nutrients

Article
Association between Gestational Weight Gain,
Gestational Diabetes Risk, and Obstetric Outcomes:
A Randomized Controlled Trial Post Hoc Analysis
David Simmons 1,2, * , Roland Devlieger 3,4,5 , Andre van Assche 3 , Sander Galjaard 3,4,6 ,
Rosa Corcoy 7,8 , Juan M. Adelantado 8 , Fidelma Dunne 9 , Gernot Desoye 10 ,
Alexandra Kautzky-Willer 1 , Peter Damm 11 , Elisabeth R. Mathiesen 11 , Dorte M. Jensen 12,13,14 ,
Lise Lotte T. Andersen 14 , Annunziata Lapolla 15 , Maria G. Dalfra 15 , Alessandra Bertolotto 16 ,
Ewa Wender-Ozegowska 17 , Agnieszka Zawiejska 17 , David Hill 18 , Frank J. Snoek 19 and
Mireille N. M. van Poppel 19,20
1 Institute of Metabolic Science, Addenbrooke’s Hospital, CB2 0QQ Cambridge, UK;
[email protected]
2 Macarthur Clinical School, Western Sydney University, Locked Bag 1797, Campbelltown, Sydney,
NSW 2760, Australia
3 Department of Development and Regeneration, KU Leuven, 3000 Leuven, Belgium;
[email protected] (R.D.); [email protected] (A.v.A.);
[email protected] (S.G.)
4 Department of Obstetrics and Gynecology, University Hospitals Leuven, 3000 Leuven, Belgium
5 Department of Obstetrics, Gynecology and Fertility, GasthuisZusters Antwerpen Sint-Augustinus,
2610 Wilrijk, Belgium
6 Department of Obstetrics and Gynaecology, Division of Obstetrics and Prenatal Medicine, Erasmus MC,
University Medical Centre, 3015 GD Rotterdam, The Netherlands
7 CIBER Bioengineering, Biomaterials and Nanomedicine, Instituto de Salud Carlos III, Zaragoza, 50018 Spain;
[email protected]
8 Institut de Recerca de l’Hospital de la Santa Creu i Sant Pau, 08025 Barcelona, Spain;
[email protected]
9 Galway Diabetes Research Centre (GDRC) and National University of Ireland, H91 CF50 Galway, Ireland;
[email protected]
10 Department of Obstetrics and Gynecology, Medizinische Universitaet Graz, A-8036 Graz, Austria;
[email protected]
11 Center for Pregnant Women with Diabetes, Departments of Endocrinology and Obstetrics, Rigshospitalet
and The Clinical Institute of Medicine, Faculty of Health and Medical Sciences, University of Copenhagen,
DK-1165 Copenhagen, Denmark; [email protected] (P.D.);
[email protected] (E.R.M.)
12 Department of Endocrinology, Odense University Hospital, DK-5000 Odense, Denmark;
[email protected]
13 Department of Gynaecology and Obstetrics, Odense University Hospital, DK-5000 Odense, Denmark
14 Department of Clinical Research, Faculty of Health Sciences, University of Southern Denmark,
DK-5230 Odense, Denmark; [email protected]
15 Department of Medical and Surgical Sciences. Università degli Studi di Padova, 35100 Padua, Italy;
[email protected] (A.L.); [email protected] (M.G.D.)
16 Department of Clinical and Experimental Medicine, University of Pisa, 56126 Pisa, Italy;
[email protected] (A.B.)
17 Department of Reproduction, Poznan University of Medical Sciences, 61-701 Poznan, Poland;
[email protected] (E.W.-O.); [email protected] (A.Z.)
18 Recherche en Santé Lawson SA, 9552 St. Gallen, Switzerland; [email protected]
19 Department of Public and Occupational Health, Amsterdam Public Health Research Institute,
Amsterdam UMC, Vrije Universiteit Amsterdam, Van der Boechorststraat 7, NL-1081 BT Amsterdam,
The Netherlands.; [email protected] (F.J.S.); [email protected] (M.N.M.v.P.)
20 Institute of Sports Sciences, University of Graz, A-8010 Graz, Austria
* Correspondence: [email protected]; Tel.: +61-2-4620-3899; Fax: +61-2-4620-3890

Nutrients 2018, 10, 1568; doi:10.3390/nu10111568 www.mdpi.com/journal/nutrients

Nutrients 2018, 10, 1568 2 of 13




Received: 27 September 2018; Accepted: 19 October 2018; Published: 23 October 2018


Abstract: Excess gestational weight gain (GWG) is associated with the development of gestational
diabetes mellitus (GDM). Lifestyle trials have not achieved much GWG limitation, and have
largely failed to prevent GDM. We compared the effect of substantial GWG limitation on maternal
GDM risk. Pregnant women with a body mass index (BMI) ≥29 kg/m2 <20 weeks gestation without
GDM (n = 436) were randomized, in a multicenter trial, to usual care (UC), healthy eating (HE),
physical activity (PA), or HE and PA lifestyle interventions. GWG over the median was associated
with higher homeostasis model assessment insulin resistance (HOMA-IR) and insulin secretion
(Stumvoll phases 1 and 2), a higher fasting plasma glucose (FPG) at 24–28 weeks (4.66 ± 0.43 vs.
4.61 ± 0.40 mmol/L, p < 0.01), and a higher rate of caesarean section (38% vs. 27% p < 0.05). The GWG
over the median at 35–37 weeks was associated with a higher rate of macrosomia (25% vs. 16%,
p < 0.05). A post hoc comparison among women from the five sites with a GWG difference >3 kg
showed no significance difference in glycaemia or insulin resistance between HE and PA, and UC.
We conclude that preventing even substantial increases in GWG after the first trimester has little
effect on maternal glycaemia. We recommend randomized controlled trials of effective lifestyle
interventions, starting in or before the first trimester.

Keywords: gestational diabetes mellitus; pregnancy; lifestyle intervention; randomised

controlled trial; healthy eating; physical activity; overweight; motivational interviewing; prevention

1. Introduction
The risk of gestational diabetes mellitus (GDM) increases with both obesity and gestational
weight gain (GWG) [1] GDM, excess GWG, and overweight/obesity are all independently associated
with an increased risk of macrosomia, operative delivery, and other adverse perinatal outcomes,
including shoulder dystocia [1]. If increased GWG is causally related to an increase in GDM incidence,
then limiting GWG should reduce the incidence of GDM. However, the results from randomized
controlled trials (RCTs) for the prevention of GDM through lifestyle have been mixed [2]. The mean
difference in GWG between the intervention and control groups has ranged between −9.07 kg and
+3.5 kg in a meta-analysis in 2012, with a mean of −2.21 kg and no significant reduction in GDM
rates [3]. A recent meta-analysis of the individual participant data from randomized trials estimated a
GWG reduction of 0.7 kg overall, with no overall reduction in GDM (unless data from studies without
individual data were included) [4]. A GDM reduction was found with interventions involving physical
activity alone (PA) [4]. The most recent Cochrane review (2017) showed no reduction in GDM or in
adverse obstetric outcomes with lifestyle interventions [5]. However, another recent meta-analysis
showed a reduction in GDM if intervention commenced in the first, but not the second trimester [6].
Moreover, prevention occurred in some lifestyle intervention studies with minimal GWG difference
(e.g., The Finnish Gestational Diabetes Prevention Study (RADIEL), where women with previous
GDM and/or obesity were studied) [7]. More information is therefore needed on the relationship
between GWG, the development of GDM, and obstetric outcomes.
The Vitamin D and Lifestyle Intervention for GDM Prevention (DALI) [8–10] is a European
multicenter RCT that tested different approaches for the reduction of GDM risk. The study was
unique in that it had the following two limbs: (1) the DALI Lifestyle Study, which compared healthy
eating (HE), PA, and mixed (HE and PA) with a control group, and (2) the DALI Vitamin D study
that compared vitamin D supplementation with and without an HE and PA intervention. The main
lifestyle RCT (the DALI Lifestyle Study) found the HE and PA intervention was associated with the
least GWG, but there was no significant reduction in the fasting glucose, GDM incidence, or insulin
sensitivity (as measured with the homeostasis model assessment insulin resistance (HOMA-IR) [10].
Nutrients 2018, 10, 1568 3 of 13

We hypothesized that the greater the GWG limitation, the greater the difference in GDM rates,
fasting glucose, HOMA, and adverse obstetric outcomes between controls and intervention subjects.
We tested this by initially comparing women above and below the median GWG, independent of
intervention. As randomization in the DALI Lifestyle Study was stratified by site, we have gone on to
treat each site within this study as a separate RCT, and tested whether the GDM risk was reduced in the
five sites with the greatest GWG limitation, using the HE and PA intervention—the only intervention
able to achieve GWG limitation.

2. Materials and Methods

2.1. Overall Study Design

The methods used in this study have been described previously [8–10]. Briefly, the DALI
Lifestyle Study is an RCT Trial registration: ISRCTN70595832 that compared three different lifestyle
approaches that could prevent GDM across the following ten European centers: Cambridge,
(United Kingdon: coordinating center), Amsterdam (Netherlands), Leuven (Belgium), Barcelona
(Spain), Galway (Ireland), Pisa/Padova (two sites; Italy), Vienna (Austria), Poznan (Poland),
Copenhagen, and Odense (Denmark). The primary outcomes were changes in the GWG,
fasting glucose, and HOMA-IR. The target GWG was 5 kg, which is at the lower end of the Institute of
Medicine’s recommendations [11].
The women were randomized, and were stratified by site to usual care (UC), HE, PA, or HE and PA.
The intervention has been described in detail previously [8,10]. After randomization, the women were
assigned a lifestyle coach who provided five face-to-face, and up to four telephone coaching sessions,
based on the principles of motivational interviewing. The coaches received standardized training
and an intervention toolkit (including, e.g., a pedometer in the PA intervention) tailored to their
culture/language. The coaching involved the discussion of seven HE and/or five PA “messages”,
and a GWG <5 kg was targeted [10]. The HE intervention promoted a food-based, lower simple
and complex carbohydrate, lower fat, higher fiber, and higher protein diet, including a focus on
portion size, and therefore a more limited intake of total calories. The PA intervention promoted
both aerobic and resistance physical activity. At least four face-to-face coaching sessions were
expected to take place before the second measurement session (24–28 weeks), and the intervention
was completed after 35 weeks of gestation. The HE intervention was associated with the self-reported
reduced carbohydrate intake, reduced portion size, increased vegetable intake, and reduced sugary
drinks intake, while the PA intervention was associated with a reduced sedentary time and increased
moderate or vigorous activity [10]. The combined intervention was the only intervention associated
with a significant reduction in GWG.
The participants were pregnant women aged ≥18 years, before 20 weeks gestation, with a
pre-pregnancy body mass index (BMI) ≥29 kg/m2 , and were recruited between January 2012 and
February 2014. All of the women underwent an oral glucose tolerance test (OGTT), and those
with GDM (International Association Diabetes Pregnancy Study Group (IADPSG)/World Health
Organisation (WHO) 2013 criteria of fasting venous plasma glucose ≥5.1 mmol/L and/or one
hour of glucose ≥10 mmol/L and/or two hours of glucose ≥8.5 mmol/L) [12] were excluded
from the study. Other exclusion criteria are reported elsewhere [8]. All of the women gave signed
informed consent. The study was approved by the relevant ethical committees and was registered
as an RCT (ISRCTN70595832). The assessments were made by the research midwife/nurse at four
antenatal time points—before 20 weeks (baseline), between 24–28 weeks (visit 2), and between 35–37
weeks (visit 3) gestation. Maternal gestational weight gain (GWG) was defined as the weight difference
between the self-reported pre-pregnancy and the DALI antenatal measurement. Randomization was
stratified by site, and the randomization method, and interventions have been described in detail
previously [8]. Those involved in taking the measurements were kept blinded to the intervention.
Nutrients 2018, 10, 1568 4 of 13

Information on the demographics, pre-pregnancy weight, smoking, alcohol consumption,

past/current medical and obstetric history, and medication use was gathered using questionnaires.
The data from the medical records were obtained regarding the co-morbidities, obstetric and
perinatal outcomes, and birth weight. The women attended the three assessments while fasting,
and undertook a standardized, sitting, 75 g OGTT, with blood samples taken at 0, 60, and 120 min
after glucose ingestion. The women completed the questionnaire and anthropometric measurements
between the blood tests. The local laboratories were used to rapidly obtain the OGTT results so as to
assess their eligibility for the study, and to support referral for clinical care where needed. The blood
samples were centrifuged and separated from the serum and plasma aliquots (1000 µL or 250 µL),
placed in microrack tubes, and stored at −20 or −80 ◦ C, until further analysis, in the central trial
laboratory in Graz, Austria, which was certified according to ISO 9001 standards.
The laboratory glucose and insulin analytical methods have been described previously [8–10].
The homeostatic model assessment-insulin resistance (HOMA-IR) was calculated as (glucose ×
insulin)/22.5 and the homeostatic model assessment-insulin secretion (HOMA-IS) (B) was calculated
using the following formula of 20 × fasting insulin (µIU/mL)/fasting glucose (mmol/mL) − 3.5 [13].
The Stumvoll first and second phase indices are the surrogates of early and late insulin secretion, and
were calculated as 1194 + 4.724 × Ins0 − 117.0 × Gluc60 + 1.414 × Ins60 for the Stumvoll first phase,
and 295 + 0.349 × Ins60 − 25.72 × Gluc60 + 1.107 × Ins0 for the Stumvoll second phase, as described
earlier by Stumvoll et al. [14,15]
The height was measured at baseline with a stadiometer (SECA 206, SECA, Birmingham, UK;
Leicester Height Measure), and the average value of the two measurements was used. The women
were weighed on calibrated electronic scales (SECA 888; SECA 877, SECA, Birmingham, UK) without
shoes and wearing light clothes, to the nearest 0.1 kg; the average value of the two measurements
was used.

2.2. Statistics
The trial data were entered into a bespoke web-based electronic database using the Microsoft.Net
development environment. The analyses were performed in SPSS22. The trial data were analyzed
according to the intention-to-treat principle. Two-sided p < 0.05 was taken as significant. The discrete
variables were described as crude %, with comparisons adjusted for sites using logistic regression,
and were reported with 95% confidence intervals. The continuous variables were compared using
generalized linear modelling, again adjusting for sites, with the mean ± SEM shown. The comparisons
were also adjusted for the other covariates, as described. The following two analyses were undertaken
in this post hoc study: The first was to compare the characteristics of the women above and below
the median GWG at each gestation. This comparison was undertaken rather than above and below
the lower limit of the Institute of Medicine (IOM) [11] targets so as to maximize the numbers in
both groups, and only a minority of women achieved the 5 kg limitation. The second analysis used
the RCT framework to compare the impact of a substantial GWG limitation on metabolic the and
obstetric outcomes. A substantial GWG limitation was defined as >50% higher than the mean DALI
Lifestyle Trial GWG difference between the UC and intervention groups (i.e., >3 kg difference). This was
only achieved with the HE and PA intervention in five sites, and hence the post hoc comparison was
only undertaken between UC, and HE and PA in these five sites.

3. Results
The gestational age on entry ranged from 8 to 19+6 weeks. The median GWG between
pre-pregnancy and 24–28 weeks was 5.5 (IQR 2.3–9.1) kg, and at 35–37 weeks it was 9.5 (IQR 5.3–14.3) kg.
Table 1 shows the baseline data for those above and below the median GWG at baseline, 24–28 weeks
overall, and at 35–37 weeks, excluding those developing GDM by 24–28 weeks gestation. The women
with the greatest GWG commenced at a lower weight and body mass index (BMI), and were more
likely to be smokers and nulliparous (35–37 weeks group only). Those with the greatest GWG at 24–28
Nutrients 2018, 10, 1568 5 of 13

weeks had a lower fasting glucose, while those with the greatest GWG at 35–37 weeks (excluding those
with GDM by 24–28 weeks) had lower fasting, and 1 h and 2 h glucose concentrations at baseline.
Table
Nutrients 2018,2 10,
shows
x FORthat
PEERthe HOMA-IR, Stumvoll phase 1, and Stumvoll phase 2 were significantly
REVIEW 5 of 12
higher among those with the greatest GWG by 24–28 weeks and 35–37 weeks. HOMA-IS was also
those with the
statistically greatest
higher GWG
in those withat the
24–28 weeksGWG
greatest and 35–37
at 24–28weeks.
weeks.TheThewomen with
fasting the greatest
glucose was higherGWG in
by 35–37
those withweeks (but not
the greatest GWG byat24–28
24–28 weeks)
weeks andhad35–37
significantly
weeks. The larger
womenbabies.
withThere were no
the greatest GWG other
by
significant
35–37 weeks obstetric
(but notdifferences. There were
by 24–28 weeks) no differences
had significantly in the
larger gender
babies. of the
There babies
were between
no other groups
significant
by GWG.differences. There were no differences in the gender of the babies between groups by GWG.
obstetric
Figure 11shows showsthethe
flowflow of participants
of the the participants
in the in
fivethe five
sites withsites with
a high GWGa high GWGthroughout
difference difference
throughout
the post hocthe post hoc
analysis. Theanalysis.
numbersThe thatnumbers that were were
were randomized randomized were Table
comparable. comparable. Tablethe
3 compares 3
compares the maternal characteristics
maternal characteristics and the pregnancy and the pregnancy
outcome measuresoutcome measures
between usualbetween
care (UC),usual
andcare
HE
(UC),
and PA.and HEwomen
The and PA. inThe women
the HE in the
and PA HE and
group PA kg
had 2.6 group
less had
GWG 2.6atkg less GWG
24–28 weeksat 24–28
and weeks
4.3 kg and
at 35–37
4.3 kg at
weeks 35–37 weeks
compared withcompared with the
the UC group. TheUC group.
glucose The glucose
levels, measures levels, measures
of insulin of insulin
resistance andresistance
secretion,
and GDM
secretion,
ratesand wereGDM rates were not
not significantly significantly
different. different. lifestyle
The combined The combined lifestyle
intervention wasintervention
associated
was associated
with a significantly withlower
a significantly lower largeage
large for gestational for (LGA)
gestational
rate. age (LGA) rate.

Figure1.1.Recruitment
Figure Recruitmentflowchart
flowchartfor
for five
five sites
sites achieving the greatest
achieving the greatestgestational
gestationalweight
weightgain
gainlimitation.
limitation.
Nutrients 2018, 10, 1568 6 of 13

Table 1. Maternal baseline characteristics in the Vitamin D and Lifestyle Intervention for Gestational Diabetes Mellitus (GDM) Prevention (DALI) participants,
above and below median gestational weight gain from pre-pregnancy to baseline, to 24–28 weeks, and to 35–37 weeks gestation. The latter excludes women with
GDM at 24–28 weeks.

<1.80 kg at Baseline ≥1.80 kg at Baseline <5.65 kg to 24–28 Weeks ≥5.65 kg to 24–28 Weeks <9.5 kg to 35–37 Weeks † ≥9.5 kg to 35–37 Weeks †
Gestational Weight Gain Group
n = 215 n = 216 n = 203 n = 201 n = 158 n = 159
Age (years) 31.5 ± 5.3 32.4 ± 5.4 32.0 ± 5.2 32.1 ± 5.4 32.7 ± 5.1 31.7±5.4
Pre-pregnancy weight (kg) 94.1 ± 14.0 91.4 ± 12.0 * 94.9 ± 14.0 90.5 ± 11.9 *** 94.7 ± 13.9 90.8 ± 11.4 **
Pre-pregnancy BMI (kg/m2 ) 34.2 ± 4.2 33.3 ± 3.6 ** 34.6 ± 4.2 33.0 ± 3.5 *** 34.4 ± 4.1 33.0 ± 3.5 **
European descent 184/215 (86%) 189/216 (88%) 177/203 (87%) 176/201 (88%) 137/158 (87%) 142/159 (89%)
Nullipara 109/215 (51%) 106/216 (49%) 100/203 (49%) 102/201 (51%) 69/158 (44%) 84/159 (53%) *
Smokers 26/215 (12%) 41/216 (19%) 19/203 (9%) 42/201 (21%) ** 14/158 (9%) 29/159 (18%) *
First degree relative with diabetes 52/215 (24%) 48/216 (22%) 48/2103 (24%) 41/201 (20%) 33/158 (21%) 38/159 (24%)
Previous GDM 7/136 (5%) 10/137 (7%) 8/133 (6%) 5/120 (4%) 4/112 (4%) 6/91(7%)
Fasting glucose (mmol/L) 4.6 ± 0.4 4.6 ± 0.4 4.7 ± 0.4 4.6 ± 0.4 4.7 ± 0.3 4.6 ± 0.4 **
1-h glucose (mmol/L) 6.9 ± 1.4 6.7 ± 1.3 6.9 ± 1.4 6.7 ± 1.4 6.9 ± 1.3 6.5 ± 1.3 **
2-h glucose (mmol/L) 5.9 ± 1.2 5.8 ± 1.0 5.9 ± 1.1 5.8 ± 1.1 5.9 ± 1.0 5.7 ± 1.1 *
HOMA-IR (IQR) 2.5 (2.0, 3.4) 2.8 (2.0, 3.7) 2.7 (2.0, 3.6) 2.6 (2.0, 2.4) 2.7 (2.1, 3.4) 2.5 (1.9, 3.4)
HOMA insulin secretion (IQR) 217 (170, 312) 256 (180, 363) 220 (166, 324) 245 (181, 346) 215 (158, 295) 251 (180, 358)
Stumvoll phase 1 (IQR) 1590 (1221, 2067) 1521 (1255, 2045) 1560 (1219, 2083) 1502 (1258, 2016) 1595 (1224, 2099) 1507 (1261, 2042)
Stumvoll phase 2 (IQR) 409 (318, 532) 388 (327, 520) 408 (319, 533) 386 (326, 519) 410 (319, 534) 388 (327, 519)
For continuous outcomes, the differences between the groups were tested using multilevel regression models (country and individual as levels), and were adjusted for gestational age at
the outcome measurement. For the dichotomous outcomes, logistic regression models were performed, and were adjusted for country and gestational age at the outcome measurement.
* p < 0.05; ** p < 0.01; *** p < 0.001. † Excluding women with GDM at 24–28 weeks, based on local glucose values. BMI—body mass index; HOMA-IR—homeostasis model assessment
insulin resistance. Numbers in bold highlight statistically significant comparisons.
Nutrients 2018, 10, 1568 7 of 13

Table 2. Metabolic status at 24–28 weeks, 35–37 weeks, and birth outcomes in DALI participants, according to gestational weight gain from pre-pregnancy.

<1.80 kg at Baseline ≥1.80 kg at Baseline <5.65 kg to 24–28 Weeks ≥5.65 kg to 24–28 Weeks <9.5 kg to 35–37 Weeks † ≥9.5 kg to 35–37 Weeks †
Gestational Weight Gain Group
n = 215 n = 216 n = 203 n = 201 n = 156 n = 158
24–28 weeks
Weight (kg) 96.5 ± 12.4 100.7 ± 12.8 ** 96.7±12.7 100.4±12.7* 96.7 ± 12.7 100.2 ± 12.6 *
Gestational weight gain (kg) # 2.3 ± 3.9 9.3 ± 4.3 *** 1.7±3.3 9.9±3.8*** 2.0 ± 3.5 9.4 ± 4.3 ***
Fasting blood glucose (BG) (mmol/l) # 4.62 ± 0.40 4.66 ± 0.42 * 4.61±0.40 4.66±0.43** 4.58 ± 0.36 4.57 ± 0.37 *
1-h glucose (mmol/l) # 7.73 ± 1.58 7.81 ± 1.66 7.74±1.59 7.83±1.68 7.52 ± 1.29 7.39 ± 1.49
2-h glucose (mmol/l) # 6.34 ± 1.23 6.21 ± 1.24 6.33±1.26 6.22±1.19 6.18 ± 1.02 5.98 ± 1.13
HOMA-IR # 2.88 (2.17, 3.83) 3.13 (2.39, 4.45) * 2.80 (2.07, 3.76) 3.15 (2.46, 4.47) *** 2.84 (2.17, 3.69) 3.03 (2.26, 4.29) ***
HOMA insulin secretion # 256 (190, 366) 285 (200, 393) 264 (189, 345) 283 (204, 407) * 269 (193, 343) 300 (208, 420)
Stumvoll phase 1 # 1757 (1292, 2308) 1888 (1404, 2347) * 1675 (1276, 2256) 1919 (1403, 2378) *** 1612 (1287, 2255) 1929 (1494, 2366) ***
Stumvoll phase 2 # 454 (337, 588) 484 (365, 598) * 433 (335, 576) 496 (368, 607) *** 418 (336, 576) 495 (380, 604) ***
35–37 weeks
Weight (kg)† 100.3 ± 12.3 104.9 ± 13.0 ** 100.1 ± 12.6 105.2 ± 12.8 ** 99.4 ± 12.8 105.7 ± 12.3 ***
Gestational weight gain (kg) #† 6.2 ± 5.3 13.4 ± 5.7 *** 5.4 ± 4.5 14.5 ± 4.9 *** 4.7 ± 3.7 14.9 ± 4.4 ***
Fasting BG (mmol/l) #$ 4.57 ± 0.45 4.60 ± 0.51 4.55 ± 0.45 4.61 ± 0.51 * 4.49 ± 0.43 4.53 ± 0.43 **
1-h glucose (mmol/l) #$ 8.20 ± 1.61 8.46 ± 1.57 8.21 ± 1.54 8.45 ± 1.67 * 8.01 ± 1.37 8.17 ± 1.47 *
2-h glucose (mmol/l) #$ 6.74 ± 1.28 6.58 ± 1.21 6.82 ± 1.31 6.51 ± 1.14 * 6.63 ± 1.21 6.43 ± 1.09
HOMA-IR #$ 3.11 (2.35, 4.46) 3.38 (2.55, 4.60) 2.96 (2,25, 4.32) 3.45 (2.62, 4.76) ** 2.66 (2.21, 4.04) 3.47 (2.70, 4.74) ***
HOMA insulin secretion #$ 345 (234, 565) 354 (250, 483) 346 (227, 478) 351 (252, 531) 324 (217, 445) 367 (256, 531)
Stumvoll phase 1 #$ 2469 (1722, 3174) 2518 (1898, 3117) 2403 (1723, 3124) 2561 (1917, 3200) ** 2383 (1732, 3026) 2684 (2162, 3354) ***
Stumvoll phase 2 #$ 629 (441, 801) 639 (497, 786) 618 (450, 790) 652 (497, 810) ** 608 (452, 766) 684 (554, 841) ***
Birth N = 198 N = 195 N = 194 N = 194 N = 154 N = 158
Gestation at birth (weeks) 39.5 ± 2.6 39.6 ± 1.7 39.8 ± 1.6 39.5 ± 1.7 39.8 ± 1.4 39.8 ± 1.3
Gender (male) 102/198 (52%) 94/195 (48%) 101/194 (52%) 94/194 (49%) 84/154 (55%) 74/158 (47%)
Birthweight 3490 ± 538 3479 ± 557 3457 ± 541 3505 ± 551 3477 ± 503 3602 ± 515
Birthweight ≥4 (kg) 37/195 (19%) 34/195 (17%) 33/192 (17%) 35/193 (18%) 25/153 (16%) 39/157 (25%) *
Birthweight <2.5 (kg) 8/195 (4%) 8/195 (4%) 8/192 (4%) 8/193 (4%) 5/153 (3%) 2/157 (1%)
Large for Gestational Age 26/187 (14%) 25/186 (13%) 23/186 (12%) 26/182 (14%) 13/150 (9%) 31/150 (21%) **
Small for Gestational Age 12/186 (7%) 16/186 (9%) 15/186 (8%) 13/182 (7%) 10/150 (7%) 9/150 (6%)
Preterm birth 8/194 (4%) 14/195 (7%) 8/192 (4%) 14/192 (7%) 2/153 (1%) 4/157 (3%)
Induction of labor or planned caesarean section 70/188 (37%) 82/187 (44%) 76/186 (41%) 75/184 (41%) 61/151 40% 60/148 41%
Caesarean section 51/190 (27%) 73/190 (38%) * 58/188 (31%) 65/187 (35%) 45/153 (29%) 53/151 (35%)
Pre-eclampsia 4/181 (2%) 10/187 (5%) 7/181 (4%) 8/183 (4%) 4/147 (3%) 5/148 (3%)
Neonatal Intensive Care Unit admission 19/170 (11%) 18/185 (10%) 19/173 (11%) 19/179 (11%) 14/145 (10%) 14/144 (10%)
GDM total 71/183 (39%) 61/175 (35%) 68/185 (37%) 63/175 (36%) 40/153 (26%) 45/155 (29%)
For the continuous outcomes, the differences between the groups were tested with multilevel regression models (country and individual as levels), and adjusted for the gestational age at
the outcome measurement. For the dichotomous outcomes, logistic regression models were performed, and were adjusted for country and gestational age at outcome measurement.
The data are adjusted for the pre-pregnancy BMI and fasting glucose at baseline. * p < 0.05; ** p < 0.01; *** p < 0.001. #—regression models were additionally adjusted either for value at
baseline or for pre-pregnancy BMI when GWG was the outcome. $—value of 24–28 weeks carried forward to 35–37 weeks when GDM was diagnosed at 24–28 weeks. †—excluding
women with GDM at 24–28 weeks, based on local glucose values. Numbers in bold highlight statistically significant comparisons.
Nutrients 2018, 10, 1568 8 of 13

Table 3. Maternal and neonatal characteristics of women at the five DALI sites with maternal gestational
weight gain >3 kg difference at 35–37 weeks between healthy eating (HE), and physical activity
intervention (HE and PA) and usual care (UC).

UC HE + PA
Baseline
n = 50 n = 53
Age (years) 32.1 ± 6.0 31.9 ± 5.0
Pre-pregnancy weight (kg) 94.7 ± 11.4 94.7 ± 13.6
Pre- pregnancy BMI (kg/m2 ) 33.6 ± 3.3 34.8 ± 4.1
Fasting (F) BG (mmol/L) 4.69 ± 0.33 4.64 ± 0.36
1-h BG (mmol/L) 6.80 ± 1.36 6.84 ± 1.23
2-h BG (mmol/L) 5.69 ± 1.25 5.72 ± 0.99
HOMA-IR 2.62 (1.94, 4.30) 2.49 (2.19, 2.82)
HOMA insulin secretion 229 (142, 365) 215 (168, 313)
Stumvoll phase 1 1498 (1211, 2144) 1489 (1236, 1948)
Stumvoll phase 2 379 (313, 554) 382 (322, 501)
European descent 45/50 90.0% 47/53 88.7%
Nullipara 27/50 46.2% 27/53 49.1%
Smokers 6/50 12.0% 5/53 9.4%
First degree relative with diabetes 13/50 26.0% 6/53 11.3%
24–28 weeks n = 50 n = 53
Weight gain from pre-pregnancy 6.7 ± 5.9 (n = 49) 4.1 ± 5.0 (n = 46) *
Fasting Blood Glucose (BG) 4.59 ± 0.40 4.57 ± 0.43
1-h BG 7.86 ± 1.64 7.93 ± 1.60
2-h BG 6.23 ± 1.31 6.09 ± 1.20
HOMA-Insulin Resistance 2.86 (2.18, 3.82) 2.54 (2.16, 3.02)
HOMA-Insulin secretion 284 (203, 394) 249 (198, 352)
Stumvoll phase 1 1762 (1386, 2371) 1943 (1311, 2327)
Stumvoll phase 2 455 (359, 606) 501 (346, 594)
GDM at 24–28 weeks 9/49 (18.4%) 9/45 (20.0%)
35–37 weeks N = 46 N = 42
Weight gain from pre-pregnancy † 11.3 ± 6.7 (n = 40) 7.0 ± 6.0 (n = 36) *
FBG & 4.53 ± 0.46 4.49 ± 0.51
1-h BG & 8.57 ± 1.31 8.32 ± 1.44
2-h BG & 6.70 ± 1.23 6.59 ± 1.01
HOMA-Insulin Resistance & 2.89 (2.07, 4.44) 2.56 (2.28, 3.84)
HOMA insulin secretion & 309 (233, 504) 337 (242, 411)
Stumvoll phase 1 & 2644 (1793, 3179) 2577 (2001, 3189)
Stumvoll phase 2 & 674 (457, 802) 654 (512, 811)
GDM at 35–37 weeks 9/39(23.1%) 5/36 (13.9%)
Birth outcomes N = 45 N = 47
Gestational age at birth 39.8 ± 1.5 39.8 ±1.2
LGA 7/42 16.7% 2/44 4.5% *
SGA 1/42 2.4% 4/44 9.1%
Birthweight 3588 ± 524 3455 ± 463
Preterm birth 2/44 4.5% 0/45 0%
Caesarean section 14/43 30.2% 14/47 29.8%
Pre-eclampsia 4/42 9.5% 2/43 4.7%
Birthweight ≥4 kg 9/45 20.0% 7/45 15.6%
Birthweight <2.5 kg 1/45 2.2% 1/45 2.2%
NICU admission 7/41 17.1% 6/44 13.6%
GDM total 15/46 32.6% 13/44 29.5%
Numbers in bold highlight statistically significant comparisons. The HE and PA comparison alone is reported, as the
intervention group used for selecting the sites, and the intervention with a significant effect on GWG limitation.
There was only one woman with previous GDM in the HE and PA group. All: mean ± SD or n (%). The differences
between the groups at baseline were tested using T-test or Mann–Whitney U-test, as appropriate. For the continuous
outcomes, the differences between groups at 24–28 and 35–37 weeks were tested in mixed models, adjusted for
baseline values, and for the outcome weight gain adjusted for pre-pregnancy BMI. For the dichotomous outcomes,
a logistic regression was performed, and was adjusted for country. * p < 0.05 ** p < 0.01 vs. UC. &—values of 24–28
weeks (T3) carried forward to 35–37 weeks (T4) when GDM was present at T3. &–excluding women with GDM
at T3.
Nutrients 2018, 10, 1568 9 of 13

4. Discussion
We have shown that GWG above the median is indeed associated with increased fasting glucose,
insulin resistance, and insulin secretion. However, there was no increased risk of GDM. In a post
hoc analysis in the five sites with an average 4.3 kg GWG reduction in the HE and PA intervention
by 35–37 weeks, there were similar rates of GDM and no difference in the glycaemia, HOMA-IR,
or measures of insulin secretion. A GWG above the median in the first and third trimesters was also
associated with more caesarean sections and greater macrosomia/LGA, respectively. The HE and PA
intervention with its associated GWG limitation showed a significant reduction in LGA—something
not shown in previous meta-analyses [4,5] with their lower GWG limitation.
No other large study has achieved this degree of GWG limitation [3]. We believe that this post hoc
study clearly demonstrates that GDM cannot be prevented in overweight/obese women with lifestyle
intervention initiated in the second trimester, even with substantial GWG limitation. Previous studies
have failed to conclusively demonstrate this lack of effect, because of the limited effect of their lifestyle
interventions on GWG [4]. There are two key questions that arise, as follows: firstly, why is it that
there was no effect on the risk of GDM, and, secondly, what other strategies might be effective?
There are several theories behind why lifestyle change, even when resulting in a significant
GWG limitation, is insufficient to prevent GDM in overweight/obese women. The first is that there is
simply insufficient time between commencing the intervention and the OGTT at 24 weeks of gestation
for the intervention to be effective. Certainly, some subjects in the DALI study only experienced 4–8
weeks of the intervention. A further possibility is that the reduction in insulin resistance achievable
with lifestyle is insufficient. While the mean insulin resistance was reduced by 10–15%, this was not
statistically significant and was unlikely to be physiologically meaningful, given the fact that the
overall increase in insulin resistance during pregnancy is two- to three-fold [16]. A parallel situation
occurs with the use of metformin for the prevention of GDM. While metformin prevents type 2 diabetes
over time [17], no reduction in the development of GDM was seen in either the Effect of Metformin
on Maternal and Fetal Outcomes in Obese Pregnant Women (EMPOWaR) or in the Metformin in
Obese Non-Diabetic Pregnant Women (MOP) RCTs of metformin for the prevention of GDM [18,19],
although MOP did show a reduction in insulin resistance [20]. However, one would expect that the
intervention would have been effective in reducing GDM in a proportion of women, but this was
not seen.
The relationship between insulin resistance and insulin secretion follows a clear hyperbola [21],
and as long as the insulin secretory capacity is sufficient, the glucose homeostasis would be predicted
to remain steady, and prevent the development of GDM. An increase in insulin resistance was shown
in our comparison of women with GWG above and below the median, and insulin secretion was
also greater, suggesting a compensatory increase. In women with previous GDM, the development
of type 2 diabetes outside of pregnancy was prevented with troglitazone, a thiazolidinedione that
also reduces insulin resistance [22]. The proposed mechanism was that the reduced insulin resistance
reduced the insulin secretion, thereby preventing/delaying the onset of beta cell exhaustion, and the
drop off from the set insulin resistance–insulin secretion hyperbola. While an estimated increase in
insulin sensitivity of 88% was seen within three months [23], prevention occurred over a median of
30 months, much longer than the time available during pregnancy, and without the need to adapt
to rapidly changing insulin requirements. One possibility is that the trajectory of insulin secretion
by the beta cell is already established by the second trimester (i.e., some form of “programming” has
occurred in the first trimester) [24].
The possibility of maternal metabolic trajectory setting in the first trimester as an explanation
for the failure to prevent GDM in DALI is supported by the recent meta-analysis by Song [6],
and another recent study introducing physical activity in the first trimester, which resulted in a
substantial reduction in GDM [25]. The possibility of trajectory setting is also supported by the study
by Hedderson et al. [26], where women with GWG in the highest tertile in the first trimester had an
Nutrients 2018, 10, 1568 10 of 13

increased risk of GDM, but a similar relationship between GWG and GDM risk was not found in the
second trimester. However, the mechanism for such a trajectory setting is unclear.
If the trajectory is set in the first trimester, then there are two possible approaches to prevent
GDM through standard (i.e., through some non-tested strategy) lifestyle changes among obese women.
The first is for interventions to commence early in the first trimester, as suggested by Song [6]. However,
for many women, the first contact with the health service is near the end of (or after) the first trimester,
and an intervention may be challenging to commence. Nevertheless, this is certainly an aspect that
warrants an RCT commencing between, for example, 4–8 weeks’ gestation. There is no evidence that
such an RCT would increase spontaneous miscarriage. This may either involve a community wide
intervention or personalized strategies, and may be before the first pregnancy or between pregnancies.
One possible recommendation would be for obese women to plan their pregnancies in a comparable
manner to women with pre-existing diabetes.
We have also shown that a substantial GWG limitation (mean 4.3 kg vs. usual care) is associated
with reduced rates of LGA. This was shown in both the comparison of GWG above and below the
median at 35–37 weeks, and with the RCT of usual care vs. the combined HE and PA intervention.
Few studies have shown such an effect in pregnancy. The LIMIT RCT found a reduction after a post hoc
analysis [27], but overall no beneficial effect has been shown [5]. The post hoc analysis undertaken here
suggests that lifestyle interventions can reduce LGA rates, but that the degree of GWG reduction needs
to be sufficient. There was a non-significantly higher rate of small for gestational age (SGA), and larger
RCTs are needed to assess whether this degree of GWG limitation can restrict growth excessively.
This study has a number of strengths. It is one of the larger RCTs on lifestyle to prevent GDM,
and it compared three different interventions. DALI was across nine European countries, with the
participants encompassing a range of lifestyles and cultures, making it more widely representative
than studies within a single site or country. The intervention was clearly effective, and being within
site randomization allowed for the post hoc analysis. On the other hand, there are some shortcomings.
In retrospect, the number of patients successfully recruited was sub-optimal to answer definitively
important questions regarding a reduction in fasting glucose with lifestyle change and its relationship to
the development of GDM. Nevertheless, the sample size provided 82% power to detect a 0.2 mmol/L
difference in the fasting glucose, representing a clinically meaningful difference in the context of
pregnancy. This is a post hoc analysis, with all of the statistical limitations that result from such an
analysis. On the other hand, achieving GWG reduction goals is remarkably difficult, and to only select
sites with an achieved GWG above the median is a novel way to test this within an RCT framework.
We have not compared above and below the IOM lower target (5 kg), as only a limited number
of women achieved this goal. Women coming into an RCT are more motivated than most women,
and some women in the control arm appeared to be motivated to improve their lifestyle as a result of
participation, thus reducing the differences between the intervention and control group in some sites.
A further issue is the exclusion of women with GDM at baseline. DALI is one of the few to include this
design feature, and while this allows a test of the impact of GWG limitation on the incidence of GDM,
those with hyperglycemia at baseline may have had greater benefit from the intervention [28,29].

5. Conclusions
In conclusion, we have shown in a lifestyle intervention study RCT cohort that GWG is
associated with an increase in insulin resistance and glycaemia and some worse obstetric outcomes,
however, a post hoc analysis of RCT sites with the greatest GWG limitation showed no reduction
in GDM or its risk, but a reduction in LGA. We conclude that lifestyle intervention in the second
trimester is too late for GDM prevention, and more RCTs are needed in the first trimester,
which includes sub-studies, so as to understand the mechanisms behind a putative “locking in”
of the metabolic trajectory.
Nutrients 2018, 10, 1568 11 of 13

Author Contributions: All of the authors contributed to the conception and/or design of the trial, read and
corrected draft versions of the manuscript, and approved the final manuscript. D.S. wrote the first draft of the
paper and M.N.M. undertook statistical analyses. D.S. is Guarantor.
Funding: The project described has received funding from the European Community’s 7th Framework Programme
(FP7/2007–2013) under grant agreement no. 242187. In the Netherlands, additional funding was provided by
the Netherlands Organization for Health Research and Development (ZonMw) (grant no. 200310013). In Poland,
additional funding was obtained from Polish Ministry of Science (grant no. 2203/7. PR/2011/2). In Denmark,
additional funding was provided by the Odense University Free Research Fund. In the United Kingdom, the DALI
team acknowledge the support received from the NIHR Clinical Research Network: Eastern, especially the local
diabetes clinical and research teams based in Cambridge. In Spain, additional funding was provided by CAIBER
1527-B-226. The funders had no role in any aspect of the study beyond funding.
Acknowledgments: We thank the participants, coaches, research midwives/nurses, and health professionals
collaborating in the recruitment.
Conflicts of Interest: The authors declare no conflicts of interest. All of the authors have completed the ICMJE
uniform disclosure form at www.icmje.org/coi_disclosure.pdf and declare no support from any organization for
the submitted work beyond that below; no financial relationships with any organizations that might have an
interest in the submitted work in the previous three years; no other relationships or activities that could appear to
have influenced the submitted work.

References
1. Simmons, D. Diabetes and obesity in pregnancy. Best Pract. Res. Clin. Obstet. Gynaecol. 2011, 25, 25–36.
[CrossRef] [PubMed]
2. Simmons, D. Prevention of gestational diabetes mellitus: Where are we now? Diabetes Obes. Metab. 2015, 17,
824–834. [CrossRef] [PubMed]
3. Oteng-Ntim, E.; Varma, R.; Croker, H.; Poston, L.; Doyle, P. Lifestyle interventions for overweight and obese
pregnant women to improve pregnancy outcome: Systematic review and meta-analysis. BMC Med. 2012, 10.
[CrossRef] [PubMed]
4. The International Weight Management in Pregnancy (i-WIP) Collaborative Group. Effect of diet and physical
activity based interventions in pregnancy on gestational weight gain and pregnancy outcomes: Meta-analysis
of individual participant data from randomised trials. BMJ 2017, 358. [CrossRef]
5. Shepherd, E.; Gomersall, J.C.; Tieu, J.; Han, S.; Crowther, C.A.; Middleton, P. Combined diet and exercise
interventions for preventing gestational diabetes mellitus. Cochrane Database Syst. Rev. 2017, 11. [CrossRef]
[PubMed]
6. Song, C.; Li, J.; Leng, J.; Ma, R.C.; Yang, X. Lifestyle intervention can reduce the risk of gestational diabetes:
A meta-analysis of randomized controlled trials. Obes. Rev. 2016, 17, 960–969. [CrossRef] [PubMed]
7. Koivusalo, S.B.; Rönö, K.; Klemetti, M.M.; Roine, R.P.; Lindström, J.; Erkkola, M.; Kaaja, R.J.;
Pöyhönen-Alho, M.; Tiitinen, A.; Huvinen, E.; et al. Gestational diabetes mellitus can be prevented by lifestyle
intervention: The Finnish Gestational Diabetes Prevention Study (RADIEL): A randomized controlled trial.
Diabetes Care 2016, 39, 24–30. [CrossRef] [PubMed]
8. Jelsma, J.G.; van Poppel, M.N.; Galjaard, S.; Desoye, G.; Corcoy, R.; Devlieger, R.; van Assche, A.;
Timmerman, D.; Jans, G.; Harreiter, J.; et al. DALI: Vitamin D and lifestyle intervention for gestational
diabetes mellitus (GDM) prevention: An European multicentre, randomised trial–study protocol.
BMC Pregnancy Childbirth 2013, 13. [CrossRef] [PubMed]
9. Simmons, D.; Jelsma, J.G.; Galjaard, S.; Devlieger, R.; van Assche, A.; Jans, G.; Corcoy, R.; Adelantado, J.M.;
Dunne, F.; Desoye, G.; et al. Results from a european multicenter randomized trial of physical activity and/or
healthy eating to reduce the risk of gestational diabetes mellitus: The DALI Lifestyle Pilot. Diabetes Care
2015, 38, 1650–1656. [CrossRef] [PubMed]
10. Simmons, D.; Devlieger, R.; van Assche, A.; Jans, G.; Galjaard, S.; Corcoy, R.; Adelantado, J.M.; Dunne, F.;
Desoye, G.; Harreiter, J.; et al. Effect of physical activity and/or healthy eating on GDM risk: The DALI
Lifestyle Study. J. Clin. Endocrinol. Metab. 2017, 102, 903–913. [CrossRef] [PubMed]
11. Institute of Medicine. Weight gain during pregnancy: Reexamining the guidelines. In Institute of Medicine
(US) and National Research Council (US) and Committee to Reexamine IOM Pregnancy Weight Guidelines;
National Academies Press (US): Washington, DC, USA, 2009.
Nutrients 2018, 10, 1568 12 of 13

12. World Health Organization. Diagnostic Criteria and Classification of Hyperglycaemia First Detected in Pregnancy;
WHO/NMH/MND/13.2; World Health Organization: Geneva, Switzerland, 2013.
13. Matthews, D.R.; Hosker, J.P.; Rudenski, A.S.; Naylor, B.A.; Treacher, D.F.; Turner, R.C. Homeostasis model
assessment: Insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations
in man. Diabetologia 1985, 28, 412–419. [CrossRef] [PubMed]
14. Stumvoll, M.; Mitrakou, A.; Pimenta, W.; Jenssen, T.; Yki-Jarvinen, H.; Van Haeften, T.; Renn, W.; Gerich, J.
Use of oral glucose tolerance test to assess insulin release and insulin sensitivity. Diabetes Care 2000, 23,
295–301. [CrossRef] [PubMed]
15. Stumvoll, M.; Van Haeften, T.; Fritsche, A.; Gerich, J. Oral glucose tolerance test indexes for insulin sensitivity
and secretion based on various availabilities of sampling times. Diabetes Care 2001, 24, 796–797. [CrossRef]
[PubMed]
16. Catalano, P.M.; Tyzbir, E.D.; Roman, N.M.; Amini, S.B.; Sims, E.A.H. Longitudinal changes in insulin release
and insulin resistance in nonobese pregnant women. Am. J. Obstet. Gynecol. 1991, 165, 1667–1672. [CrossRef]
17. Ratner, R.E.; Christophi, C.A.; Metzger, B.E.; Dabelea, D.; Bennett, P.H.; Pi-Sunyer, X.; Fowler, S.; Kahn, S.E.
The Diabetes Prevention Program Research Groupa. Prevention of diabetes in women with a history of
gestational diabetes: Effects of metformin and lifestyle interventions. J. Clin. Endocrinol. Metab. 2008, 93,
4774–4779. [CrossRef] [PubMed]
18. Chiswick, C.; Reynolds, R.M.; Denison, F.; Drake, A.J.; Forbes, S.; Newby, D.E.; Walker, B.R.; Quenby, S.;
Wray, S.; Weeks, A.; et al. Effect of metformin on maternal and fetal outcomes in obese pregnant women
(EMPOWaR): A randomised, double-blind, placebo-controlled trial. Lancet Diabetes Endocrinol. 2015, 3,
778–786. [CrossRef]
19. Syngelaki, A.; Nicolaides, K.H.; Balani, J. Metformin versus placebo in obese pregnant women without
diabetes mellitus. N. Engl. J. Med. 2016, 374, 434–443. [CrossRef] [PubMed]
20. Balani, J.; Hyer, S.; Syngelaki, A.; Akolekar, R.; Nicolaides, K.H.; Johnson, A.; Shehata, H. Association
between insulin resistance and preeclampsia in obese non-diabetic women receiving metformin. Obstet. Med.
2017, 10, 170–173. [CrossRef] [PubMed]
21. Kahn, S.E.; Prigeon, R.L.; McCulloch, D.K.; Boyko, E.J.; Bergman, R.N.; Schwartz, M.W.; Neifing, J.L.;
Ward, W.K.; Beard, J.C.; Palmer, J.P.; et al. Quantification of the relationship between insulin sensitivity and
-cell function in human subjects: Evidence for a hyperbolic function. Diabetes 1993, 42, 1663–1672. [CrossRef]
[PubMed]
22. Buchanan, T.A.; Xiang, A.H.; Peters, R.K.; Kjos, S.L.; Marroquin, A.; Goico, J.; Ochoa, C.; Tan, S.; Berkowitz, K.;
Hodis, H.N.; et al. Preservation of pancreatic beta-cell function and prevention of type 2 diabetes by
pharmacological treatment of insulin resistance in high-risk hispanic women. Diabetes 2002, 51, 2796–2803.
[CrossRef] [PubMed]
23. Berkowitz, K.; Peters, R.; Kjos, S.L.; Goico, J.; Marroquin, A.; Dunn, M.E.; Xiang, A.; Azen, S.; Buchanan, T.A.
Effect of troglitazone on insulin sensitivity and pancreatic -cell function in women at high risk for
non-insulin-dependent diabetes mellitus. Diabetes 1996, 45, 1572–1579. [CrossRef] [PubMed]
24. Banerjee, R.R. Piecing together the puzzle of pancreatic islet adaptation in pregnancy. Ann. N. Y. Acad. Sci.
2018, 1411, 120–139. [CrossRef] [PubMed]
25. Wang, C.; Wei, Y.; Zhang, X.; Zhang, Y.; Xu, Q.; Sun, Y.; Su, S.; Zhang, L.; Liu, C.; Feng, Y.; et al. A randomized
clinical trial of exercise during pregnancy to prevent gestational diabetes mellitus and improve pregnancy
outcome in overweight and obese pregnant women. Am. J. Obstet. Gynecol. 2017, 216, 340–351. [CrossRef]
[PubMed]
26. Hedderson, M.M.; Gunderson, E.P.; Ferrara, A. Gestational weight gain and risk of gestational diabetes
mellitus. Obstet. Gynecol. 2010, 115, 597–604. [CrossRef] [PubMed]
27. Dodd, J.M.; Turnbull, D.; McPhee, A.J.; Deussen, A.R.; Grivell, R.M.; Yelland, L.N.; Crowther, C.A.; Wittert, G.;
Owens, J.A.; Robinson, J.S. Antenatal lifestyle advice for women who are overweight or obese: LIMIT
randomised trial. BMJ 2014, 348, g1285. [CrossRef] [PubMed]
28. Crowther, C.A.; Hiller, J.E.; Moss, J.R.; McPhee, A.J.; Jeffries, W.S.; Robinson, J.S. Effect of treatment of
gestational diabetes mellitus on pregnancy outcomes. N. Engl. J. Med. 2005, 352, 2477–2486. [CrossRef]
[PubMed]
Nutrients 2018, 10, 1568 13 of 13

29. Landon, M.B.; Spong, C.Y.; Thom, E.; Carpenter, M.W.; Ramin, S.M.; Casey, B.; Wapner, R.J.; Varner, M.W.;
Rouse, D.J.; Thorp, J.M., Jr.; et al. A multicenter, randomized trial of treatment for mild gestational diabetes.
N. Engl. J. Med. 2009, 361, 1339–1348. [CrossRef] [PubMed]

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).