Exercise During Pregnancy Protects Against Hypertension and Macrosomia: Randomized Clinical Trial
Exercise During Pregnancy Protects Against Hypertension and Macrosomia: Randomized Clinical Trial
Exercise During Pregnancy Protects Against Hypertension and Macrosomia: Randomized Clinical Trial
org
OBSTETRICS
Exercise during pregnancy protects against hypertension
and macrosomia: randomized clinical trial
Ruben Barakat, PhD; Mireia Pelaez, PhD; Yaiza Cordero, PhD; Maria Perales, PhD; Carmina Lopez, MSc;
Javier Coteron, PhD; Michelle F. Mottola, PhD, FACSM
BACKGROUND: The prevalence of all pregnancies with some STUDY DESIGN: A randomized controlled trial was used. Women were
form of hypertension can be up to 10%, with the rates of diagnosis randomized into an exercise group (N ¼ 382) or a control group (N ¼ 383)
varying according to the country and population studied and the receiving standard care. The exercise group trained 3 d/wk (50-55 min/
criteria used to establish the diagnosis. Prepregnancy obesity and session) from gestational weeks 9-11 until weeks 38-39. The 85 training
excessive gestational weight gain (GWG) of all body mass index (BMI) sessions involved aerobic exercise, muscular strength, and flexibility.
categories have been associated with maternal hypertensive disor- RESULTS: High attendance to the exercise program regardless of BMI
ders and linked to macrosomia (>4000 g) and low birthweight showed that pregnant women who did not exercise are 3 times more likely
(<2500 g). No large randomized controlled trial with high adherence to develop hypertension (odds ratio [OR], 2.96; 95% confidence interval
to an exercise program has examined pregnancy-induced hyper- [CI], 1.29e6.81, P ¼ .01) and are 1.5 times more likely to gain excessive
tension and these associated issues. We investigated whether weight if they do not exercise (OR, 1.47; 95% CI, 1.06e2.03, P ¼ .02).
women adherent (!80% attendance) to an exercise program initi- Pregnant women who do not exercise are also 2.5 times more likely to give
ated early showed a reduction in pregnancy-induced hypertension birth to a macrosomic infant (OR, 2.53; 95% CI, 1.03e6.20, P ¼ .04).
and excessive GWG in all prepregnancy BMI categories, and CONCLUSION: Maternal exercise may be a preventative tool for hy-
determined if maternal exercise protected against macrosomia and pertension and excessive GWG, and may control offspring size at birth
low birthweight. while reducing comorbidities related to chronic disease risk.
OBJECTIVE: We sought to examine the impact of a program of su-
pervised exercise throughout pregnancy on the incidence of pregnancy- Key words: exercise, gestational weight gain, hypertension, interven-
induced hypertension. tion, outcome, pregnancy
Exercise intervention
The randomization was performed in
waves so that each wave had between
10-12 participants in the exercise group,
and 10-12 women in the control group.
The exercise group trained 3 d/wk
(50-55 min/session), from weeks 9-11 of
pregnancy, to the end of the third
Barakat et al. Early maternal exercise prevents hypertension. Am J Obstet Gynecol 2016. trimester (weeks 38-39). An average of
85 training sessions was originally plan-
ned for each participant in the event of
induced hypertension and that exercise in 2000. The research protocol was no preterm delivery. The intervention
will protect against the initiation of this reviewed and approved by the ethics re- involved aerobic exercise, aerobic dance,
disease in women of all prepregnancy view board of Hospital Severo Ochoa muscular strength, and flexibility, and
BMI categories, while also protecting (Madrid, Spain). The onset of patient met the standards of the American
against excessive GWG. In addition, we enrollment was November 2012. Congress of Obstetricians and Gyne-
hypothesized that exercise protects cologists.19 Women used a heart rate
against macrosomia (>4000 g) and low Participants and randomization monitor (Accurex Plus; Kempel,
birthweight (<2500 g) and other preg- A total of 1100 Spanish-speaking Finland) during the training sessions
nancy complications. (Caucasian) pregnant women from pri- (heart rate was consistently <70% of
mary care medical centers (Figure) were age-predicted maximum) and the rating
Materials and Methods assessed for eligibility. Women with of perceived exertion scale ranged from
The present RCT (identifier: singleton and uncomplicated pregnan- 12-14 (somewhat hard).20
NCT01723098) was conducted from cies (no type 1, type 2, or gestational Each exercise session was preceded
December 2011 through January 2015 diabetes mellitus [GDM] at baseline) and followed by a gradual warm-up and
following the ethical guidelines of the with no history or risk of preterm cool-down period (both 10-12 minutes’
Declaration of Helsinki, last modified delivery were included. Women not duration) and consisted of walking and
light static stretching of most muscle occupational activity, previous physical fetal outcomes from the medical
groups. The cool-down period included activity habits, smoking status, previous records.
relaxation and pelvic floor exercises. preterm birth, and previous miscarriage
The main exercise session (25-30 mi- were obtained at the first prenatal Statistical analysis
nutes) included moderate resistance ex- visit either by reviewing the medical Power calculations for the primary
ercise performed through the full range records or by questionnaire. Inclusion/ outcome (diagnosis of hyperten-
of motion and engaged major muscle exclusion criteria were determined at this sion)23,24 used a prevalence of w4% in
groups (pectoral, back, shoulder, upper initial visit by the attending obstetrician. the intervention group and 10% in the
and lower limb muscles). One set (10-12 usual care group. Under these as-
repetitions) was conducted using bar- Outcomes sumptions, a 2-sample comparison (c2)
bells (2 kg/exercise) or low-to-medium Primary outcome with a 5% level of significance and a
resistance (elastic) bands (Therabands, Diastolic and systolic arterial BP were statistical power of 0.90 gave a study
Hygenic Corp., Catalonia, Spain). Exer- measured at every visit to the obstetri- population of 378 patients in each
cises in the supine position were not cian as part of standard care (once each group. Assuming a maximum lost to
performed for >2 minutes. trimester) and were obtained from follow-up of 10%, approximately 416
medical records. Criteria for measuring women were needed for each group at
Usual care (control) group BP3 were as follows: baseline.
Women randomly assigned to the con- For treatment group comparisons, we
trol group received general advice from 1) Measured in the sitting position with analyzed continuous and nominal data
their health care provider about the the arm at the level of the heart using with a Student t test for unpaired data
positive effects of physical activity. Par- an appropriately sized cuff; and c2 tests, respectively. We used
ticipants in the control group had their 2) Korotkoff phase V was used to logistic regression analysis to examine
usual visits with health care providers designate diastolic BP; the interaction between study group
during pregnancy, which were equal to 3) Diagnosis of hypertension was (training and control) on the likelihood
the exercise group. Women were not defined as a diastolic BP of !90 mm of developing hypertension (primary
discouraged from exercising on their Hg and a systolic BP !140 mm Hg,4 outcome), gaining excessive gestational
own. However, women in the control based on the average of at least 2 weight (secondary outcome), developing
group were asked by telephone about measurements, using the same arm GDM (secondary outcome), delivering a
their exercise once each trimester using a and recorded in the medical file. preterm infant (secondary outcome),
decision algorithm. and modifying other pregnancy out-
Question 1: Since the beginning of The primary outcome was the num- comes (length of newborn, Apgar scores
pregnancy, have you exercised in your ber (percentage/incidence) of women at 1 and 5 minutes after delivery, and
leisure time, in a supervised program, or who developed hypertension during cord blood pH; secondary outcomes) as
on your own? pregnancy. separate endpoints after controlling for
a. Answer: No. maternal age, parity, smoking status,
b. Answer: Yes. Secondary outcomes occupational activity during pregnancy,
Question 2 (if the previous response Total GWG was calculated on the basis of prepregnancy exercise habits, and
was “b”): Given 7 days a week, how many the weight at the last clinic visit before prepregnancy BMI. We also used logistic
days per week did you exercise? delivery minus the pregravid weight regression analysis to examine the
(from hospital records) and stratified by interaction between study group
a. Answer: <3 days.
prepregnancy BMI categories based on (training and control) and birthweight
b. Answer: !3 days.
the Institute of Medicine (IOM) guide- categories as separate endpoints on the
Question 3 (if the previous response lines.21 Excessive body weight gain was probability of having a newborn with
was “b”): Taking into account the total determined by IOM guidelines21 for macrosomia (>4000 g) and on the like-
duration of physical exercise continuously, prepregnancy BMI categories for each lihood of having a low-birthweight
how long did you exercise every day? woman: >18 kg for underweight; >16 (<2500 g) infant (secondary outcomes)
a. Answer: <20 minutes each day. kg for normal weight; >11.5 kg for after the same adjustment. We
b. Answer: !20 minutes each day. overweight; and >9 kg for obese. Diag- conducted statistical analyses using
Interpretation of the decision algo- nosis of GDM was also included from software (SPSS, Version 18.0; IBM Corp,
rithm: Pregnant women in the control medical records. Armonk, NY, and SAS, Version 9.3; SAS
group who reached level “b” of these 3 Birthweight was recorded from hos- Institute, Cary, NC). The level of signif-
questions were excluded from the study. pital perinatal records. Newborns were icance was set to #.05.
classified as having macrosomia when
Participant demographics birthweight was >4000 g and low Results
Demographic and other information birthweight was defined as <2500 g.22 A total of 840 pregnant women met the
(pregravid weight and height), parity, We obtained other maternal and criteria. After randomization, 38 women
TABLE 3
Effect of exercise on hypertension and other pregnancy outcomes by prepregnancy body mass index categories
Body mass index categories
Underweight Normal weight Overweight Obese
Control Exercise Control Exercise Control Exercise Control Exercise
Variable (n ¼ 20) (n ¼ 10) (n ¼ 256) (n ¼ 257) (n ¼ 78) (n ¼ 90) (n ¼ 29) (n ¼ 25) P
Maternal hypertension 0 0 15/5.9 2/0.8 6/7.7 4/4.4 1/3.5 2/8.0 .02
Excessive maternal weight gain 4/20.0 1/10.0 71/27.7 39/15.2 41/52.6 50/55.6 15/51.7 11/44.0 .01
Gestational diabetes 3/15.0 0 13/5.1 6/2.3 4/5.1 2/2.2 1/3.5 1/4.0 .03
Preterm delivery, <37 wk 1/5.0 0 21/8.2 19/7.4 13/16.7 7/7.8 2/6.9 3/12.0 .31
Newborn birthweight categories
Adequate 2500e4000 g 18/90.0 8/80.0 232/90.6 240/93.4 63/80.8 88/97.8 27/93.1 23/92.0 .01
Low <2500 g 2/10.0 1/10.0 14/5.5 12/4.7 7/9.0 2/2.2 2/6.9 1/4.0 .15
Macrosomia >4000 g 0 1/10.0 10/3.9 5/1.9 8/10.3 0 0 1/4.0 .03
2
Data are expressed as n/%, unless otherwise indicated. P values are based on c analyses.
Barakat et al. Early maternal exercise prevents hypertension. Am J Obstet Gynecol 2016.
lowered by the exercise intervention. If an pregnant women. Exercise may also pro- the ability to identify those women in the
exercise program started early in gestation tect against preeclampsia by reducing control group who did not remain
in asymptomatic low-risk women pre- maternal byproducts of oxidative stress, sedentary built into the study design.
vented hypertension (and potentially preventing endothelial dysfunction, and Interestingly, none of the control women
preeclampsia), which may be linked to stimulating vascularity and placental were excluded. It may be that without an
excessive GWG, then a healthy lifestyle growth.27 Furthermore, exercise has been exercise intervention and weekly
initiated preconception and in early shown to reduce excessive GWG in accountability, few pregnant women will
gestation may perhaps be the key issues to normal-weight,28,29 overweight, and continue exercising on their own. One
preventing chronic disease risk in both obese women,30 and to improve maternal limitation of our study is that we did not
mother and baby. Epidemiological studies health perception31 and mood.32,33 assess nutrition or energy intake, how-
suggest that women who are physically Women with !2 pregnancies may also ever, all women had standard care and
active are less likely to develop gestational benefit from maternal exercise because information regarding a healthy lifestyle
hypertension, based on retrospective our results would suggest that these during pregnancy, as the only difference
questionnaires, and none examined the women may be more susceptible to between the 2 groups was the initiation
interaction of exercise and prevention of excessive GWG than women with fewer of the exercise program. In addition, we
excessive GWG.12-15 We believe we are the pregnancies. In addition, exercise has did not assess occupational job stress in
first to link an early exercise intervention been linked to beneficial fetal and preg- our participants, which may also be
with high adherence and prevention of nancy outcomes.34-36 Our large RCT linked to baby size at birth42 although we
excessive weight gain to reducing the confirms that a healthy lifestyle inter- did control for occupation.
incidence of maternal chronic disease risk vention with high adherence that pre- A potential weakness may be the
(hypertension and GDM). In any RCT vents excessive GWG may indeed provide utility and applicability of our findings to
examining the effects of an exercise pro- a healthy environment to prevent future other clinical settings. However, it was
gram, high adherence and early preven- chronic disease risk in both mother and our intent to offer this exercise program
tion of excessive GWG may be the key offspring. Excessive GWG, macrosomia, within a hospital setting, with “buy-in”
issues to the efficacy of the trial. and low-birthweight babies have been from the hospitals involved. Many
Maternal exercise has many benefits linked with childhood obesity37-40 and pregnant women do not engage in
and has been associated with lowering of other offspring chronic disease risks, physical activity unless advised by their
BP25 and an increase in aerobic and car- including cardiovascular disease.41 physician/obstetrician to do so and with
diovascular conditioning.26 Engaging in an intervention placed within the hos-
exercise may be particularly important Strengths and weaknesses pital setting, our intervention women
for nulliparous women, as we found the The major strengths of our study are the were motivated to attend at least 80% of
incidence of hypertension and preterm large RCT with high adherence (!80% the offered sessions. Because of the suc-
delivery was reduced in our first-time attendance) in our exercise group and cess of our program, perhaps more
complications (review). Cochrane Database gain in normal weight pregnant women. Med Sci 38. Seiga-Riz AM, Evenson KR, Dole N. Preg-
Syst Rev. 2006 Apr 19;(2):CD005942. Review. Sports Exerc 2012;44:1419-26. nancy-related weight gainea link to obesity?
PMID: 16625645. 29. Ruiz JR, Perales M, Pelaez M, Lopez C, Nutr Rev 2004;62:S105-11.
19. American College Obstetricians Gynecolo- Lucia A, Barakat R. Supervised exercise-based 39. Schack-Neilsen L, Michaelsen KF,
gists. Exercise during pregnancy and the post- intervention to prevent excessive gestational Gamborg M, Mortensen EL, Sørensen TI. Gesta-
partum period. ACOG Committee opinion no. weight gain: a randomized controlled trial. Mayo tional weight gain in relation to offspring body
267, January 2002 (reaffirmed 2009). Obstet Clin Proc 2013;88:1388-97. mass index and obesity from infancy through
Gynecol 2002;99:171-3. 30. Mottola MF, Giroux I, Gratton R, et al. adulthood. Int J Obes (Lond) 2010;34:67-74.
20. O’Neill ME, Cooper KA, Mills CM, Boyce ES, Nutrition and exercise prevents excess weight 40. Ruchat SM, Mottola MF. Preventing long-
Hunyor SN. Accuracy of Borg’s ratings of gain in pregnant overweight women. Med Sci term risk of obesity for two generations: prena-
perceived exertion in the prediction of heart rates Sports Exerc 2010;42:265-72. tal physical activity is part of the puzzle.
during pregnancy. Br J Sports Med 1992;26: 31. Barakat R, Pelaez M, Lopez C, Montejo R, J Pregnancy 2012;2012:470247.
121-4. Coteron J. Exercise during pregnancy 41. Benyshek DC. The developmental origins of
21. Institute of Medicine. Weight gain during reduces the rate of cesarean and instrumental obesity and related health disorderseprenatal and
pregnancy: reexamining the guidelines. Wash- deliveries: results of a randomized controlled perinatal factors. Coll Antropol 2007;31:11-7.
ington (DC): National Academies Press; 2009: trial. J Matern Fetal Neonatal Med 2012;25: 42. Lee BE, Ha M, Park H, et al. Psychosocial
324. 2372-6. work stress during pregnancy and birthweight.
22. Stetzer BP, Thomas A, Amini SB, 32. Polman R, Kaiseler M, Borkoles E. Effect of a Paediatr Perinat Epidemiol 2011;25:246-54.
Catalano PM. Neonatal anthropometric mea- single bout of exercise on the mood of pregnant
surements to predict birth weight by ultrasound. women. J Sports Med Phys Fitness 2007;47:
J Perinatol 2002;22:397-402. 103-11.
Author and article information
23. Moher D, Dulberg CS, Wells GA. Statistical 33. Poudevigne MS, O’Connor PJ. Physical From the Physical Activity and Sports in Specific Pop-
power, sample size, and their reporting in ran- activity and mood during pregnancy. Med Sci ulations (AFIPE) Research Group, Faculty of Physical
domized controlled trials. JAMA 1994;272: Sports Exerc 2005;37:1374-80. Activity and Sports Sciences-Facultad de Ciencias de la
122-4. 34. Ferraro ZM, Gaudet L, Adamo KB. The po- Actividad Fı́sica y el Deporte, Technical University of
24. Schulz KF, Grimes DA. Sample size calcu- tential impact of physical activity during preg- Madrid, Madrid, Spain (Drs Barakat, Pelaez, Perales,
lations in randomized trials: mandatory and nancy on maternal and neonatal outcomes. Lopez, and Coteron); Catholic University of Murcia,
mystical. Lancet 2005;365:1348-53. Obstet Gynecol Surv 2012;67:99-110. Murcia, Spain (Dr Cordero); and R. Samuel McLaughlin
25. Yeo S. Prenatal stretching exercise and 35. Barakat R, Pelaez M, Montejo R, Luaces M, Foundation-Exercise and Pregnancy Laboratory, School
autonomic responses: preliminary data and a Zakynthinaki M. Exercise during pregnancy im- of Kinesiology, Faculty of Health Sciences, Department of
model for reducing preeclampsia. J Nurs proves maternal health perception: a random- Anatomy and Cell Biology, Schulich School of Medicine
Scholarsh 2010;42:113-21. ized controlled trial. Am J Obstet Gynecol and Dentistry, Children’s Health Research Institute, Uni-
26. Ruchat SM, Davenport MH, Giroux I, et al. 2011;204:402.e1-7. versity of Western Ontario, London, Ontario, Canada
Walking program of low or vigorous intensity 36. May LE, Suminski RR, Langaker MD, (Dr Mottola).
during pregnancy confers an aerobic benefit. Int Yeh HW, Gustafson KM. Regular maternal ex- Received Oct. 8, 2015; revised Nov. 23, 2015;
J Sports Med 2012;33:661-6. ercise dose and fetal heart outcome. Med Sci accepted Nov. 30, 2015.
27. Falcao S, Bisotto S, Michel C, et al. Exercise Sports Exerc 2012;44:1252-8. The authors report no conflict of interest.
training can attenuate preeclampsia-like fea- 37. Oken E, Kleinman KP, Belfort MB, This work was partially supported by the program
tures in an animal model. J Hypertension Hammitt JK, Gillman MW. Associations of AL14-PID-39, AL15-PID-06, Technical University of
2010;28:2446-53. gestational weight gain with short- and longer- Madrid, Spain.
28. Ruchat SM, Davenport MH, Giroux I, et al. term maternal and child health outcomes. Am Corresponding author: Michelle F. Mottola, PhD,
Nutrition and exercise reduce excessive weight J Epidemiol 2009;170:173-80. FACSM. [email protected]