DIABETICMedicine DOI: 10.1111/dme.12548
Research: Metabolism Metabolic effects of lifestyle intervention in obese pregnant women. Results from the randomized controlled trial ‘Lifestyle in Pregnancy’ (LiP) C. A. Vinter1, J. S. Jørgensen1, P. Ovesen2, H. Beck-Nielsen3, A. Skytthe4 and D. M. Jensen3 1
Department of Gynecology and Obstetrics, Odense University Hospital, Odense, 2Department of Gynecology and Obstetrics, Aarhus University Hospital, Skejby, Department of Endocrinology, Odense University Hospital, Odense and 4Epidemiology, Institute of Public Health, University of Southern Denmark, Odense, Denmark 3
Accepted 30 June 2014
Abstract Aims The Lifestyle in Pregnancy intervention in obese pregnant women resulted in significantly lower gestational weight gain compared with the control group, but without improvement in rates of clinical pregnancy complications. The impact of the lifestyle intervention on metabolic measurements in the study participants is now reported.
The Lifestyle in Pregnancy study was a randomized controlled trial among 360 obese women (BMI 30–45 kg/m2) who were allocated in early pregnancy to lifestyle interventions with diet counselling and physical activities or to the control group. Fasting blood samples, including plasma glucose, insulin, lipid profile and capillary blood glucose during a 2-h oral glucose tolerance test were carried out three times throughout pregnancy. Insulin resistance was estimated with the homeostasis model assessment of insulin resistance. Methods
Three hundred and four women (84%) were followed until delivery. Women in the intervention group had a significantly lower change in insulin resistance (HOMA-IR) from randomization to 28–30 weeks’ gestation compared with control subjects (mean SD: 0.7 1.3 vs. 1.0 1.3, P = 0.02). Despite a significantly lower gestational weight gain in the intervention group, there was no difference between the groups with respect to total cholesterol, HDL, LDL or triglycerides.
Results
Conclusions Lifestyle intervention in obese pregnant women resulted in attenuation of the physiologic pregnancy-induced insulin resistance. Despite restricted gestational weight gain, there were no changes in glucose or lipid metabolism between the groups.
Diabet. Med. 31, 1323–1330 (2014)
Introduction Obesity in pregnancy has become highly prevalent worldwide and is associated with adverse maternal and neonatal outcomes [1,2]. The underlying pathophysiology linking maternal obesity and adverse outcomes is not fully understood, but seems to involve alterations in glucose and lipid metabolism [3]. Pregnancy is characterized by substantial changes in maternal metabolism that can lead to gestational diabetes mellitus and increase the risk of other adverse outcomes, especially in obese women [4,5]. However, our knowledge is limited with respect to the metabolic changes that occur in obese pregnant women, as well as the metabolic
Correspondence to: Christina Anne Vinter. E-mail: [email protected] (Clinical Trials Registry No; NCT 00530439)
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
impact of lifestyle changes during pregnancy. Normal pregnancy is characterized by a significant increase in insulin resistance [6]. In obese women, the insulin resistance of pregnancy is superimposed on the underlying pre-pregnancy insulin resistance and these changes may lead to high maternal glucose levels, increased risk of gestational diabetes and neonatal macrosomia [7]. Pregnancy is characterized by marked increases in plasma lipid concentrations with increasing gestational age [3,4]. The objective of this clinical trial was to investigate whether lifestyle intervention during pregnancy could improve the metabolic status and subsequently improve pregnancy outcomes in obese women. The current publication is an extension of a previous report where the major focus was the clinical implications concerning obstetric and neonatal outcomes [8].
1323
DIABETICMedicine
What’s new? • This study is among the largest of the randomized controlled trials in obese pregnant women to investigate the metabolic impact of a lifestyle intervention during pregnancy. • Lifestyle intervention in obese pregnant women resulted in attenuation of the physiologic pregnancy-induced insulin resistance. • Despite restricted gestational weight gain in the intervention group, there was no major effect in glucose or lipid metabolism.
Patients and methods This was a randomized controlled trial with a description of the biochemical changes in obese pregnant women during a lifestyle intervention programme. The Lifestyle in Pregnancy (LiP) study ran from October 2007 to October 2010 in two university hospitals in Denmark: Odense and Aarhus University Hospitals, Skejby. The project was approved by the local ethics committee of the Region of Southern Denmark (S-20070058) and registered at clinicaltrials.gov as NCT 00530439. Obese women aged 18–40 years were recruited and included at 10–14 weeks of gestation after referral to the Department of Gynecology and Obstetrics. The inclusion criteria required a BMI of 30–45 kg/m2 (calculated from the pre-pregnancy weight or first measured weight in pregnancy), and women were excluded if they had prior serious obstetric complications, major medical disorders—including pre-gestational diabetes—were non-Danish speaking, abuse of alcohol or a multiple pregnancy. A more detailed description of the study and the procedure for randomization has been published elsewhere [8]. The study was designed as a non-blinded randomized controlled trial with two arms: intervention and control. After receiving written and oral information and giving written consent, the 360 participants were randomized 1:1 by computer-generated numbers in closed, opaque envelopes. A flow chart for participation and randomization is shown in the Supporting Information (Fig. S1). Throughout pregnancy the intervention group received four separate diet counselling sessions and an exercise programme consisting of weekly aerobic classes, free fitness membership during pregnancy and exercise motivating initiatives. More detailed information about the randomization process and details about the intervention are found in the online Supporting Information (Appendix S1). During pregnancy the intervention and control groups were monitored using fasting blood samples, oral glucose tolerance tests, sonar fetal biometry and measurements of maternal weight and blood pressure. Gestational diabetes was diagnosed if the 2-h oral glucose tolerance test capillary
1324
Metabolic effects of lifestyle intervention C. A. Vinter et al.
blood glucose concentration was ≥ 9 mmol/l as according to Danish national recommendations. A baseline questionnaire provided demographic information about previous pregnancies, dietary habits and physical activities. Information on daily physical activities during work or leisure time was based on the validated Saltin–Grimby Physical Activity Level Scale (SGPALS) [9,10]. A short fitness test (the Danish step test) was performed at baseline and at the last visit before delivery to measure physical fitness [11].
Outcomes
On three occasions during pregnancy [gestational ages 12–15 weeks (baseline), 28–30 and 34–36 weeks] following an overnight fast, blood samples were collected from the ante-cubital vein and a 2-h 75-g oral glucose tolerance test was performed. Fasting plasma glucose was measured using enzymatic reference method with hexokinase (Integra 700; Roche, Mannheim, Germany). Blood-glucose values as part of a 2-h oral glucose tolerance test were measured using capillary blood and analysed photometrically in a HemoCue € analyser (HemoCue, Angelholm, Sweden). Serum levels of insulin were analysed by time-resolved fluoro-immunoassay (AutoDELFIA; Wallac Oy, Turku, Finland). For insulin, the total coefficient of variation was 6.5%. Insulin was measured in pmol/l and converted to mU/l by dividing with the conversion factor of 6 [12]. Insulin resistance was estimated using the homeostasis model assessment of insulin resistance (HOMA-IR) according to Matthews et al. [13] and calculated with the following formula: fasting plasma insulin in mU/ml 9 fasting plasma glucose in mmol/ml)/22.5. Plasma concentrations of total cholesterol, HDL cholesterol, LDL cholesterol and triglycerides were determined (Modular; Roche Diagnostics, Basel, Switzerland). Gestational weight gain was calculated as weight at the 35-week visit minus weight measured at recruitment to the study.
Statistical analyses
All analyses were conducted using Stata version 10.0 software (StataCorp, College Station, TX, USA). Differences between groups were analysed with the v2-test for categorical variables. The Student t-test was used for continuous variables with normal distribution; otherwise the Mann– Whitney U-test was used. A significance level of 0.05 (two-sided) was used. Results of metabolic data were analysed with and without log transformation of continuous variables. Sub-analyses according to the Institute of Medicine (IOM) gestational weight gain recommendations [14] were performed using one-way analysis of variance (ANOVA) or v2-test. The underlying power calculations were based on an expectation of finding a difference of 5-kg gestational weight gain between intervention and control groups in the Lifestyle in Pregnancy study and with five major clinical outcomes
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
DIABETICMedicine
combined to a score from 0 to 5 as the primary outcomes. To detect a significant difference in this score (Wilcoxon rank-sum test, 85% power), 180 women were required in each group (P < 0.05).
Results A total of 360 women were included in this study. Because of a number of dropouts during pregnancy for different reasons, we have results from 150 women in the intervention group and 154 in the control group. The intervention and control groups were comparable in all baseline maternal characteristics and represent an ethnically homogenous group of Danish, Caucasian obese women (Table 1). All participants had an oral glucose tolerance test performed at baseline. Among the women included and randomized, a total of 12 women were excluded after entry to the study because of a diagnosis of gestational diabetes after the oral glucose tolerance test between 12 and 15 weeks’ gestation. Another 17 women developed gestational diabetes during pregnancy: 6.0% in the intervention group and 5.2% in the control group (P = 0.76). Nine of the 17 women were diagnosed at gestational age 34–36 weeks and the others prior to this. The intervention group had significantly lower gestational weight gain (kg) compared with the control group (mean SD): 7.4 4.6 vs. 8.6 4.4, P = 0.01). We were not able to detect any significant differences in the clinical outcomes with respect to pre-eclampsia or pregnancy-induced hypertension, gestational diabetes, Caesarean section, large for gestational age and admission to the neonatal intensive care unit. The oral glucose tolerance test was not repeated in women who were diagnosed with gestational diabetes in week 28–30. Consequently, oral glucose tolerance test Table 1 Baseline maternal characteristics Intervention group n = 150 Baseline Maternal age (years) Maternal BMI (kg/m2) Obesity class I BMI (30–34.9 kg/m2) Obesity class II BMI (35–39.9 kg/m2) Obesity class III BMI (40–45 kg/m2) Smokers Primiparous School ≥ 12 years Further education ≥ 3 years In gainful employment
Control subjects n = 154
29 (27–32) 29 (26–31) 33.4 (31.7–36.5) 33.3 (31.7–36.9) 95 (63.3) 102 (66.2) 42 (28)
45 (29.2)
13 (8.7)
7 (4.6)
11 79 111 75
(7.3) (52.7) (74) (50)
103 (68.7)
18 84 100 67
(11.7) (54.6) (64.9) (43.5)
106 (68.8)
Data are given as median (interquartile range) or number (%). There were no statistically significant differences in any variables between intervention and control groups.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
results at gestational age 28–30 weeks are the most reliable (Table 2). Self-reported physical activity level and eating habits are given in Table 3. Compliance with the training classes was low. The mean attendance for the 20 planned aerobic classes during pregnancy was 10.4 h, and 56% of women in the intervention group attended the aerobic classes for at least half of the sessions. In terms of attendance to the diet counselling compliance was very satisfactory. Ninety-two per cent of the women completed all four counselling sessions and a total of 98% completed at least three sessions. According to the self-reported level of physical activity, women in the intervention group increased the weekly time spent with physical activities from baseline to gestational week 35 (Table 3). The self-reported leisure time physical activity level was significantly improved in the intervention group. The percentage of women spending at least 2 h per week with leisure time physical activities making them break sweat or be short of breath increased from 29% to 56% during pregnancy. In the control group this physical activity level decreased from 34% to 25% and the difference in the third trimester was statistically significant (P < 0.001). Still, the time spent on walking or cycling, as a means of getting to and from work, shopping, etc. was unchanged between the groups. Self-reported eating habits improved from baseline to the third trimester of pregnancy. The percentage of women in the intervention group who reported themselves as eating the healthiest diet was significantly higher in the intervention group in the third trimester. Twenty per cent of women in the control group reported that they believed their participation in the study in itself had improved their lifestyle, despite no active intervention. The fitness score in late pregnancy was significantly higher in the intervention group compared with the control group, but the number of women completing the Steptest at 34– 36 weeks’ gestation was low because of pregnancy-related discomfort. The difference in systolic and diastolic blood pressure between groups and throughout pregnancy did not differ significantly.
Glucose metabolism and insulin sensitivity
Changes in fasting as well as the 2-h glucose concentration during the oral glucose tolerance test were limited during pregnancy (Table 2). Although baseline (gestational weeks 10–14) 2-h oral glucose tolerance test result was higher in the control group, the changes in 2-h glucose concentration during the oral glucose tolerance test were comparable in the groups throughout pregnancy. There was no difference in the rate of women diagnosed with gestational diabetes between intervention and control groups (6.0% vs. 5.2%, P = 0.760), according to current Danish criteria (2-h capillary blood glucose concentration of ≥ 9.0 mmol/l). When using a diagnostic 2-h cut-off of ≥ 8.5 mmol/l, as suggested in the International Association
1325
Metabolic effects of lifestyle intervention C. A. Vinter et al.
DIABETICMedicine
Table 2 Biomarkers of metabolic outcomes at three different time points during pregnancy in the intervention and control group and changes in blood pressure, VO2max, and gestational weight gain from baseline to gestational age 34–36 weeks Intervention, n = 150 Fasting glucose (mmol/l) Gestational age, 12–15 weeks Gestational age, 28–30 weeks Gestational age, 34–36 weeks 2-h oral glucose tolerance test (mmol/l) Gestational age, 12–15 weeks Gestational age, 28–30 weeks Gestational age, 34–36 weeks Fasting-insulin (mU/l) Gestational age, 12–15 weeks (1) Gestational age, 28–30 weeks (2) Gestational age, 34–36 weeks (3) Delta 1? 2 Delta 1? 3 HOMA-IR Gestational age, 12–15 weeks (1) Gestational age, 28–30 weeks (2) Gestational age, 34–36 weeks (3) Delta 1? 2 Delta 1? 3 Fasting cholesterol (mmol/l) Gestational age, 12–15 weeks (1) Gestational age, 28–30 weeks (2) Gestational age, 34–36 weeks (3) Fasting-HDL (mmol/l) Gestational age, 12–14 weeks Gestational age, 28–30 weeks Gestational age, 34–36 weeks Fasting LDL (mmol/l) Gestational age, 12–14 weeks Gestational age, 28–30 weeks Gestational age, 34–36 weeks Fasting triglycerides (mmol/l) Gestational age, 12–15 weeks (1) Gestational age, 28–30 weeks (2) Gestational age, 34–36 weeks (3) Systolic blood pressure (mmHg) Gestational age, 12–15 weeks Gestational age, 34–36 weeks Diastolic blood pressure (mmHg) Gestational age, 12–15 weeks Gestational age, 34–36 weeks VO2max (ml kg1 min1)* Gestational age, 12–15 weeks Gestational age, 34–36 weeks Gestational weight gain (kg)
Control, n = 154
P
4.9 0.4, 148 4.9 0.4, 145 4.9 0.5, 134
5.0 0.3, 154 5.0 0.5, 150 4.9 0.5, 143
0.114 0.060 0.431
6.2 1.0, 144 6.4 1.2, 141 6.4 1.3, 127
6.5 1.1, 148 6.5 1.2, 142 6.5 1.2, 134
0.023 0.459 0.723
10.7 13.7 15.7 3.1 5.2
4.9, 6.9, 7.9, 4.9, 6.0,
148 145 131 143 129
10.8 15.0 16.9 4.3 6.2
5.1, 6.7, 6.8, 4.9, 5.3,
151 147 142 144 139
0.895 0.040 0.065 0.015 0.063
2.4 3.1 3.5 0.7 1.2
1.2, 1.8, 2.1, 1.3, 1.7,
147 144 131 141 128
2.4 3.4 3.7 1.0 1.4
1.2, 1.8, 1.7, 1.3, 1.4,
151 147 142 144 139
0.585 0.032 0.062 0.022 0.079
5.2 0.9, 148 6.5 1.1, 146 6.7 1.2, 134
5.2 0.9, 152 6.3 1.0, 150 6.5 1.1, 143
0.613 0.332 0.484
1.8 0.4, 148 1.8 0.4, 146 1.7 0.4, 134
1.8 0.4, 152 1.8 0.4, 149 1.7 0.4, 143
0.946 0.781 0.871
2.9 0.8, 148 3.9 1.0, 146 4.0 1.1, 134
2.9 0.7, 152 3.6 0.9, 149 3.8 1.0, 143
0.970 0.148 0.183
1.3 0.5, 148 2.2 0.6, 146 2.7 0.9, 134
1.5 0.6, 152 2.4 0.9, 149 2.8 0.9, 143
0.063 0.385 0.399
124 10, 150 123 9, 142
123 10, 154 123 11, 148
0.857 0.693
80 7, 150 83 9, 142
81 8, 154 84 8, 148
0.267 0.263
25 5, 149 23 5, 90 7.4 4.6, 144
25 5, 149 22 4, 76 8.6 4.4, 148
0.412 0.049 0.014
Data are given as mean SD and number (n). The Mann–Whitney U-test was used as data were not within normal distribution. Where n < 150 in the intervention group and n < 154 in the control group, it is because of incomplete data or missing values during pregnancy. Those with a diagnosis of gestational diabetes in gestational age 28–30 weeks did not take an oral glucose tolerance test again in gestational age 34–36 weeks. *VO2max as an indicator of physical fitness. HOMA-IR, homeostasis model assessment of insulin resistance.
of Diabetes and Pregnancy Study Group (IADPSG) criteria for gestational diabetes [15], the number of women with gestational diabetes increased to 10.0% in the intervention group and 9.7% in the control group (P = 0.940). When women with a fasting value (venous plasma glucose) of ≥ 5.1 mmol/l were added, we had high rates of gestational diabetes: 39% in the intervention group and 44% in the control group (P = 0.394). Fasting serum insulin increased
1326
by approximately 1.5-fold in both groups throughout pregnancy. However, women in the intervention group had a significantly lower change (d-value) in fasting insulin from randomization to 28 weeks compared with the control group (mean SD: 3.1 4.9 vs. 4.3 4.9 mU/l, P = 0.015). Similarly, the physiological increase in insulin resistance (d-HOMA) was lower in the intervention group compared with control subjects at gestational 28–30 weeks
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
DIABETICMedicine
Table 3 Self-reported physical activity level and eating habits at baseline and at gestational age 35 weeks according to questionnaires Intervention
Control
P-value
Baseline Physical activity ≥ 2 h/week Physical activity making the woman break sweat or be short of breath ≥ 2 h/week The women considered themselves as in the most healthy eating habit groups*
Intervention
Control
P-value
Gestational age 35 weeks
72% 29%
76% 34%
0.815 0.171
88% 56%
67% 25%
< 0.001 < 0.001
76%
73%
0.495
94%
81%
< 0.001
Data are given as percentages. Differences between groups were analysed with the v2-test for categorical variables using a significance level of 0.05 (two-sided). *Category 1 or category 2 are the two most healthy eating habit groups: category 1, healthy eating; category 3, traditional Danish eating; category 2 is an intermediate between categories 1 and 3.
(mean SD: 0.7 1.3 vs. 1.0 1.3, P = 0.022). At 34–36 weeks’ gestation, insulin resistance was higher in the control group, but this was not statistically significant. Delta values for both fasting insulin and HOMA-IR between baseline and 34–36 weeks tended to be lower in the intervention group (Table 2).
Lipid metabolism
There was an increase in total plasma cholesterol and plasma triglycerides during pregnancy (Table 2). Overall, the increase in plasma cholesterol during pregnancy was approximately 25%. The twofold increase in plasma triglycerides during pregnancy was similar in the two groups. LDL increased approximately 30% and HDL was almost unchanged. There was no significant difference at any time during pregnancy in any of these four lipid variables when comparing the intervention with the control group.
Subanalysis according to Institute of Medicine criteria
As no major effect was seen between intervention and the control group, we have analysed the data, irrespective of treatment group, according to success in controlling gestational weight gain according to Institute of Medicine recommendations: < 5 kg, 5–9 kg and ≥ 9 kg. Women exceeding 9 kg in gestational weight gain had an increased risk of Caesarean section (32% vs. 21%, P = 0.038) and large-for-gestational-age infants (20% vs. 9%, P = 0.09). Gaining less than 5 kg did not increase the risk of small-for-gestational-age infants compared with normal or excess weight gain (5.7% vs. 7.7%, P = 0.58). Fasting insulin and HOMA-IR differed significantly in the three Institute of Medicine groups at gestational week 34–36, with the lowest values in the low weight-gain group: Fasting insulin mean SD: 13.2 5.6 vs. 15.8 6.2 vs. 18.4 8.4 mU/l, P < 0.001) and HOMA-IR mean SD: 2.8 1.4 vs. 3.6 1.7 vs. 4.1 2.2, P = 0.001. Fasting insulin and HOMA-IR earlier in pregnancy were not influenced by Institute of Medicine category. ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Discussion In this randomized controlled trial in obese pregnant women, we found that lifestyle intervention during pregnancy could restrict gestational weight gain and limit the physiological increase in insulin resistance during pregnancy. The intervention did not significantly affect clinical obstetric outcomes, overall oral glucose tolerance test results or lipid profiles between groups. To our knowledge, this is the largest randomized controlled trial among obese pregnant women to investigate the metabolic impact of a lifestyle intervention programme during pregnancy. The strengths of this study are the large sample size and the randomized design. Furthermore, women were included as early as 10–14 weeks of pregnancy. The study was performed in a well-defined homogenous population of obese Caucasian women. However, there are some limitations that complicate our intention to analyse the potential effect of the intervention performed. The women0 s participation in this study was based on their willingness to change lifestyle. Hence, women in the control group were as motivated as those in the intervention group. The control group was followed up more closely than obese women outside the study and received the same information about aims and content of the study. It is likely that their behaviour would have changed towards a healthier lifestyle compared with those not participating. Consequently, the control group could be characterized as a ‘passive intervention group’ and the crossover from the control group could explain the small differences between the groups and thus underestimate the potential effect of the intervention. According to the initial power calculations, 180 women were needed in each arm to have a power of 85%, but we ended up with approximately 150 participants, reducing the power to 77%. Power calculations were based on the expectation of a larger difference in gestational weight gain between groups than we actually found. This should be considered in the conclusions regarding the clinical effects of intervention.
1327
DIABETICMedicine
Glucose metabolism
The physiologic increase in insulin resistance during pregnancy in both women of normal weight and obese women [16] was significantly attenuated in women from our intervention group. This improvement in insulin resistance may be attributable to increased physical activity and restricted weight gain. Although of no major clinical relevance, this finding might explain some of the metabolic modifications that occur when restricting gestational weight gain during lifestyle intervention. The gold standard to measure insulin sensitivity is considered to be the high-cost and time-consuming hyperinsulinaemic–euglycaemic clamp method [6]. In our study, we used the HOMA-IR as a surrogate estimation of insulin sensitivity, based on fasting glucose and insulin. HOMA-IR correlates with the clamp method, but might be more reflective of hepatic insulin sensitivity compared with the peripheral skeletal muscles, as is the target with exercise. Estimates using glucose and insulin measures during a 2-h oral glucose tolerance test (ISOGTT) seems to reflect both hepatic and peripheral insulin sensitivity more precisely in pregnant women [17]. For logistical reasons, such measures were unfortunately not performed during the oral glucose tolerance test in our study. An oral glucose tolerance test primarily reflects changes in metabolism during the night and in the morning and under standardized conditions. Individual lifestyle habits may result in various changes in the normal course of the day both preand post-meals. As women with pre-gestational diabetes were excluded, we did not expect to find any significant changes in self-monitored glucose profiles and therefore these were not carried out. There was a minor effect on fasting glucose by the end of second trimester and no significant effect in the 2-h oral glucose tolerance test result. We would expect higher 2-h oral glucose tolerance test concentrations at gestational age 34–36 weeks if women diagnosed with gestational diabetes had been included, but they did not have repeated oral glucose tolerance tests. The 2-h oral glucose tolerance test result in the remaining participants was unchanged or decreased, but the values were comparable in the two groups, as was the number of women developing gestational diabetes. A small study from Australia measured the effect of 10 weeks home-based exercise and found attenuation of the decline of glucose tolerance in obese pregnant women [18]. They reported that women who completed the exercise programme maintained a similar blood glucose response to the oral glucose tolerance test throughout the exercise intervention period, while those in the control group experienced worsening of glucose tolerance. In a study of 50 obese pregnant women, Callaway et al. reported lower fasting glucose levels at the end of second trimester in the intervention group with a physical activity programme, but no difference was seen between groups with respect to insulin resistance, despite some
1328
Metabolic effects of lifestyle intervention C. A. Vinter et al.
improvements in physical activity [19]. Wolff et al. [20], in an intensively individualized dietary intervention programme in 50 Danish obese women, reported gestational weight gain of 6.6 kg in the intervention group compared with 13.3 kg in the control group and showed a significant reduction in serum insulin during pregnancy. The HOMA-IR was not reported, but the intervention restricted serum insulin. In the FitFor2 intervention study among 121 women at increased risk of gestational diabetes, Oostdam et al. found no effect on fasting blood glucose or insulin sensitivity between women allocated to 2 h of weekly exercises compared with a control group [21]. It was concluded that the main reason for this lack of metabolic differences was attributable to low compliance with the intervention.
Lipid metabolism
It has been described that total plasma cholesterol concentration increases by 25–50% and plasma triglyceride concentration increases two- to fourfold during pregnancy [22]. We found a similar 25% increase in plasma total cholesterol and an approximate 2-fold increase in plasma triglycerides, although the increment was small. Despite significantly lower gestational weight gain in the intervention group, we did not detect any difference in the lipid profile between groups. A possible explanation could be the rather low gestational weight gain in both groups, reflecting lower fat deposits in pregnancy. Previous studies reporting lipid profiles in pregnancy are small. A randomized controlled trial from Finland [23] with 54 women at high risk for gestational diabetes evaluated the effect of dietary therapy on gestational weight gain, gestational diabetes and metabolism. They found no significant effect on weight or clinical outcomes and changes in lipids were similar to ours. Ramsay et al. compared 23 obese and 24 lean pregnant women in a cross-sectional study and found a significant adverse lipid profile and hyperinsulinaemia in obese women [22]. Based on our findings of attenuated insulin resistance in the intervention group, it might be suggested that dietary changes and exercise can improve the metabolic changes in pregnancy and hence, at least theoretically, clinical complications such as gestational diabetes, pre-eclampsia and macrosomia [24,25]. However, as suggested in systematic reviews [26,27], the impact of antenatal intervention may not be sufficient to overcome the presence of pre-gestational obesity.. The intensity, duration and nature of such interventions should be further investigated to facilitate more efficient interventional strategies. Intervention prior to conception should also be considered. We found that insulin resistance increased in parallel with gestational weight gain and that adherence to Institute of Medicine criteria decreased the risk of Caesarean section and large for gestational age infants. The debate is ongoing concerning possible risk of small-for-gestational-age infants when suggesting weight loss or weight gain below 5 kg in
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
obese pregnant women. This concern was not confirmed in our study. The limited evidence in the literature to support clinical effects of a lifestyle intervention programme in obese pregnant women [26–28] might be attributable to difficulties in design and lack of power in studies performed so far. However, even minor improvements in metabolic status during pregnancy might influence the programming in utero, even if no beneficial effect is shown on short-term neonatal outcomes. Prospective follow-up studies in offspring from mothers who participated in antenatal intervention programmes should be performed to address this issue. Large interventional studies in obese women are ongoing [29,30] and should provide further evidence on how to manage obesity in relation to pregnancy.
Funding sources
The study was supported by Trygfonden, The Health Insurance Foundation (Sygekassernes Helsefond), The Faculty of Health Sciences, University of Southern Denmark, The Danish Diabetes Association, Odense University Hospital, The Novo Foundation, The Danish Medical Association Research Foundation, Aase og Ejnar Danielsens Fond, CMA Medico, Ferrosan A/S.
Competing interests
None declared.
Acknowledgements
Assistance with data collection and contributions to the study from midwife and physiotherapist P. Ingerslev and dietician C. Wolff (Department of Gynecology and Obstetrics, Aarhus University Hospital) and midwife C. Thomsen, dieticians B. Knold and R. Nestor, physiotherapist M. Bødker and D. Larsen (Department of Gynecology and Obstetrics, Odense University Hospital) is gratefully acknowledged. The authors acknowledge the editorial assistance of Prof. Ronald F. Lamont, University of Southern Denmark.
References 1 Nelson SM, Matthews P, Poston L. Maternal metabolism and obesity: modifiable determinants of pregnancy outcome. Hum Reprod Update 2010; 16: 255–275. 2 Ovesen P, Rasmussen S, Kesmodel U. Effect of prepregnancy maternal overweight and obesity on pregnancy outcome. Obstet Gynecol 2011; 118: 305–312. 3 Huda SS, Brodie LE, Sattar N. Obesity in pregnancy: prevalence and metabolic consequences. Semin Fetal Neonatal Med 2010; 15: 70–76. 4 Lain KY, Catalano PM. Metabolic changes in pregnancy. Clin Obstet Gynecol 2007; 50: 938–948.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
DIABETICMedicine
5 Catalano PM. Obesity, insulin resistance, and pregnancy outcome. Reproduction 2010; 140: 365–371. 6 Catalano PM, Tyzbir ED, Roman NM, Amini SB, Sims EA. Longitudinal changes in insulin release and insulin resistance in non-obese pregnant women. Am J Obstet Gynecol 1991; 165: 1667–1672. 7 Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U, Coustan DR et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008; 358: 1991–2002. 8 Vinter CA, Jensen DM, Ovesen P, Beck-Nielsen H, Jorgensen JS. The LiP (Lifestyle in Pregnancy) Study: a randomized controlled trial of lifestyle intervention in 360 obese pregnant women. Diabetes Care 2011; 34: 2502–2507. 9 Saltin B, Grimby G. Physiological analysis of middle-aged and old former athletes. Comparison with still active athletes of the same ages. Circulation 1968; 38: 1104–1115. 10 Rodjer L, Jonsdottir IH, Rosengren A, Bjorck L, Grimby G, Thelle DS et al. Self-reported leisure time physical activity: a useful assessment tool in everyday health care. BMC Public Health 2012; 12: 693. 11 Aadahl M, Zacho M, Linneberg A, Thuesen BH, Jorgensen T. Comparison of the Danish step test and the watt-max test for estimation of maximal oxygen uptake: the Health 2008 study. Eur J Prev Cardiol 2013; 20: 1088–1094. 12 Volund A. Conversion of insulin units to SI units. Am J Clin Nutr 1993; 58: 714–715. 13 Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 412–419. 14 Institute of Medicine. Weight Gain During Pregnancy: Re-examining the Guidelines. Washington, DC: The National Academy Press, 2009. 15 Metzger BE, Gabbe SG, Persson B, Buchanan TA, Catalano PA, Damm P et al. International Association of Diabetes and Pregnancy Study Group’s recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care 2010; 33: 676–682. 16 Catalano PM, Huston L, Amini SB, Kalhan SC. Longitudinal changes in glucose metabolism during pregnancy in obese women with normal glucose tolerance and gestational diabetes mellitus. Am J Obstet Gynecol 1999; 180: 903–916. 17 Kirwan JP, Huston-Presley L, Kalhan SC, Catalano PM. Clinically useful estimates of insulin sensitivity during pregnancy: validation studies in women with normal glucose tolerance and gestational diabetes mellitus. Diabetes Care 2001; 24: 1602–1607. 18 Ong MJ, Guelfi KJ, Hunter T, Wallman KE, Fournier PA, Newnham JP. Supervised home-based exercise may attenuate the decline of glucose tolerance in obese pregnant women. Diabetes Metab 2009; 35: 418–421. 19 Callaway LK, Colditz PB, Byrne NM, Lingwood BE, Rowlands IJ, Foxcroft K et al. Prevention of gestational diabetes: feasibility issues for an exercise intervention in obese pregnant women. Diabetes Care 2010; 33: 1457–1459. 20 Wolff S, Legarth J, Vangsgaard K, Toubro S, Astrup A. A randomized trial of the effects of dietary counseling on gestational weight gain and glucose metabolism in obese pregnant women. Int J Obes (Lond) 2008; 32: 495–501. 21 Oostdam N, van Poppel MN, Wouters MG, Eekhoff EM, Bekedam DJ, Kuchenbecker WK et al. No effect of the FitFor2 exercise programme on blood glucose, insulin sensitivity and birthweight in pregnant women who were overweight and at risk for gestational diabetes: results of a randomised controlled trial. BJOG 2012; 119: 1098–1107. 22 Ramsay JE, Ferrell WR, Crawford L, Wallace AM, Greer IA, Sattar N. Maternal obesity is associated with dysregulation of metabolic,
1329
DIABETICMedicine
23
24
25
26
27
28
vascular, and inflammatory pathways. J Clin Endocrinol Metab 2002; 87: 4231–4237. Korpi-Hyovalti E, Schwab U, Laaksonen DE, Linjama H, Heinonen S, Niskanen L. Effect of intensive counselling on the quality of dietary fats in pregnant women at high risk of gestational diabetes mellitus. Br J Nutr 2012; 108: 910–917. Chu SY, Callaghan WM, Kim SY, Schmid CH, Lau J, England LJ et al. Maternal obesity and risk of gestational diabetes mellitus. Diabetes Care 2007; 30: 2070–2076. Nohr EA, Vaeth M, Baker JL, Sorensen TI, Olsen J, Rasmussen KM. Combined associations of prepregnancy body mass index and gestational weight gain with the outcome of pregnancy. Am J Clin Nutr 2008; 87: 1750–1759. Dodd JM, Grivell RM, Crowther CA, Robinson JS. Antenatal interventions for overweight or obese pregnant women: a systematic review of randomised trials. BJOG 2010; 117: 1316–1326. Quinlivan JA, Julania S, Lam L. Antenatal dietary interventions in obese pregnant women to restrict gestational weight gain to Institute of Medicine recommendations: a meta-analysis. Obstet Gynecol 2011; 118: 1395–1401. Thangaratinam S, Rogozinska E, Jolly K, Glinkowski S, Duda W, Borowiack E et al. Interventions to reduce or prevent obesity in
1330
Metabolic effects of lifestyle intervention C. A. Vinter et al. pregnant women: a systematic review. Health Technol Assess 2012; 16: iii–iv, 1–191. 29 Poston L, Briley AL, Barr S, Bell R, Croker H, Coxon K et al. Developing a complex intervention for diet and activity behaviour change in obese pregnant women (the UPBEAT Trial); assessment of behavioural change and process evaluation in a pilot randomised controlled trial. BMC Pregnancy Childbirth 2013; 13: 148. 30 Jelsma JG, van Poppel MN, Galjaard S, Desoye G, Corcoy R, Devlieger R et al. DALI: Vitamin D and Lifestyle Intervention for gestational diabetes mellitus (GDM) prevention: a European multi-centre, randomised trial—study protocol. BMC Pregnancy Childbirth 2013; 13: 142.
Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Flow chart for participation. Appendix S1. Research design and methods.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research: Metabolism Metabolic effects of lifestyle intervention in obese pregnant women. Results from the randomized controlled trial ‘Lifestyle in Pregnancy’ (LiP) C. A. Vinter1, J. S. Jørgensen1, P. Ovesen2, H. Beck-Nielsen3, A. Skytthe4 and D. M. Jensen3 1
Department of Gynecology and Obstetrics, Odense University Hospital, Odense, 2Department of Gynecology and Obstetrics, Aarhus University Hospital, Skejby, Department of Endocrinology, Odense University Hospital, Odense and 4Epidemiology, Institute of Public Health, University of Southern Denmark, Odense, Denmark 3
Accepted 30 June 2014
Abstract Aims The Lifestyle in Pregnancy intervention in obese pregnant women resulted in significantly lower gestational weight gain compared with the control group, but without improvement in rates of clinical pregnancy complications. The impact of the lifestyle intervention on metabolic measurements in the study participants is now reported.
The Lifestyle in Pregnancy study was a randomized controlled trial among 360 obese women (BMI 30–45 kg/m2) who were allocated in early pregnancy to lifestyle interventions with diet counselling and physical activities or to the control group. Fasting blood samples, including plasma glucose, insulin, lipid profile and capillary blood glucose during a 2-h oral glucose tolerance test were carried out three times throughout pregnancy. Insulin resistance was estimated with the homeostasis model assessment of insulin resistance. Methods
Three hundred and four women (84%) were followed until delivery. Women in the intervention group had a significantly lower change in insulin resistance (HOMA-IR) from randomization to 28–30 weeks’ gestation compared with control subjects (mean SD: 0.7 1.3 vs. 1.0 1.3, P = 0.02). Despite a significantly lower gestational weight gain in the intervention group, there was no difference between the groups with respect to total cholesterol, HDL, LDL or triglycerides.
Results
Conclusions Lifestyle intervention in obese pregnant women resulted in attenuation of the physiologic pregnancy-induced insulin resistance. Despite restricted gestational weight gain, there were no changes in glucose or lipid metabolism between the groups.
Diabet. Med. 31, 1323–1330 (2014)
Introduction Obesity in pregnancy has become highly prevalent worldwide and is associated with adverse maternal and neonatal outcomes [1,2]. The underlying pathophysiology linking maternal obesity and adverse outcomes is not fully understood, but seems to involve alterations in glucose and lipid metabolism [3]. Pregnancy is characterized by substantial changes in maternal metabolism that can lead to gestational diabetes mellitus and increase the risk of other adverse outcomes, especially in obese women [4,5]. However, our knowledge is limited with respect to the metabolic changes that occur in obese pregnant women, as well as the metabolic
Correspondence to: Christina Anne Vinter. E-mail: [email protected] (Clinical Trials Registry No; NCT 00530439)
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
impact of lifestyle changes during pregnancy. Normal pregnancy is characterized by a significant increase in insulin resistance [6]. In obese women, the insulin resistance of pregnancy is superimposed on the underlying pre-pregnancy insulin resistance and these changes may lead to high maternal glucose levels, increased risk of gestational diabetes and neonatal macrosomia [7]. Pregnancy is characterized by marked increases in plasma lipid concentrations with increasing gestational age [3,4]. The objective of this clinical trial was to investigate whether lifestyle intervention during pregnancy could improve the metabolic status and subsequently improve pregnancy outcomes in obese women. The current publication is an extension of a previous report where the major focus was the clinical implications concerning obstetric and neonatal outcomes [8].
1323
DIABETICMedicine
What’s new? • This study is among the largest of the randomized controlled trials in obese pregnant women to investigate the metabolic impact of a lifestyle intervention during pregnancy. • Lifestyle intervention in obese pregnant women resulted in attenuation of the physiologic pregnancy-induced insulin resistance. • Despite restricted gestational weight gain in the intervention group, there was no major effect in glucose or lipid metabolism.
Patients and methods This was a randomized controlled trial with a description of the biochemical changes in obese pregnant women during a lifestyle intervention programme. The Lifestyle in Pregnancy (LiP) study ran from October 2007 to October 2010 in two university hospitals in Denmark: Odense and Aarhus University Hospitals, Skejby. The project was approved by the local ethics committee of the Region of Southern Denmark (S-20070058) and registered at clinicaltrials.gov as NCT 00530439. Obese women aged 18–40 years were recruited and included at 10–14 weeks of gestation after referral to the Department of Gynecology and Obstetrics. The inclusion criteria required a BMI of 30–45 kg/m2 (calculated from the pre-pregnancy weight or first measured weight in pregnancy), and women were excluded if they had prior serious obstetric complications, major medical disorders—including pre-gestational diabetes—were non-Danish speaking, abuse of alcohol or a multiple pregnancy. A more detailed description of the study and the procedure for randomization has been published elsewhere [8]. The study was designed as a non-blinded randomized controlled trial with two arms: intervention and control. After receiving written and oral information and giving written consent, the 360 participants were randomized 1:1 by computer-generated numbers in closed, opaque envelopes. A flow chart for participation and randomization is shown in the Supporting Information (Fig. S1). Throughout pregnancy the intervention group received four separate diet counselling sessions and an exercise programme consisting of weekly aerobic classes, free fitness membership during pregnancy and exercise motivating initiatives. More detailed information about the randomization process and details about the intervention are found in the online Supporting Information (Appendix S1). During pregnancy the intervention and control groups were monitored using fasting blood samples, oral glucose tolerance tests, sonar fetal biometry and measurements of maternal weight and blood pressure. Gestational diabetes was diagnosed if the 2-h oral glucose tolerance test capillary
1324
Metabolic effects of lifestyle intervention C. A. Vinter et al.
blood glucose concentration was ≥ 9 mmol/l as according to Danish national recommendations. A baseline questionnaire provided demographic information about previous pregnancies, dietary habits and physical activities. Information on daily physical activities during work or leisure time was based on the validated Saltin–Grimby Physical Activity Level Scale (SGPALS) [9,10]. A short fitness test (the Danish step test) was performed at baseline and at the last visit before delivery to measure physical fitness [11].
Outcomes
On three occasions during pregnancy [gestational ages 12–15 weeks (baseline), 28–30 and 34–36 weeks] following an overnight fast, blood samples were collected from the ante-cubital vein and a 2-h 75-g oral glucose tolerance test was performed. Fasting plasma glucose was measured using enzymatic reference method with hexokinase (Integra 700; Roche, Mannheim, Germany). Blood-glucose values as part of a 2-h oral glucose tolerance test were measured using capillary blood and analysed photometrically in a HemoCue € analyser (HemoCue, Angelholm, Sweden). Serum levels of insulin were analysed by time-resolved fluoro-immunoassay (AutoDELFIA; Wallac Oy, Turku, Finland). For insulin, the total coefficient of variation was 6.5%. Insulin was measured in pmol/l and converted to mU/l by dividing with the conversion factor of 6 [12]. Insulin resistance was estimated using the homeostasis model assessment of insulin resistance (HOMA-IR) according to Matthews et al. [13] and calculated with the following formula: fasting plasma insulin in mU/ml 9 fasting plasma glucose in mmol/ml)/22.5. Plasma concentrations of total cholesterol, HDL cholesterol, LDL cholesterol and triglycerides were determined (Modular; Roche Diagnostics, Basel, Switzerland). Gestational weight gain was calculated as weight at the 35-week visit minus weight measured at recruitment to the study.
Statistical analyses
All analyses were conducted using Stata version 10.0 software (StataCorp, College Station, TX, USA). Differences between groups were analysed with the v2-test for categorical variables. The Student t-test was used for continuous variables with normal distribution; otherwise the Mann– Whitney U-test was used. A significance level of 0.05 (two-sided) was used. Results of metabolic data were analysed with and without log transformation of continuous variables. Sub-analyses according to the Institute of Medicine (IOM) gestational weight gain recommendations [14] were performed using one-way analysis of variance (ANOVA) or v2-test. The underlying power calculations were based on an expectation of finding a difference of 5-kg gestational weight gain between intervention and control groups in the Lifestyle in Pregnancy study and with five major clinical outcomes
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
DIABETICMedicine
combined to a score from 0 to 5 as the primary outcomes. To detect a significant difference in this score (Wilcoxon rank-sum test, 85% power), 180 women were required in each group (P < 0.05).
Results A total of 360 women were included in this study. Because of a number of dropouts during pregnancy for different reasons, we have results from 150 women in the intervention group and 154 in the control group. The intervention and control groups were comparable in all baseline maternal characteristics and represent an ethnically homogenous group of Danish, Caucasian obese women (Table 1). All participants had an oral glucose tolerance test performed at baseline. Among the women included and randomized, a total of 12 women were excluded after entry to the study because of a diagnosis of gestational diabetes after the oral glucose tolerance test between 12 and 15 weeks’ gestation. Another 17 women developed gestational diabetes during pregnancy: 6.0% in the intervention group and 5.2% in the control group (P = 0.76). Nine of the 17 women were diagnosed at gestational age 34–36 weeks and the others prior to this. The intervention group had significantly lower gestational weight gain (kg) compared with the control group (mean SD): 7.4 4.6 vs. 8.6 4.4, P = 0.01). We were not able to detect any significant differences in the clinical outcomes with respect to pre-eclampsia or pregnancy-induced hypertension, gestational diabetes, Caesarean section, large for gestational age and admission to the neonatal intensive care unit. The oral glucose tolerance test was not repeated in women who were diagnosed with gestational diabetes in week 28–30. Consequently, oral glucose tolerance test Table 1 Baseline maternal characteristics Intervention group n = 150 Baseline Maternal age (years) Maternal BMI (kg/m2) Obesity class I BMI (30–34.9 kg/m2) Obesity class II BMI (35–39.9 kg/m2) Obesity class III BMI (40–45 kg/m2) Smokers Primiparous School ≥ 12 years Further education ≥ 3 years In gainful employment
Control subjects n = 154
29 (27–32) 29 (26–31) 33.4 (31.7–36.5) 33.3 (31.7–36.9) 95 (63.3) 102 (66.2) 42 (28)
45 (29.2)
13 (8.7)
7 (4.6)
11 79 111 75
(7.3) (52.7) (74) (50)
103 (68.7)
18 84 100 67
(11.7) (54.6) (64.9) (43.5)
106 (68.8)
Data are given as median (interquartile range) or number (%). There were no statistically significant differences in any variables between intervention and control groups.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
results at gestational age 28–30 weeks are the most reliable (Table 2). Self-reported physical activity level and eating habits are given in Table 3. Compliance with the training classes was low. The mean attendance for the 20 planned aerobic classes during pregnancy was 10.4 h, and 56% of women in the intervention group attended the aerobic classes for at least half of the sessions. In terms of attendance to the diet counselling compliance was very satisfactory. Ninety-two per cent of the women completed all four counselling sessions and a total of 98% completed at least three sessions. According to the self-reported level of physical activity, women in the intervention group increased the weekly time spent with physical activities from baseline to gestational week 35 (Table 3). The self-reported leisure time physical activity level was significantly improved in the intervention group. The percentage of women spending at least 2 h per week with leisure time physical activities making them break sweat or be short of breath increased from 29% to 56% during pregnancy. In the control group this physical activity level decreased from 34% to 25% and the difference in the third trimester was statistically significant (P < 0.001). Still, the time spent on walking or cycling, as a means of getting to and from work, shopping, etc. was unchanged between the groups. Self-reported eating habits improved from baseline to the third trimester of pregnancy. The percentage of women in the intervention group who reported themselves as eating the healthiest diet was significantly higher in the intervention group in the third trimester. Twenty per cent of women in the control group reported that they believed their participation in the study in itself had improved their lifestyle, despite no active intervention. The fitness score in late pregnancy was significantly higher in the intervention group compared with the control group, but the number of women completing the Steptest at 34– 36 weeks’ gestation was low because of pregnancy-related discomfort. The difference in systolic and diastolic blood pressure between groups and throughout pregnancy did not differ significantly.
Glucose metabolism and insulin sensitivity
Changes in fasting as well as the 2-h glucose concentration during the oral glucose tolerance test were limited during pregnancy (Table 2). Although baseline (gestational weeks 10–14) 2-h oral glucose tolerance test result was higher in the control group, the changes in 2-h glucose concentration during the oral glucose tolerance test were comparable in the groups throughout pregnancy. There was no difference in the rate of women diagnosed with gestational diabetes between intervention and control groups (6.0% vs. 5.2%, P = 0.760), according to current Danish criteria (2-h capillary blood glucose concentration of ≥ 9.0 mmol/l). When using a diagnostic 2-h cut-off of ≥ 8.5 mmol/l, as suggested in the International Association
1325
Metabolic effects of lifestyle intervention C. A. Vinter et al.
DIABETICMedicine
Table 2 Biomarkers of metabolic outcomes at three different time points during pregnancy in the intervention and control group and changes in blood pressure, VO2max, and gestational weight gain from baseline to gestational age 34–36 weeks Intervention, n = 150 Fasting glucose (mmol/l) Gestational age, 12–15 weeks Gestational age, 28–30 weeks Gestational age, 34–36 weeks 2-h oral glucose tolerance test (mmol/l) Gestational age, 12–15 weeks Gestational age, 28–30 weeks Gestational age, 34–36 weeks Fasting-insulin (mU/l) Gestational age, 12–15 weeks (1) Gestational age, 28–30 weeks (2) Gestational age, 34–36 weeks (3) Delta 1? 2 Delta 1? 3 HOMA-IR Gestational age, 12–15 weeks (1) Gestational age, 28–30 weeks (2) Gestational age, 34–36 weeks (3) Delta 1? 2 Delta 1? 3 Fasting cholesterol (mmol/l) Gestational age, 12–15 weeks (1) Gestational age, 28–30 weeks (2) Gestational age, 34–36 weeks (3) Fasting-HDL (mmol/l) Gestational age, 12–14 weeks Gestational age, 28–30 weeks Gestational age, 34–36 weeks Fasting LDL (mmol/l) Gestational age, 12–14 weeks Gestational age, 28–30 weeks Gestational age, 34–36 weeks Fasting triglycerides (mmol/l) Gestational age, 12–15 weeks (1) Gestational age, 28–30 weeks (2) Gestational age, 34–36 weeks (3) Systolic blood pressure (mmHg) Gestational age, 12–15 weeks Gestational age, 34–36 weeks Diastolic blood pressure (mmHg) Gestational age, 12–15 weeks Gestational age, 34–36 weeks VO2max (ml kg1 min1)* Gestational age, 12–15 weeks Gestational age, 34–36 weeks Gestational weight gain (kg)
Control, n = 154
P
4.9 0.4, 148 4.9 0.4, 145 4.9 0.5, 134
5.0 0.3, 154 5.0 0.5, 150 4.9 0.5, 143
0.114 0.060 0.431
6.2 1.0, 144 6.4 1.2, 141 6.4 1.3, 127
6.5 1.1, 148 6.5 1.2, 142 6.5 1.2, 134
0.023 0.459 0.723
10.7 13.7 15.7 3.1 5.2
4.9, 6.9, 7.9, 4.9, 6.0,
148 145 131 143 129
10.8 15.0 16.9 4.3 6.2
5.1, 6.7, 6.8, 4.9, 5.3,
151 147 142 144 139
0.895 0.040 0.065 0.015 0.063
2.4 3.1 3.5 0.7 1.2
1.2, 1.8, 2.1, 1.3, 1.7,
147 144 131 141 128
2.4 3.4 3.7 1.0 1.4
1.2, 1.8, 1.7, 1.3, 1.4,
151 147 142 144 139
0.585 0.032 0.062 0.022 0.079
5.2 0.9, 148 6.5 1.1, 146 6.7 1.2, 134
5.2 0.9, 152 6.3 1.0, 150 6.5 1.1, 143
0.613 0.332 0.484
1.8 0.4, 148 1.8 0.4, 146 1.7 0.4, 134
1.8 0.4, 152 1.8 0.4, 149 1.7 0.4, 143
0.946 0.781 0.871
2.9 0.8, 148 3.9 1.0, 146 4.0 1.1, 134
2.9 0.7, 152 3.6 0.9, 149 3.8 1.0, 143
0.970 0.148 0.183
1.3 0.5, 148 2.2 0.6, 146 2.7 0.9, 134
1.5 0.6, 152 2.4 0.9, 149 2.8 0.9, 143
0.063 0.385 0.399
124 10, 150 123 9, 142
123 10, 154 123 11, 148
0.857 0.693
80 7, 150 83 9, 142
81 8, 154 84 8, 148
0.267 0.263
25 5, 149 23 5, 90 7.4 4.6, 144
25 5, 149 22 4, 76 8.6 4.4, 148
0.412 0.049 0.014
Data are given as mean SD and number (n). The Mann–Whitney U-test was used as data were not within normal distribution. Where n < 150 in the intervention group and n < 154 in the control group, it is because of incomplete data or missing values during pregnancy. Those with a diagnosis of gestational diabetes in gestational age 28–30 weeks did not take an oral glucose tolerance test again in gestational age 34–36 weeks. *VO2max as an indicator of physical fitness. HOMA-IR, homeostasis model assessment of insulin resistance.
of Diabetes and Pregnancy Study Group (IADPSG) criteria for gestational diabetes [15], the number of women with gestational diabetes increased to 10.0% in the intervention group and 9.7% in the control group (P = 0.940). When women with a fasting value (venous plasma glucose) of ≥ 5.1 mmol/l were added, we had high rates of gestational diabetes: 39% in the intervention group and 44% in the control group (P = 0.394). Fasting serum insulin increased
1326
by approximately 1.5-fold in both groups throughout pregnancy. However, women in the intervention group had a significantly lower change (d-value) in fasting insulin from randomization to 28 weeks compared with the control group (mean SD: 3.1 4.9 vs. 4.3 4.9 mU/l, P = 0.015). Similarly, the physiological increase in insulin resistance (d-HOMA) was lower in the intervention group compared with control subjects at gestational 28–30 weeks
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
DIABETICMedicine
Table 3 Self-reported physical activity level and eating habits at baseline and at gestational age 35 weeks according to questionnaires Intervention
Control
P-value
Baseline Physical activity ≥ 2 h/week Physical activity making the woman break sweat or be short of breath ≥ 2 h/week The women considered themselves as in the most healthy eating habit groups*
Intervention
Control
P-value
Gestational age 35 weeks
72% 29%
76% 34%
0.815 0.171
88% 56%
67% 25%
< 0.001 < 0.001
76%
73%
0.495
94%
81%
< 0.001
Data are given as percentages. Differences between groups were analysed with the v2-test for categorical variables using a significance level of 0.05 (two-sided). *Category 1 or category 2 are the two most healthy eating habit groups: category 1, healthy eating; category 3, traditional Danish eating; category 2 is an intermediate between categories 1 and 3.
(mean SD: 0.7 1.3 vs. 1.0 1.3, P = 0.022). At 34–36 weeks’ gestation, insulin resistance was higher in the control group, but this was not statistically significant. Delta values for both fasting insulin and HOMA-IR between baseline and 34–36 weeks tended to be lower in the intervention group (Table 2).
Lipid metabolism
There was an increase in total plasma cholesterol and plasma triglycerides during pregnancy (Table 2). Overall, the increase in plasma cholesterol during pregnancy was approximately 25%. The twofold increase in plasma triglycerides during pregnancy was similar in the two groups. LDL increased approximately 30% and HDL was almost unchanged. There was no significant difference at any time during pregnancy in any of these four lipid variables when comparing the intervention with the control group.
Subanalysis according to Institute of Medicine criteria
As no major effect was seen between intervention and the control group, we have analysed the data, irrespective of treatment group, according to success in controlling gestational weight gain according to Institute of Medicine recommendations: < 5 kg, 5–9 kg and ≥ 9 kg. Women exceeding 9 kg in gestational weight gain had an increased risk of Caesarean section (32% vs. 21%, P = 0.038) and large-for-gestational-age infants (20% vs. 9%, P = 0.09). Gaining less than 5 kg did not increase the risk of small-for-gestational-age infants compared with normal or excess weight gain (5.7% vs. 7.7%, P = 0.58). Fasting insulin and HOMA-IR differed significantly in the three Institute of Medicine groups at gestational week 34–36, with the lowest values in the low weight-gain group: Fasting insulin mean SD: 13.2 5.6 vs. 15.8 6.2 vs. 18.4 8.4 mU/l, P < 0.001) and HOMA-IR mean SD: 2.8 1.4 vs. 3.6 1.7 vs. 4.1 2.2, P = 0.001. Fasting insulin and HOMA-IR earlier in pregnancy were not influenced by Institute of Medicine category. ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Discussion In this randomized controlled trial in obese pregnant women, we found that lifestyle intervention during pregnancy could restrict gestational weight gain and limit the physiological increase in insulin resistance during pregnancy. The intervention did not significantly affect clinical obstetric outcomes, overall oral glucose tolerance test results or lipid profiles between groups. To our knowledge, this is the largest randomized controlled trial among obese pregnant women to investigate the metabolic impact of a lifestyle intervention programme during pregnancy. The strengths of this study are the large sample size and the randomized design. Furthermore, women were included as early as 10–14 weeks of pregnancy. The study was performed in a well-defined homogenous population of obese Caucasian women. However, there are some limitations that complicate our intention to analyse the potential effect of the intervention performed. The women0 s participation in this study was based on their willingness to change lifestyle. Hence, women in the control group were as motivated as those in the intervention group. The control group was followed up more closely than obese women outside the study and received the same information about aims and content of the study. It is likely that their behaviour would have changed towards a healthier lifestyle compared with those not participating. Consequently, the control group could be characterized as a ‘passive intervention group’ and the crossover from the control group could explain the small differences between the groups and thus underestimate the potential effect of the intervention. According to the initial power calculations, 180 women were needed in each arm to have a power of 85%, but we ended up with approximately 150 participants, reducing the power to 77%. Power calculations were based on the expectation of a larger difference in gestational weight gain between groups than we actually found. This should be considered in the conclusions regarding the clinical effects of intervention.
1327
DIABETICMedicine
Glucose metabolism
The physiologic increase in insulin resistance during pregnancy in both women of normal weight and obese women [16] was significantly attenuated in women from our intervention group. This improvement in insulin resistance may be attributable to increased physical activity and restricted weight gain. Although of no major clinical relevance, this finding might explain some of the metabolic modifications that occur when restricting gestational weight gain during lifestyle intervention. The gold standard to measure insulin sensitivity is considered to be the high-cost and time-consuming hyperinsulinaemic–euglycaemic clamp method [6]. In our study, we used the HOMA-IR as a surrogate estimation of insulin sensitivity, based on fasting glucose and insulin. HOMA-IR correlates with the clamp method, but might be more reflective of hepatic insulin sensitivity compared with the peripheral skeletal muscles, as is the target with exercise. Estimates using glucose and insulin measures during a 2-h oral glucose tolerance test (ISOGTT) seems to reflect both hepatic and peripheral insulin sensitivity more precisely in pregnant women [17]. For logistical reasons, such measures were unfortunately not performed during the oral glucose tolerance test in our study. An oral glucose tolerance test primarily reflects changes in metabolism during the night and in the morning and under standardized conditions. Individual lifestyle habits may result in various changes in the normal course of the day both preand post-meals. As women with pre-gestational diabetes were excluded, we did not expect to find any significant changes in self-monitored glucose profiles and therefore these were not carried out. There was a minor effect on fasting glucose by the end of second trimester and no significant effect in the 2-h oral glucose tolerance test result. We would expect higher 2-h oral glucose tolerance test concentrations at gestational age 34–36 weeks if women diagnosed with gestational diabetes had been included, but they did not have repeated oral glucose tolerance tests. The 2-h oral glucose tolerance test result in the remaining participants was unchanged or decreased, but the values were comparable in the two groups, as was the number of women developing gestational diabetes. A small study from Australia measured the effect of 10 weeks home-based exercise and found attenuation of the decline of glucose tolerance in obese pregnant women [18]. They reported that women who completed the exercise programme maintained a similar blood glucose response to the oral glucose tolerance test throughout the exercise intervention period, while those in the control group experienced worsening of glucose tolerance. In a study of 50 obese pregnant women, Callaway et al. reported lower fasting glucose levels at the end of second trimester in the intervention group with a physical activity programme, but no difference was seen between groups with respect to insulin resistance, despite some
1328
Metabolic effects of lifestyle intervention C. A. Vinter et al.
improvements in physical activity [19]. Wolff et al. [20], in an intensively individualized dietary intervention programme in 50 Danish obese women, reported gestational weight gain of 6.6 kg in the intervention group compared with 13.3 kg in the control group and showed a significant reduction in serum insulin during pregnancy. The HOMA-IR was not reported, but the intervention restricted serum insulin. In the FitFor2 intervention study among 121 women at increased risk of gestational diabetes, Oostdam et al. found no effect on fasting blood glucose or insulin sensitivity between women allocated to 2 h of weekly exercises compared with a control group [21]. It was concluded that the main reason for this lack of metabolic differences was attributable to low compliance with the intervention.
Lipid metabolism
It has been described that total plasma cholesterol concentration increases by 25–50% and plasma triglyceride concentration increases two- to fourfold during pregnancy [22]. We found a similar 25% increase in plasma total cholesterol and an approximate 2-fold increase in plasma triglycerides, although the increment was small. Despite significantly lower gestational weight gain in the intervention group, we did not detect any difference in the lipid profile between groups. A possible explanation could be the rather low gestational weight gain in both groups, reflecting lower fat deposits in pregnancy. Previous studies reporting lipid profiles in pregnancy are small. A randomized controlled trial from Finland [23] with 54 women at high risk for gestational diabetes evaluated the effect of dietary therapy on gestational weight gain, gestational diabetes and metabolism. They found no significant effect on weight or clinical outcomes and changes in lipids were similar to ours. Ramsay et al. compared 23 obese and 24 lean pregnant women in a cross-sectional study and found a significant adverse lipid profile and hyperinsulinaemia in obese women [22]. Based on our findings of attenuated insulin resistance in the intervention group, it might be suggested that dietary changes and exercise can improve the metabolic changes in pregnancy and hence, at least theoretically, clinical complications such as gestational diabetes, pre-eclampsia and macrosomia [24,25]. However, as suggested in systematic reviews [26,27], the impact of antenatal intervention may not be sufficient to overcome the presence of pre-gestational obesity.. The intensity, duration and nature of such interventions should be further investigated to facilitate more efficient interventional strategies. Intervention prior to conception should also be considered. We found that insulin resistance increased in parallel with gestational weight gain and that adherence to Institute of Medicine criteria decreased the risk of Caesarean section and large for gestational age infants. The debate is ongoing concerning possible risk of small-for-gestational-age infants when suggesting weight loss or weight gain below 5 kg in
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
obese pregnant women. This concern was not confirmed in our study. The limited evidence in the literature to support clinical effects of a lifestyle intervention programme in obese pregnant women [26–28] might be attributable to difficulties in design and lack of power in studies performed so far. However, even minor improvements in metabolic status during pregnancy might influence the programming in utero, even if no beneficial effect is shown on short-term neonatal outcomes. Prospective follow-up studies in offspring from mothers who participated in antenatal intervention programmes should be performed to address this issue. Large interventional studies in obese women are ongoing [29,30] and should provide further evidence on how to manage obesity in relation to pregnancy.
Funding sources
The study was supported by Trygfonden, The Health Insurance Foundation (Sygekassernes Helsefond), The Faculty of Health Sciences, University of Southern Denmark, The Danish Diabetes Association, Odense University Hospital, The Novo Foundation, The Danish Medical Association Research Foundation, Aase og Ejnar Danielsens Fond, CMA Medico, Ferrosan A/S.
Competing interests
None declared.
Acknowledgements
Assistance with data collection and contributions to the study from midwife and physiotherapist P. Ingerslev and dietician C. Wolff (Department of Gynecology and Obstetrics, Aarhus University Hospital) and midwife C. Thomsen, dieticians B. Knold and R. Nestor, physiotherapist M. Bødker and D. Larsen (Department of Gynecology and Obstetrics, Odense University Hospital) is gratefully acknowledged. The authors acknowledge the editorial assistance of Prof. Ronald F. Lamont, University of Southern Denmark.
References 1 Nelson SM, Matthews P, Poston L. Maternal metabolism and obesity: modifiable determinants of pregnancy outcome. Hum Reprod Update 2010; 16: 255–275. 2 Ovesen P, Rasmussen S, Kesmodel U. Effect of prepregnancy maternal overweight and obesity on pregnancy outcome. Obstet Gynecol 2011; 118: 305–312. 3 Huda SS, Brodie LE, Sattar N. Obesity in pregnancy: prevalence and metabolic consequences. Semin Fetal Neonatal Med 2010; 15: 70–76. 4 Lain KY, Catalano PM. Metabolic changes in pregnancy. Clin Obstet Gynecol 2007; 50: 938–948.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
DIABETICMedicine
5 Catalano PM. Obesity, insulin resistance, and pregnancy outcome. Reproduction 2010; 140: 365–371. 6 Catalano PM, Tyzbir ED, Roman NM, Amini SB, Sims EA. Longitudinal changes in insulin release and insulin resistance in non-obese pregnant women. Am J Obstet Gynecol 1991; 165: 1667–1672. 7 Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U, Coustan DR et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008; 358: 1991–2002. 8 Vinter CA, Jensen DM, Ovesen P, Beck-Nielsen H, Jorgensen JS. The LiP (Lifestyle in Pregnancy) Study: a randomized controlled trial of lifestyle intervention in 360 obese pregnant women. Diabetes Care 2011; 34: 2502–2507. 9 Saltin B, Grimby G. Physiological analysis of middle-aged and old former athletes. Comparison with still active athletes of the same ages. Circulation 1968; 38: 1104–1115. 10 Rodjer L, Jonsdottir IH, Rosengren A, Bjorck L, Grimby G, Thelle DS et al. Self-reported leisure time physical activity: a useful assessment tool in everyday health care. BMC Public Health 2012; 12: 693. 11 Aadahl M, Zacho M, Linneberg A, Thuesen BH, Jorgensen T. Comparison of the Danish step test and the watt-max test for estimation of maximal oxygen uptake: the Health 2008 study. Eur J Prev Cardiol 2013; 20: 1088–1094. 12 Volund A. Conversion of insulin units to SI units. Am J Clin Nutr 1993; 58: 714–715. 13 Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 412–419. 14 Institute of Medicine. Weight Gain During Pregnancy: Re-examining the Guidelines. Washington, DC: The National Academy Press, 2009. 15 Metzger BE, Gabbe SG, Persson B, Buchanan TA, Catalano PA, Damm P et al. International Association of Diabetes and Pregnancy Study Group’s recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care 2010; 33: 676–682. 16 Catalano PM, Huston L, Amini SB, Kalhan SC. Longitudinal changes in glucose metabolism during pregnancy in obese women with normal glucose tolerance and gestational diabetes mellitus. Am J Obstet Gynecol 1999; 180: 903–916. 17 Kirwan JP, Huston-Presley L, Kalhan SC, Catalano PM. Clinically useful estimates of insulin sensitivity during pregnancy: validation studies in women with normal glucose tolerance and gestational diabetes mellitus. Diabetes Care 2001; 24: 1602–1607. 18 Ong MJ, Guelfi KJ, Hunter T, Wallman KE, Fournier PA, Newnham JP. Supervised home-based exercise may attenuate the decline of glucose tolerance in obese pregnant women. Diabetes Metab 2009; 35: 418–421. 19 Callaway LK, Colditz PB, Byrne NM, Lingwood BE, Rowlands IJ, Foxcroft K et al. Prevention of gestational diabetes: feasibility issues for an exercise intervention in obese pregnant women. Diabetes Care 2010; 33: 1457–1459. 20 Wolff S, Legarth J, Vangsgaard K, Toubro S, Astrup A. A randomized trial of the effects of dietary counseling on gestational weight gain and glucose metabolism in obese pregnant women. Int J Obes (Lond) 2008; 32: 495–501. 21 Oostdam N, van Poppel MN, Wouters MG, Eekhoff EM, Bekedam DJ, Kuchenbecker WK et al. No effect of the FitFor2 exercise programme on blood glucose, insulin sensitivity and birthweight in pregnant women who were overweight and at risk for gestational diabetes: results of a randomised controlled trial. BJOG 2012; 119: 1098–1107. 22 Ramsay JE, Ferrell WR, Crawford L, Wallace AM, Greer IA, Sattar N. Maternal obesity is associated with dysregulation of metabolic,
1329
DIABETICMedicine
23
24
25
26
27
28
vascular, and inflammatory pathways. J Clin Endocrinol Metab 2002; 87: 4231–4237. Korpi-Hyovalti E, Schwab U, Laaksonen DE, Linjama H, Heinonen S, Niskanen L. Effect of intensive counselling on the quality of dietary fats in pregnant women at high risk of gestational diabetes mellitus. Br J Nutr 2012; 108: 910–917. Chu SY, Callaghan WM, Kim SY, Schmid CH, Lau J, England LJ et al. Maternal obesity and risk of gestational diabetes mellitus. Diabetes Care 2007; 30: 2070–2076. Nohr EA, Vaeth M, Baker JL, Sorensen TI, Olsen J, Rasmussen KM. Combined associations of prepregnancy body mass index and gestational weight gain with the outcome of pregnancy. Am J Clin Nutr 2008; 87: 1750–1759. Dodd JM, Grivell RM, Crowther CA, Robinson JS. Antenatal interventions for overweight or obese pregnant women: a systematic review of randomised trials. BJOG 2010; 117: 1316–1326. Quinlivan JA, Julania S, Lam L. Antenatal dietary interventions in obese pregnant women to restrict gestational weight gain to Institute of Medicine recommendations: a meta-analysis. Obstet Gynecol 2011; 118: 1395–1401. Thangaratinam S, Rogozinska E, Jolly K, Glinkowski S, Duda W, Borowiack E et al. Interventions to reduce or prevent obesity in
1330
Metabolic effects of lifestyle intervention C. A. Vinter et al. pregnant women: a systematic review. Health Technol Assess 2012; 16: iii–iv, 1–191. 29 Poston L, Briley AL, Barr S, Bell R, Croker H, Coxon K et al. Developing a complex intervention for diet and activity behaviour change in obese pregnant women (the UPBEAT Trial); assessment of behavioural change and process evaluation in a pilot randomised controlled trial. BMC Pregnancy Childbirth 2013; 13: 148. 30 Jelsma JG, van Poppel MN, Galjaard S, Desoye G, Corcoy R, Devlieger R et al. DALI: Vitamin D and Lifestyle Intervention for gestational diabetes mellitus (GDM) prevention: a European multi-centre, randomised trial—study protocol. BMC Pregnancy Childbirth 2013; 13: 142.
Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Flow chart for participation. Appendix S1. Research design and methods.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
