Clinical Case
A 38-year old G2P1001 presented for a second opinion at 34 weeks of gestational age.
Her obstetric history included a term vaginal delivery of a 3,460 g infant complicated
by a postpartum hemorrhage requiring the administration of oxytocin, methergine, hemabate,
misoprostol, and curettage for resolution. Her current pregnancy was a spontaneous
dichorionic diamniotic twin pregnancy. The fetuses were concordantly grown and the
pregnancy had otherwise been uncomplicated. She sought a new obstetrician as her physician
had recommended a cesarean for breech/breech presentation and declined to offer a
vaginal delivery in this circumstance.
The risks and benefits of a primary cesarean delivery, an attempted version of the
first twin, and an attempt at breech/breech vaginal delivery were discussed. Potential
complications were discussed with the patient to include cord prolapse, fetal distress,
head entrapment, fetal death, and neonatal developmental delay. The potential need
for Dührssen's incisions, forcep delivery of the aftercoming fetal head and the unlikely
event of twin locking should the second fetus become cephalic were discussed. The
American College of Obstetricians and Gynecologists (ACOG) guidelines recommending
cesarean in her situation were reviewed, as was the paucity of trial data regarding
breech/breech twin deliveries. The provider's experience with breech vaginal delivery
was discussed with the patient. She was given an approximate 30% risk of requiring
emergency cesarean delivery. The patient was advised that the provider thought her
chance of requiring forceps for the delivery of one or both twins was likely also
approximately 30% and informed that these estimates were based on the provider's experience
with the proposed procedure. The patient decided to pursue vaginal delivery.
The patient developed severe cholestasis of pregnancy with her total bile acids returning
at a value of 96 umol/L at 36 5/7 weeks of gestation. Delivery was recommended. The fetuses were in complete breech
(Twin A) and transverse (Twin B) presentations. She chose an induction of labor. Her
starting Bishop score was 8. An oxytocin infusion was commenced. During a cervical
examination at 6 cm dilatation, 70% effacement, and −1 station, the patient's membranes
ruptured with confirmation of fetus A in a complete breech presentation with the right
fetal foot, the fetal genitalia and the sacrum presenting. No umbilical cord was palpable.
Epidural anesthesia was administered. The patient progressed over the next several
hours to complete dilatation and was allowed to “labor down” for approximately 45
minutes at which time she was transferred to the operating room for delivery. She
delivered Twin A, a complete breech fetus, without the need for assistance with forceps.
Newborn was a 3,090 g male with Apgar scores of 7 and 8. A breech extraction, in combination
with maternal expulsive efforts, was performed of twin B which presented as a frank
breech. Again forceps were not required. The newborn was a 2,750 g female with Apgar
scores of 9 and 9. In spite of active management of the third stage, the uterus did
not readily contract. At placental separation the patient began bleeding and uterine
atony was diagnosed. At 5 minutes after placental delivery, the provider estimated
blood loss to be at least 1,000 mL. The atony was refractory to oxytocin bolus, hemabate,
methergine, misoprostol, tranexamic acid, and continued bimanual massage. The anesthesiologist
called to have blood products available and a second obstetrician was made aware that
a postpartum hemorrhage had been diagnosed. A Bakri balloon was placed and inflated
to 500 mL. The estimated blood loss was 1,300 mL at this time and bleeding resolved.
Laboratory testing was performed and a stable maternal fibrinogen was noted at 861
mg/dL. The total hemoglobin drop over serial laboratory examinations over 6 hours
was 5 (from 15.4 to 10 g/dL). The Bakri was removed the following morning without
recurrence of bleeding. Both babies and mother were clinically stable and were discharged
from the hospital on day 2 after delivery.
This case highlights a quandary: a patient makes a request for vaginal delivery and
seeks a provider willing to support her in her decision. The immediate question is
whether this is a reasonable request (is practice outside of guidelines in this case
acceptable)? Secondarily, which party's autonomy should be paramount in this situation,
the physician's or the patient's?
Discussion
“Planned Caesarean Section versus Planned Vaginal Birth for Breech Presentation at
term: A Randomized Multicenter Trial,” published in the year 2000[1] has had profound and at times unintended consequences on women and their infants.
The investigators randomized 2088 women in 26 countries with singleton fetuses in
frank or complete breech presentations to a planned cesarean or planned attempt at
vaginal birth. Most women in the trial remained in the group to which they were assigned
and outcomes were assessed on 2078 of the participants. The trial was a major achievement
and one of its size will likely never be repeated in relation to the question of the
safety of breech vaginal birth. Women randomized to planned vaginal delivery had a
36.1% chance of cesarean delivery if they attempted a vaginal delivery, were more
likely to experience a cord prolapse, fetal heart rate abnormalities, difficult delivery
of the fetal head, and most importantly, a stillbirth or a neonatal death. Maternal
outcomes were equivalent whether they were delivered via cesarean or vaginally. An
obvious difference is that all women delivering via cesarean were exposed to a surgical
procedure, with its attendant risks, whereas those who were allocated to vaginal delivery
only had a 36.1% chance of this outcome.
In spite of the high integrity of the study, there were limitations. The providers
participating in the study had to self-identify as experienced in breech vaginal delivery
with the agreement of the head of their department. This study led to many institutions
in developed countries increasing the number of breech deliveries they did as they
were participating in the protocol which has been a criticism of the work (the suspicion
being that perhaps providers who would have otherwise recommended a cesarean were
allowing an attempt at vaginal delivery if a patient was randomized into that group).[2] The investigators noted that the neonatal benefits of cesarean delivery were greater
in high resource countries.[1] There are multiple potential reasons for this; however, a consideration would be
that providers in low-resource settings may in fact have more experience with breech
vaginal delivery.[2] Conversely, cesarean may be less safe in low-resource settings.
The conclusion reached by many after review of this rigorously-implemented trial was
that cesarean delivery should be recommended in cases of breech presentation. Indeed
ACOG recommended that women undergo cesarean delivery in the setting of a term breech
fetus in 2001. Many other high and middle income countries followed, with the rates
of breech vaginal delivery falling from 80 to 50% in the Netherlands shortly after
the trial was published.[3] It is curious that one trial changed practice so dramatically.
In light of follow-up publications by trial authors and multiple others regarding
the outcomes of women and neonates after breech vaginal birth and the limitations
of the study,[2]
[4]
[5]
[6]
[7] ACOG made new recommendations in 2006 regarding the acceptability of breech vaginal
delivery. The specific findings that seemed to sway opinion were that when many of
the children delivered in the initial trial by Hannah et al were reassessed at 2 years
of age, there were not differences in outcomes irrespective of mode of delivery.[5] Additionally, several authors published series of patients that had delivered vaginally
without the same apparent risk of mortality for the fetus as seen in the Hannah trial.[6]
[7]
The 2006 ACOG committee opinion, recently updated in 2018,[8] recommends several criteria which have not been rigorously tested in trials. The
author suggests that several of them never should be, rather that there is likely
an elevated risk with breech vaginal delivery, even in experienced hands. Providers
should explain this and divulge their experience and confidence with breech vaginal
deliveries to patients and allow them to exercise their autonomy regarding their preferred
mode of delivery. The ACOG recommendations are discussed below:
(1) “The decision regarding the mode of delivery should consider patient wishes and
the experience of the health care provider.”
The safety of breech vaginal delivery in relation to provider experience has never
been (and likely will never be) prospectively studied as the primary outcome in any
study. The level of experience may mean different things. For example, a provider
may have delivered 10 breech fetuses, none with the need for forceps. This provider
may be less equipped to deal with fetal head entrapment than a provider who has delivered
only five breech fetuses, three of them with Piper forceps. The individual dexterity
of the provider cannot be controlled for. For some providers five breech deliveries
may be enough, whereas for others, 20 is insufficient. In our case, the provider supervised
a resident during these deliveries, continuing the education in the method of breech
vaginal delivery. This opportunity to deliver two breech fetuses consecutively is
a rare one and likely provides valuable tactile learning for when an unexpected breech
presents later in that resident's career. That is to say, the experience of delivering
a vaginal breech is valuable, even in trainees who will go into practice not planning
to deliver breeches vaginally, as will the majority of graduates currently in training.
The author notes that difficult delivery of the fetal head and other problems, such
as a nuchal arm can occur at cesarean also,[1] so many of the maneuvers required for safe delivery of a breech apply to both types
of delivery.
(2) “Obstetrician-gynecologists and other obstetric care providers should offer external
cephalic version as an alternative to planned cesarean for a woman who has a term
singleton breech fetus, desires a planned vaginal delivery of a vertex-presenting
fetus, and has no contraindications. External cephalic version should be attempted
only in settings in which cesarean delivery services are readily available.”
The author agrees with this statement and indeed performs the procedure. It is one
of the three options offered to patients[9] and some patients elect to undergo an attempt at a vaginal delivery without an external
cephalic version which the author believes is an acceptable choice. ACOG also recommends
referral to other providers in the case where patients wish to undergo a procedure
the clinician is uncomfortable to provide (primary elective cesarean delivery without
a medical indication, or referring for abortion care, for example).[10] If a provider is uncomfortable performing an external cephalic version or a breech
vaginal delivery and the patient desires one of these procedures, referral so that
the patient may access the desired procedure should be considered.[10]
(3) “Planned vaginal delivery of a term singleton breech fetus may be reasonable under
hospital-specific protocol guidelines for eligibility and labor management.”
This recommendation seems based on large patient cohorts published after the trial
by Hannah.[6]
[7] Several studies since have assisted in predicting which deliveries are more likely
to entail neonatal complications.[11] While the author agrees that clinical judgement should be used, the underlying assumption
is that if the patient falls outside of the protocol, they have no option other than
a cesarean delivery. Most patients are willing to undergo a cesarean if it is recommended.
However, we should include in our considerations of each individual patient that no
protocol is specifically engineered for them. The patient in the case discussed here,
for example, likely would not have met criteria for a hospital protocol and would
effectively have been compelled to have a cesarean. Additionally, though good results
have been noted in cohorts when induction of labor is not permitted and labor must
progress in a timely fashion, this does not mean that less favorable results would
occur with induced labor or more patience with the labor in general. Multiple studies
have now indicated that women with cephalic-presenting fetuses should be given more
time to labor.[12]
[13] The same may be true for breech fetuses. Indeed, permitting more liberal induction
of labor may actually decrease fetal and neonatal morbidity and mortality as it may
result in breech fetuses delivering when the most skilled attendants are available,
rather than when whoever is on call is available. It also may make providers with
experience more willing to perform the deliveries rather than being constantly available
for weeks as they await the patient's spontaneous labor. Lastly, the reality is that
many hospitals have financial interests that would introduce bias into decisions regarding
higher-risk deliveries and the creation of protocols. It may be that insurers and
hospitals would prefer to take on the legal and financial risks of a planned cesarean
over a breech vaginal delivery. A recommendation regarding using a hospital protocol
may be more about the comfort of risk managers and insurers than the promotion of
the health of women and their babies. A more valid question may be whether or not
insurers should be able to influence decisions made by a patient and physician.
(4) “If a vaginal breech delivery is planned, a detailed informed consent should be
documented, including risks that perinatal or neonatal mortality or short-term serious
neonatal morbidity may be higher than if a cesarean delivery is planned.”
The author agrees that informed consent should be sought for every procedure, including
breech vaginal delivery.
An additional recommendation regarding breech delivery of second twins (discussing
the safety and appropriateness of this approach as desired) was included in the 2006
ACOG committee opinion which has now been removed.[8] As far as the author is aware, there are no trials specifically assessing mode of
delivery for breech/breech presentations in twins and suspects that one will not be
performed. However, the author does have experience with breech vaginal delivery and
with breech-presenting first twins. Aside from the unlikely possibility of twin locking
that would require version of the second twin to a cephalic presentation, it seems
that the risks of breech/breech vaginal delivery would apply mostly to the first,
or presenting twin, making it no less safe that singleton breech vaginal delivery.
The author notes that ACOG guidelines recommend an attempt at vaginal delivery only
if the presenting twin is vertex.[14]
The decision to act in the best interest of a patient may be outside of ACOG guidelines,[15] as this may be preferable to the alternative, which is in many states is a high-risk
home birth. This patient accessed the opportunity for a vaginal delivery only by seeking
out a physician who would provide it through word-of-mouth (the author believes that
this patient would have ultimately unhappily accepted a cesarean if forced to choose
between surgery and a home birth, but not all patients will). Conversely, physicians
frequently are put in a position of considering their own autonomy in these situations.
Are they willing to take the risk of the personal and professional consequences of
a delivery that does not go well when they are practicing differently, even with good
documentation? Many physicians understandably find the risks to themselves to be too
high and exercise their own professional autonomy, declining to respect the autonomy
of the patient.
If trained providers in medical facilities are unwilling to provide care that is acceptable
to their patients, alternative routes will be sought. This has in fact been seen in
this provider's state. In Arizona, home birth is permissible by licensed midwives.
Legally, a midwife may also perform breech vaginal deliveries in a home setting, along
with vaginal birth after cesarean delivery, though delivery of multiples is considered
a contraindication to home delivery.[16] Lawmakers are not subject to following ACOG guidelines and hence, legislated to
allow home birth in situations specifically cited to be contraindications to home
birth by ACOG.[17] This is not to say that all providers that offer home birth will agree to perform
these deliveries (the author has had several midwives bring patients laboring breeches
to the hospital for a physician-supervised delivery). However, permissive laws in
combination with restrictive guidelines can have devastating effects for women and
their families. Home birth is known to increase the risk of neonatal morbidity and
mortality.[17]
[18] Importantly in the instance of maternal postpartum hemorrhage access to emergency
surgery and blood products are vital to prevent maternal mortality, and neither of
these options are available with home birth. Tragically, mothers can die from hemorrhage
in home birth settings awaiting the arrival of emergency medical care.[19] And while there are benefits of the interventions in our hospitals, cesarean deliveries
(what was recommended to this patient) more than triple the risk of maternal mortality
and have long-term consequences, especially in repeat pregnancies.[20]
[21]
[22] It is acceptable for a patient to weigh risks with all information available and
request a vaginal delivery in the safest setting possible. It is ethically required
that we as providers allow autonomy and minimize risk. In the case presented here,
the patient had a life-threatening hemorrhage and the author is skeptical that she
would have arrived at the hospital in time to avoid severe maternal morbidity or mortality.
Lastly, the management of this patient should be open to discussion. Procedures are
learned during an apprenticeship of sorts (for the author, a residency and fellowship
during which they had multiple mentors who taught twin, forcep and breech vaginal
deliveries). Several parts of the patient management were based on combinations of
styles of the author's predecessors. Letting a breech “labor down” to decrease the
risks of a residual cervical lip, the correct approach to placement of forceps, the
desire to deliver all breeches and twins in the operating room were all elements of
training that the author accepted as safe practice. However, that is not to say that
this is the only or safest approach. Many of these elements will likely never be prospectively
studied. As we face a climate of increasing litigation, our approach to teaching our
apprentices may need to change, and this article does not mean to provide a solution
to the question of how effective continuing education of breech vaginal deliveries
should be performed.
No provider should be put in the situation of performing a procedure that they do
not feel is safe or that they worry will create an unacceptable risk of litigation.
However, the cesarean delivery rate we see is likely at least in part related to self-protection
on the part of physicians. We do need to respect that patients have the right to make
their own decisions, even if we perceive them to be a bad decisions.[23] We are much more comfortable with this, however, when active management is not involved
(refusal of a medically-indicated induction, for example, does not actually require
the physician to perform a procedure). In this case, the provider thought that the
patient was making a considered decision and agreed that the attempt at a vaginal
delivery was reasonable. It is fortunate that she delivered in the operating room
of a hospital with availability of appropriate treatment for her postpartum hemorrhage.
Providers and hospitals should consider that in protecting our own interests, refusal
to provide coverage for higher risk procedures can place patients at risk.
