Research www. AJOG.

org

OBSTETRICS
Resistance exercise and glycemic control
in women with gestational diabetes mellitus
Marcelo C. de Barros, MSc; Marco A. B. Lopes, MD, PhD; Rossana P. V. Francisco, MD, PhD;
Andreia D. Sapienza, MD; Marcelo Zugaib, MD, PhD

OBJECTIVE: The objective of the study was to evaluate the effect of a
resistance exercise program with an elastic band on insulin require-
ment and glycemic control in patients with gestational diabetes mellitus
(GDM).
STUDY DESIGN: Sixty-four patients with gestational diabetes mellitus
were randomly assigned into 2 groups: an exercise group (EG; n ⫽ 32)
and a control group not submitted to the exercise program (CG; n ⫽
32).
RESULTS: A significant reduction in the number of patients who re-
quired insulin was observed in the EG (7/32) compared with the CG
group (18/32) (P ⫽ .005). The percentage of time spent within the pro-
posed target glucose range (of at least 80% of weekly measurements
below the limits preestablished for the disease) was significantly higher
in EG compared with the CG group (EG ⫽ 0.63 ⫾ 0.30; CG ⫽ 0.41 ⫾
0.31; P ⫽ .006).
CONCLUSION: The resistance exercise program was effective in reduc-
ing the number of patients with GDM who required insulin and in im-
proving capillary glycemic control in this population.
Key words: gestational diabetes mellitus, insulin, pregnancy,
resistance exercise

Cite this article as: de Barros MC, Lopes MAB, Francisco RPV, et al. Resistance exercise and glycemic control in women with gestational diabetes mellitus. Am J
Obstet Gynecol 2010;203:xx-xx.

G estational diabetes mellitus (GDM)

is defined as any degree of glucose
intolerance that first occurs or is first di-
agnosed during pregnancy.1 Adequate
glycemic control of patients with GDM is
necessary to minimize complications re-
sulting from the disease. In this respect,
diet therapy is the first intervention,
which, if unsuccessful, should be accom-
panied by insulin treatment.2 Although
effective in reducing the risk of fetal mac-
rosomia,3 insulin therapy does not act on
the origin of the problem (ie, increased
peripheral resistance to the action of in-
sulin in the second half of gestation).4
From the Department of Obstetrics and
Gynecology, University of Sao Paulo School
of Medicine, Sao Paulo, Brazil.
Received Dec. 28, 2009; revised May 3, 2010;
accepted July 15, 2010.
Reprints: Marcelo Costa de Barros, MSc,
Instituto Central do Hospital das Clínicas da
Universidade de São Paulo, Avenida Doutor
Enéas de Carvalho Aguiar, 255, 10o Andar,
Sala 10.086, CEP 05403-000, São Paulo, SP,
Brasil. barrosmarcelo@uol.com.br.
0002-9378/$36.00
© 2010 Mosby, Inc. All rights reserved.
doi: 10.1016/j.ajog.2010.07.015
Because skeletal muscles represent the
main site of insulin resistance observed
during pregnancy,5 both the American
College of Sports Medicine6 and the
American College of Obstetricians and
Gynecologists7 recommend exercise as
an effective and safe supporting therapy
for the treatment of GDM because sev-
eral lines of evidence have shown the ef-
fectiveness of aerobic exercise in reduc-
ing the frequency of women using
insulin. On the other hand, little is
known about the therapeutic application
of resistance exercise (RE) to women
with GDM. This modality of exercise
(RE) consists of the voluntary contrac-
tion of skeletal muscle of a certain body
segment against some type of external re-
sistance.8 The scientific literature re-
garding the use of this type of physical
activity as a coadjuvant for the treatment
of GDM is scarce. To our knowledge,
there is only 1 study on this subject con-
ducted by Brankston et al9 in 2004.
Because RE might be an important
tool for glycemic control in patients with
GDM, the objective of the present study
was to evaluate the impact of an RE pro-
gram with elastic band on insulin re-
quirement and on the adequacy of
capillary glycemic control in pregnant
patients with GDM. We hypothesized
that RE would reduce the number of
women who require insulin and would
improve glycemic control in these
patients.
M ATERIALS AND M ETHODS
A randomized controlled trial was con-
ducted involving patients with GDM
who were under prenatal follow-up at
the Obstetric Clinic of the University
Hospital, University of Sao Paulo School
of Medicine, between October 2006 and
November 2008 (study period). The
study was approved by the Institutional
Review Board of the University Hospital,
University of Sao Paulo, and written in-
formed consent for participation was
obtained from each patient. For data
analysis, the patients were randomized
into 2 groups: an exercise group (EG),
which underwent an RE program until
the end of gestation, and a control group
(CG), which was not submitted to this
program.
Patients with a diagnosis of GDM
(made by the 3 hour oral glucose toler-
ance test [OGTT] after a 100 g glucose
overload or by the 2 hour OGTT after a
75 g overload)10 who fulfilled the follow-
ing criteria were included in the study:

MONTH 2010 American Journal of Obstetrics & Gynecology 1.e1

Research Obstetrics
FIGURE 1
Elastic band
de Barros. Resistance exercise and gestational diabetes mellitus. Am J Obstet Gynecol 2010.
sedentary patients according to the In-
ternational Physical Activity Question-
naire (IPAQ),11 nonsmokers, age range
18-45 years, no physical factor or disease
limiting exercise, singleton pregnancy
and absence of fetal malformation upon
ultrasound, gestational age ranging from
24 to 34 weeks, and no risk factors for
preterm delivery. Exclusion criteria were
the occurrence of clinical or obstetric
complications contraindicating exercise
during pregnancy and loss to follow-up.
All subjects were instructed not to start
any new type of physical activity after the
randomization.
On the first visit after the diagnosis of
GDM, the patients received instructions
regarding capillary glucose monitoring
through self-monitoring of fingertip
capillary blood glucose 4 times per day
(after an overnight fast, 2 h after break-
fast, 2 h after lunch, and 2 h after dinner).
The objective was to maintain fasting
glucose levels of 95 mg/dL or less and
postprandial levels of 120 mg/dL or less
according to American Diabetes Associ-
ation (ADA) guidelines.10
All pregnant women with a diagnosis
of GDM were systematically evaluated
and received diabetic dietary instruc-
tions from a nutritionist. The diet was
divided into 7 servings, containing 35
1.e2 American Journal of Obstetrics & Gynecology MONTH 2010
kcal/kg ideal weight–1/day–1, and 300
kcal/day were added in the second and
third trimesters of gestation.
After the first medical visit, the patient
was informed about the opportunity to
participate in this study. Women admit-
ted to the study were randomized using a
computer-generated random series pro-
duced by a person not related to the pro-
tocol.12 The main researcher (M.C.D.B.)
received the next on a numbered series of
sealed opaque envelopes and was re-
sponsible for allocating each patient to 1
of the 2 groups (EG or CG) according to
the randomization list and instructed the
patients regarding RE. However, this re-
searcher did not participate, at any time,
in the decision about the introduction or
adjustment of the insulin dose. The ob-
stetricians responsible for clinical and
prenatal care and data recording was un-
aware to which group the patients be-
longed, and only the main researcher
(M.C.D.B.) questioned the patients with
respect to the exercise practice.
RE was performed with an elastic band
(Figure 1). The patients were advised to
start the program about 90 minutes after
consuming a meal of their preference
(breakfast, lunch, or dinner) and after
the measurement of capillary glycemia
with an Accu-Chek Advantage glucose
www.AJOG.org
meter (Roche Diagnostics, Indianapolis,
IN). If capillary glucose levels were be-
tween 100 and 250 mg/dL, EG patients
started the program with a stretching se-
quence. If capillary glycemia was below
this range, the patients were instructed to
do the RE program on the next day to
prevent hypoglycemia. If the values were
above this range, the patients were in-
structed not to undergo RE to prevent
the occurrence of ketoacidosis and to
contact the responsible obstetrician.
The RE program consisted of a circuit
type resistance training,8 elaborated in
such a way that the main muscle groups
of the patients would be exercised (chest,
back, biceps, triceps, deltoid, quadriceps,
thigh, and calf muscles). A circuit series
was defined as a sequence of these eight
exercises (stations). The women per-
formed 15 repetitions of each exercise
(station), with a minimum resting pe-
riod of 30 seconds and a maximum of 1
minute between each one. In the first and
second week of follow-up, the women
underwent 2 circuit series, followed by 3
circuit series from the third week of
inclusion in the study to the end of
gestation.
Exercise intensity was controlled by
the women themselves using a perceived
exertion scale for RE13 provided to them.
The women were advised to maintain an
exercise intensity close to 5 or 6, which
corresponds to a “somewhat heavy” ex-
ercise perception. The use of the elastic
band facilitated the adaptation of exer-
cise intensity to the level of conditioning
of the patients because only its length
needed to be adjusted. The main re-
searcher (M.C.D.B.) was responsible for
the adjustments, which were always
made on the weekly return visits of the
patients to the outpatient clinic.
The patients received written guide-
lines of how to perform each exercise and
were instructed to undergo the program
on 3 nonconsecutive days a week (twice a
week at home). The participants were
contacted by telephone at least once a
week to verify adherence to the program.
The other session was performed during
the weekly return visit, always under the
supervision of the main researcher
(M.C.D.B.). On that occasion, EG pa-
tients were asked whether they had per-
www.AJOG.org Obstetrics Research

formed any type of physical activity

other than RE. In addition, the adequate TABLE 1
execution of RE at the correct intensity Maternal characteristics
was verified. Control group Exercise group
The prenatal routine care of CG pa- Variable (n ⴝ 32) (n ⴝ 32) Pa
tients was kept unchanged. The patients Maternal age, y 32.40 ⫾ 5.40 31.81 ⫾ 4.87 .646
..............................................................................................................................................................................................................................................
attended weekly return outpatient visits Gestational age at diagnosis, wks 27.53 ⫾ 3.33 28.37 ⫾ 2.52 .258
..............................................................................................................................................................................................................................................
and, on that occasion, were questioned Gestational age at first visit, wks 31.06 ⫾ 2.30 31.56 ⫾ 2.29 .391
whether they had started some type of ..............................................................................................................................................................................................................................................

physical activity. In addition, the pa- Pregestational BMI, kg/m 2

25.39 ⫾ 3.81 25.34 ⫾ 4.16 .953
..............................................................................................................................................................................................................................................

tients again responded to the IPAQ.11 First-visit BMI, kg/m 2

29.43 ⫾ 4.01 29.61 ⫾ 4.13 .857
..............................................................................................................................................................................................................................................
The research visits and care visits were OGTT, 75 g (n ⫽ 5) (n ⫽ 6)
.....................................................................................................................................................................................................................................
scheduled weekly at the same day for all Fasting, mg/dL 93.0 ⫾ 14.0 92.3 ⫾ 17.2 .946
patients (CG and EG). .....................................................................................................................................................................................................................................

The glycemic profile of each patient 1 hour, mg/dL 177.5 ⫾ 29.0 189.5 ⫾ 13.4 .648
.....................................................................................................................................................................................................................................

was determined weekly, and insulin was 2 hour, mg/dL 161.4 ⫾ 20.4 169.1 ⫾ 43.8 .725
..............................................................................................................................................................................................................................................
introduced when more than 30% of the OGTT, 100 g (n ⫽ 27) (n ⫽ 26)
.....................................................................................................................................................................................................................................
glucose measurements were above the Fasting, mg/dL 95.93 ⫾ 14.84 94.53 ⫾ 23.38 .775
recommended value or when 20-30% of .....................................................................................................................................................................................................................................

the measurements indicated hypergly- 1 hour, mg/dL 194.17 ⫾ 28.97 186.96 ⫾ 24.97 .320
.....................................................................................................................................................................................................................................

cemia and fetal weight was above the 2 hour, mg/dL 170.63 ⫾ 26.25 170.91 ⫾ 26.76 .966
.....................................................................................................................................................................................................................................
75th percentile. The estimated fetal 3 hour 143.48 ⫾ 25.24 143.2 ⫾ 24.92 .945
..............................................................................................................................................................................................................................................
weight percentile was determined once a Values are the mean ⫾ SD.
month.14 The initial insulin dose (NPH) BMI, body mass index; OGTT, oral glucose tolerance test.
was calculated at 0.4 IU/kg–1 per day–1, a
Student t test.
divided into half at breakfast, one-quar- de Barros. Resistance exercise and gestational diabetes mellitus. Am J Obstet Gynecol 2010.

ter at lunch, and one-quarter at 10:00

10
PM. Regular insulin was added at meals
to improve the glycemic control if the ex-
clusive use of NPH insulin was not suffi-
cient to maintain the glycemias at ade-
quate levels.10
Comparison of the percentage of
weeks spent within the target glucose
range of at least 80% of weekly measure-
ments considered to be normal for ges-
tation according to ADA guidelines10
was used as a parameter to evaluate the
adequacy of glycemic control. Through-
out the follow-up period (time between
study inclusion and delivery), the weekly
mean of all capillary glucose was calcu-
lated for both groups. The mean weekly
values of fasting and postprandial capil-
lary blood glucose levels were also ob-
tained. The analysis of the mean of
means weekly values previously estab-
lished10 was performed for comparison
age of patients in EG who required insu- Variables measured at diagnosis, at in-
lin to 20%, with approximately 50% of clusion in the study, and during delivery
patients with GDM using insulin accord- were compared between groups by the
ing to the databank of the Endocrinopa- Student t test.15 The type of delivery is
thy and Pregnancy Unit (obstetric clin- reported as absolute and relative fre-
ic’s department responsible for the care quency for each group, and a possible as-
of pregnant women affected by endo- sociation of this variable with the groups
crine disturbances, such as thyroidopa- was subsequently tested by the ␹2 test.16
thies and GDM). Assuming a power of A graph of mean glycemic profiles and
the test of 80% and a level of significance their respective standard errors was con-
of 5%, the calculated sample size in each structed for the analysis of glucose levels
group was 30 patients. over the weeks of gestation,17 and the re-
TABLE 2
Outcomes
Control group Exercise group
Variable (n ⴝ 32) (n ⴝ 32) P
Required insulin 18 (56.3%) 7 (21.9%) .005a
..............................................................................................................................................................................................................................................
Amount of insulin required, U/kg 0.49 ⫾ 0.14 0.44 ⫾ 0.11 .401 b
..............................................................................................................................................................................................................................................
Latency to insulin requirement, wks 2.11 ⫾ 1.64 1.85 ⫾ 1.21 .715 b

between groups. ..............................................................................................................................................................................................................................................

Mean glucose levels, mg/dL 102.89 ⫾ 7.88 100.30 ⫾ 9.37 .084 b
..............................................................................................................................................................................................................................................

Statistical analysis Percentage of weeks spent within the 0.41 ⫾ 0.31 0.63 ⫾ 0.30 .006 b

All variables were submitted to compar- target glucose rangec

..............................................................................................................................................................................................................................................
ative analysis using the SPSS for Win- Values are the mean ⫾ SD.

dows 15.0 program (SPSS, Inc, Chicago,

IL). For calculation of the sample size,
the objective was to reduce the percent-
a
␹2 test; b Student t test; c At least 80% of weekly capillary glucose measurements within American Diabetes Association
guidelines.
de Barros. Resistance exercise and gestational diabetes mellitus. Am J Obstet Gynecol 2010.

MONTH 2010 American Journal of Obstetrics & Gynecology 1.e3

Research Obstetrics
FIGURE 2
Mean capillary glucose profile and SEs at different times of day
Mean capillary glucose profile and SEs obtained at different times of day for the exercise and control
groups throughout the follow-up period.
P, analysis of variance for repeated measures with Huynh-Feldt correction.
de Barros. Resistance exercise and gestational diabetes mellitus. Am J Obstet Gynecol 2010.
sults were submitted to analysis of vari-
ance for repeated measures with Huynh-
Feldt correction.18 Mean values of each
patient per time of the day (after an over-
night fast, 2 h after breakfast, 2 h after
lunch, and 2 h after dinner) were used for
analysis. Gestational body mass index
(BMI) was also compared between
groups and periods of gestation using
analysis of variance for repeated mea-
sures with Huynh-Feldt correction.18
Bonferroni’s multiple comparisons were
performed to determine between which
periods glucose levels differed.18
The unpaired Student t test was used
to compare the mean percentage of time
(weeks) spent within the target glucose
range during the follow-up period be-
tween groups.15 The existence of an as-
sociation between groups for the use or
not of insulin was evaluated by the ␹2
test.16 Data were analyzed using the in-
tention-to-treat principle. The level of
significance was set at 5% for all tests.
R ESULTS
The sample consisted of 64 patients who
were equally divided into the 2 groups.
After randomization, 1 patient reported
lack of time to perform RE (EG) and an-
1.e4 American Journal of Obstetrics & Gynecology MONTH 2010
other started to use metformin for glyce-
mic control (CG). None of these patients
were excluded. An intention-to-treat
analysis was chosen to know not just the
efficacy of exercise but also the efficacy of
the program overall. As shown in Table
1, the groups were similar in terms of the
variables measured at the time of inclu-
sion in the study (P ⬎ .05). The fasting
glucose values in the diagnostic test were
95mg/dL or greater in 19 of 32 patients in
the CG (59.3%), against 18 of 32 patients
in the EG (56.2%).
The EG participated in an average of
2.36 ⫾ 0.4 sessions of RE per week. Each
RE session lasted 30-40 minutes. The RE
program outcomes are shown in Table 2.
A significant decrease in the number of
patients who required insulin was ob-
served in EG compared with CG. Glyce-
mic control was significantly better in
EG compared with CG. In EG, the per-
centage of weeks spent within the target
glucose range (80% of weekly capillary
glucose measurements within preestab-
lished guideline values11) was signifi-
cantly higher when compared with CG.
No significant difference in mean (⫾
SD) glucose levels was observed between
EG and CG. In addition, mean glucose
www.AJOG.org
levels measured by the patients at differ-
ent times of the day throughout the fol-
low-up period were lower in EG com-
pared with CG, but the difference was
not significant (Figure 2).
The 2 groups did not differ signifi-
cantly in terms of the amount of insulin
(international units per kilogram) re-
quired by the patients or in the time in-
terval (weeks) between inclusion in the
study and insulin use (Table 2).
EG patients who used insulin contin-
ued to present adequate glycemic con-
trol according to the target established
for a longer percent period of weeks than
control patients who used insulin (EG ⫽
0.40 ⫾ 0.24 vs CG ⫽ 0.25 ⫾ 0.23), but
the difference was not significant (P ⫽
.173). No significant difference in mean
glucose levels was observed between pa-
tients of the 2 groups who used insulin
(CG: 106.83 ⫾ 7.45 vs EG: 109.83 ⫾ 9.04
mg/dL; P ⫽ .342). There were no cases of
postexercise hypoglycemia or capillary
glycemia higher than 250 mg/dL.
The 2 groups were similar in terms of
the variables measured at the time of de-
livery (P ⬎ .05) (Table 3). No difference
in the frequency of cesarean section was
observed between groups (n ⫽ 21 of
32 in EG vs n ⫽ 24 of 32 in CG; P ⫽ .412).
Newborn birthweight greater than
4000 g was observed in 1 EG case and 3
CG cases. There were 3 cases of preterm
delivery in each group (gestational age at
birth ranging from 35 to 36 weeks).
C OMMENT
The present study permits us to conclude
that RE with an elastic band is an effec-
tive therapeutic alternative for patients
with GDM, reducing the number of pa-
tients who will require insulin and im-
proving glycemic control in this popula-
tion. In the present investigation, a
significant decrease in the number of pa-
tients who required insulin was observed
in the group submitted to RE.
In contrast, Brankston et al9 did not
find a significant difference in the num-
ber of women using insulin between the
group undergoing RE and the control
group (43.8% vs 56.3%; P ⫽ .48). These
distinct results might be explained by
differences in the design of the present
www.AJOG.org
study and that of Brankston et al.9 The
sample size was practically double in the
present investigation and different in-
clusion criteria were used in the 2 studies
(different gestational ages at inclusion,
exclusion of patients older than 40 years
in the study of Brankston et al9).
With respect to the characteristics of
the RE program proposed in the 2 stud-
ies, there were differences in the choice
of exercises to be performed. In addition,
in the present study, the RE program
was evaluated weekly by the researcher
(M.C.D.B.), who monitored the ade-
quacy of the exercise intensity, with ad-
justment of the length of the elastic band
whenever necessary, and the correct ex-
ecution of the exercises proposed.
Despite the lack of observation of a re-
duction in the number of women requir-
ing insulin in the group submitted to RE,
Brankston et al9 found a significant de-
crease in the time interval between inclu-
sion in the study and the onset of insulin
treatment (control group: 1.11 ⫾ 0.8
weeks; RE group: 3.71 ⫾ 3.1 weeks; P ⬍
.05) as well as in the mean (⫾ SD)
amount of insulin used at the end of the
study by patients of the RE group (con-
trol group 0.48 ⫾ 0.3 IU/kg; RE group:
0.22 ⫾ 0.2 IU/kg; P ⬍ .05). In the present
study, no differences in the time interval
between inclusion in the study and insu-
lin use (P ⫽ .715), amount of insulin re-
quired (P ⫽ .401), capillary mean glu-
cose levels (P ⫽ .342), or percentage of
weeks spent within the proposed target
glucose range were observed between EG
and CG patients who used insulin (P ⫽
.173).
In the present study, the percentage of
days spent within the target glucose
range was significantly higher in EG
compared with CG (Table 3). Although
not statistically significant, the mean
blood glucose levels measured by the pa-
tients at different times of the day
throughout the follow-up period were
lower in the EG compared with CG (Fig-
ure 2). The mean glucose levels were also
lower in the EG compared with CG, but
statistically significant (Table 2). Per-
haps with a larger number of patients, it
was possible to observe a statistically sig-
nificant difference between groups re-
garding capillary glucose levels.
Obstetrics Research
TABLE 3
Maternal and newborn characteristics at birth
Control group Exercise group
Variable (n ⴝ 32) (n ⴝ 32) Pa
Delivery BMI, kg/m2 29.92 ⫾ 4.12 30.38 ⫾ 4.36 .600
..............................................................................................................................................................................................................................................
Pregnancy weight gain, kg 11.28 ⫾ 5.63 12.64 ⫾ 5.29 .324
..............................................................................................................................................................................................................................................
Gestational age at delivery, wks 38.57 ⫾ 1.24 38.61 ⫾ 1.13 .883
..............................................................................................................................................................................................................................................
Newborn weight, kg 3.30 ⫾ 0.49 3.23 ⫾ 0.45 .531
..............................................................................................................................................................................................................................................
Values are the mean ⫾ SD.
BMI, body mass index.
a
Student t test.
de Barros. Resistance exercise and gestational diabetes mellitus. Am J Obstet Gynecol 2010.
With respect to safety, RE does not currence of GDM as observed for aerobic
seem to interfere with pregnancy out- exercise.19-22 f
comes. There were no cases of postexer-
cise hypoglycemia or capillary glycemia
ACKNOWLEDGMENT
higher than 250 mg/dL. In addition, no
We thank the Coordenação de Aperfeiçoa-
significant differences in BMI, preg- mento de Pessoal de Nível Superior for the
nancy weight gain, gestational age at de- master’s fellowship granted.
livery, or number of cesarean sections
were observed between the groups stud-
ied. The level of adherence to the RE pro-
REFERENCES
gram in pregnancies complicated by
1. Proceedings of the 4th International Work-
shop-Conference on Gestational Diabetes Mel-
GDM was satisfactory, with only 1 pa- litus. Chicago, Illinois, USA. March 14-16, 1997.
tient being lost to follow-up in the Diabetes Care 1998;21(Suppl 2):B1-167.
present investigation and in the study of 2. Artal R. Exercise: the alternative therapeutic
Brankston et al.9 The fact that the
intervention for gestational diabetes. Clin Ob-
stet Gynecol 2003;46:479-87.
women can do the exercises at home may 3. Coustan DR, Imarah J. Prophylactic insulin
have contributed to this result. treatment of gestational diabetes reduces the
incidence of macrosomia, operative delivery,
The better glycemic control observed
and birth trauma. Am J Obstet Gynecol
in pregnant women with GDM who
1984;150:836-42.
regularly underwent RE may reduce
4. Coustan DR. Pharmacological management
treatment costs and assist public health of gestational diabetes: an overview. Diabetes
policies. An elastic band (with an ap- Care 2007;30(Suppl 2):S206-8.
5. Leturque A, Burnol AF, Ferre P, Girard J.
proximate cost of US $8) was chosen for
Pregnancy-induced insulin resistance in the rat:
RE to reduce costs and to permit access
assessment by glucose clamp technique. Am J
to this type of physical activity by the Physiol 1984;246:E25-31.
largest possible number of patients. 6. Pate RR, Pratt M, Blair SN, et al. A recom-
The present study provided evidence mendation from the Centers for Disease Control
and Prevention and the American College of
that RE may contribute to the care of pa-
Sports Medicine. JAMA 1995;273:402-7.
tients with GDM. Further well-designed 7. Artal R, O’Toole M. Guidelines of the Ameri-
and controlled studies investigating the can College of Obstetricians and Gynecologists
effects of RE on metabolic control in for exercise during pregnancy and the postpar-
GDM should compare RE with aerobic tum period. Br J Sports Med 2003;37:6-12.
8. Fleck SJ, Kraemer WJ. Designing resistance
exercise and should determine whether
exercise programs, 3rd ed. Champaign; IL: Hu-
the adoption of both types of exercise man Kinetics; 2003.
may confer additional benefits. In addi- 9. Brankston GN, Mitchell BF, Ryan EA, Okun
tion, the combined action of RE and NB. Resistance exercise decreases the need
metformin should be analyzed as well as for insulin in overweight woman with gestational
diabetes mellitus. Am J Obstet Gynecol 2004;
whether the molecular mechanisms un- 190:188-93.
derlying the disease are modified by RE 10. Metzger BE, Buchanan TA, Coustan DR, et
and whether RE is able to prevent the oc- al. Summary and recommendations of the Fifth
MONTH 2010 American Journal of Obstetrics & Gynecology 1.e5
Research Obstetrics
International Workshop-Conference on Ges-
tational Diabetes Mellitus. Diabetes Care
2007;30(Suppl 2):S251-60.
11. Matsudo S, Araújo T, Matsudo V, Andrade
D, et al. Questionário internacional de atividade
física (IPAQ): estudo de validade e reprodutibi-
lidade no Brasil. RBAFS 2001;6:5-18.
12. Wichmann BA, Hill ID. Algorithm AS 183. An
efficient and portable pseudo-random number
generator. Appl Stat 1982;31:188-90.
13. Robertson RJ, Goss FL, Rutkowski J, et al.
Concurrent validation of the OMNI perceived
exertion scale for resistance exercise. Med Sci
Sports Exerc 2003;35:333-41.
14. Hadlock FP, Harrist RB, Martinez-Poyer J.
In utero analysis of fetal growth: a sonographic
weight standard. Radiology 1991;181:129-33.
1.e6 American Journal of Obstetrics & Gynecology MONTH 2010
15. Bussab WO, Morettin PA. Estatística
Básica, 4th ed. São Paulo, Brazil: Atual; 1987.
16. Kirkwood BR, Sterne JAC. Essential medi-
cal statistics, 2nd ed. Boston, MA: Blackwell
Science; 2006.
17. Singer JM, Andrade DF. Analysis of longitu-
dinal data. In: Sen PK, Rao CR, eds. Handbook
of statistics. Volume 18: Bio-environmental and
public health statistics. Amsterdam: North Hol-
land; 2000.
18. Neter J, Kutner MH, Nachtsheim CJ, Was-
serman W. Applied linear statistical models, 4th
ed. Homewood, IL: Richard D. Irwing; 1996.
19. Dempsey JC, Sorensen TK, Williams MA,
et al. Prospective study of gestational diabe-
tes mellitus risk in relation to maternal recre-
ational physical activity before and during
www.AJOG.org
pregnancy. Am J Epidemiol 2004;159:
663-70.
20. Zhang C, Solomon CG, Manson JE, Hu FB.
A prospective study of pregravid physical activ-
ity and sedentary behaviors in relation to the risk
for gestational diabetes mellitus. Arch Intern
Med 2006;166:543-8.
21. Rudra CB, Williams MA, Lee IM, Miller RS,
Sorensen TK. Perceived exertion in physical ac-
tivity and risk of gestational diabetes mellitus.
Epidemiology 2006;17:31-7.
22. Oken E, Ning Y, Rifas-Shiman SL, Radesky
JS, Rich-Edwards JW, Gillman MW. Associa-
tions of physical activity and inactivity before
and during pregnancy with glucose tolerance.
Obstet Gynecol 2006;108:1200-7.