Professional Documents
Culture Documents
Penurunan BB DGN Aktifitas
Penurunan BB DGN Aktifitas
Penurunan BB DGN Aktifitas
x
General obstetrics
www.bjog.org
Objective To examine the effect of an exercise and dietary Results A total of 190 pregnant women, 88 in the control group
intervention during pregnancy on excessive gestational weight gain and 102 in the intervention group, completed the study.
(EGWG), dietary habit and physical activity in pregnant women. Decreased daily intakes of calorie, fat, saturated fat and cholesterol
were detected in participants in the intervention group at
Design Randomised controlled trial.
2 months after enrolment compared with the control group
Setting Community-based study. (P < 0.01). Participants in the intervention group had higher
physical activity 2 months after enrolment compared with the
Population Nondiabetic urban-living pregnant women (<26 weeks
control group (P < 0.01). The lifestyle intervention during
of gestation).
pregnancy reduced the prevalence of EGWG in the intervention
Methods Participants in the intervention group were provided group compared with the control group (P < 0.01) according to
with community-based group exercise sessions, instructed home the guidelines of the Institute of Medicine.
exercise and dietary counselling between 20 and 36 weeks of
Conclusion The findings suggest that lifestyle intervention during
gestation. Participants in both groups received physical activity
pregnancy increased physical activity, improved dietary habits and
and food intake surveys at enrolment and 2 months after the
reduced EGWG in urban-living pregnant women.
enrolment.
Keywords Dietary intervention, exercise, gestational weight gain,
Main outcome measures Prevalence of EGWG and measures of
pregnancy.
physical activity and food intakes between the two groups.
Please cite this paper as: Hui A, Back L, Ludwig S, Gardiner P, Sevenhuysen G, Dean H, Sellers E, McGavock J, Morris M, Bruce S, Murray R, Shen G.
Lifestyle intervention on diet and exercise reduced excessive gestational weight gain in pregnant women under a randomised controlled trial. BJOG
2012;119:70–77.
70 ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG
Exercise and dietary intervention on gestational weight gain
Enrolment
Eligible
(n = 227)
ical activity have been recognised as important contribut-
ing factors to the epidemics of obesity and type II diabetes
mellitus.8 The American Congress of Obstetricians and Enrolled
(n = 224)
Excluded (n = 3)
Medical (n = 1)
Gynecologists has recommended mild-to-moderate exercise Obstetrical (n = 2)
Allocation
cise.9 Several groups examined the impact of lifestyle inter- (n = 224)
Follow up
EGWG according to the 2009 IOM guidelines remains
unclear. Discontinued (n = 23) Discontinued (n = 7)
Relocation (n = 1)
We hypothesise that a newly developed exercise and die- Relocation (n = 2)
Loss to follow up (n = 6)
Work or study (n = 8)
tary intervention during pregnancy19 may improve dietary Loss to follow up (n = 13)
Analysis
efficacy of the lifestyle intervention on EGWG, food intake Analysed
Analysed
(n = 88)
(n = 102)
and physical activity in urban-living pregnant women
through a randomised controlled trial (RCT). Figure 1. Consorted chart of the study.
Methods
Registration of clinical trial
Participants The enrolment for the trial started in July 2004 and
Nondiabetic pregnant women (<26 weeks of pregnancy, the trial was retrospectively registered with http://clinical
n = 227) living in Winnipeg were recruited on a voluntary trials.gov on 13 June 2007. Updates or changes were made
basis between July 2004 and February 2010 from prenatal eight times as required. Besides the addition of funding
classes or community clinics through posters and local sources and the website of the study, the major changes in
newspaper advertisements in Winnipeg. All applicants read the design of the trial are summarised as follows: the study
and signed an informed consent. Three applicants were start and end dates were changed from 2006–12 to 2004–15
excluded by physicians because of medical or obstetric con- (the initial recruitment of pilot participants in 2004–05
traindications to exercise during pregnancy, the remaining were added to the trial; the assessment of EGWG will end
224 pregnant women were approved and randomised into in 2014); the estimated enrolment was changed from 1600
control or intervention groups. Randomisation was per- to 500 based on our achieved enrolment during the last
formed using a computer-generated randomisation alloca- 6 years.
tion table by a staff member without involvement in the
study design. After randomisation, participants received a Instructed exercise
sealed envelope labelled with the assigned randomisation A community-based exercise programme specifically
number, which contained instructions for participants. The designed for pregnant women developed in our previous
nature of the study meant that participants and study staff studies19 was provided to participants in the intervention
were not blinded to the types of interventions. After group. Recommended exercise included walking, mild-to-
recruitment, four participants had miscarriages before the moderate aerobic, stretching and strength exercises. An
start of intervention (one in the control group and three in exercise regimen, three to five times per week (including a
the intervention group). Thirty participants withdrew from weekly exercise session and multiple home sessions) of
the study (23 in the control group and seven in the inter- mild-to-moderate exercise for 30–45 minutes/session was
vention group) because of relocation, time commitment recommended to participants in the intervention group.
(work or study) or loss of contact for undefined reasons. It was recommended that the exercise programme started
A total of 190 participants completed the programme, 88 between 20 and 26 weeks of pregnancy and ended at
in the control group and 102 in the intervention group, 36 weeks of pregnancy. The group exercise sessions were
and their data were analysed (Figure 1). held in air-conditioned gymnasia in community centres
ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG 71
Hui et al.
located in residential areas. Participants had options for Society of Exercise Physiology, which was validated previ-
attending daytime or evening exercise sessions. Floor aero- ously using peak oxygen consumption.26 Unfit (physical
bic, stretching and strength exercises were led by licensed activity index = 0) during pregnancy was defined as recrea-
fitness trainers in group sessions for participants to learn tional activity <1–2 times/week plus <20 minutes/time.
skills for exercise during pregnancy. An exercise instruc- Active (physical activity index = 1) was defined as recrea-
tion video designed for pregnant women by exercise phy- tional activity 1–2 times/week, >20 minutes/time or
siologists was provided to participants in the intervention >2 times/week but <20 minutes/time. Fit (physical activity
group to assist their home exercise. Participants in the index = 2) was defined as recreational activity >2 times/
intervention group were taught to record daily physical week plus >20 minutes/time. Food intakes of all partici-
activities in activity logbooks, which were collected weekly pants were assessed using 3-day food records27 and the
by the project coordinator from the participants in the method was previously validated.28 The results of food
intervention group. intake were analysed using NutriBase 6.0 software (Cyber-
Soft, Inc., Phoenix, AZ, USA).
Dietary intervention
Dietary interviews and counselling were provided twice to Data collection
each participant in the intervention group by registered Data on delivery route, maternal weight at delivery room,
dietitians, the first at enrolment and the second 2 months birthweight and birthweight-related obstetric procedures
after enrolment. The dietary interview was assisted with a (induction, forceps or caesarean section) were collected
Food Choice Map, a computerised dietary interview tool. from hospital medical charts. The diagnosis of gestational
The Food Choice Map consists of a specifically designed diabetes mellitus was made according to the 2008 guide-
board or ‘map’, 91 magnetic stickers with pictures of com- lines of the Canadian Diabetes Association.29 Large-for-
mon foods and bar codes, and software20 modified for gestational-age was determined based on birthweight and
pregnant women. Food Choice Map assessment for nutri- gestational age as previously described.30 EGWG was calcu-
tional intake was recently validated in women of reproduc- lated by subtracting the upper limit of normal weight gain
tive age.21 During interviews, participants recalled their for corresponding pre-pregnancy BMI following the 1990
food intakes in a typical week. Participants and dietitians or 2009 guidelines of IOM3,5 from the actual weight gain
jointly placed stickers on the map. Bar codes and locations (difference between pre-pregnancy weight and bodyweight
of stickers on the map represent the frequency, types and at delivery room).
quantity of food intakes, which were scanned into the com-
puter at the end of each interview.20 Participants’ daily Outcomes
intakes for total calorie and nutrients were instantly analy- The primary outcome of the present study was the preva-
sed. Dietitians provided personalised dietary counselling to lence of EGWG. Other measured variables included food
participants based on their Food Choice Map interview intake, physical activity, the prevalence of large-for-gesta-
results, pregnancy week, weight gain and the Health Can- tional-age, gestational diabetes mellitus, weight-related
ada guidelines for food intake in pregnancy.22,23 The dieti- obstetric procedures, gestational weight gain and birth-
tian assessed the changes in dietary pattern in the weights.
participants in the intervention group during the second
dietary interview at 2 months after enrolment and provided Power assessment and statistical analysis
further advice if it was indicated. The results of our pilot study detected EGWG in seven of
19 participants in the control group, but two out of 20 in
Control group the intervention group. With an estimated dropout rate of
Participants in the control group received standard prenatal 16%, the sample size for detecting a significant difference
care recommended by the Society of Obstetricians and in the primary outcome, EGWG, between the control and
Gynecologists of Canada24 and were provided with a pack- intervention groups was 86/group (a = 0.05) with 80%
age of up-to-date information on physical activity and power for the chi-square test. The effect size for sample size
nutrition healthy pregnancy from the Health Canada.25 estimation for EGWG based on the results of the pilot
Exercise instruction or dietary intervention was not pro- study was 0.37 with Phi coefficient assessment. Demo-
vided to participants in the control group. graphic variables (age, pre-pregnancy weight, height, BMI
and annual family income), dietary intake, physical activity
Assessment of physical activity and food intake index, birthweights and gestational weight gain were analy-
Recreational physical activities of all participants were sed using the Student’s t-test. Bonferroni correction was
semi-quantified using an activity questionnaire on a used for data with multiple variables. The prevalence of
PARMed-X form for Pregnancy designed by the Canadian gestational diabetes mellitus, EGWG, First Nations status
72 ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG
Exercise and dietary intervention on gestational weight gain
and large-for-gestational-age infants was assessed using the 3-day food records at baseline and 2 months after enrol-
chi-squared test. Data analyses were conducted by statisti- ment. The remaining participants submitted the first, but
cians without knowledge of participants. not the second, food records. At the baseline, no significant
difference in total calorie, protein, carbohydrate, fat, satu-
rated fat, fibres or cholesterol intake was detected between
Results
participants in the control and intervention groups who
A total of 190 participants (88 in the control group and completed both food records (Table 2). Participants in the
102 in the intervention group) completed their programme intervention group at 2 months after enrolment had signifi-
and delivered babies before 31 July 2010. No significant cantly lower intake of total calorie, fat, saturated fat and
difference was detected in age, pre-pregnancy weight, cholesterol compared with that in the control group
height, BMI, First Nations status, or annual family income (P = 0.00004–0.002). The ratios of daily carbohydrate, pro-
between the two groups (Table 1). tein or fat intake versus total calorie intake between the
Over half (55.8%) of participants (53 in the control two groups at baseline were not significantly different.
group and 53 in the intervention group) completed both At 2 months after enrolment, participants in the interven-
tion group had a significantly lower ratio of fat compared
with those in the control group (27.0% versus 31.5%,
P = 0.001) (Table 2). No significant difference after Bonfer-
Table 1. Demographic information of pregnant women in control
roni correction in the intake of starch, very lean meat, lean
and intervention groups
meat, medium-fat meat, high-fat meat, vegetables, fruits,
Control Intervention skim milk, 1–2% fat milk, whole fat milk or dietary fats
(n = 88) (n = 102) (oil and fats) was detected between the control and inter-
vention groups at baseline or 2 months after enrolment
Age (years) 28.7 ± 5.9 30.1 ± 5.2 (Table 3).
Pre-pregnancy weight (kg) 70.7 ± 15.7 68.0 ± 16.9 In participants who completed the programme, 179
Height (cm) 165.8 ± 7.3 165.3 ± 6.4
(93.7%) (84 in the control group and 95 in the interven-
Pre-pregnancy BMI (kg/m2) 25.7 ± 5.1 24.9 ± 5.4
tion group) submitted both physical activity questionnaires
Family annual income ($) 48,602 ± 29,628 50,833 ± 23,792
First Nations (number/%) 22/25.0 19/17.4 at baseline and 2 months after enrolment, and the other
11 participants submitted one of the two questionnaires.
BMI, body mass index; First Nations, Canadian Aboriginals with First The baseline levels of physical activity index of participants
Nations status.
in the intervention group were not significantly different
Values are expressed in mean ± SD or number versus percentage in
the group.
from that in the control group. All participants in the
intervention group had self-reported home exercise for
Table 2. Daily intakes of nutrients of pregnant women in control and intervention groups
Total calorie 2139 ± 580 2416 ± 848 2095 ± 661 1991 ± 458 0.71 0.002
Carbohydrate (g) 288 ± 76.2 324 ± 126 291 ± 83.4 283 ± 71.0 0.86 0.04
Protein (g) 90.2 ± 34.9 94.6 ± 34.4 86.2 ± 26.1 85.5 ± 21.0 0.50 0.11
Fat (g) 71.7 ± 27.3 86.8 ± 36.2 70.4 ± 36.6 62.5 ± 24.4 0.84 0.0001
Saturated fat (g) 25.1 ± 10.5 29.2 ± 13.2 21.1 ± 10.9 19.7 ± 9.2 0.06 0.00004
Cholesterol (mg) 270 ± 173 323 ± 220 219 ± 118 208 ± 104 0.08 0.001
Fibre (g) 22.2 ± 10.1 23.3 ± 11.8 25.2 ± 10.1 24.3 ± 9.9 0.13 0.63
Carbohydrate ratio (%) 53.3 ± 6.6 52.6 ± 7.7 55.1 ± 6.5 55.8 ± 5.8 0.14 0.02
Protein ratio (%) 16.6 ± 3.7 15.7 ± 3.3 16.4 ± 3.6 17.1 ± 3.7 0.85 0.04
Fat ratio (%) 30.3 ± 6.0 31.5 ± 7.5 28.6 ± 7.6 27.0 ± 6.5 0.20 0.001
a, Control versus Intervention at baseline; b, Control versus Intervention at 2 months after enrolment; Ratio, % of daily calorie from macro-
nutrient.
Values are expressed in mean ± SD and analysed using Student’s t test. Underlining indicates P values with statistical significance after Bonferroni
correction.
ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG 73
Hui et al.
Table 3. Daily servings in food groups in pregnant women in control and intervention groups
Starch 7.43 ± 3.63 7.41 ± 4.13 7.56 ± 4.55 7.11 ± 2.51 0.87 0.66
Very lean meat 1.46 ± 2.03 1.71 ± 2.28 2.02 ± 2.35 1.54 ± 1.67 0.19 0.66
Lean meat 1.78 ± 2.36 1.35 ± 1.71 1.53 ± 2.05 1.87 ± 2.11 0.55 0.17
Medium-fat meat 1.14 ± 1.44 1.69 ± 2.51 0.71 ± 0.87 0.76 ± 1.00 0.07 0.01
High-fat meat 1.14 ± 1.92 0.75 ± 1.09 0.68 ± 0.89 0.62 ± 0.95 0.12 0.50
Vegetables 2.86 ± 3.58 2.14 ± 2.38 2.59 ± 2.92 2.51 ± 2.47 0.67 0.43
Fruits 2.53 ± 2.12 2.66 ± 1.55 2.93 ± 2.38 2.95 ± 1.89 0.36 0.39
Skim milk 0.96 ± 1.30 0.87 ± 1.28 1.11 ± 1.20 1.43 ± 1.26 0.54 0.02
1–2% fat milk 0.75 ± 1.15 0.74 ± 1.15 0.29 ± 0.53 0.29 ± 0.75 0.01 0.02
Whole fat milk 0.12 ± 0.33 0.16 ± 0.52 0.09 ± 0.23 0.06 ± 1.15 0.60 0.15
Oil and fats 5.17 ± 3.63 5.51 ± 3.47 5.14 ± 5.08 4.08 ± 2.85 0.97 0.02
a, Control versus Intervention at baseline; b, Control versus Intervention at 2 months after enrolment; Oil and fats, dietary fats including butter,
cooking oil, dressing and cream.
Values are expressed in mean ± SD and analysed using Student’s t test (no significant difference in above variables was detected between control
and intervention groups after Bonferroni correction).
3–5 times/week between 20 and 36 weeks of gestation. No participants in the control group and two in the interven-
serious adverse effect of exercise (syncope, chest pain, tion group received caesarean sections. The prevalence of
shortness of breath, vaginal bleeding or miscarriage) was gestational diabetes mellitus or caesarean section between
detected in participants in the intervention group during the two groups was not significantly different in the sample
the study. At 2 months after the enrolment, the levels of size of the present study (Table 5).
physical activity index were significantly higher in the Weight gains of participants between the two groups
intervention group compared with baseline (P = 0.0001) or (control group: 15.2 ± 5.9 kg and intervention group:
that in the control group at a comparable pregnancy period 14.1 ± 6.0 kg) were not significantly different. According to
(P = 0.00002). No significant difference in the levels of the 2009 IOM guidelines,5 54.5% of participants in the
physical activity index was detected between the control control group (48 out of 88) had EGWG. The participants
and intervention groups at baseline (Table 4).
Gestational weeks, birth weights of offspring and the
prevalence of large-for-gestational-age infants were not sig- Table 5. Pregnancy outcomes of pregnant women in control and
intervention groups
nificantly different between the two groups. Three partici-
pants in the control group (3.3%) and two in the Variables Control Intervention P value
intervention group (1.8%) developed gestational diabetes (n = 88) (n = 102)
mellitus during the study. Participants with gestational dia-
betes mellitus received standard treatment following the Gestational weeks (week) 39.4 ± 1.5 39.6 ± 1.2 0.33
guidelines of the Canadian Diabetes Association.29 Three Gestational weight 15.2 ± 5.9 14.1 ± 6.0 0.28
gain (kg)
EGWG (2009 IOM 48 36 0.008
guidelines)
Table 4. Physical activity data Baby birthweight (g) 3,516 ± 530 3,490 ± 509 0.73
Large-for-gestational-age 15/17.0 12/11.8 0.41
Control Intervention P value
(n/%)
(n = 84) (n = 95)
Gestational diabetes (n/%) 3/3.4 2/2.0 0.87
Cesarean section (n/%) 3/3.4 2/2.0 0.87
Baseline 1.62 ± 0.64 1.51 ± 0.73 0.27
2 months after 1.45 ± 0.72 1.85 ± 0.44 0.00002 EGWG, excessive gestational weight gain; IOM, Institute of Medi-
enrolment cine.
Values are expressed in mean ± SD and analysed using Student’s
Values are expressed as mean ± SD and analysed using Student’s t test and chi-square assay. Underlining indicates P values with sta-
t test. Underlining indicates P values with statistical significance. tistical significance.
74 ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG
Exercise and dietary intervention on gestational weight gain
ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG 75
Hui et al.
dietary habit probably contributes to the reduced preva- project, data interpretation and manuscript preparation;
lence of EGWG in participants in the intervention group in RM and SB provided advice on statistical analysis.
the present study. The improvement in the prevalence of
EGWG in the pregnant women in the intervention group Details of ethics approval
may justify the intensity of the dietary intervention. The study protocol and the informed consent form were
Multiple factors may influence the effect of exercise on approved by the Health Research Ethics Board in the
EGWG, including the types or intensity of lifestyle inter- University of Manitoba (the study was first approved on
vention, monitoring, compliance, and socioeconomic and 28 June 2004 and the most recent renewal was received
environmental factors.33 Home exercise is the main stream on 28 January 2011, Reference No. H2004-002).
of the exercise programme recommended to pregnant
women in the present study. A specifically designed video Funding
exercise instruction was provided to participants in the The study was supported by operating grants from the
intervention group in the present study to assist their home Lawson Foundation, the Canadian Institutes of Health
exercise. The aerobic and strength exercises demonstrated Research and the Public Health Agency of Canada.
in the video instruction require minimal space or facilities,
making them feasible for city-living pregnant women in Acknowledgements
various weather and socioeconomic conditions. Weekly The authors are grateful for help with recruitment from
group exercise led by a professional trainer helped partici- the Healthy Start for Mom & Me Prenatal Program in
pants to acquire and validate knowledge and skills for exer- Winnipeg and the Winnipeg Regional Health Authority. j
cise during pregnancy. Group exercise may also help to
develop acceptance and adhesion of pregnant women to
References
the healthy lifestyle programme. The activity logbook and
follow-up visits helped the monitoring of physical activity 1 Writing Group for the SEARCH for Diabetes in Youth Study Group,
Dabelea D, Bell RA, D’Agostino RB Jr, Imperatore G, Johansen JM,
in participants. The levels of physical activity index were
et al. Incidence of diabetes in youth in the United States. JAMA
significantly increased in participants in the intervention 2007;297:2716–24.
group compared with that in the control group at 2 Tremblay MS, Katzmarzyk PT, Willms JD. Temporal trends in over-
2 months after enrolment. This suggests that the majority weight and obesity in Canada, 1981–1996. Int J Obes Relat Metab
of participants in the intervention group followed the rec- Disord 2002;26:538–43.
3 Harris SB, Gittelsohn J, Hanley A, Barnie A, Wolever TM, Gao J,
ommended exercise programme during the study. Increased
et al. The prevalence of NIDDM and associated risk factors in native
physical activity may also contribute to the reduced EGWG Canadians. Diabetes Care 1997;20:185–7.
in participants in the intervention group. 4 Boney CM, Verma A, Tucker R, Vohr BR. Metabolic syndrome in
In conclusion, the present study demonstrated that a childhood: association with birth weight, maternal obesity and ges-
community-based exercise and dietary intervention reduced tational diabetes mellitus. Pediatrics 2005;115:e290–6.
5 Parker JD, Abrams B. Prenatal weight gain advice: an examination of
the prevalence of EGWG in urban-living pregnant women,
the recent prenatal weight gain recommendations of the Institute of
which was associated with increased physical activity and Medicine. Obstet Gynecol 1992;79:664–9.
improved dietary habit compared with the control group. 6 Rasmussen KM, Yaktine AL. Weight Gain During Pregnancy; Reex-
Making lifestyle changes could be physically and mentally amine the Guidelines. Committee to Reexamine IOM Pregnancy
challenging. Pregnant women tend to be more motivated to Weight Guidelines. Washington, DC: Institute of Medicine; National
Research Council. USA, 2009.
suggestions on lifestyle modification compared with other
7 Scholl TA, Hediger ML, Schall JI, Ances IG, Smith WK. Gestational
stages of their lives, which is possibly because of their con- weight gain, pregnancy outcome, and postpartum weight retention.
cerns for the health of the fetus.34 A healthy lifestyle adopted Obstet Gynecol 1995;86:423–7.
during pregnancy potentially exerts an extended impact on 8 Ravussin E, Valencia ME, Esparza J, Bennett PH, Schulz LO. Effects
the health of mothers and their children, which remains to of a traditional lifestyle on obesity in Pima Indians. Diabetes Care
1994;17:1067–74.
be examined through long-term follow-up study.
9 ACOG Committee Opinion No 267. Exercise during pregnancy and
the postpartum period. Obstet Gynecol 2002;99:171–3.
Disclosure of interests 10 Borberg C, Gillmer MD, Brunner EJ, Gunn PJ, Oakley NW, Beard
The authors do not have any conflict of interest regarding RW. Obesity in pregnancy: the effect of dietary advice. Diabetes
the content of results presented in the text. Care 1980;3:476–81.
11 Olson CM, Strawderman MS, Reed RG. Efficacy of an intervention
to prevent excessive gestational weight gain. Am J Obstet Gynecol
Contribution to authorship 2004;191:530–6.
AH and LB were the main contributors for data and were 12 Claesson IM, Sydsjo G, Cedergren M, Jeppsson A, Nystrom F, Sydsjo A,
involved in the manuscript preparation; GXS, SL, PG, GS, et al. Weight gain restriction for obese pregnant women: a case–
HD, ES, JM, RM and MM contributed to the design of the control intervention study. BJOG 2008;115:44–50.
76 ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG
Exercise and dietary intervention on gestational weight gain
ª 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2011 RCOG 77