PDF hosted at the Radboud Repository of the Radboud University

Nijmegen

The following full text is a publisher's version.

For additional information about this publication click this link.

http://hdl.handle.net/2066/86666

Please be advised that this information was generated on 2017-12-06 and may be subject to

change.

THE INTERACTIONS OF EXERCISE AND PREGNANCY: A

REVIEWFREDERIK K. LOTGERING, M.D., Ph.D.

RAYMOND D. GILBERT, Ph.D. and

LAWRENCE D. LONGO, M.D.Loraa Linda, California

From the Division of Perinatal Biology, Departments of Physiology and Obstetrics and Gynecology, School

of Medicine, Loma Linda University.

Reprinted from AMERICAN JOURNAL OF OBSTETRICS

AND GYNECOLOGY St. Louis

Vol. 149, No. 5, pp. 560-568, Ju ly 1, 1984 (Copyright© 1984, by The C.V. Mosby Company)

(Printed in the U.S.A.)

The interactions of exercise and pregnancy: A review

F rederik K. Lotgering, M.D., Ph.D ., R aym ond D. G ilbert, Ph.D ., and L aw rence D. Longo, M.D.Loma Linda, California

Increasing numbers of women engage in relatively strenuous exercise during pregnancy. The interaction of the increased metabolic demands of physical activity with those of pregnancy is poorly understood. We review what is known and what is not known of the extent to which pregnancy affects a woman's ability to perform strenuous activity and the degree to which exercise affects the pregnant woman, the fetus, and the infant. (Am. J. O b s t e t . G y n e c o l . 149:560, 1984.)

W hen the stress o f strenuous physical activity is su­perim posed upon that o f pregnancy, the metabolic dem ands o f the gravid u te ru s may com e in conflict with those o f the exercising muscles. This may result in re­duced exercise perform ance o f the m other an d /o r ad ­verse effects, including acute fetal distress. It is also possible that the repeated stress o f daily exercise results in fetal grow th retardation .

Because little is known about this field we will review presen t knowledge o f the physiology o f exercise du rin g pregnancy. T h e two m ain questions which we will con­sider concern first the extent to which pregnancy af­fects a w om an’s ability to perform strenuous activity and then the ex ten t to which exercise affects the preg­n an t woman, the fetus, an d the infant. Ancillary ques­tions include the following: (1) T o what ex ten t does total m aternal oxygen consum ption d u rin g physical activity d iffer from that in nonp regnan t individuals? (2) Is one’s physical w orking capacity affected by p reg­nancy? (3) T o what ex ten t are u terine blood flow and u terine oxygen consum ption altered by exercise? (4) W hat is the significance o f tem pera tu re changes for the m other and fetus? (5) T o what degree does exercise affect m aternal and fetal resp ira tory blood gases? (6) Is there evidence for acute fetal distress d u rin g exercise? (7) Is fetal outcom e affected by repeated strenuous physical activity d u rin g pregnancy? In this review we

From the Division of Perinatal Biology, Departments of Physiology and Obstetrics and Gynecology, School of Medicine, Loma Linda University.

Supported in part by United States Public Health Service Grants HD-03807 and HD-13949 and NATO Grant N95-119 awarded by The Netherlands Organization for the Advancement of Pure Research (ZWO) to Dr. Lotgering.

Reprint requests: Lawrence D. Longo, M.D., Division of Perinatal Biology, School of Medicine, Loma Linda University, Loma Linda, California 92350.

will briefly consider these questions. A m ore com plete exposition is given elsew here.1

Before dealing with these issues, however, we would like to point ou t some o f the problem s associated with the study o f the physiologic effects o f exercise du ring pregnancy. First, pregnancy affects m aternal body weight, dim ensions, com position, and , consequently, the baseline values o f m any physiologic variables rele­vant to the study o f exercise. Second, these changes also affect the physiologic b u rd en o f a given exercise regim en on the individual. T his is obvious in w eight­bearing exercise (e.g., treadm ill, step test). However, the am oun t o f energy requ ired to pedal a bicycle e r­gom eter (non-w eigh t-bearing exercise) may also in ­crease because o f fluid accum ulation in the legs and perhaps because o f o the r factors. T h ird , the physio­logic response to a fixed exercise regim en shows a wide variation betw een individuals. By standardizing the exercise level to a percentage o f the individual’s maxi­mal oxygen consum ption ra th e r than to a fixed ex ter­nal task this ap p a ren t variation can be m inim ized. Al­though this standardization is com m only used in phys­iologic studies, no one has used it in p regnan t women. F ourth , because m any variables affect both pregnancy an d exercise, few, if any, studies on the com bined sub­je c t can be considered well controlled. Fifth, only a lim ited num ber o f studies have been rep o rted un d er the most strenuous circum stances, i.e., exhaustive exercise near term . T herefo re , it is no t known to what ex ten t exercise responses d iffe r as a result o f p reg ­nancy p er se. For sim ilar reasons there are no data on fetal outcom e u n d e r these circum stances. Sixth, most reliable physiologic data are derived from anim al stud­ies. However, the results o f such studies may be o f lim ited applicability to hum ans because quad ru p ed s (in which most such studies are perfo rm ed) are less subject

560

Volume 149Number 5

Interactions of exercise and pregnancy 561

Table I. Oxygen consum ption d u rin g bicycle exercise in the sitting position

Author, year Subjects

Oxygen consumption

Rest Exercise Exercise-rest

Absolute (ml ■ m in '1)

Pregnant-nonpregnant Absolute

(ml ■ min~')

Pregnant-nonpregnant

(%)Absolute

(ml • min 1}

Pregnant-nonpregnant

(%)

Ueland et al.,’ 1973 Pregnant 330 17 836 15 506 13Nonpregnant 281 728 447

Knuttgen and Emerson,5 1974 Pregnant 249 30 1060 3 811 - 3Nonpregnant 191 1030 839

Pernoll et al.,a 1975 Pregnant 331 33 1167 15 836 8Nonpregnant 248 1019 771

Lehmann and Regnat,11 1976 Pregnant 340 16 1265 11 925 9N onpregnant 293 1141 848

Edwards et al.," 1981 Pregnant 243 27 883 4 640 - 2N onpregnant 191 847 656

to venous pooling than are hum ans, tend to elim inate heat by a d iffe ren t m echanism (panting versus sweat­ing), and cannot necessarily be m otivated to perform exhaustive exercise in the absence o f o the r stress, as are hum ans. Seventh, the perform ance o f such studies in hum ans presents ethical and legal problem s. T h e re ­fore, while knowledge o f the interactions o f pregnancy and exercise m ust be derived from both clinical studies in hum an volunteers and laboratory studies in experi­m ental animals, all such studies presen t problem s and have limitations.

Maternal oxygen consumptionD uring pregnancy resting oxygen consum ption in­

creases w'ith advancing gestational age to a m axim um value near term o f 16% to 32% above nonpregnan t values.2, 3 This h igher value results largely from the increased u terine tissue mass, including that o f the fetus. M aternal oxygen consum ption in sheep is in­creased only 4% above that o f the nonp regnan t state afte r subtracting fo r the total u te rine contents,4 and there is no evidence th a t this is any d iffe ren t in hu ­mans. T hus, the metabolic rate o f the o ther m aternal tissues is virtually unaffected by pregnancy despite a slight increase in cardiac and resp ira tory work.

Because oxygen consum ption increases with both exercise level and gestational age, the most p ronounced increases in oxygen consum ption can be expected d u r­ing maximal exercise near term . Subm axim al exercise du rin g late gestation is associated with approxim ately 10% higher absolute values fo r oxygen consum ption than is exercise d u rin g the n o n p reg n an t state, both d u r­ing w eight-bearing exercise in hum ans’’ and goats” and du ring non-w eigh t-bearing exercise.2, 3# 5- 7-11

T h e am ount o f oxygen requ ired fo r exercise can be calculated by subtracting the resting oxygen consum p­tion from the total oxygen consum ption d u rin g and

following exercise. D uring pregnancy h igher values have been rep o rted fo r treadm ill exercise,5, 12 as would be expected because o f the pregnancy wreight increase. In contrast, non-w eigh t-bearing exercise d u rin g p reg­nancy does not consistently increase the oxygen re ­qu irem ents (Table I). This suggests that exercise efficiency is not greatly affected by pregnancy, either by body position and com position o r by the metabolic and endocrinologic changes.

Because all “norm al” w eight-bearing activities du ring pregnancy require a h igher energy ou tpu t, some tra in ­ing effect seems inevitable unless a m ore sedentary life style is adopted . In the nonp reg n an t individual physi­cal train ing increases maximal oxygen consum ption up to 33%. T o w hat extent maximal oxygen consum ption is affected by pregnancy is largely unknow'n. In p reg ­nan t ewes oxygen consum ption can increase fivefold to sixfold w'ith maximal exercise,13 bu t w hether this increase is g rea ter than that in nonpregnan t sheep is unknow n. A controlled study in rats show'ed a nonsig­nificant increase o f about 8% in maxim al oxygen con­sum ption in both sedentary rats and in anim als trained before pregnancy .14 Rats that were trained both p rio r to and d u rin g pregnancy had 23% h igher maximal oxygen consum ption values than sedentary rodents, b u t the train ing effect p rio r to pregnancy accounted fo r 13% o f the total.

In addition , two case reports suggest that maximal oxygen consum ption increases about 20% in p regnan t women who m aintain fairly strenuous activity d u ring pregnancy .15, IK

Physical working capacityIf one could extrapolate from the above observations

at subm axim al w'ork levels in hum ans, a given maximal task d u rin g pregnancy would require about a 10% higher absolute value for maximal oxygen consum p-

562 Lotgering, Gilbert, and Longo July 1, 1984Am. J. Obstet. Gynecol.

Fig. 1. Physiologic changes for several functions in response to 40 minutes o f exercise at 70% maximal oxygen consumption in pregnant sheep, a: Uterine blood flow (percentage of con­trol value); b: hematocrit (percentage o f control value); c: uter­ine oxygen consumption (percentage o f control value); d: ma­ternal and fetal tem peratures. Values are means ± SEM (n = 8 in a to c and 6 in d).

tion, which seems possible fo r otherw ise sedentary in­dividuals. This suggests tha t objectively such women should be able to m aintain the same tasks as they did p rio r to pregnancy. However, w orking capacity is af­fected not only by m axim al oxygen consum ption but also by a variety o f conditions, including somatic fac­tors, environm ental factors, wrork characteristics, train ­ing and adaptation , and psychic factors, including motivation. This makes it difficult, if n o t impossible, to draw definitive conclusions as to physical w orking ca­pacity d u rin g pregnancy. Because o f the hem odynam ic changes d u rin g pregnancy, including increased blood volume, it is conceivable not only tha t maxim al oxygen consum ption may increase in sedentary individuals bu t th a t tra in ing may increase maximal oxygen consum p­tion beyond its n o n p reg n an t limits. F u rth e r study o f

this im portan t variable is clearly indicated, particularly in hum ans. .

Uterine oxygen consumptionU terine blood flow increases with advancing gesta­

tional age, although flow decreases som ewhat per kilo­gram o f total u te rine contents. T h e flow increase results from a decrease in u terine vascular resistance secondary to vasodilatation associated with increasing concentrations o f estrogens p roduced by the fetopla­cental unit and prostaglandins (F.2 and I2) p roduced by the vessel w'all.17 A lthough the dilated u terine vascula­tu re is less sensitive to the effects o f vasoconstrictive agents d u rin g pregnancy than in the n o n p reg n an t state, it responds to circulating prostaglandins and catecholam ines and to sym pathetic stim ulation. U ter­ine blood flow may decrease spontaneously up to 20% ,18 an d reductions up to 50% have been rep o rted in response to alkalosis and hypertherm ia19 and to a vari­ety o f o th e r stresses.20 It has been suggested th a t these vasoconstrictive occurrences are catecholam ine-m edi­ated, but this has not yet been dem onstrated .

D uring exercise the redistribution o f cardiac ou tpu t has been though t to result from vasodilatation in exer­cising muscles, m ediated by local metabolic factors, and from sym pathetic vasoconstriction in tissues with a high resting flow and low oxygen extraction, such as the splanchnic bed and nonw'orking muscles.21 Because the u terine vasculature d u rin g pregnancy is sensitive to sym pathetic stim ulation and catecholam ine release, one would also expect a reduction in u terine blood flow' d u rin g m aternal exercise. Using the disappearance o f sodium 24 injected into the m yom etrium , M orris et al.22 suggested a 25% reduction in flow to the p reg ­n an t hum an u te rus d u rin g mild short-term bicycle exercise in the supine position. However, this probably represen ts an overestim ate. In the supine position the presence o f a large p regnan t u te ru s may well affect u te rine blood flow. In addition, anim al data show that the m yom etrium is m ore sensitive to catecholam ines than is the placenta,2'1 resulting in flow redistribution within the u terus, favoring cotyledonary flow at the expense o f m yom étrial flow.24- 25

Lotgering et al.2® studied u terine blood flow at d if­feren t levels (percentage o f maximal oxygen consum p­tion) and durations o f exercise in sheep. As shown in Fig. 1, a, u te rine blood flow decreased im m ediately at the onset o f exercise, was significantly below1 control values th roughou t the exercise period, and re tu rn ed to control levels within 10 m inutes o f recovery. Flow' d e­creased 13% d u rin g a 10-m inute exercise period at 70% maximal oxygen consum ption, 17% d u rin g a 10-minute exercise period at 100% maximal oxygen consum ption, and 24% n ea r the end o f a 40-m inute exercise period at 70% maximal oxygen consum ption.

Volume 149Number 5

Interactions of exercise and pregnancy 563

Regression analysis show ed a significant decrease in flow w'ith exercise time, and flow varied linearly w'ith h eart rate , w here heart ra te is linearly related to the level o f exercise. T hus, it is likely tha t u terine blood flow decreases with both the level and the duration o f exercise. H ohim er et al.,25 C lapp,27 and C hand ler and Bell28 also have rep o rted decreases in u terine blood flow o f u p to 36% d u rin g exercise. O th er inves­tigators24, 29 have concluded that u te rine blood flow in sheep rem ains constant d u rin g treadm ill exercise. However, the ir m easurem ents w'ere m ade shortly after, ra th e r than du ring , exercise, and u terine flow re tu rns rapidly to control levels when exercise is discon­tinued .26

A lthough a reduction in u terine blood flow suggests a reduction in the supply o f oxygen and nu trien ts to the u terus an d /o r a reduction in oxygen consum ption, this result is not necessarily tru e d u rin g exercise, be­cause exercise is associated with a m arked rise in m a­ternal hem atocrit. D uring exercise plasma filtrate is forced across the capillary m em brane in exercising muscles, resulting in a decrease o f u p to 14% in plasma volume in m an30 and up to 20% in p reg n an t sheep,26 w'hile the red blood cell mass rem ains constant.26 This hem oconcentration is associated with an increase in hem oglobin concentration (show'n as hem atocrit in Fig. 1, b) and thus o f blood oxygen-carrying capacity. C on­sequently, the reduction in oxygen delivery (oxygen content x flow') to the u te rus is m uch sm aller than the decrease in blood flow w'ould suggest.26 In addition , in sheep the blood flow is red istribu ted w ithin the uterus, favoring the placental cotyledons at the expense o f the m yom etrium ,24 , 25 and oxygen extraction is increased.31 T h e net result o f these com pensatory m echanism s is a constant oxygen up take by the u te ru s as a w’hole (Fig. 1, c)27, 28, 31 and by the fetus.27

Temperature changesD uring exercise the total heat production may in­

crease as m uch as 20 times resting values. Only 20% to 25% o f the added energy expend itu re is used for ex­ternal work, while the rem aining 75% to 80% is trans­form ed into heat. A lthough m ost o f the heat is lost to the environm ent, some is stored, resulting in increased body tem peratu re . Body tem pera tu re increases with the level and d u ra tion o f exercise. T his is accom panied by m arked circulatory changes, including increased blood flow’ to the skin to provide adequate cooling. I f heat loss is reduced because o f high am bient tem pera­tu re an d /o r hum idity, the w orking capacity o f an indi­vidual will be m arkedly decreased.21

U nder norm al resting conditions the tem pera tu re of the fetus is about 0.5° C h igher than that o f the m other in hum ans32 and sheep.31, 33 Most o f the fetal heat is transfe rred to the m other across the placenta and a

sm aller p roportion is transfe rred across the fetal skin, am niotic fluid, and u terine wall.32',33 Recent theoretical studies34 suggest that the m aternal body tem pera tu re is the m ajor determ inan t o f fetal tem peratu re , while changes in u terine blood flow o r fetal metabolism are quantitatively less im portant. .

In studies o f exercising sheep from o u r laboratory,31 the fetal tem pera tu re lagged behind the rapidly chang­ing m aternal tem p era tu re at the onset and cessation o f exercise (Fig. 1, d). Consequently the fetal-m aternal tem pera tu re d ifference d u rin g the onset o f exercise w'as reduced o r even reversed, while a la rger tem pera­tu re d ifference existed following exercise. T hese changes were m ore p ronounced with h igher levels o f exercise du rin g which the m aternal tem p era tu re in­creases m ore rapidly. R etu rn o f the fetal tem pera tu re to control values w'as slow, requ iring m ore than 1 hou r following prolonged (40 minutes) exhaustive exercise at 70% maximal oxygen consum ption (Fig. 1, d).

A m ong the possible physiologic implications o f in­creased body tem pera tu re are increased metabolism (Qio effect*), rightw ard shifts o f the m aternal and fetal oxyhem oglobin dissociation curves, and a reduction o f u terine blood flow. However, the quantitative aspects o f these changes are not yet fully understood and re­qu ire fu rth e r study.1

Respiratory blood gasesKnowledge o f the m aternal and fetal tem peratu res is

essential fo r the correct in te rp re tation o f fetal blood gas m easurem ents. Blood obtained anaerobically and analyzed for resp ira tory blood gases at a tem peratu re below that o f the body shows a rise in pH and a fall in oxygen and carbon dioxide tensions.35 T h e failure to correct fo r a 1° C increase has been estim ated to result in about a 1.9 and 2.7 to rr underestim ate fo r fetal oxy­gen and carbon dioxide tensions, respectively, and a pH that is falsely 0.015 units too high.31 All factors which change m aternal tem peratu re , u te rine o r um bil­ical blood flow, o r fetal metabolism may affect the tem ­p era tu re gradient, especially in the n o n -s tead y state. U nfortunately , this consideration is often not taken into account in studies o f fetal blood gas values.

W hen the p ro p er tem pera tu re corrections are m ade, m aternal oxygen tension and oxygen conten t increase and carbon dioxide tension decreases as a result o f exercise-induced hyperventilation and hem oconcen­tration . As shown in Fig. 2, in sheep exercised to

*The interdependent physical, chemical, and metabolic processes o f the hum an body may vary with small differences in tem perature. This change in activity with tem perature is usually expressed as Q J0. This is a factor by which the velocity o f a reaction at given tem perature is multiplied to give the velocity of that reaction at a tem perature o f 10° C higher. The Q ,0 o f physical reactions is very nearly 1.

564 Lotgering, Gilbert, and Longo July 1,1984Am. J. Obstet. Gynecol.

40sO 20

5 0 O ce u.uiO -20z <Xo

-40 L

Fig. 2. Percent changes in maternal and fetal respiratory blood gas concentrations during prolonged (40 minutes) exercise at 70% maximal oxygen consumption in sheep (after Lotgering et al.31). Values are means ± SEM (n = 12).

exhaustion at 70% maximal oxygen consum ption, ma­ternal oxygen tension increased 13% and oxygen con­ten t increased 25%, while carbon dioxide tension de­creased 28%.31 Consequently the oxygen-carrying capacity o f the m aternal blood was increased and the oxygen transport across the placenta was enhanced.

A lthough the reduced u terine blood flow du rin g exercise will tend to lower the placental oxygen trans­port, both theoretical analysis36 an d observations in ou r laboratory d u rin g acute em bolization o f the u terine vascular bed suggest only m inim al decreases in fetal oxygen tension when u terine blood flow is approx i­mately 60% o r m ore o f its norm al resting value. Several studies in exercising sheep13- 27- ” • 37 have repo rted re­ductions in fetal arterial oxygen and carbon dioxide tensions o f as m uch as 25%. Failure to correct the blood gas values for the tem p era tu re changes largely explains the differences in results am ong these studies and the sm aller changes observed in o u r studies.31 In these studies31 fetal arterial oxygen and carbon dioxide ten­sions and oxygen conten t decreased with the level and the duration o f exercise, but the values d iffered signif­icantly from control only when the ewes were ru n to exhaustion. D uring prolonged (40 m inutes) exhaustive exercise at 70% maximal oxygen consum ption fetal aortic oxygen tension decreased 3.0 to rr (from 26.2 to 23.2 torr); carbon dioxide tension, 4.5 to rr (from 54.1 to 49.6 torr); and oxygen conten t, 1.5 ml • dl~' (from 5.8 to 4.3 ml ■ d l_1), while pH increased 0.02 units. T h e percen t changes are shown in Fig. 2. Theoretical calcu­lations suggest that about 30% o f the decrease in oxy­gen saturation can be accounted for by the tem pera­tu re and B ohr shifts o f the oxyhem oglobin dissociation curve, w hereas the rem ain ing 70% o f the decrease in oxygen saturation is associated with the 3 to rr decrease

M*t*rn«l FetalCoj] Pcoj [w] Roj [ o J PcOj M

■ f t

f t

-

+4-

H-

+

in oxygen tension.31 A lthough fetal oxygen tension and conten t decrease with exercise, this decrease does not necessarily indicate tha t the fetal metabolic dem ands are not m et o r tha t the fetus is in “distress.”

Other fetal responsesFetal hypoxia an d /o r “distress" may be associated with

changes in any o f the following fetal variables: heart rate , blood pressure, cardiac o u tp u t d istribution,38 he­m atocrit,39 an d catecholam ine concentra tion .40

Several au thors41-50 have studied the fetal heart rate p rio r to and following a mild to m odera te exercise stress o f short du ration . T hey noted only small changes in m ean fetal heart rate and heart rate patterns, changes tha t were inconclusive as to possible “distress.” However, it is conceivable that the fetal responses to m aternal exercise recover rapidly following such a m in­imal stress. Recendy Dale et al.31 rep o rted transient fetal b radycardia for about 3 m inutes d u rin g short­term , m oderately strenuous exercise in th ree subjects. In contrast, Sibley et al.52 repo rted a slightly h igher fetal hea rt rate d u rin g short-term treadm ill exercise as com pared to baseline and recovery values, but the m ean values w ere not com pared statistically. This is also tru e fo r a study53 in which h igher fetal heart rates were found d u rin g the recovery from m oderately strenuous exercise (1.5 miles o f jogging) in seven women. In neither study did the fetal heart rate pat­te rn show any signs o f “distress.” O ne study rep o rted on the changes in fetal h ea rt rate both d u rin g and fol­lowing 25 m inutes o f bicycle exercise at about 70% maximal oxygen consum ption.54 T h e auscultated heart rate dem onstrated only a m ean increase o f 4 bpm d u r­ing an d following exercise, which is physiologically in­significant. In sheep, one investigator21 observed an in­creased fetal h ea rt rate , but o thers have repo rted no significant changes d u rin g e ither short-term 31 o r p ro ­longed exhaustive exercise.31, **’ 56

Animal studies show th a t the o ther fetal cardiovascu­lar variables are also largely unaffected by m aternal exercise. Fetal arterial blood pressure is unaffected by short-term 31 as well as prolonged exercise in sheep.9- 31- 37- 55 C ardiac o u tp u t and cardiac o u tp u t dis­tribution are unaffected by exhaustive short-term exercise,31 bu t cardiac o u tp u t has not been studied d u r­ing p ro longed exercise in sheep. O ne study27 reported a 10% reduction in umbilical blood flow d u rin g pro­longed exhaustive exercise. However, this observation needs to be confirm ed because the rep o rted resting value o f 354 ml • m in_1kg~‘ was well above accepted norm al values. In addition , fetal hem atocrit28- 31 and red blood cell and plasm a volumes31 are unaffected by pro longed exercise.

T h e num ber o f studies o f fetal cardiovascular re ­sponses to m aternal exercise is lim ited, but the available

Volume 149Number 5

Interactions of exercise and pregnancy 565

evidence suggests the absence o f hypoxia o r “stress.” In addition, fetal catecholam ine concentrations are not significantly increased in fetal sheep d u rin g either short-term 31 o r pro longed exercise.31' °5, 56 However, ex trapolating these responses from fetal sheep to the hum an fetus may not be legitim ate fo r the reasons de­tailed above.

Fetal outcomeT h e effect on fetal outcom e o f a single factor such as

exercise is easily obscured by the wide norm al variation in outcom e caused by a m ultitude o f variables, includ­ing genetic and socioeconomic factors, nutrition , en ­vironm ental factors, “stress," and so forth . T hus, the question o f w hether physical activity affects fetal o u t­com e can be answ ered only by large, well-controlled prospective epidem iologic studies. We know o f no such study in p regnan t women.

However, several studies concerning this subject are nonetheless worthy o f discussion. C lapp and Dick- stein57 prospectively studied the effects o f m aternal exercise on fetal outcom e in 336 pregnan t wom en, di­vided into five activity groups on the basis o f interview' data. T hey concluded tha t preconceptional exercise had no effect on fetal outcom e bu t that regu lar vigor­ous exercise d u rin g pregnancy increased the incidence o f lo w -b irth weight infants. However, as the au thors pointed out, the findings o f the ir study m ust be in te r­p reted with caution because o f the lim ited accuracy o f the data. In contrast, a high level o f voluntary daily exercise o r a high “physical fitness score” was found not to be associated with low A pgar scores o r low birth weights in hum ans,58- 59 and adverse effects on fetal outcom e w ere not observed in tw'o small prospective studies in which the expectant m others participated in mild exercise program s.52- 54 T h e effects o f strenuous exercise d u rin g hum an pregnancy have been studied mainly in women who were highly physically active p rio r to pregnancy ,16- 60-87 an d all o f these studies re ­port norm al o r im proved fetal outcom es. A lthough these studies were retrospective an d /o r uncontrolled , with the w'omen in excellent condition p rio r to p reg ­nancy, they suggest the absence o f m ajor negative ef­fects o f strenuous exercise on fetal outcom e in healthy W'omen. S trenuous exercise is norm al d u rin g la ter ges­tation in m any large gam e m am m als and the ir p re­dators as well as in some dom estic anim als.68 This sug­gests that strenuous activity d u ring pregnancy does not adversely affect fetal outcom e in healthy individuals o f m any species.

In contrast, adverse effects o f m aternal physical ac­tivity on fetal outcom e have been rep o rted in both hu ­m ans and laboratory animals. T h e most com monly m entioned negative effect is low birth weight. Several studies have repo rted a decrease o f as m uch as 400 gm

Positive

LU

Fig. 3. Diagrammatic representation o f how exercise intensity and duration might have positive or negative effects on cer­tain physiologic functions o r outcomes during pregnancy. (See text for details.)

and an increased incidence o f low -b irth w eight infants in offspring o f w orking m others.69-72 How'ever, this may reflect the poo r nutritional status o f these wom en,70 o r o th e r selection factors ra th e r than rep re ­sent the effect o f physical activity per se. W eight re­duction has also been rep o rted in p regnan t laboratory anim als that were forced to exercise strenuously d u rin g pregnancy. An 8% reduction was rep o rted in forcefully exercised mice73 an d guinea pigs.74 Fetal weight in rats forced to swim was 6% lower than norm al75; however, rats forced to ru n d id not show growth re ta rda tion .76 A 12% to 20% reduction in fetal weight was also repo rted in exercised pygmy goats,6 but this study may not have been w'ell controlled. In contrast, no reduction in birth weights was observed in swine m oderately exercised th ro u g h o u t gestation.77 Because fetal growth re ta rd a ­tion has been associated not only with increased m ater­nal physical activity bu t also with restriction o f activ­ity,78 one cannot exclude the possibility tha t the “stress” o f fear and handling related to forced exercise in a laboratory environm ent ra th e r than the exercise p e r se contribu ted to the growth retardation . A variety o f o th e r adverse effects has also been suggested, in­cluding increased perinatal m ortality and increased incidence o f p rem aturity .69 However, fu rth e r well- m atched prospective studies are necessary to confirm any such effects. T his is also true fo r the possible te ra ­togenic effect o f increased tem peratu re d u rin g exer­cise in early pregnancy.

It is ap p aren t that exercise m ight have positive or negative health effects d u rin g pregnancy. Fig. 3 p re ­sents a scheme o f the m anner in which this m ight oc­cur. T h e o rd inate indicates such health effects w ithout num erical values. T h e p roduct o f exercise intensity and du ration as a rough approxim ation o f overall work

T raining

\ Exercise Intensity \ y

\ /

HypoxiaMalnutritionD isease

Negative

566 Lotgering, Gilbert, and Longo July 1, 1984Am. J. Obstet. Gynecol.

o r e ffo rt is shown on the abscissa. Line A indicates those effects which are beneficial and tend to increase as a function o f exercise intensity and duration . Line B indicates the opposite effects. Line C indicates that for some functions the re may be a positive effect at low to m oderate exercise levels but that this decreases o r re ­verses a t h igher levels. Finally, line-D suggests that for some functions there is no discernible effect. In add i­tion, the figure suggests that train ing may shift the curves upw ard and to the right, while, in contrast, o the r stresses such as hypoxia, m alnutrition , o r certain disease states may com pete fo r the subject’s physiologic reserves, shifting the curve dow nw ard and to the left.

CommentAnimal studies show tha t physical activity du rin g

pregnancy results in m arked cardiovascular ad just­m ents in the m other, including a reduction in u terine blood flow. However, because o f sim ultaneous hem o- concentration and increased oxygen extraction, u te rine oxygen consum ption rem ains constant. In contrast to relatively p ro found physiologic changes in the m other, the changes in the fetus are small. T h e oxygen tension and oxygen conten t in the fetal arterial blood decrease slightly, but o th e r fetal variables, including catechol­am ine concentrations, h eart rate, blood pressure, car­diac ou tpu t, blood flow' distribution, and blood volume, rem ain virtually constant. This suggests that acute exercise does not rep resen t a m ajor hypoxic o r o the r stress to the fetus. Little is known about the physiologic effects o f chronic exercise on the fetus. T h e most likely possible effect is a small reduction in b irth weight, at least in some species. Again we wish to stress that one m ust be careful in ex trapolating these anim al data to hum ans. Large, well-controlled prospective epidem io­logic studies are necessary to establish possible positive o r negative effects in hum ans. Finally, additional physiologic studies are needed fo r a m ore com plete understand ing o f the rem arkably effective hom eostatic mechanism s in both m other and fetus d u rin g m aternal exercise.

We wish to thank Mrs. B. K reutzer, S. Taylor, and A. Bijl for assistance in p reparing the m anuscript.

R E F E R E N C E S

1. Lotgering FK, Gilbert RD, Longo LD. Maternal and fetal responses to exercise during pregnancy. Physiol Rev (in press).

2. Pernoll ML, Metcalfe J, Schlenker TL, Welch JE, Mat- sumoto JA. Oxygen consumption al rest and during exercise in pregnancy. Respir Physiol 1975;25:285-93.

3. Ueland K, Novy MJ, Metcalfe J. Cardiorespiratory re­sponses to pregnancy and exercise in normal women and patients with heart disease. A m J O b s t e t G y n e c o l 1973;115:4-10.

4. Clapp JF III. Cardiac output and uterine blood flow in the pregnant ewe. A m J O b s t e t G y n e c o l 1978; 130:419­23.

5. Knuttgen HG, Emerson K Jr . Physiological response to pregnancy at rest and during exercise. J Appl Physiol 1974;36:549-53.

6. Dhindsa DS, Metcalfe J, Hummels DH. Responses to exercise in the pregnant pygmy goat. Respir Physiol 1978;32:299-311.

7. Bergamaschi P, Capodaglio E, Fre B. Aspetti cardiore­spiratorie metabolici del lavora muscolare in gravidanza. Minerva Ginecol 1964:16:831-8.

8. Edwards MJ, Metcalfe J , Dunham MJ, Paul MS. Acceler­ated respiratory response to moderate exercise in late pregnancy. Respir Physiol 1981;43:229-41.

9. Eismayer G, Pohl A. Untersuchungen über den Kreislauf und den Gasstoffwechsel in der Schwangerschaft bei Ar­beitsversuchen. Arch Gynaekol 1934;156:428-53.

10. Guzman CA, Caplan R. Cardiorespiratory response to exercise during pregnancy. A m I O b s t e t G y n e c o l 1970;108:600-5. '

11. Lehmann V, Regnat K. U ntersuchung zur körperlichen Belastungsfahigkeit schwangeren Frauen. Der Einfluss standardisierter Arbeit auf Herzkreislaufsystem, Ven­tilation, Gasaustausch, Kohlenhydratstoffwechsel und Säure-Basen-Haushalt. Z Geburtshilfe Perinatol 1976; 180:279-89.

12. Teruoka G. Labour physiological studies on the pregnant women. Arb Physiol 1933;7:259-79.

13. Longo LD, Hewitt CW, Lorijn RHW, Gilbert RD. To what extent does maternal exercise affect fetal oxygena­tion and uterine blood flow? Fed Proc 1978;37:905.

14. Wilson NC, Gisolfi CV. Effects o f exercising rats during pregnancy. J Appl Physiol 1980;48:34-40.

15. Dressendorfer RH. Physical training during pregnancy and lactation. Phys Sports Med 1978;6:74-80.

16. Ruhling RO, Cameron J , Sibley L, Christensen C, Bolen T . Maintaining aerobic fitness while jogging through a pregnancy! A case study. Med Sei Sports Exerc 1981; 13: 93.

17. Wallenburg HCS. Modulation and regulation of uteroplacental blood flow. Placenta 1981;2(suppl. 1): 45-64.

18. Lanz E, Caton D, Schlereth H, Barron DH. Die Wirkung von Lokalanaesthetika auf D urchblutung und 0 2-Ver- brauch des U terus von schwangeren Schafen. Anaes- thesist 1977;26:403-10.

19. Oakes GK, Walker AM, Ehrenkranz RA, Cefalo RC, Chez RA. Uteroplacental blood flow during hypertherm ia with and out respiratory alkalosis. J Appl Physiol 1976;41: 197-201.

20. Shnider SM, Wright RG, Levinson G, Roizen MF, Wallis KL, Rolbin SH, Craft JB. U terine blood flow and plasma norepinephrine changes during m aternal stress in the pregnant ewe. Anesthesiology 1979;50:524-7.

21. Rowell LB. Human cardiovascular adjustments to exer­cise and thermal stress. Physiol Rev 1974;54:75-159.

22. Morris N, Osborn SB, Wright HP, Hart A. Effective u ter­ine blood-flow during exercise in normal and pre­eclamptic pregnancies. Lancet 1956;2:481-4.

23. Greiss FC Jr. Differential reactivity of the myoendome- trial and placental vasculatures: adrenergic responses. A m J O b s t e t G y n e c o l 1972;112:20-30.

24. Curet LB, O rr JA, Rankin JH G , U ngerer T . Effect of exercise on cardiac output and distribution o f uterine blood flow in pregnant ewes. J Appl Physiol 1976;40: 725-8.

25. H ohim er AR, McKean TA, Bissonnette JM, Metcalfe J. Maternal exercise reduces myoendometrial blood flow in the pregnant goat. Fed Proc 1982:41:1490.

26. Lotgering FK, Gilbert RD, Longo LD. Exercise responses in pregnant sheep: oxygen consumption, uterine blood flow, and blood volume. J Appl Physiol 1983;55:834-41.

27. Clapp JF III. Acute exercise stress in the pregnant ewe. A m J O b s t e t G y n e c o l 1980;136:489-94.

28. Chandler KD, Bell AW. Effects o f maternal exercise on

Volume 149Num ber 5

Interactions of exercise and pregnancy 567

fetal and maternal respiration and nutrient metabolism in the pregnant ewe. J Dev Physiol 1981;3:161-76.

29. O rr J, U ngererT , Will J , Wernicke K, Curet LB. Effect of exercise stress on carotid, uterine, and iliac blood flow in pregnant and nonpregnant ewes. A m J O b s t e t G y n e c o l 1972;114:213-7.

30. Greenleaf JE, Convertino VA, Stremel RW, Bernauer EM, Adams WC, Vignau SR, Brock PJ. Plasma (Na+), (Ca2+), and volume shifts and thermoregulation during exercise in man. J Appl Physiol 1977;43:1026-32.

31. Lotgering FK, Gilbert RD, Longo LD. Exercise responses in pregnant sheep: blood gases, tem peratures and fetal cardiovascular system. J Appl Physiol 1983;55:842-50.

32. Adamsons K Jr . T he role o f thermal factors in fetal and neonatal life. Pediatr Clin North Am 1966;13:599-619.

33. Abrams R, Caton D, Clapp J, Barron DH. Therm al and metabolic features o f life in utero. Clin Obstet Gynecol 1970;13:549-64.

34. Schroder H, Gilbert RD, Power GG. Fetal heat dissipa­tion: A com puter model and some preliminary experi­mental results from fetal sheep. In: Scientific Abstracts of the Twenty-ninth Annual Meeting o f the Society for Gy­necologic Investigation. Dallas: for Gynecologic Society Investigation, 1982:113.

35. Severinghaus JW . Blood gas calculator. J Appl Physiol 1966;2:1108-16.

36. Longo LD, Hill EP, Power GG. Theoretical analysis o f factors affecting placental 0 2 transfer. Am J Physiol 1972;222:730-9.

37. Emmanouilides GC, Hobel CJ, Yashiro K, Klyman G. Fetal responses to maternal exercise in the sheep. A m J O b s t e t G y n e c o l 1972;112:130-7.

38. Peeters LLH, Sheldon RE, Jones MD Jr, Makowski EL, Meschia G . Blood flow to fetal organs as a function of arterial oxygen content. Am J O b s t e t G y n e c o l 1979; 135:637-46.

39. Shelley HJ. T he use o f chronically catheterized foetal lambs for the study o f foetal metabolism. In: Comline RS, Cross KW, Dawes GS, Nathanielsz PW, eds. Foetal and Neonatal Physiology. London: Cambridge University Press, 1973:360-81.

40. Jones CT. Circulating catecholamines in the fetus, their origin, actions and significance. In: Parvez H, Parvez S, eds. Biogenic Amines in Development. Amsterdam: Elsevier-North Holland Press, 1980:63-86.

41. Eisenberg De Smoler P, Armass-Dominguez J , Domin- guez-Munguia C, Karchmer S. Ejercicio materno y fre­cuencia cardiaca materno fetal. Arch Invest Med (Mex) 1974;5:595-602.

42. Eisenberg De Smoler P, Karchmer S, Castelazo Ayala L, Armas Dominquez J. El electrocardiograma fetal durante el ejercicio materno. Ginecol Obstet Mex 1974;35:521- 34.

43. Hon EH, Wohlgemuth R. T he electronic evaluation of fetal heart rate. IV. The effect o f maternal exercise. Am J O b s t e t G y n e c o l 1961;81:361-71.

44. Kuhlmann G, Mosler KH, Schwalm H. Ergometric strain test at the end o f gravidity and resulting fetal heart rate and course o f labor. Z Geburtshilfe Perinatol 1974; 178: 279-84.

45. Marsal K, Lofgren O, Gennser G. Fetal breathing move­ments and m aternal exercise. Acta Obstet Gynecol Scand 1979;58:197-201.

46. Pernoll ML, Metcalfe J, Paul M. Fetal cardiac response to maternal exercise. In: Longo LD, Reneau DD, eds. Fetal and Newborn Cardiovascular Physiology', vol. 2, New York: Garland Press, 1978:389-98.

47. Pokorny J , Rous J. T he effect o f m other’s work on foetal heart sounds. In: Horsky J, Stembera ZK, eds. In tra­Uterine Dangers to the Foetus, Amsterdam: Excerpta Medica, 1967:354-7.

48. Pomerance JJ, Gluck L, Lynch VA. Maternal exercise as a screening test for uteroplacental insufficiency. Obstet Gynecol 1974;44:383-7.

49. Soiva K, Salmi A, Grönroos M, Peltonen T. Physical work­ing capacity during pregnancy, and effect o f physical work tests on foetal heart rate. Ann Chir Gynaecol Fenn 1963;53:187-96.

50. Stembera ZK, H odr J. T he "exercise test" as an early di­agnostic aid for foetal distress. In: Horsky J, Stembera ZK, eds. Intra-uterine Dangers to the Foetus. Amster­dam: Exerpta Medica 1967:349-53.

51. Dale E, Mullinax KM, Bryan DH. Exercise during preg­nancy: Effect on the fetus. Can J Appl Sport Sci 1982;7: 98-103.

52. Sibley L, Ruhling RO, Cameron-Foster J, Christensen C, Bolen T . Swimming and physical fitness during preg­nancy. J Nurs Midwife 1981;26:3-12.

53. H auth JC, Gilstrap LC III, Widmer K. Fetal heart rate reactivity before and after maternal jogging during the third trimester. A m J O b s t e t G y n e c o l 1982; 142: 545-7.

54. Collings CA, Curet LB, Mullin JP. Maternal and fetal responses to a maternal aerobic exercise program. A m J O b s t e t G y n e c o l 1983;145:702-7.

55. Hobel CJ, Artal R, Emmanouilides G, Lam RW, Fischer DA. Maternal and fetal catecholamine responses to ma­ternal exercise in the normal and compromised pregnant ewe. In: Scientific Abstracts, Twenty-fifth Annual Meet­ing of the Society o f Gynecologic Investigation. Atlanta: Society for Gynecologic Investigation, 1978:5.

56. Palmer SM, Óakes GK, Hobel CJ, Hawkins PM, Fisher DA. The role o f hypocapnea on acute exercise (EX) in­duced catecholamine (CAT) release in chronically catheterized ovine pregnancy. In: Scientific Abstracts o f the Twenty-seventh Annual Meeting of the Society for Gynecologic Investigation. Denver: Society for Gyneco­logic Investigation, 1980:62.

57. Clapp JF III, Dickstein S. Maternal exercise performance and pregnancy outcome (personal communication).

58. Erkkola R. The physical work capacity o f the expectant m other and its effect on pregnancy, labor and the new­born. Int J Gynaecol Obstet 1976;14:153-9.

59. Pomerance JJ, Gluck L, Lynch VA. Physical fitness in pregnancy: Its effect on pregnancy outcome. Am J O b s t e t G y n e c o l 1974;119:867-76.

60. Casper H. Der Einfluss sportlichen Trainings au f den Geburtsverlauf. Dtsch Med Wochenschr 1928;54:821-3.

61. Düntzer E, Hellendall M. Einwirkungen der Leibesübun­gen au f weibliche Konstitution, Geburt und Menstruation, Munch Med Wochenschr 1929;76:1835-8.

62. Erdelyi GJ. Gynecological survey of female athletes. J Sports Med Phys Fitness 1962;2:174-9.

63. Ferrarese C, Fraccaroli G. Attività sportiva e maternità. Minerva Med 1980;71:1321-8.

64. Korcok M. Pregnant jogger; What a record! JAMA 1981; 246:201.

65. Novotna J, Pros JR, Tucková H. Prubeh tehotenstvi vyk- onnostnich sportovkyn se zretelem k disciplinam a nek- terym psychologickym faktorum. Cesk Gynekol 1977;42: 730-3.

66. Pfeifer WA. Neuere Untersuchungen über Geburtsver­lauf, Fertilität und Konstitution bei Spitzenkonnerinnen der Leichtathletik. Zbl Gynaekol 1951;73:17-23.

67. Zaharieva E. Olympic participation by women. Effects on pregnancy and childbirth. JAMA 1972;221:992-5.

68. Bruser M. Sporting activities during pregnancy. Obstet Gynecol 1968;32:721-5.

69. Fox ME, Harris RE, Brekken AL. The active-duty mili­tary pregnancy: A new hieh-risk category. Am I O b s t e t G y n e c o l 1977;129:705-7.

70. Naeye RL, Peters EC. Working during pregnancy, effects on the fetus. Pediatrics 1982;69:721-7.

71. Rodríguez-Escudero FJ, Belausteguigoitia Abando FJ, Gutierrez Martínez S. Implicaciones perinatales del trabajo y el descanso en la gestante. An Esp Pediatr 1980;13:456-76.

72. Tafari N, Naeye RL, Gobezie A. Effects o f maternal

568 Lotgering, Gilbert, and Longo July 1, 1984Am. J. Obstet. Gynecol.

under-nutrition and heavy physical work during preg­nancy on birth weight. Br J Obstet Gynaecol 1980;87: 222-6 .

73. T erada M. Effect o f physical activity before pregnancy on fetuses o f mice exercised forcibly during pregnancy. Teratology 1974;10:141-4.

74. Nelson PS, Gilbert RD, Longo LD. Fetal growth and pla­cental diffusing capacity in guinea pigs following long­term maternal exercise. J Dev Physiol 1983;5:1-10.

75. Levitsky LL, Kimber A, Marchichow JA , Uehara J. Meta­bolic response to fasting in experimental intrauterine

growth retardation induced by surgical and nonsurgical maternal stress. Biol Neonate 1977;31:311-5.

76. Blake CA, Hazelwood RL. Effect o f pregnancy and exer­cise on actomysin, nucleic acid, and glycogen content of the rat heart. Proc Soc Exp Biol Med 1971;136:632-6.

77. Hale OM, Booram CV, McCormick WC. Effects o f forced exercise during gestation on farrowing and weaning per­formance of swine. J Anim Sci 1981;52:1240-3.

78. Karasawa K, Kimura S. Voluntary exercise during preg­nancy and lactation and its effect on lactational per­formance in mice. J N utr Sci Vitaminol 1981;27:333-9.