Student Projects
Evidence Based Clinical Practice Guideline

Intermittent auscultation of the fetal heart rate during labor

Patti Valasek

Patti R. Valasek, CNM, MS
Education Program Associates, 1997
Philadelphia University, Master of Science with Major in Midwifery, May 2000


Electronic fetal heart rate monitoring (EFM) has become the standard of
care in hospital obstetrics since the mid 1970's.  However, this
technology has contributed to the dramatic increase in cesarean section
birth, while the perinatal morbidity and mortality rate have not declined
with routine electronic monitoring.  Women are more likely to be
dissatisfied with their care in labor if they are continuously monitored,
since their comfort and mobility is reduced, and they are less likely to
receive emotionally supportive care and the therapeutic presence of their
providers and nursing staff.  Midwives believe that EFM should not be
forced on every laboring woman in the hospital, and support intervention
during birth only when necessary.  When free to move around in labor,
most women change positions often, and may use warm water in a shower or
whirlpool tub for pain relief.  Interventions are less likely to be used
if the woman is not continuously attached to EFM, and maternal
satisfaction is greater if her choices are honored in labor.   Research
from the past 25 years has shown that intermittent auscultation (IA) of
the fetus in labor is as safe as electronic monitoring, and yields
equally good perinatal outcomes.  ACOG's 1997 guidelines state that both
EFM and IA can be appropriate choices for women in labor.   Sixteen
randomized controlled clinical trials were reviewed, and an
evidence-based clinical practice guideline developed to encourage the
standard use of intermittent auscultation in labor for most women.  These
evidence-based clinical practice guidelines for the use of intermittent
auscultation of the fetus in labor will guide the practitioner in
correctly matching the level of technology best suited to the laboring
woman's needs.   Intermittent auscultation should be encouraged by health
care providers as an equally safe alternative in birth to electronic
fetal monitoring in normal, healthy childbearing women.


Women have given birth to their children for centuries without the
knowledge of how the baby was doing in labor.  Very little effort was
made to understand how the fetus was withstanding the forces of birth, or
whether the end result could be changed.  Today however, an enormous
difference exists in the way that healthy, low risk women are allowed to
labor.  Concern for the unborn baby's condition, anxiety about legal
liability, along with increased convenience for the caregivers has caused
most healthy women in the late 1990's to be continuously monitored in
labor by an electronic fetal monitor.  Out of the approximately four
million women who gave birth in the United States in 1988, at least two
thirds of the women ( 2.5 million) had continuous electronic fetal
monitoring (EFM) in their labor (Flamm, 1994), many of whom were
unnecessarily confined to a technology that does not statistically
improve the outcomes for mothers or babies.


In theory, electronic fetal monitoring should give physicians and
midwives information about the fetal status, with the ability to perform
an emergency delivery if complications occur. However, the information
received from EFM must be interpreted, and the interpretation about
monitor strips and their accompanying fetal status is often open to
debate.  Serious questions have been asked about significant increases in
maternal morbidity and mortality in the 30 years that fetal monitors have
been widely used.  The decreases in perinatal morbidity and mortality
that have occurred are felt to be mostly due to advances made in neonatal
medicine during this time span (Prentice, 1987).  Midwives and others who
use intervention only when needed in the labors of healthy women, believe
that the routine use of fetal monitors in our society has introduced many
negative influences. Some of the negative influences are an exaggerated
reliance on technology, rather than respecting what the laboring woman
wants, as well as less time and emotional support from nursing staff.
Intermittent auscultation (IA) is an alternative that can be safely used
instead of electronic monitors, resulting in freedom to move in labor,
less unnecessary intervention, and greater satisfaction for the woman
giving birth.

Another of the complications that is associated with EFM is the
significant increase in cesarean and operative birth.  Cesarean section
is an effective tool in obstetrics, and has saved many mothers and babies
in true emergencies.  Forceps and vacuum extractors can be similarly
helpful in labors where the fetus must be delivered quickly.  However,
the routine use of continuous fetal monitoring results in many false
positives, where fetal distress is assumed and a cesarean section is
performed, though no true distress ever really occurred.  It has been
reported that between 71 - 95 % of operative deliveries for fetal
distress result in a baby that is not "clinically" distressed at birth,
based on Apgar scores and umbilical artery pH (Prentice, 1987).   One
study has shown that routine EFM increases the cesarean rate by 21% and
the use of forceps and vacuums by 11%, with a total increased risk of 23%
(Thacker, 1995), as compared to IA.  Another study reports that patients
who were continuously monitored had a cesarean rate that was 53% higher
than the similar group who had auscultation in labor (Vintzileos, 1995).

When the cesarean section rate increases, the maternal mortality and
morbidity rate also increases, due to the complications of surgery such
as increased infection and blood loss, along with anesthesia problems
(Sandmire, 1990).  The costs are also higher for this type of birth,
along with generally decreased rates of satisfaction with their
deliveries for most women (Sandmire, 1990).  A committee of the Institute
of Medicine recently concluded that the increased cesarean rate was
primarily due to concerns about medicolegal liability, along with an
unnecessary dependence on EFM (Sandmire, 1990).  

Another reason for concern about monitoring most low risk women is that
their mobility is then restricted.  The best tracing on an external
monitor occurs when a woman remains in the same position for an
indefinite length of time.  Women need to frequently change their
positions in labor in order to be comfortable, and decrease pain.  In
addition, the supine position that allows the best monitoring increases
pressure on maternal blood vessels (the inferior vena cava and aorta),
which decreases the oxygen available to the fetus (Varney, 1997).  When
attached to a fetal monitor, women are prevented from changing positions,
walking, emptying their bladders, or using a whirlpool tub or shower for
relaxation or pain relief.  Midwives believe that mobility will increase
the strength of uterine contractions and cause the labor to be shorter.
When made immobile by a fetal monitor,  a "cascade of medical
interventions" (Albers, 1994) may then begin, possibly after an
inadequate labor pattern is observed, followed by Pitocin augmentation
and other interventions to hurry the labor along.  Even when telemetry
monitoring (which allows the woman to ambulate while remaining on the
monitor) becomes more common,  the woman is still made uncomfortable by
the presence of belts around her abdomen, and an overreliance on
technology continues among health care providers.

Rather than managing labor in an aggressive manner, midwives believe in
honoring the natural progress of labor.  Midwifery philosophy would
support a woman's choice to have as little intervention as possible, as
long as her cervix was progressively changing, and her baby was coping
well.  Women are less likely to feel empowered as healthy, normal women,
when unnecessarily confined to bedrest on a monitor.  Routine monitor use
may also increase maternal stress, due to her discomfort, which can
negatively affect the fetus (Albers, 1994)  It may also alter the
perception of pain, and increase the need for analgesia in labor, which
can then have undesirable side effects on the fetus, such as apnea
secondary to narcotic use (Blackburn & Loper, 1992).

Along with the discomfort reported by women who are forced to be confined
to bed with monitor use, there is likely to be less nursing care when EFM
is used.  Many studies have shown that having a caring support person in
labor decreases fetal distress, prolonged labor, operative delivery,
newborn complications, and postpartum depression (Albers, 1994).
Unfortunately, the use of continuous fetal monitoring in labor tends to
decrease the amount of personal contact that nurses have with their
patients (Albers, 1994).  A study done in 1985 in Dublin, Ireland
confirms that women who were continuously monitored were statistically
more likely to be left alone for short periods of time.  When these women
were questioned about their feelings about the absence of their
caregivers, most reported negative feelings about the time that they were
left alone (Garcia, J., Corry, M., MacDonald, D., Elbourne, D., Grant,
A., 1985)

Significance of the EB-CPG to midwifery

This inquiry will help to determine the conditions under which
intermittent auscultation in labor results in equally satisfactory
results as continuous electronic fetal monitoring.  A thorough literature
review was done, and both the maternal and neonatal mortality and
morbidity rates were compared, along with complications noted with both
methods.  The goal is that the use of intermittent auscultation for
healthy women will be greatly expanded in midwifery practice.  

A second goal for this inquiry is to develop a research based protocol
for performing intermittent auscultation for low risk women in labor.  It
is strongly felt by the author that women's needs and comforts in labor
are routinely ignored as a result of anxiety felt by the hospital nurses
regarding fetal well-being.  Since the literature shows that intermittent
auscultation is statistically safe in labor for most women, a clinical
practice guideline was developed that will insure the well-being of the
baby, but that will allow maximal mobility of the woman in labor.  

This inquiry has significant importance to the practice of midwifery.  As
legal liability grows in midwifery, some are reacting to the threat by
using more and more technology in a misguided effort to prevent lawsuits
resulting from a bad outcome.  Instead, midwifery needs to remember its
roots, where it was distinguished from medicine by the constant presence
and respect that a midwife has for her client in labor.  Since
intermittent auscultation is as safe as electronic monitoring for most
women, more midwives and hospital nurses will feel empowered to use the
monitor only as a screening device for normal, healthy women.  Once a
reassuring 20 minute monitor strip is obtained, the laboring woman would
be encouraged to ambulate freely, assume any position that she desires,
and be respected for the way that she chooses to give birth.  As a new
century is beginning, midwives should be bold in challenging the unproven
beliefs about how to safely care for women in labor, and learn to trust
their skills in birth, rather than technology.

Gaps between current practice and evidence-based practice

Electronic fetal monitoring (EFM) is used continuously in labor for most
healthy women with low risk pregnancies, despite research evidence
suggesting that it does not improve outcomes, and that it contributes to
higher rates of operative birth.  In the past 30 years, it has become
part of the standard of care in hospital births.  The original studies
from the 1970's used small numbers of women, but suggested better
perinatal outcomes than with those monitored traditionally with
intermittent auscultation (IA).  The technology then became widespread
before rigorous studies were done to show EFMs efficacy and safety.  A
great deal of research has been done in the past 25 years in an attempt
to determine whether EFM actually reduces perinatal mortality and
morbidity, and what its effect is on maternal complications.  This
research has demonstrated that routine EFM has no impact on perinatal
morbidity and mortality, as compared to IA (Elison, Foster,
Sheridan-Pereira, & MacDonald, 1991; Haverkamp, Thompson, McFee, &
Centrulo, 1976; Haverkamp et al., 1979; Kelso et al., 1977, Luthy et al.,
1987; MacDonald, Grant, Sheridan-Pereira, Boylan, & Chalmers, 1985; Shy
et al., 1990; Thacker, 1986; Thacker, Stoup, & Peterson, 1995; Vintzileos
et al., 1995; Wood et al., 1981).  However, the rate of cesarean delivery
was noted to be higher in women with EFM (relative risk 1.33) (Thacker,
Stroup, & Peterson, 1995), and it has been estimated that 71-95% of
cesareans done for fetal distress resulted in an infant that was
clinically not depressed (based on cord gases and Apgar scores) (Prentice
& Lind, 1987).   ACOG's 1997 guidelines state that both EFM and IA can be
appropriate choices for women in labor.  Intermittent auscultation should
be encouraged by health care providers as an equally safe alternative in
birth to electronic fetal monitoring in normal, healthy childbearing
Theoretical and operational definition of terms

1)  Neonatal outcome:  Neonatal is a newborn infant until 6 weeks of age,
and outcome is the end result of patient care. Therefore, the theoretical
definition of neonatal outcome is any end result of care provided to a
newborn baby from birth to 6 weeks of age.  A good outcome is defined as
an Agpar score of 7 or above at 5 minutes of age, and a bad outcome is
defined as an Apgar score of 6 or below at 5 minutes of age.  This entire
concept is operationalized by the number of infants (or percentage of the
total) with 5 minute Apgar scores of  7 or above, as obtained from the
chart information.

2)  Intermittent FHT auscultation:  Intermittent is the same as not
continuous, or done at intervals.  Auscultation is listening to sounds
(FHTs ) within the uterus.  Therefore, the theoretical definition of IA
is regular listening to FHTs in a woman in labor, without strapping the
fetal monitor on for more than 10% of the total time.  This is
operationalized by the number of women (or percentage of the total) who
have their fetus evaluated at specific intervals in labor, without using
the EFM for more than 10% of the time in labor.  

Assumptions and philosophical approach

The reason that a debate even exists over EFM is that a good outcome in
labor and delivery is vitally important to the woman and to the provider
who has responsibility for the birth.  The two concepts of : 1) neonatal
outcome and 2) intermittent auscultation are linked, because a great
volume of research has demonstrated that IA is as effective as EFM in
obtaining a good outcome for a woman and her family (Herbst &
Ingemarsson, 1994; MacDonald et al.,1985; Shy et al., 1990; Thacker et
al., 1995).  

In midwifery philosophy,  a good outcome in pregnancy is vital, but it is
also extremely important to consider the manner in which the woman gives
birth.  It must be remembered that most birthing women are healthy and
normal (Guidelines for perinatal care, 1997).  Those who support
midwifery and its philosophy find it offensive that EFM is forced upon
nearly every laboring woman in a hospital setting (Flamm, 1994), despite
the lack of clear evidence to demonstrate its efficacy and safety
(Thacker et al., 1995).  The majority of studies that have been done in
the last 15 years show that IA of the fetus in labor is as safe as
continuous electronic monitoring (Thacker et al., 1995).  Guidelines have
existed since 1989 from the American College of Obstetricians and
Gynecologists, which advise that intermittent auscultation is one of two
appropriate options for surveillance of the fetus in labor  (Guidelines
for perinatal care, 1997).  Therefore, IA should be used much more widely
in labor for several reasons, apart from its safety record over the

 IA should be used routinely in hospital labors, so that women can assume
any position that they want.  EFM requires that women remain in bed while
they labor, and this immobility is limiting in several ways.
Unfortunately, the supine position that produces the clearest monitor
strip increases pressure on the maternal inferior vena cava and aorta,
which ultimately decreases the oxygen available to the baby (Varney,
1997).  Women in labor should always avoid extended periods of time spent
on their backs to decrease fetal compromise.  Also, walking in labor may
shorten the length of the woman's labor, since it increases the strength
of the uterine contractions (Albers, 1994; Varney, 1997).  By being
forced to remain in bed on a monitor, the contractions may never get
strong enough for cervical effacement and dilation to occur.
Intervention, such as Pitocin augmentation, may then be introduced in an
effort to speed up labor.  Being confined to bed is described by Albers
(1994) as precipitating a "cascade of medical intervention" (p. 108) that
could likely be avoided if she were allowed freedom of movement during
her labor.   If the laboring woman was encouraged to walk and move around
freely,  her uterine contractions would be at their most efficient
(Albers, 1994).

In addition, women's perception of pain is altered by the experience of
being unable to freely change positions. Women are usually uncomfortable
when they are on the fetal monitor for a long time, since the best
tracing requires that they remain in one position.  In addition, they are
more likely to ask for pharmacologic pain relief when they are monitored
with continuous EFM (Albers, 1994; Varney, 1997).  Though considered safe
in limited amounts, narcotic pain medicines used in labor do have
potential negative effects on the woman and her unborn baby (Albers,
1994; Blackburn & Loper, 1992), and might be avoided if the laboring
woman were free to move around.  When women are not immobilized on a
monitor, many lean on their partner during contractions, use comfort
measures such as massage, or use a warm shower or whirlpool bath for
relaxation and pain relief in labor.  

IA samples the fetal heart rate, rather than obtaining it continuously,
as in EFM.  By sampling the heart rate, inferences are made about the
overall quality of fetal health, since the variability and rate of the
fetal heart tells observers a great deal about the presence or absence of
a healthy fetal nervous system.  As long as the sample is representative
of the overall heart rate of the baby (such as IA being done with and
after the stress of a uterine contraction), the inferences made about
fetal health will be accurate.  This measurement provides a way to
accurately know the status of the baby during labor, without having to
record each beat of it's heart rate for many hours, as in EFM.  

Midwives have an inherent respect for the women that they care for in
labor. Midwives believe in offering women choices in their birth
experience, so that the decisions that are made are ideally shared
between the woman and her midwife. When IA is used in labor, the woman's
choices about birth options are usually more likely to be honored.
Often, these requests include the desire to be in any position that she
desires during labor and birth, the desire to avoid pain medications and
episiotomy, and the desire to be actively involved with her birth.  It is
much more difficult for a laboring woman to reach down and lift her own
baby up onto her belly when she has been a passive observer of the rest
of the labor.  When a woman successfully gets through labor without the
medical interventions that she hoped to avoid, her confidence soars, and
her satisfaction with the birth experience increases.  

Midwives also have a great respect for the wisdom of the birthing woman,
so that her opinions and thoughts about how to successfully birth her
baby are respected and considered.  In contrast, when care providers get
used to dependence upon EFM to give them information, they lose the
ability to believe that the laboring woman can give birth without the
"help" of a machine that tells the staff when she is contracting, and how
strong the contractions are.  IA forces the care providers to rely on the
use of their hands to determine how strong the contractions are, and
encourages them to trust the patient's perceptions and her experience.
Thus, the woman is made the center of the birth experience rather than
the monitor, and the nurses and midwives who care for her are reminded
that their support and attention needs to be focused on the one giving
birth, not on obtaining a good tracing (Albers, 1994).


IA -> ability to move in labor -> greater maternal comfort and
confidence-> decreased fetal compromise from compression on inferior vena
cava-> equally good outcomes with continuous EFM

Limitations of the state of the science in this area  

Although much research has been done in the past 3 decades on electronic
fetal monitoring, a great deal of information is still unknown.  For
instance, a great number of babies who are born with significant hypoxia
may begin to experience that problem days and weeks before the onset of
labor.  Even if an immediate cesarean section were done upon their
admission to the labor and delivery unit, the outcome would not be
improved, since the compromise occurred long before the mother entered
the hospital (Albers, 1994; Paneth, Bommarito, & Stricker, 1993).
Research needs to take place which would both identify fetuses who are
likely to experience hypoxia, and a plan of intervention begun long
before the damage is done.  

Other knowledge gaps that exist in the literature are how to identify
groups of women who would benefit the most from intensive intrapartum
monitoring, and how to use this technology to get the maximum number of
good outcomes (Thacker et al., 1995).  In addition,  those who support
intermittent auscultation commonly use EFM for an initial strip upon
admission to the labor and delivery area.   Many clinicians require a
low-risk woman to be evaluated upon admission to a labor and delivery
unit by a 15-20 minute EFM strip.  If the tracing is judged to be
reassuring, the fetus is then evaluated periodically with IA throughout
her labor, as long as no concerning patterns are noted.  Unfortunately,
the safety and effectiveness of this approach, has not been evaluated
statistically (Spencer, 1994).  Randomized clinical trials are needed to
establish that this compromise is one that can be defended by scientific
studies (Neilson, 1994).  In addition, all the complexities of a cost /
benefit analysis of EFM compared to IA (including the increased maternal
mortality and morbidity associated with EFM ) has yet to be done (Tallon,


Various results are seen from the literature on this topic.  One study
that was quite controversial was published in the June, 1993 issue of
Obstetrics and Gynecology (Vintzileos et al., 1993).  The author 's goal
was to determine whether continuous EFM decreased perinatal mortality and
morbidity as compared to IA, and predicted that a two thirds reduction
would take place in perinatal mortality in the EFM group.  Women from two
different Greek hospitals were assigned to either continuous EFM or IA by
the flip of a coin.  Before the study was begun, intermittent
auscultation was the standard practice at these hospitals for fetal
monitoring, and only 20% of the patients had continuous EFM.  The results
of this study showed an impressively large reduction in perinatal
mortality from 13 per 1000 women in the IA group to 2.6 per 1000 women in
the group with continuous EFM.  Interestingly, this same study showed no
difference between the two groups in Apgar scores, amount of fetal
acidosis measured by scalp pH, need for resuscitation after birth,
admissions to the neonatal intensive care unit (NICU), or neonatal
complications.  Additionally, in this group of women, the rates of
induction or augmentation with oxytocin were markedly different.  In the
group monitored by IA, the rates of induction and augmentation were 7%
and 38.1% respectively, compared to 15.6% and 52.4% of the EFM women who
received synthetic oxytocin in their labor.  Finally, the authors
conceded that the cesarean rate increased significantly when continuous
EFM was used.  The rate was 11.2% for the EFM group, compared to 4.8% of
women whose babies were monitored with IA (Vintzileos et al., 1993).

This study caused quite an uproar when it was published, since it was the
first study since the poorly done studies of the 1970's (MacDonald,
Grant, Sheridan-Pereira, Boylan, & Chalmers, 1985) to statistically
suggest that EFM was superior to IA in bringing about better neonatal
outcomes.  Since the study involved a relatively few number of women
(1428) giving birth in two different hospitals in Greece, it was felt by
some that the results could not be generalized to women in the United
States and Canada (Keirse, 1994).  Criticisms were made about whether the
higher rates of oxytocin use in the EFM group could have skewed the
results (Keirse, 1994).  Also, it was felt that the method of
randomization (a coin toss) was not ideal, since it does not divide women
into two groups in a fair and equal manner.  Since the numbers were quite
different in the two groups (682 for IA vs 745 for EFM), a suggestion was
made that unbiased randomization did not occur with the coin toss
(Neilson, 1994).  In addition, concern has been voiced that cultural
factors in obstetric care may have influenced the results.  Greek
hospitals have a reputation for aggressive obstetrical practices (Keirse,
1994) which could explain why 57% of all the women were either induced or
augmented with oxytocin, with this number rising to almost 70% of the
women with EFM.  Greece has also been noted to have a higher rate of
perinatal mortality than other European countries (Keirse, 1994), so the
extreme differences in mortality rates between the EFM and IA groups may
not be reproduced in other places.  

The author of the Greek study then had two subsequent studies published
in 1995 (Vintzileos et al.,1995a; Vintzileos et al.,1995b)  which both
support the findings from the first study that continuous EFM reduces
perinatal mortality.   In the first study, a meta-analysis was done of
all previously performed, randomized published trials to determine if
continuous EFM improved pregnancy outcomes as compared to intermittent
auscultation (Vintzileos, 1995a).  This retrospective study used MEDLINE
to retrieve nine studies, which included a total of 18,561 women in a
variety of locations.  The results from this combination of nine
randomized clinical trials show that there was no difference in the
perinatal mortality rate among the women in groups with continuous EFM as
compared to IA (.42% mortality in the EFM group, compared to .49% in the
IA group).  However, the author chose to downplay this result, and
highlight that the perinatal mortality rate due only to hypoxia was
markedly decreased (.41 odds ratio) in the continuously monitored group
(Vintzileos, 1995).    Like other studies, the cesarean rate was shown to
be significantly higher (1.53 odds ratio) in the EFM group, as compared
to those with IA.  Theoretically, the women in both groups were
comparable to each other, and no significant differences existed between
the groups.  

The second 1995 study published by the same author (Vintzileos et al.,
1995b) had the goal of comparing fetal acidemia at birth between groups
of women in either a continuous EFM group or IA.  Interestingly, his
study used the very same 1428 Greek women from the original,
controversial 1993 study, where a coin was flipped to assign them to EFM
or IA.  Both groups of patients had a one to one nurse / patient ratio
for their labor.  After delivery, a section of the umbilical cord was
clamped, and blood gases obtained from both the artery and the vein, and
the acid-base measurement was performed within 10 minutes of delivery.
Although his first study showed no difference between rates of fetal
acidosis in the two groups (Vintzileos et al., 1993), this later
retrospective study of the same infants showed a better sensitivity and
greater positive predictive value (37% vs 22%) of EFM as compared to IA
in detecting fetal acidemia at birth (defined as arterial cord blood less
than 7.15) (Vintzileos, 1995b).  This is difficult to understand, and
both the method of randomization (Neilson, 1994) and the results of the
neonatal acidemia measured are somewhat suspect.  No explanation was
given by the author to explain this apparent discrepancy.


In discussing acidemia, it is important to review the process of fetal
asphyxia, since the lack of oxygen and the possibility of brain damage or
death is the reason that fetal heart rates are even evaluated in labor.
When the placenta is not exchanging oxygen and carbon dioxide in a normal
way (secondary to abruption, IUGR, etc.) the fetus begins to accumulate
carbon dioxide, and the baby becomes hypoxic. The fetus redistributes its
blood flow during a hypoxic episode, in response to chemoreceptors
activated in the autonomic nervous system,  so that the brain, heart, and
adrenals receive the most oxygen.  Rapid gasping occurs in utero, as the
fetus attempts to blow off carbon dioxide and increase its oxygen levels.
 The blood pressure initially rises, and primary apnea results.  The
respiratory efforts then cease, and the heart rate and blood pressure
both fall, activating baroreceptors in the process.  However, if provided
with tactile stimulation, the fetus would likely respond with respiratory
effort or an increase in the heart rate (Blackburn & Loper, 1992;
Prentice & Lind, 1987).  

After 2-4 minutes of asphyxia, another stage (secondary apnea) begins as
the fetus begins slow and deep gasping attempts.  The heart and blood
pressure continue to fall, and respiratory efforts stop completely after
7-8 minutes.  Tactile stimulation would not increase the fetal heart rate
at this point, and if born during this time, the infant would need
medications and positive pressure ventilation to recover.  After 8
minutes of complete asphyxia, brain damage begins, and death usually
follows after about 13 minutes (Blackburn & Loper, 1992).  Fortunately
however, when a fetus is appropriately monitored with EFM or IA, and
corrective actions are taken, asphyxia rarely occurs.    


Contrary to Vintzileos' studies (Vintzileos et al., 1993; Vintzileos et
al., 1995a; Vintzileos et al., 1995b) the majority of high quality,
published studies that have been done on EFM in the past fifteen years
have not shown improved perinatal outcomes with the use of continuous
electronic monitoring (Herbst & Ingemarsson, 1994; MacDonald et al.,1985;
Shy et al., 1990; Thacker et al., 1995).  A large randomized, prospective
study (MacDonald et al., 1985) from Dublin was done in the early 1980's,
and compared women in labor using continuous EFM to those monitored by
IA.  The intent of the study was to conduct research that was
methodologically superior from the observational studies that had been
done in the 1970's, so that better information could be obtained about
the outcome of EFM, compared to IA.  A large sample was obtained for this
study (12, 964), and randomization was used to allocate women to the EFM
or IA group by opening a serially numbered envelope.   In this setting in
Ireland, there was strict criteria for inclusion in the study.  Women
were not admitted to the labor ward until they were determined to
actually be in labor, and a one to one nursing ratio was always provided
for each patient.  Within an hour of admission, the amniotic fluid was
assessed, with an amniotomy performed if necessary.  The 95% of women who
did not have significant meconium were then included in the study, and
monitored either with EFM or IA.   Fetal scalp sampling was also
performed in both groups, when indicated.  The results showed no
difference between the number of neonatal deaths, low Apgar scores, and
need for infant resuscitation in both groups.  Only in the IA group was
there an increased number of neonatal seizures noted (1.0 odds ratio
compared to .45 for the EFM groups) (MacDonald et al., 1985).  However,
the hospitals in Ireland have long been known for their aggressive
obstetrical style (oxytocin used in approximately 23% of both groups). In
addition, the very low cesarean (2.4% with EFM, 2.2% with IA) and
epidural rates (approximately 3.1% in both groups) (MacDonald et al.,
1985) may mean that this otherwise helpful study would not apply in every

Another recent prospective study (Herbst & Ingemarsson, 1994) proposed to
evaluate the difference between groups of low to moderate risk women in
labor with either continuous EFM or intermittent electronic fetal
monitoring (rather than auscultation with a Doppler or stethoscope), in
order to compare each method's efficacy in detecting fetal hypoxia.  4044
women in Sweden were asked to participate in the study when they entered
the hospital, and were randomized to each group by opening an envelope
from a package of previously prepared shuffled envelopes.   During the
first stage of labor, the intermittent group was placed on the monitor
for 10 to 30 minutes every 2 to 2 1/2 hours, with auscultation by a
midwife every 15 to 30 minutes that the patients were off the monitor.
Both groups were evaluated by continuous EFM in the second stage of
labor.  The results showed no significant difference in any neonatal
outcome (cord arterial pH < 7.10  1.9% for IA vs 2.8% for EFM) , ominous
fetal heart rate patterns, or the cesarean rate (1.8% for IA vs 2.4% for
EFM).  The authors conclude that intermittent use of the fetal monitor is
as safe in labor for women as continuous EFM.  They further suggest that
there are advantages in encouraging low to moderate risk women to be
intermittently monitored in labor, and that individually assessing a
woman's risk factors as to the appropriateness of intermittent evaluation
would serve her better than rigid policies that require continuous EFM
(Herbst & Ingemarsson, 1994).  Unfortunately, both groups used electronic
fetal monitors, and it is unknown whether these same reassuring results
would have occurred with intermittent auscultation as the alternative to

In the ongoing debate over continuous EFM vs periodic auscultation,
proponents of routine electronic monitoring claim that EFM likely saves
many infants from neurologic insults that would have occurred if they had
been monitored intermittently.  A randomized, prospective clinical trial
was done to compare neurologic outcome (specifically cerebral palsy) of
preterm infants who were monitored either by EFM or IA (Shy et al.,
1990).  The study compared 189 children who were born with a weight of
1750 grams or less, and compared their mental and psychomotor development
at 4, 8 and 18 months of age.  The outcome was that cerebral palsy was
much higher in the group who received continuous EFM (20%), compared to
the group who was monitored by IA (8%).  In addition, the mental and
motor development scores were higher in the IA group (104.9 and 98.3),
compared to the babies who were electronically monitored (100.5 and 94)
(Shy et al., 1990).  The conclusion of the authors is that "clinicians
should be attentive to the potentially adverse outcomes associated with
electronic fetal monitoring" (Shy et al., 1990, p. 593).  However, this
was a rather small sample of preterm infants, and it is difficult to
understand how cerebral palsy would occur more frequently (20% vs 8%) as
an outcome of monitoring alone.

Perhaps the most impressive recent retrospective study is one with a very
large sample size of 58,855 pregnant women.  The authors wanted to
explore the contradiction between routine and widespread use of EFM with
the recent expert recommendations to limit routine use of this technology
(Thacker, Stroup, & Peterson, 1995).  Twelve randomized clinical trials
on EFM vs IA were obtained, and data from these trials were used to
obtain relative risk (RR) for each of eight outcomes .  These outcomes
were 1 minute Apgar scores less than 7, 1 minute Apgar scores less than
4, neonatal seizures, neonatal intensive care unit admissions,
stillbirths, neonatal deaths, cesarean delivery, and operative vaginal
delivery.  Cumulative meta-analysis was done on all the data in order to
maximize accuracy.    

Except for a reduction of neonatal seizures with EFM (RR 0.5), the
results of this very large study showed that routine EFM has no impact on
perinatal morbidity and mortality (Thacker et al., 1995).  Only in non-US
studies was there a very slight protective effect shown for EFM and Apgar
scores less than 4 (RR 0.82).  No significant difference was seen in the
1 minute Apgar scores less than 7 (RR 1.02  for the EFM groups) or
perinatal deaths (RR 0.83 for EFM).  Cesarean rates were higher in the
EFM group of women (RR 1.33),  and the risk of cesarean delivery was
highest in low-risk women (RR 2.05).  The authors conclude that EFM was
introduced into widespread clinical practice before data from controlled
trials were available.  Since efficacy and safety of routine continuous
EFM has not been supported by most research,  they believe that "the
benefits once claimed for EFM are clearly more modest than once believed
and appear to be primarily in the prevention of neonatal seizures"
(Thacker et al., 1995, p. 619).  However, since neonatal seizures may be
associated with poor outcomes such as long-term disabilities and eventual
neonatal deaths, the study may be presenting a more positive picture than
is actually true concerning the safety of IA.



These comparative studies were researched during an extensive literature
search for information on neonatal and maternal outcomes with both
methods of fetal surveillance.  MEDLINE was consulted in an Internet
search for appropriate articles on EFM and IA in labor.  Once the
abstracts were obtained, the actual full-text articles were retrieved by
a medical librarian at Longmont United Hospital in Longmont, Colorado.
In addition, the medical library at the University of Colorado Health
Sciences Center was visited, and many research articles obtained there,
which had not been available on MEDLINE.  These journal articles then
were analyzed with regards to the inquiry question of : are there
differences between the outcomes of auscultated labors versus electronic
monitoring?  What practice guidelines could be developed for intermittent
auscultation that would be based on the best research available on the

Evidence considered

As previously described, a literature search was performed on the topics
of electronic fetal monitoring and intermittent auscultation of the fetus
in labor.  In all, 31 articles were obtained on the subject.  Sixteen
original clinical studies were selected for consideration in developing
an evidence-based clinical practice guideline.  These studies are
summarized in an evidence table, Appendix A.

Criteria established for effectiveness

In order to be considered for use in this project, the study had to be
either an original randomized, controlled clinical trial, or a meta
analysis of several clinical trials done previously.  In addition, the
study had to include both of the outcomes of interest (1. intermittent
auscultation as compared to other types of fetal monitoring,  2. neonatal
outcome), and had to provide a rigorous scientific comparison between the
outcomes of electronic fetal monitoring and intermittent auscultation.
Clinical perspectives represented

The studies that were analyzed for this protocol were all performed by
physicians, some of whom tested the hypothesis that EFM was better than
IA on neonatal indices, and others who hoped to demonstrate the safety of
intermittent auscultation in labor.  Other resources that were utilized
in the development of this project were written by registered nurses and
certified nurse-midwives.  Thus, the clinical perspectives included
medicine (obstetrics and gynecology), nursing, and nurse-midwifery,
making the evidence interdisciplinary.  


The clinical practice guidelines that are presented in this paper are
applicable to all women who are laboring in a hospital, and therefore
might have a fetal monitor used at some time during their labor.  The
guidelines for intermittent auscultation might also be utilized in an
out-of-hospital birth center or for a home delivery, although in both of
those situations, fetal monitors are not normally used, unless a transfer
was anticipated to a hospital.  Exceptions to the guidelines would
concern those whose high risk history make continuous fetal monitoring a
good choice.  This will be discussed in more detail in the following

Synthesis of evidence and summary statement of findings

Since most low risk women in labor are routinely placed on continuous
fetal monitoring in labor, a significant problem exists, since the
technology of EFM has not been shown to be effective in reducing neonatal
neurologic impairment (Shy et al., 1990) or perinatal mortality (Thacker
et al., 1995).  The efficacy and safety of any intervention should be
shown by randomized clinical trials before it becomes routine practice,
which in the case of EFM, has been made part of the standard of care in
obstetrics from the 1970's until now (Thacker et al., 1995). In the early
1970's, physicians failed to insist on large, rigorous randomized
clinical trials of EFM before the technology became widespread, and EFM
was made a part of the standard of care for most women (Tallon, 1996).
In addition, the early EFM trials (though flawed) appeared to be
successful in reducing neonatal mortality, so that little effort was made
to define exact criteria as to what constituted fetal distress (Cibils,
1996; Tallon, 1996).  Today,  many clinicians use EFM in an effort to
avoid lawsuits, believing that if EFM may do no good, at least it does no
harm.  However, this is not the case, since the rates of cesarean and
operative birth are consistently higher when women are managed by EFM,
and the perinatal outcomes are no better (Prentice & Lind, 1987).

Intermittent auscultation has been shown in multiple randomized clinical
trials over the past 25 years to be as safe as continuous fetal
monitoring ( Elison, Foster, Sheridan-Pereira, & MacDonald, 1991;
Haverkamp, Thompson, McFee, & Centrulo, 1976; Haverkamp et al., 1979;
Kelso et al., 1977, Luthy et al., 1987; MacDonald, Grant,
Sheridan-Pereira, Boylan, & Chalmers, 1985; Shy et al., 1990; Thacker,
1986; Thacker, Stoup, & Peterson, 1995; Vintzileos et al., 1995; Wood et
al., 1981.)  Unfortunately, a perception exists that intermittent
auscultation always requires a one: one nursing ratio, and thus is many
times made unavailable to laboring patients (Sandmire & DeMott, 1995).  

Outline of Evidence-based Clinical Practice Guidelines / Algorithm

Please see Appendix B


The evidence-based clinical practice guidelines contained in this project
are based upon the best available research over the past 25 years.  As a
result of this guideline and other similar projects, intermittent
auscultation will be used more freely for most healthy, low-risk women in
labor.  As further research is performed in the years to come, more data
will be added to the body of knowledge that is available regarding
neonatal outcomes and surveillance of the fetus in labor. This
evidence-based clinical practice guideline will be reviewed annually, in
order to remain current with the best research available.  As more
information is collected by midwives, nurses, and physicians, the
clinical guidelines will be modified in order to better honor the choice
of laboring women, while maintaining the good outcomes for which midwives
are known.


Intermittent auscultation is one of two recommended methods of monitoring
fetal well-being in labor, even if the woman has risk factors
(Guidelines for perinatal care, 1997)  Unfortunately, care providers have
not paid attention to recent research demonstrating the safety of IA.
According to Albers (1994), "it is unclear whether they do not read,
understand, or believe the research, or whether they are unwilling to
challenge the systems in which they provide care" (p. 109).  Care for
laboring women should be evidence-based, rather than relying on old
traditions and routines that provide comfort for the clinician, rather
than the laboring woman.  Laboring women and their families need
information about their options, so that they can make an informed choice
about IA (Sandmire,1990; Sandmire & Demott, 1995).  It is time that
practice guidelines for intermittent auscultation be used, promoted, and
published extensively in the obstetric world, so that cesarean rates may
be dramatically decreased, and women's needs may be honored by those
attending her birth.  


      Albers, L.(1994). Clinical issues in electronic fetal monitoring.
Birth, 21 (2), 108-110.

      Blackburn, S., Loper, D. (1992).  Maternal, fetal and neonatal
physiology: a clinical perspective.  Philadelphia: W.B.Saunders.        
      Cibils, L.(1996).  On intrapartum fetal monitoring.  American
Journal of Obstetrics and Gynecology, 174 (4), 1382-1389.

      Elison, Foster, Sheridan-Pereira, MacDonald. (1991).  Electronic
fetal heart monitoring, auscultation, and neonatal outcome.  American
Journal of Obstetrics and Gynecology, 164 (1), 1281-1289.

      Flamm, B. (1994).  Electronic fetal monitoring in the United
States.  Birth, 21 (2), 105-106.
      Freeman, R. (1990).  Intrapartum fetal monitoring - a disappointing
story.  The New England Journal of Medicine, 322 (9), 624-626.      

      Garcia, J., Corry, M., MacDonald, D., Elbourne, D., Grant,
A.(1985). Mother's views of continuous electronic fetal heart monitoring
and intermittent auscultation in a randomized controlled trial.  Birth,
12 (2), 79-86.

      Guidelines for perinatal care (4th ed.).  (1997) . Washington, DC:
American College of Obstetricians and Gynecologists.

      Haverkamp, Orleans, Langendoerfer, McFee, Murphy, Thompson. (1979).
 A controlled trial of the differential effects of intrapartum
monitoring.  American Journal of Obstetrics and Gynecology, 134 (4),

      Haverkamp, Thompson, McFee, Centrulo. (1976).  The evaluation of
continuous fetal heart rate monitoring in high-risk pregnancy.  American
Journal of Obstetrics and Gynecology, 125 (3), 310-320.      

      Herbst, A., Ingemarsson, I. (1994).  Intermittent versus continuous
electronic monitoring in labour: a randomised study.  British Journal of
Obstetrics and Gynaecology, 101, 663-668.      

      Keirse, M. (1994).  Electronic monitoring: who needs a trojan
horse?  Birth, 21 (2), 111-113.

      Kelso, Parsons, Lawrence, Arora, Edmonds, Cooke. (1977).  A
controlled trial of the differential effects of intrapartum fetal
monitoring.  American Journal of Obstetrics and Gynecology, 131 (5),
      Levano, Cunningham, Nelson, Roark, Williams, Guzicle, Dowling,
Rosenfeld, Buckley. (1986).  A prospective comparison of selective and
universal electronic fetal monitoring in 34,995 pregnancies.  The New
England Journal of Medicine, 315 (10), 615-619.      

      Luthy, Shy, van Belle, Larson, Hughes, Benedetti, Brown, Effer,
King, Stenchever. (1987).  A randomized trial of electronic fetal
monitoring in preterm labor.  Obstetrics and Gynecology, 69 (5), 687-695.
      MacDonald, D., Grant, A., Sheridan-Pereira, M., Boylan, P.,
Chalmers, I. (1985).  The dublin randomized controlled trial of
intrapartum fetal heart rate monitoring.  American Journal of Obstetrics
and Gynecology, 152 (5),  524-539.

      Orleans, M. (1985).  Lessons from the dublin study of electronic
fetal monitoring.  Birth, 12 (2), 86.      

      Prentice, A., Lind, T. (1987).  Fetal heart rate monitoring during
labour - too frequent intervention, too little benefit?  The Lancet, 2

      Renou, Chang, Anderson, Wood. ( 1976).  American Journal of
Obstetrics and Gynecology, 126 (4), 470-476.    

      Sandmire, H.(1990).  Whither electronic fetal monitoring?
Obstetrics and Gynecology, 76 (6), 1130-1134.

      Sandmire, H., DeMott, R. (1995).  Auscultation of the fetal heart
presents advantages over electronic monitoring.  Wisconsin Medical
Journal, 94 (12), 661-663.

      Shy, K., Luthy, D., Bennett, B., Whitfield, M., Larson, E., Van
Belle, G., Hughes, J., Wilson, J., Stenchever, M. (1990).  Effects of
electronic fetal-heart-rate monitoring, as compared with periodic
auscultation, on the neurologic development of premature infants.  The
New England Journal of Medicine, 322 (9), 588-593.

      Spencer, J. (1994).  Electronic fetal monitoring in the united
kingdom.  Birth, 21 (2), 106-107.

      Tallon, R. (1996).  Electronic fetal monitoring.  Nursing
Management, June, 1996, 49-51.
      Thacker, S. (1987).  The efficacy of intrapartum electronic fetal
monitoring.  American Journal of Obstetrics and Gynecology, 156 (1),

      Thacker, S., Stroup, D., Peterson, H.(1995).  Efficacy and safety
of intrapartum electronic fetal monitoring: an update.  Obstetrics and
Gynecology, 86 (4), 613-620.  

      Varney, H. (1997).  Varney's midwifery (3rd edition).  Sudbury,
Massachusetts :  Jones and Bartlett.

     Vintzileos, A., Antsaklis, A., Varvarigos, I., Papas, C., Sofatzis,
I., Montgomery, J. (1993).  A randomized trial of intrapartum electronic
fetal heart rate monitoring versus intermittent auscultation.  Obstetrics
and Gynecology, 81 (6), 899-907.        

      Vintzileos, A., Nochimson, D., Guzman, E., Knuppel, R., Lake, M.,
Schifrin, B. (1995a).  Intrapartum electronic fetal heart rate monitoring
versus intermittent auscultation:  a meta-analysis.  Obstetrics and
Gynecology, 85 (1), 149-155.

      Vinzileos, A., Nochimson, D., Antsaklis, A., Varvarigos, I.,
Guzman, E., Knuppel, R.  (1995b).  Comparison of intrapartum electronic
fatal heart rate monitoring versus intermittent auscultation in detecting
fetal acidemia at birth.  American Journal of Obstetrics and Gynecology,
173 (4), 1021-1024.

      Wood, Renou, Oats, Farrell, Beischer, Anderson.  (1981).  A
controlled trial of fetal heart rate monitoring in a low-risk obstetric
population.   American Journal of Obstetrics and Gynecology, 141 (5),
. . . . . . .