Fetal Heart Rate Response

to Maternal Exertion
Marshall W. Carpenter, MD; Stanley P. Sady, PhD; Bente Hoegsberg, MD; Mina A. Sady, MA; Barbara Haydon, BSN;
Eileen M. Cullinane, MS; Donald R. Coustan, MD; Paul D. Thompson, MD

Doppler monitoring of fetal heart rates during maternal exertion has suggested ments and distribution of brochures.
that fetal bradycardia occurs frequently during vigorous exercise, causing Women were excluded from the study if
concern for fetal safety. Doppler determination of fetal heart rate during vigorous they gave a history of medical problems
maternal effort is difficult. To avoid motion artifact, we observed fetal heart rate (such as asthma) or previous abnormal
using two-dimensional ultrasound and determined the incidence of fetal brady- pregnancies (such as preterm labor)
cardia in 45 pregnant women (age, 29.0\m=+-\3.7years [mean \m=+-\SD]; gestational that might compromise their ability to
age, 25.2 \m=+-\3.0 weeks) during 85 submaximal and 79 maximal cycle ergometer
complete the exercise test sessions. All
tests. Average fetal heart rate did not change during exercise. A single episode subjects provided written informed
consent following an orientation visit to
of fetal bradycardia (heart rate <110 beats per minute for \m=ge\10s) occurred the exerciselaboratory. The protocol
during submaximal exertion during a maternal vasovagal episode. Sixteen was approved by the institutional re¬
episodes of fetal bradycardia were noted within three minutes after cessation of view boards of all participating in¬
exercise, 15 of which followed maximal maternal effort. We conclude that brief stitutions.
submaximal maternal exercise up to approximately 70% of maximal aerobic Procedure
power (maternal heart rate \m=le\148beats per minute) does not affect fetal heart
rate. In contrast to submaximal maternal exertion, maximal exertion is common- Subjects were studied twice on sepa¬
ly followed by fetal bradycardia. This may indicate inadequate fetal gas ratedays, using an identical protocol
each time. Exercise tests were per¬
exchange. formed on an electrically braked cycle
(JAMA 1988;259:3006-3009) ergometer (Collins Pedal Mate, Brain-
tree, Mass). At each exercise test ses¬
sion, resting measurements were ob¬
VIGOROUS exertion in pregnant ewes rate decelerations are associated with
tained with subjects seated on the
reduces uterine blood flow and fetal ar¬ fetal hypoxia, such results raise concern ergometer for ten minutes (Fig 1). Sub¬
terial oxygen tension1,2 and may produce about fetal safety during vigorous exer¬ jects then cycled at 0, 30, and 60 W for
six minutes at each work load to obtain
significant fetal hypoxia. Recent human cise. Doppler measurement of fetal
submaximal exercise measurements.
studies using Doppler monitors to esti¬ heart rates during exercise is difficult,
mate fetal heart rate have reported pro¬
This was followed by five minutes of
however, because of artifact associated cool-down cycling against minimal re¬
longed fetal bradycardia even during with maternal motion.8'9 Consequently,
sistance and ten minutes of rest. The
low-intensity maternal exercise.3"5 Oth¬ we used two-dimensional ultrasonic im¬
maximal exercise session started with
ers have noted increased fetal heart rate aging to record the fetal heart rate re¬
immediately following cycling6 and run¬ sponse to maternal exercise. Our pur¬ approximately 60 W of resistance and
consisted of two-minute stages of incre¬
ning7 exercise. Since episodic fetal heart pose was to determine the effect of mental exertion until subjects were un¬
submaximal and maximal maternal ex¬
ertion on fetal heart rate and the inci¬ able or unwilling to exercise further.
From the Department of Obstetrics and Gynecology, dence of fetal bradycardia. Subjects continued to cycle for five min¬
utes against minimal resistance. They
Women and Infants Hospital (Drs Carpenter, Hoegs-
berg, and Coustan and Ms Haydon), and the Depart- METHODS then assumed a semirecumbent position
ment of Medicine, Miriam Hospital (Drs Sady and for an additional seven minutes of ultra¬
Thompson and Mss Sady and Cullinane), Brown Uni- Subjects sound observation followed by a non-
versity Program in Medicine, Providence, RI.
Reprint requests to Women and Infants Hospital, 101 Pregnant women were recruited into stress test. This latter test documents
Dudley St, Providence, RI 02905 (Dr Carpenter). the study through television announce- fetal heart rate acceleration in response

Downloaded From: http://jama.jamanetwork.com/ by a New York University User on 06/13/2015


100
75
50
25
Resti Submaximal Exertion
10 16 22 28
Exercise Period and
Fig 1.—Exercise protocol: duration of rest and exercise
aerobic power.
Table 1.—Exercise Protocol Completion by Subjects (
Submaximal
and Maximal
Session Exertion
1 42
2 37
to fetal movement. Such a response is
physiological and is accepted as exclud¬
ing fetal asphyxia.10
Maternal blood pressure was mea¬
sured with a mercury sphygmomanom-
eter at each work load and every minute
for the five minutes following maximal
exertion. Maternal heart rate and oxy¬
gen uptake were measured during the
fourth and fifth minutes of each submax¬
imal exercise stage and during each
minute of the maximal exertion part of
the protocol. Maternal heart rate was
measured using an electrocardiogram
(Marquette Instruments, Milwaukee).
Oxygen uptake was determined using
standard open-circuit spirometrie tech¬
niques.11 Expired volume was measured
with a Fleisch pneumotach attached to a
differential pressure transducer (Vali-
dyne MP-45, Northridge, Calif), and ex¬
pired gas concentrations were deter¬
mined with a mass spectrometer
(Perkin Elmer MGA 1100, Pomona,
Calif).
Fetal cardiac activity was continu¬
ously monitored and videotaped during
exercise and rest periods using a linear
array two-dimensional ultrasound sys¬
tem (Toshiba SLA-77A, Nasu, Japan).
Videotapes were subsequently re¬
viewed in entirety by an experienced
Downloaded From: http://jama.jamanetwork.com/ by a New York University User on 06/13/2015
Maximal
Rest 2 Exertion Rest 3
-1t—
33 43 ~53 —58
Elapsed Time, min
periods and associated percentage of maximal
=
45)
No. of Subjects
observer (M.W.C.) who determined the
fetal heart rate during each minute of
the exercise session. This was calcu¬
lated from the elapsed time for the first
observed 20 cardiac cycles within each
minute of the protocol. Fetal bradycar¬
dia was defined as a heart rate below 110
beats per minute for more than 10 s at
any time during the testing session.
Bradycardia duration was timed from
onset to return of heart rate to pre-
exercise values.
To determine the reproducibility of
fetal heart rate estimations, we re¬
peated heart rate measurements ten
times on the same videotape segment
for five randomly selected subjects at
each of the five rest and exercise peri¬
ods. The mean coefficient of variation
was 2.2% or less for each period.
Data
Analysis
Satisfactory submaximal and maxi¬
mal exercise data were obtained during
both exercise sessions in 37 subjects
(Table 1). Three of the first sessions and
three of the second sessions included
only the submaximal portion of the pro¬
tocol. Five subjects did not attempt a
second exercise session: one because of
uterine contractions, one because of
exercise-associated hypotension, and
three because of subject concerns about
fetal bradycardia. Consequently, 79
maximal and 85 submaximal exercise
tests among 45 subjects form the basis
of this report. Statistical tests were per¬
formed using Student's t test for inde¬
pendent means or the Wilcoxon rank-
sum test if a normal distribution was not
present. Fisher's exact test was used to
determine significant associations be¬
tween categorical variables.
RESULTS
Forty-five women were enrolled. The
mean age of the women was 29.2 ±3.7
years, and their mean gestation was
25.2 ±3.0 weeks (range, 20 to 34
weeks). All subjects had uncomplicated
pregnancies. None smoked cigarettes
or took any medications except for pre¬
natal vitamins and iron. Twenty-one re¬
ported that they had exercised vigor¬
ously more than once a week prior to
pregnancy, and 16 reported that they
engaged in regular exercise during
pregnancy. Only three of the 16 exercis¬
ers performed prolonged exercise (96 to
112 km of running per week in one and
>8 km per week of swimming in two).
The others participated in recreational
exercise of undetermined intensity.
Peak maternal oxygen uptake (V02)
among completed protocols averaged
25.9 ±4.2 mL-kg'-min' (mean ± SD)
(range, 18.0 to 39.0 mL-kg'-min"'), and
peak maternal heart rate was 180 ±8.6
beats per minute (range, 156 to 197
beats per minute). In contrast, at the
end of the submaximal portion of the
protocol, the V02 was 17.2 ±2.4
mL-kg~'-mnT' and maternal heart rate
was 148 ± 16 beats per minute. This in¬
dicates that during the submaximal por¬
tion of the protocol, subjects attained
VO2 values equaling 67.4% ±11.8% of
their measured maximal aerobic power
(V02max).
Women who exercised regularly dur¬
ing pregnancy had a higher mean peak
V02 (29.6±3.9 mL-kg'-min1) than less
active subjects (24.1 ± 3.2 mL-kg'-min ')
(P<.01). In 57 maximal exercise ses¬
sions (72%), an increase in work rate
elicited less than a 150-mL/min increase
in V02, thus meeting widely accepted
criteria for maximum oxygen uptake
(Vo2 max).12 There was no difference in
peak V02 or peak maternal heart rate
between subjects who did and did not
satisfy this VOçjnax criterion (26.0 ±3.5
vs 25.7 ± 5.8 mL-kg'-min and 180 ± 8 vs
'
179 ± 9 beats per minute, respectively).
This confirms that exercise effort was
maximal or near maximal.
There were no significant differences
in mean fetal heart rates between the
preexercise period and subsequent ex¬
ercise and rest periods (Table 2). Com-
 Table 2.—Mean Fetal Heart Rate During Rest and
Exercise Periods
No. of Average Fetal
Sessions Heart Rate,
Completely Beats min,
Condition Recorded Mean ± SD
Resti 81 147±6
Exercise
OW 80 148±7
30 W 80 148±7
60 W 80 148±8
Rest 2 80 148±7
Maximal
exercise 77 148 + 7
Rest 3 76 143±14
parison of mean fetal heart rates during
first and second exercise sessions for
each subject also showed no significant
differences.
Eighteen episodes of fetal bradycar-
dia were associated with the 85 exercise
test sessions, but only one episode oc¬
curred during exercise. This subject
sustained vasovagal hypotension dur¬
ing the second submaximal work load.
Two episodes of bradycardia in the same
fetus occurred on different exercise
days; one prior to and one following the
submaximal portion of the exercise pro¬
tocol. Therefore, 15 of 16 postexercise
fetal bradycardias occurred after maxi¬
mal maternal exertion, an incidence of
16.2% compared with 1.2% after sub-
maximal exertion (P .0003).
=
Peak maternal oxygen uptake was
higher during maximal exercise ses¬
sions that were followed by fetal
bradycardia (29.0 ±5.3 vs 25.1 ±3.6
mL-kg-'-min"1) (P .015).
=
Subjects
achieved a plateau of oxygen uptake in a
similar proportion of sessions (12/15
[80%] with and 44/63 [70%] without bra¬
dycardia). Similarly, there were no dif¬
ferences in peak maternal heart rate
between sessions with (180 ±8 beats
per minute) and without (180 ± 10 beats
per minute) bradycardia. A maternal
history of regular exercise during preg¬
nancy was noted in 21 (33%) of 64
sessions without fetal bradycardia and
in five (33%) of 15 sessions with
bradycardia.
Maternal and fetal age did not differ
in sessions with and without fetal brady¬
cardia (28.9 ±4.0 vs 29.4 ±3.8 years
[P>.05] and 26.4 ±3.4 vs 25.9±3.0
weeks [P>.05], respectively). Similar¬
ly, the duration of the maximal exercise
period, maternal blood pressure during
exercise, and the fall in maternal blood
pressure after maximal exertion did not
differ between these groups (Table 3).
Fetal bradycardia occurred 96 ± 51 s
(range, 31 to 180 s) after maximal exer¬
tion and lasted a median of 90 s (range,
13 to 561 s). The lowest fetal heart rate
during each postmaximal exertion bra¬
dycardia determined from ten cardiac
cycles averaged 73 ± 21 beats per min-
Downloaded From: http://jama.jamanetwork.com/ by a New York University User on 06/13/2015
Table 3.—Exercise Variables in Sessions Where
Fetal Bradycardia Was Present or Absent After
Maximal Exercise
Fetal
Bradycardia, Mean ± SD
Present Absent
Variable (n 15)
=
(n=64)
Peak maternal
heart rate,
beats/min 180.0 ±9.7 180.0 ±8.4
Peak maternal
oxygen uptake,
mL-kg-^min-1 29.0 ±5.3* 25.1 ±3.6
Duration of maximal
exercise period, min 7.4 ±2.5 6.9 ±1.8
Systolic pressure,
mm Hgt 39.0 ±18.7 36.0 ±15.6
Diastolic pressure,
mm Hgt 18.0±21.8 10.0 ±12.0
Mean arterial
pressure, mm Hgt 21.0±14.4 17.0 ±10.5
*P<.01, unpaired r test.
tRefers to blood pressure at maximal exertion minus
lowest pressure during recovery and includes 12 ses¬
sions with and 58 without fetal bradycardia.
ute (range, 43 to 102 beats per minute).
All episodes showed a rapid decelera¬
tion and most rapidly returned to base¬
line (Fig 2). Marked short-term varia¬
tion of heart rate characterized most
bradycardic episodes, although the two
episodes lasting greater than nine min¬
utes returned to baseline only slowly.
Nevertheless, reactive nonstress tests
(fetal cardiac acceleration with fetal
movement) were observed within 30
minutes after conclusion of all exercise
sessions.
Fetal bradycardia following maximal
exertion occurred during both exercise
test sessions in three subjects, only dur¬
ing the first session in seven subjects,
and only during the second session in
two subjects. Three subjects, however, 12 3 4 5 6 7 8 9101112
elected not to perform a second maximal
test becuse of fetal bradycardia during Time, min
the initial session. Consequently, three
of the seven subjects with bradycardia
Fig 2.—Fetal heart rate following maximal exertion
during the first exercise session who during 14 episodes of fetal bradycardia. Fetal heart
exercised twice had a repeated episode rates were averaged over ten cardiac cycles every
after their second session. 30 s during postexercise period. One tape was lost.
Subjects gave birth at 40.0 ±0.8 Predeceleration baseline fetal heart rate and nadir
weeks of gestation, and none had pre- fetal heart rate are noted for each deceleration. Zero
time equals time of cessation of maximal exertion.
term labor or bleeding. Fetal bradycar¬
dia following maximal maternal exer¬
tion did not predict future perinatal
problems. By history from the attend¬ cell tumor of the brain and died at 5
ing obstetricians, there was only one weeks of age. A third child suffered sud¬
abnormal fetal heart rate pattern intra- den infant death syndrome at 5 weeks of
partum, in an infant born with group age. None of these infants demonstrat¬
ß-hemolytic streptococcal sepsis who ed fetal bradycardia during this study.
died the day of birth. There were no
reported cord entanglements. Forty- COMMENT
three ofthe 45 infants had normal Apgar Renal13 and blood flow
splanchnic1,14
scores and normal physical examination decreases during exercise
in the non-
results at birth. Of those with abnormal pregnant and pregnant animal. This re¬
physical examination results, one was duction is proportional to the duration
the infant with streptococcal sepsis, and and intensity of exercise and is medi¬
the other had an undifferentiated germ ated by sympathetic vasoconstriction.15
 Pregnant ewes exercised to exhaustion
on a treadmill manifest a significant fall
in fetal oxygen tension without a de¬
creased fetoplacental oxygen extraction
or net lactate production.2 Mild bicycle
exercise in humans in the supine posi¬
tion reduced sodium 24 washout from
the myometrium in one study,'6 sug¬
gesting a 25% reduction in perfusion.
Heavy maternal exercise, therefore,
may challenge fetal respiratory homeo-
stasis and result in fetal hypoxia.
Studies of maternal exercise in
humans have examined changes in fetal
heart rate as an indicator of distur¬
bances of fetal gas exchange. Fetal
heart rate observations during and
after exercise have used Doppler moni¬
tors of the type used with inactive
women during labor. In some studies,
this technique has revealed increases in
fetal heart rate during maternal recov¬
ery following 30 minutes of submaximal
exertion.6,7 Others, using Doppler moni¬
toring of women in the third trimester,
have reported incidence of fetal bra¬
an
dycardia during exercise of 11% to
100%, even at low work loads.3"5
We and others,58,9 however, have ob¬
served that Doppler fetal heart rate
monitoring during exercise is difficult
because of artifact related to maternal
movement. Consequently, the inci¬
dence of fetal bradycardia during ma¬
ternal exercise and the risk of maternal
exertion have been questioned.8,9 Two-
dimensional ultrasound imaging, in con¬
trast, appears to be accurate and
reliable.
In our study, we found the mean fetal
cardiac rate to be stable during and af¬
ter maternal exertion and independent
of exercise intensity. The duration of
our exercise protocol, however, was
References
1. Lotgering FK, Gilbert RD, Longo LD: Exercise
responses in pregnant sheep: Blood gases, tem-
peratures, and fetal cardiovascular system. J Appl
Physiol 1983;55:842-850.
2. Clapp JF: Acute exercise stress in the pregnant
ewe. Am J Obstet Gynecol 1980;136:489-494.
3. Artal R, Romem Y, Paul RH, et al: Fetal brady-
cardia induced by maternal exercise. Lancet
1984;2:258-260.
4. Jovanovic L, Kessler A, Peterson CM: Human
maternal and fetal response to graded exercise. J
Appl Physiol 1985;58:1719-1722.
5. Artal R, Rutherford S, Romem Y, et al: Fetal
heart rate responses to maternal exercise. Am J
Obstet Gynecol 1986;155:729-733.
6. Collings C, Curet LB: Fetal heart rate response
to maternal exercise. Am J Obstet Gynecol
1985;151:498-501.
7. Hauth JC, Gilstrap LC, Widmer K: Fetal heart
Downloaded From: http://jama.jamanetwork.com/ by a New York University User on 06/13/2015
less than the 30 minutes previously re¬
sulting in postexercise fetal tachycar¬
dia.6,7 Our studies also contrast with pri¬
or observations in finding no
unexplained fetal bradycardia during
maternal exertion, despite work loads
at or near maximal maternal oxygen
uptake.
Subjects who had higher peak Vo2
during maximal voluntary exertion
were more likely to demonstrate post-
exercise fetal bradycardia. This asso¬
ciation suggests that higher exercise
capacity may predispose to fetal brady¬
cardia after maximal exertion. Subjects
reporting frequent recreational exer¬
cise had no higher incidence than seden¬
tary individuals, but we did not quantify
the level of physical activity in subjects
outside the laboratory. Both maternal
heart rate and Vo2 values among
subjects with and without postexercise
fetal bradycardia indicate that both
groups had peak efforts at or near their
VOjinax.
The physiology of postexertional fetal
bradycardia must account for the ob¬
served absence of fetal heart rate
changes during maternal exercise, the
prompt onset of fetal bradycardia fol¬
lowing cessation of maximal maternal
effort, the steep slope of deceleration of
fetal heart rate, and the usually rapid
return of fetal heart rate to baseline
values. A significant reduction in total
uterine blood flow with heavy exertion
has been observed.1,2 This reduction
may be exacerbated by the abrupt fall in
maternal stroke volume found after ces¬
sation of upright exercise late in human
pregnancy.17 The combined effects of re¬
duced uterine perfusion during vigor¬
ous exertion and the decrease in cardiac
output after exercise may compromise
rate reactivity before and after maternal jogging
during the third trimester. Am J Obstet Gynecol
1982;545-547.
8. Paolone AM, Shangold M: Artifact in the record-
ing of fetal heart rates during maternal exercise. J
Appl Physiol 1987;62:848-849.
9. O'Neill ME, Cooper KA, Hunyor SN, et al:
'Pseudo' fetal bradycardia during maternal exer-
cise. J Appl Physiol 1987;62:849-850.
10. Keegan KA, Paul RH: Antepartum fetal heart
rate testing: IV. The nonstress test as a primary
approach. Am J Obstet Gynecol 1980;136:75-80.
11. Wilmore JH, Costill DL: A semiautomated sys-
tems approach to the assessment of oxygen uptake
during exercise. J Appl Physiol 1974;36:618-620.
12. Taylor HL, Buskirk E, Henschel A: Maximal
oxygen uptake as an objective measure of cardio-
respiratory performance. J Appl Physiol 1955;8:
73-80.
homeostatic mechanisms that previous¬
ly maintained fetal oxygen extraction.
Consistent with this hypothesis are the
observations that moderate hypoxia
produces vagally mediated bradycardia
without antecedent tachycardia in un-
anesthetized fetal and newborn lambs'8
and our experience that most fetal bra¬
dycardia occurred within two minutes
of exercise cessation. Vagai slowing of
fetal heart rate may result from barore-
ceptor responses as well as hypoxic
stimulus of aortic and carotid chemore-
ceptors. We assume that possible fetal
hypoxia after maternal exercise is mild
and transient, since all fetuses had nor¬
mal nonstress test results within 30
minutes of exercise cessation.
Our findings suggest that unex¬
plained fetal bradycardia during mater¬
nal exertion of any intensity is rare.
Similarly, fetal bradycardia was rarely
observed after maternal exercise re¬
quiring an average of 67% of the
mother's VOjinax or a mean maternal
heart rate of 148 beats per minute. Ma¬
ternal exertion approaching maximal
aerobic capacity, however, may be fol¬
lowed by fetal bradycardia. The signifi¬
cance of this bradycardia is, as of yet,
unknown. However, it seems prudent
to recommend that pregnant women
limit their vigorous exercise to activi¬
ties requiring heart rates of 150 beats
per minute or less and conclude with a
gentle and continuous slowing of effort
during recovery. Exercise in the pres¬
ent study was brief, so we cannot com¬
ment on the effect of exercise duration
on the fetal heart rate response.
This study was supported by National Institutes
of Health grant P50HD11343.
13. Hohimer AR, Smith OA: Decreased renal
blood flow in the baboon during mild dynamic leg
exercise. Am J Physiol 1979;236:H141-H150.
14. Horvath SM: Review of energetics and blood
flow in exercise. Diabetes 1979;28(suppl 1):33-38.
15. Rowell LB: Human cardiovascular adjust-
ments to exercise and thermal stress. Physiol Rev
1974;54:75-119.
16. Morris N, Osborn SB, Wright HP, et al: Effec-
tive uterine blood-flow during exercise in normal
and pre-eclamptic pregnancies. Lancet 1956;2:481\x=req-\
484.
17. Morton MJ, Paul MS, Campos GR, et al: Exer-
cise dynamics in late gestation:Effects of physical
training. Am J Obstet Gynecol 1985;152:91-97.
18. Walker AM, Cannata JP, Dowling MH, et al:
Age-dependent pattern of autonomic heart rate
control during hypoxia in fetal and newborn lambs.
Biol Neonate 1979;35:198-208.