to see if cord has stopped pulsing and cut non-
pulsing cords. At four minutes: if the placenta has not
delivered yet, assist the mother into a squatting position.
The mother is encouraged to push out placenta with or
without a contraction. At five minutes: if the placenta is
not yet born, assist the cord to come out further by gently
pulling it down another 5–20cm in length in order to bring
the placenta low enough to give the woman the urge to
push. At ten minutes: the uterus is massaged to check
for clots.
A bowl can be placed under the woman when she gets into
a squatting position to measure blood loss. Immediately
after delivery of the placenta, the mother is given a
sanitary pad, assisted into bed, and immediately given
the baby.
If bleeding is flowing from the vagina in a 1.5–2cm wide
stream, rather than dripping or spotting the pad, an
intramuscular shot of 10u Pitocin (oxytocin) or 0.2 mg
methergine is given at ten minutes postpartum. Early
suckling at the breast is encouraged.
To refer to the three minute timing, 99% of cords have
stopped pulsing by three minutes (McDonald 2007). The
timing of cord clamping does not affect the incidence of
PPH (McDonald 2007); however, having the baby
connected to the mother often delays the mother getting
into a squatting position, for which timing is critical. If the
MIDIRS Midwifery Digest 20:3 2010
350
Towards a physiological management of the third stage that prevents postpartum haemorrhage
MIDIRS Midwifery Digest 20:3 2010
351
Towards a physiological management of the third stage that prevents postpartum haemorrhage
labour & birth
cord is not cut, delivery of the placenta may be delayed
by the juggling of the position of the baby relative to
the mother.
Outcomes
The 3, 4, 5, 10 minute third stage protocol has now been
u
sed to deliver 350 women with only two occasions of
PPH (0.6%). Both of these were in high-risk women —
one with severe vulvodynia who had an unusual tear,
ripping a 1 inch strip of cartilage-like tissue which bled
profusely (500ml), and the other who had severe
emotional difficulties around the time of the birth and
wished to die (800ml). Among 350 births using the 3, 4,
5, 10 minute protocol, 347 (99%) delivered between five
and six minutes. Two women delivering their ninth babies
delivered at 15 minutes postpartum — in these cases no
tension was put on the cord to pull it down because of the
multiparous uteri, and one first birth had a retained
placenta that was separated by hand and then removed at
30 minutes. Seven per cent of women required a shot of
Pitocin or methergine at 10 minutes.
Discussion
New evidence is presented here indicating that 99% of
placentas in mothers who get into the squatting position
are separated and delivered around five minutes after
birth. This is subsequently associated with a low PPH rate,
with blood loss accurately measured in a bowl.
Several theories for this observation, or a combination of
the three, can be offered:
1. Squatting facilitates separation occurring more quickly.
2. Gravity and abdominal and diaphragm muscles speed
the delivery of an already separated placenta.
3. The increased speed of delivery could be mediated by an
unknown neurochemical pathway, such as the mother’s
confidence in her ability to birth.
A trial in various populations of women is recommended.
It is possible this protocol unnecessarily delays placental
delivery. Waiting to get the woman into a squatting
position at four minutes was chosen to allow for the cord
to stop pulsing at three minutes, for the convenience of
cutting the cord before squatting, and to allow some time
for the woman to examine her newborn. My tendency is
to stay with 3, 4, 5 because neither of the two PPHs were
likely to have been prevented with faster delivery of the
placenta, since one was from the tear, and one was
mediated by a psychological desire to die. There is no
evidence, however, for not getting the woman to squat at
t
wo or three minutes, for example.
It may surprise some that the 3, 4, 5 protocol contradicts
the practice of waiting for the four classic signs of
separation. Watching for signs of separation only causes
unnecessary delay which leads to an increase in PPH.
Caregivers still have to look for bleeding. An early gush of
blood before four minutes means the placenta has
separated right away and is ready for delivery before four
minutes. Once the woman is squatting several centimetres
above the floor, a gush of blood and cord extension are
poorly visualised and squatting makes it is impossible to
palpate a firm, globular uterus or see whether the uterus
rises in the abdomen. Waiting for the woman to feel
contractions only leads to unnecessary delay.
Perhaps other practitioners have tried similar protocols.
I would be very happy to correspond with any readers
about the protocol, with the aim of finding a third stage
protocol that prevents PPH.
Judy Slome Cohain is a private practitioner and
independent researcher working in the USA. Email:
judyslome@hotmail.com.
References
C
ohain JS (2010). A proposed protocol for third stage management. Birth 37(1):84-5.
Combs CA, Murphy EL, Laros RK (1991). Factors associated with postpartum
hemorrhage with vaginal birth. Obstetrics and Gynecology 77(1):69-76.
Dombrowski MP, Bottoms SF, Saleh AA et al (1995). Third stage of labor: analysis of
duration and clinical practice. American Journal of Obstetrics and Gynecology
172(4,part 1):1279-84.
Fahy FM (2009). Third stage of labour care for women at low risk of postpartum
haemorrhage. Journal of Midwifery & Women’s Health 54(5):380-6.
Herman A, Zimerman A, Arieli S et al (2002). Down-up sequential separation of the
placenta. Ultrasound in Obstetrics & Gynecology 19(3):278-81.
Hofmeyr GJ, Abdel-Aleem H, Abdel-Aleem MA (2008). Uterine massage for
preventing postpartum haemorrhage. Cochrane Database of Systematic Reviews,
Issue 3.
Janssen PA, Lee SK, Ryan EM et al (2002). Outcomes of planned home births versus
planned hospital births after regulation of midwifery in British Columbia. CMAJ
166(3):315-23.
Janssen PA, Saxell L, Page LA et al (2009). Outcomes of planned home birth with
registered midwife versus planned hospital birth with midwife or physician. CMAJ
181(6-7):377-83.
Magann EF, Evans S, Chauhan SP et al (2005). The length of the third stage of labor
and the risk of postpartum hemorrhage. Obstetrics and Gynecology 105(2):290-3.
McDonald S (2007). Management of the third stage of labor. Journal of Midwifery &
Women’s Health 52(3):254-61.
McDonald S, Abbott JM, Higgins SP (2004). Prophylactic ergometrine-oxytocin versus
oxytocin for the third stage of labour. Cochrane Database of Systematic Reviews, Issue 1.
Mo A, Rogers MS (2008). Sonographic examination of uteroplacental separation
during the third stage of labor. Ultrasound in Obstetrics & Gynecology 31(4):427-31.
Williams JW, Pritchard JA, MacDonald PC (1980). Williams Obstetrics. 16th ed. New
York: Appleton-Century-Crofts.
Cohain JS. MIDIRS Midwifery Digest, vol 20, no 3, September 2010,
pp 348–351.
Original article. © MIDIRS 2010.
Editor’s note:
A recent Cochrane review has been published which looks again at the evidence for management of the third stage:
Begley CM, Gyte GML, Murphy DJ et al (2010). Active versus expectant management for women in the third stage of
labour. The Cochrane Database of Systematic Reviews, issue 7.
A MIDIRS review of this is planned for the December issue of Essentially MIDIRS.
