Gestational Weight Gain Implications of an Antenatal Lifestyle Intervention to Cornelis and Hampus Örebro Studies in Medicine 146

ANN-KRISTIN RÖNNBERG

ANN-KRISTIN RÖNNBERG

Gestational Weight Gain Implications of an Antenatal Lifestyle Intervention Abstract

Ann-Kristin Rönnberg (2016): Gestational Weight Gain. Implications of an Antenatal Lifestyle Intervention. Örebro studies in Medicine 146.

Background: Excessive gestational weight gain (GWG) is common in developed countries and is associated with an increased risk of maternal and offspring morbidity. Evidence regarding efficacy and safety of antenatal lifestyle intervention is limited in terms of both systematic reviews and original trials. This thesis is based on the need to further explore this research area. Objectives: To assess and grade current evidence and evaluate short and long-term effects of an antenatal lifestyle intervention on women and their offspring Materials: Controlled trials of intervention published before August 2009 were systematically searched and reviewed. A randomized controlled trial (RCT) including 445 healthy women aged >18 years with a body mass index (BMI) ≥19 and ≤16 weeks pregnant and their offspring was performed during 2007-2015 in Örebro Region, Sweden. Methods: The Grading of Recommendations, Assessment, Development and Evaluation (GRADE) system was used for review. Our RCT (called the VIGA trial) compared standard care with a composite intervention consisting of education, application of a personalized weight graph, prescription of exercise and more frequent monitoring of weight. Standardized measures of weight and height in offspring waere analysed based on World Health Organization (WHO) Child Growth Standards. Results: Quality of evidence across the studies published pre-August 2009 was concluded to be very low. Our intervention significantly reduced mean GWG (kg) but the proportion of women with excessive GWG, according to recommendations, was not significantly reduced. Short- term postpartum weight retention (PPWR) was significantly lower after the intervention but no significant © Ann-Kristin Rönnberg, 2016 difference remained 1 year after delivery. Offspring mean BMI z-scores or proportion of obesity did not differ between study groups at either birth or age 5. Conclusions: Title: Gestational Weight Gain. Implications of an Antenatal Lifestyle Intervention. The antenatal lifestyle intervention reduced mean GWG and short-term PPWR but Publisher: Örebro University 2016 no long-term effects on maternal weight retention or offspring obesity were seen. www.oru.se/publikationer-avhandlingar Alternative modes and timing of intervention should be considered in future research. Reducing the prevalence of pre-conception obesity must still be considered the primary means to improve maternal and fetal outcome. Print: Örebro University, Repro 08/2016

ISSN 1652-4063 Keywords: Gestational weight gain, maternal health, pregnancy, prevention of obesi- ISBN 978-91-7529-148-2 ty, lifestyle intervention, childhood obesity.

Ann-Kristin Rönnberg, School of Health and Medical Sciences, Örebro University, SE-701 82 Örebro, Sweden, [email protected] Abstract

Ann-Kristin Rönnberg (2016): Gestational Weight Gain. Implications of an Antenatal Lifestyle Intervention. Örebro studies in Medicine 146.

ISSN 1652-4063 Keywords: Gestational weight gain, maternal health, pregnancy, prevention of obesi- ISBN 978-91-7529-148-2 ty, lifestyle intervention, childhood obesity.

Ann-Kristin Rönnberg, School of Health and Medical Sciences, Örebro University, SE-701 82 Örebro, Sweden, [email protected] LIST OF PUBLICATIONS TABLE OF CONTENTS

I Ronnberg AK, Nilsson K. Interventions during pregnancy to reduce exces- INTRODUCTION ...... 11 sive gestational weight gain: a systematic review assessing current clinical evidence using the Grading of Recommendations, Assessment, Develop- BACKGROUND ...... 14 ment and Evaluation (GRADE) system. BJOG 2010; 117:1327–1334. Composition and timing of weight gain during pregnancy ...... 14 Gestational weight gain from a Swedish perspective ...... 15 II Ronnberg AK, Ostlund I, Fadl H, Gottvall T, Nilsson K. Intervention Long-term maternal effects ...... 19 during pregnancy to reduce excessive gestational weight gain — a ran- Trans-generational effects ...... 20 domised controlled trial. BJOG 2014; DOI: 10.1111/1471-0528.13131. Lifestyle intervention in relation to pregnancy ...... 21 Reviewing scientific evidence – the GRADE system ...... 22 III Ronnberg AK, Hanson U, Ostlund I, Nilsson K. Effects on postpartum ETHICAL CONSIDERATIONS ...... 24 weight retention after antenatal lifestyle intervention – a secondary analysis from a randomized controlled trial. AOGS 2016; DOI: 10.1111/ AIMS...... 25 aogs.12910. MATERIALS AND METHODS ...... 26 Paper I ...... 26 IV Ronnberg AK, Hanson U, Nilsson K. Effects of an antenatal lifestyle in- Papers II-IV ...... 27 tervention on childhood obesity – a five year follow-up of a randomised Intervention ...... 27 controlled trial. Submitted Randomization ...... 29 Statistical methods, Papers II-IV ...... 29 Sample size ...... 30 Child Growth Standards applied in Paper IV ...... 30 All previously published papers were reproduced with permission from the publishers. RESULTS ...... 31 Paper I ...... 31 Conclusion Paper I ...... 34 Paper II ...... 34 Conclusion Paper II ...... 37 Paper III ...... 37 Conclusion Paper III ...... 40 Paper IV ...... 41 Conclusion Paper IV ...... 45 DISCUSSION ...... 46 Summary of main findings ...... 46 Strength and limitations of included studies ...... 46 Interpretation and implications of study results ...... 48 Future research recommendations ...... 50 CONCLUSION ...... 51 ACKNOWLEDGEMENTS ...... 52 LIST OF PUBLICATIONS TABLE OF CONTENTS

SUMMARY IN SWEDISH AGA Appropriate for gestational age SAMMANFATTNING PÅ SVENSKA ...... 60 BMI Body mass index Viktökning under graviditet (VIGA) ...... 60 CI Confidence interval APPENDIX ...... 62 CS Caesarean section FaR Fysisk aktivitet på recept (Akronym for prescribed physical activity in Swedish) FBW Fetal birth weight GDM Gestational diabetes mellitus GWG Gestational weight gain HT Hypertension IOM Institute of Medicine LGA Large for gestational age LMP Last menstrual period OR Odds ratio PE Preeclampsia PHT Pregnancy induced hypertension PP Postpartum PPWR Postpartum weight retention SGA Small for gestational age SD Standard deviation VIGA Vikt Intervention under Graviditet (Trial akronym in Swedish) WHO World Health Organization DEFINITIONS INTRODUCTION Pre-pregnancy weight The first measured weight in pregnancy if first visit Excessive weight gain during pregnancy is common in developed countries and has occurred at <16 weeks of pregnancy according to been linked to an increased risk of complications during pregnancy, delivery and the LMP or early ultrasound dating. postpartum period for both mother and child. High gestational weight gain (GWG) is primarily linked to morbidity associated with high fetal birth weight (FBW) (1) but Gestational weight gain Bodyweight at delivery minus bodyweight at women with excessive GWG are also more likely to have high postpartum weight <16 w of pregnancy. retention (PPWR) (2) and their offspring have been reported to have an increased risk of becoming overweight or obese (3, 4). An effective intervention in maternal health Postpartum weight retention Bodyweight at postpartum visit minus bodyweight care aimed at avoiding or limiting excessive GWG, could theoretically improve not at <16 w of pregnancy. only obstetric outcome but also long term-health in mother and child by reducing the risk of developing obesity. Preterm delivery Birth < 37 weeks’ gestation. The maternal and fetal risk associated with excessive weight gain however, has to be balanced against the risk associated with inadequate GWG. Low GWG or ges- Perinatal mortality Stillbirths and deaths during the first seven days tational weight loss (GWL) has been linked to negative obstetric outcome such as postpartum independent of gestational age. preterm birth, fetal growth restriction, and failure to initiate breast-feeding as well as increased infant mortality (5-7). Maternal pre-pregnancy BMI modifies the association between GWG and many of these health outcomes (Figure 1) so that determining the optimal range of GWG for women in different BMI categories, balancing the risks associated with low and high gain for mother and child, has proved to be a challenge. In 1990 the Institute of Medicine (IOM) issued the first evidence-based guidelines for GWG in relation to maternal pre-pregnancy BMI. Primarily it was concern for women with pre-pregnancy underweight and low GWG that spurred the creation of the first IOM guidelines (8). Promted by an increased prevalence of overweight and obesity in women of reproductive age and the observation that an increasing number of women were gaining excessively during pregnancy, the IOM reconvened and revised their guidelines in 2009, adding an upper limit for GWG in women with pre-pregnancy obesity and at the same time adopting the World Health Organization (WHO) classifications for BMI (9) (Table 1).

BMI GWG (kg) Mean rate of GWG (kg/week) 2nd and 3rd trimester <18.5 12.5-18 0.51 18.5-24.9 11.5-16 0.51 25-29.9 7-11.5 0.28 >30 5-9 0.22

Table 1. Institute of Medicine (IOM) recommendations of 2009 for gestational weight gain (GWG) based on maternal pre-pregnancy body mass index (BMI).

Ann-Kristin rönnberg Gestational Weight Gain 11 DEFINITIONS INTRODUCTION Pre-pregnancy weight The first measured weight in pregnancy if first visit Excessive weight gain during pregnancy is common in developed countries and has occurred at <16 weeks of pregnancy according to been linked to an increased risk of complications during pregnancy, delivery and the LMP or early ultrasound dating. postpartum period for both mother and child. High gestational weight gain (GWG) is primarily linked to morbidity associated with high fetal birth weight (FBW) (1) but Gestational weight gain Bodyweight at delivery minus bodyweight at women with excessive GWG are also more likely to have high postpartum weight <16 w of pregnancy. retention (PPWR) (2) and their offspring have been reported to have an increased risk of becoming overweight or obese (3, 4). An effective intervention in maternal health Postpartum weight retention Bodyweight at postpartum visit minus bodyweight care aimed at avoiding or limiting excessive GWG, could theoretically improve not at <16 w of pregnancy. only obstetric outcome but also long term-health in mother and child by reducing the risk of developing obesity. Preterm delivery Birth < 37 weeks’ gestation. The maternal and fetal risk associated with excessive weight gain however, has to be balanced against the risk associated with inadequate GWG. Low GWG or ges- Perinatal mortality Stillbirths and deaths during the first seven days tational weight loss (GWL) has been linked to negative obstetric outcome such as postpartum independent of gestational age. preterm birth, fetal growth restriction, and failure to initiate breast-feeding as well as increased infant mortality (5-7). Maternal pre-pregnancy BMI modifies the association between GWG and many of these health outcomes (Figure 1) so that determining the optimal range of GWG for women in different BMI categories, balancing the risks associated with low and high gain for mother and child, has proved to be a challenge. In 1990 the Institute of Medicine (IOM) issued the first evidence-based guidelines for GWG in relation to maternal pre-pregnancy BMI. Primarily it was concern for women with pre-pregnancy underweight and low GWG that spurred the creation of the first IOM guidelines (8). Promted by an increased prevalence of overweight and obesity in women of reproductive age and the observation that an increasing number of women were gaining excessively during pregnancy, the IOM reconvened and revised their guidelines in 2009, adding an upper limit for GWG in women with pre-pregnancy obesity and at the same time adopting the World Health Organization (WHO) classifications for BMI (9) (Table 1).

BMI GWG (kg) Mean rate of GWG (kg/week) 2nd and 3rd trimester <18.5 12.5-18 0.51 18.5-24.9 11.5-16 0.51 25-29.9 7-11.5 0.28 >30 5-9 0.22

Table 1. Institute of Medicine (IOM) recommendations of 2009 for gestational weight gain (GWG) based on maternal pre-pregnancy body mass index (BMI).

Ann-Kristin rönnberg Gestational Weight Gain 11 Gestational weight gain (GWG) within IOM ranges was consistently associated with Observational studies have continued to provide ambiguous results on the subject. good maternal and fetal outcomes. Evidence at the time was judged inadequate to Several large population-based cohort studies published after the release of the IOM provide specific guidelines by grade of obesity or to support recommendations on guidelines considered it safe to set more restrictive weight gain limitations for obese GWG < 5 kg or gestational weight loss (GWL) for obese wome women (11-14). A systematic review published in 2010 states that overweight and obese women who gain less weight than the ranges recommended by the IOM do 70% Underweight 70% Normal Weight

60% 60% not have an increased risk of having a low birth weight infant (15). Other researchers

50% 50% have argued that even the present IOM recommendations may be too restrictive for severely obese women and may be associated with increased rates of preterm births, 40% 40% small-for-gestational-age (SGA) infants and perinatal mortality (7). Future studies 30% 30% may provide sufficient evidence to lead to customized recommendations on GWG 20% 20% in relation not only to grade of pre-pregnancy obesity but also to individual maternal Absolute Adjusted Risk 10% 10% metabolic status and genetic disposition for obesity. 0% 0% < 5 kg 5-9 kg 10-15 kg 16-19 kg 20-24 kg ≥ 25 kg < 5 kg 5-9 kg 10-15 kg 16-19 kg 20-24 kg ≥25 kg An increasing awareness among obstetricians about the high prevalence of exces- Gestational Weight Gain Gestational Weight Gain sive GWG among Swedish women and the growing base of evidence regarding associations with negative short-term and long-term maternal and fetal outcome prompted this thesis which encompasses research in the field of antenatal lifestyle intervention 70% Overweight 70% Obese with the aim to limit excessive GWG. 60% 60%

50% 50%

40% 40%

30% 30%

20% 20%

Absolute Adjusted Risk 10% 10%

0% 0% < 5 kg 5-9 kg 10-15 kg 16-19 kg 20-24 kg ≥25 kg < 5 kg 5-9 kg 10-15 kg 16-19 kg 20-24 kg ≥ 25 kg Gestational Weight Gain Gestational Weight Gain

70% Extremely Obese Small-for-gestational-age 60% Large-for-gestational-age 50%

40% EmergencEmergencyy caesareaCesareann Delidelivery

30% Postpartutumm Wweeightightretention Retention

20%

Absolute Adjusted Risk 10%

0% < 5 kg 5-9 kg 10-15 kg 16-19 kg 20-24 kg ≥ 25 kg Gestational Weight Gain Figure 1. Gestational weight gain (GWG)-specific absolute risks for fetal and maternal outcome based on pre-pregnancy body mass index (BMI) (Weight gain during pregnancy: reexamining the guidelines. Washington, DC: National Academies Press; 2009, Appendix G, Reprinted with permission from publisher) (10).

12 Ann-Kristin rönnberg Gestational Weight Gain 13 ANN-KrIStIN röNNBerg Gestational Weight Gain Gestational weight gain (GWG) within IOM ranges was consistently associated with Observational studies have continued to provide ambiguous results on the subject. good maternal and fetal outcomes. Evidence at the time was judged inadequate to Several large population-based cohort studies published after the release of the IOM provide specific guidelines by grade of obesity or to support recommendations on guidelines considered it safe to set more restrictive weight gain limitations for obese GWG < 5 kg or gestational weight loss (GWL) for obese wome women (11-14). A systematic review published in 2010 states that overweight and obese women who gain less weight than the ranges recommended by the IOM do 70% Underweight 70% Normal Weight

60% 60% not have an increased risk of having a low birth weight infant (15). Other researchers

50% 50%

40% 40%

30% 30%

20% 20%

Absolute Adjusted Risk 10% 10%

0% 0% < 5 kg 5-9 kg 10-15 kg 16-19 kg 20-24 kg ≥25 kg < 5 kg 5-9 kg 10-15 kg 16-19 kg 20-24 kg ≥ 25 kg Gestational Weight Gain Gestational Weight Gain

70% Extremely Obese Small-for-gestational-age 60% Large-for-gestational-age 50%

40% EmergencEmergencyy caesareaCesareann Delidelivery

30% Postpartutumm Wweeightightretention Retention

20%

Absolute Adjusted Risk 10%

12 Ann-Kristin rönnberg Gestational Weight Gain 13 ANN-KrIStIN röNNBerg Gestational Weight Gain BACKGROUND of cardiovascular disease, and the metabolic syndrome (19). Further understanding of the composition of excess weight gained during pregnancy is important since the potential metabolic impact and contribution to long-term risk is likely to vary depending Composition and timing of weight gain during pregnancy on amount of fat mass accrued as well as the site of the fat depot (19). A certain amount of weight gain is expected during pregnancy and considered During the first trimester of pregnancy approximately 0.5-2 kg weight gain is ex- essen-tial to provide the nutritional support for fetal growth and also to prepare for pected (20). The initial gain is mainly due to early placental development and ex- lactation. The placenta, growing uterus and fetus, amniotic fluid, expansion of pansion of maternal blood volume. Fetal weight does not contribute significantly to maternal blood volume, breast tissue as well as maternal adipose tissue and weight gain in the first trimester. The potential difference in fetal and maternal out- extracellular fluid all contribute to total maternal GWG (Figure 2) (16). come depending on the timing of excessive weight gain is still unclear. Early excessive weight gain has been associated with impaired maternal glucose tolerance later 12 in pregnancy (21, 22) and greater infant adiposity at birth (22, 23) independent of total maternal GWG. Early excessive weight gain has also been suggested to be pre- 10 dictive of total GWG. The predictive value differs however, depending on pre-pregnancy BMI, according to a recent study (22). In that study, women with a normal 8 Fetus pre-pregnancy BMI had a 70% probability of excess total GWG if they gained excess Placenta weight already in first trimester, while overweight and obese women had a 90% prob-

Kg ability of excess total GWG (22) if they gained excessively early in pregnancy. Fur- Amniotic uid 6 ther knowledge on the relevance of early GWG has implications for pre-conception Extracellular uid counselling since very few women are admitted to maternal health care and advised 4 Other tissue (fat/lean) on GWG in the first trimester of pregnancy.

Uterus and Breast Gestational weight gain from a Swedish perspective 2 Blood The IOM guidelines have not been systematically implemented in Swedish maternal health care. Regional differences in addressing issues of maternal weight gain are re- 8 w 20 w 28 w 36 w 40 w ported and a variety of efforts are being made to help women achieve healthy weight Figure 2. Composites of gestational weight gain (GWG) (kg). goals before, during and after their pregnancies. According to data from the Swedish Pregnancy Register, excessive GWG is main- Excess weight that is not attributed to products of conception contributes to an in- ly prevalent among women with pre-pregnancy overweight (BMI 25-29) or obesity crease in maternal energy stores (17). Knowledge about the degree to which pregnant Grade I (BMI 30-35) . In 2013, mean GWG among women with pre-pregnancy BMI women store excess energy in adipose tissue, and the sites where they do so, is 25-29 was above IOM recommendations in all (21 out of 21) Swedish health care still limited but excess GWG has primarily been associated with maternal fat mass regions. In all but one region, mean GWG among obese (BMI >30) women was accrual and not with lean mass accrual (18). Studies have so far focused mainly on above IOM recommendations. Despite various efforts to limit excessive GWG there women with pre-pregnancy overweight and obesity, but metabolic rate as well as the has been a tendency during 2013-2015 towards increasing GWG in all BMI groups disposition to accumulate fat mass may however not be similar among normal or except underweight women (24). underweight women. The maternal gain in adipose tissue can be deposited into visceral (central) or sub- cutaneous (peripheral) adipose tissue depots. Factors influencing site of fat deposition are likely to be similar with non-pregnant women but evidence here is also still lacking. Site of fat accrual is considered of importance because fat in visceral depots and around central organs such as the liver is associated with insulin resistance, risk

14 Ann-Kristin rönnberg Gestational Weight Gain 15 ANN-KrIStIN röNNBerg Gestational Weight Gain BACKGROUND of cardiovascular disease, and the metabolic syndrome (19). Further understanding of the composition of excess weight gained during pregnancy is important since the potential metabolic impact and contribution to long-term risk is likely to vary depending Composition and timing of weight gain during pregnancy on amount of fat mass accrued as well as the site of the fat depot (19). A certain amount of weight gain is expected during pregnancy and considered During the first trimester of pregnancy approximately 0.5-2 kg weight gain is ex- essen-tial to provide the nutritional support for fetal growth and also to prepare for pected (20). The initial gain is mainly due to early placental development and ex- lactation. The placenta, growing uterus and fetus, amniotic fluid, expansion of pansion of maternal blood volume. Fetal weight does not contribute significantly to maternal blood volume, breast tissue as well as maternal adipose tissue and weight gain in the first trimester. The potential difference in fetal and maternal out- extracellular fluid all contribute to total maternal GWG (Figure 2) (16). come depending on the timing of excessive weight gain is still unclear. Early excessive weight gain has been associated with impaired maternal glucose tolerance later 12 in pregnancy (21, 22) and greater infant adiposity at birth (22, 23) independent of total maternal GWG. Early excessive weight gain has also been suggested to be pre- 10 dictive of total GWG. The predictive value differs however, depending on pre-pregnancy BMI, according to a recent study (22). In that study, women with a normal 8 Fetus pre-pregnancy BMI had a 70% probability of excess total GWG if they gained excess Placenta weight already in first trimester, while overweight and obese women had a 90% prob-

14 Ann-Kristin rönnberg Gestational Weight Gain 15 ANN-KrIStIN röNNBerg Gestational Weight Gain 15 13.4 12.7 12.8*

12 10.9*

8.8 9 7.1

Mean GWG (kg) 3

0 Obesity grade I Underweight Normalweight Overweight Obesity grade III Obesity grade II

Figure 3. Mean gestational weight gain (GWG) (kg) by pre-pregnancy body mass index (BMI) category, in 21 Swedish health care regions (Swedish Pregnancy Register 2014) (24). *Mean GWG above IOM recommendations. Among pregnant women with Grade II-III obesity (BMI >35) the majority of health care regions reported a mean GWG within recommended ranges (5-9 kg). Inadequate GWG was reported among underweight women in 9 out of 21 regions, setting focus on different issues needing to be addressed when discussing implementation of lifestyle intervention regarding GWG among underweight women. Johansson et al recently published z-score charts of GWG for gestational age and maternal BMI using data from pregnancies in a geographically defined area in south- ern Sweden (Stockholm and Gotland) (25).

Figure 4. Smoothed percentiles and standard deviations (SDs), of gestational weight gain (GWG) among 141,767 women who gave birth in the Stockholm and Gotland regions of Sweden in 2008–2014. Reported by pre-pregnancy BMI group and with Institute of Medicine (IOM) recommendations on range of GWG in shaded areas (25). (Reprinted with permission from publisher

16 Ann-Kristin rönnberg Gestational Weight Gain 17 ANN-KrIStIN röNNBerg Gestational Weight Gain 15 13.4 12.7 12.8*

12 10.9*

8.8 9 7.1

Mean GWG (kg) 3

0 Obesity grade I Underweight Normalweight Overweight Obesity grade III Obesity grade II

16 Ann-Kristin rönnberg Gestational Weight Gain 17 ANN-KrIStIN röNNBerg Gestational Weight Gain Median weight gains in this Swedish population were higher than IOM recommenda- Long-term maternal effects tions for many BMI categories (from overweight to obesity Grade II). Median GWG Excessive GWG can affect the woman’s subsequent weight-related health. There are among women with obesity Grade III (BMI >40) was however, within the IOM rec- a number of observational studies linking excessive GWG to excessive PPWR (26) ommendations, but 10% of women in this group gained essentially no weight (0.5 kg) and increased risk of subsequent obesity (27). An extensive meta-analysis by Neh- during pregnancy (Figure 4). Among women with normal BMI, the median GWG ring et al concluded that excess weight gained in pregnancy was still retained >15 was within the IOM recommendations, whereas in underweight women, the median years postpartum (28). The degree of weight retention is, however, highly variable was at the lower limit. The observed distributions of GWG in this Swedish population and women with pre-pregnancy overweight or obesity are considered more likely to were considerably broader than the range of the IOM recommendations. experience a high degree of PPWR (15, 29). The proportion of Swedish women with overweight or obesity (BMI>25) at ad- Entering a potential subsequent pregnancy with a pre-pregnancy BMI >30 has im- mission to maternal health care has increased from 25.4% in 1992 to 38.1% in 2014. plications for a number of maternal and fetal obstetric outcomes (30). Complications The study area for the VIGA trial, discussed in this thesis, had similar distribution during pregnancy, delivery and postpartum, associated with maternal obesity will not of prepregnancy BMI at admission to maternal health care during the study period be covered in this thesis other than to be mentioned here as a consequence of previous (Figure 5). excessive GWG. The long-term impact of PPWR on subsequent pregnancies is, how- 50% 48% ever, an important factor motivating intervention to reach GWG within recommended range among women entering first pregnancy with normal BMI.

40% Excessiv GWG Second Long during first pregnancy term pregnancy 30% 26%

 Prepregnancy BMI  Prepregancy BMI 20%  PPWR  Fat mass  Excessiv GWG  Fat mass  Visceral fat  Fat mass  Visceral fat  LGA  Visceral fat  Type 2 DM  GDM  CVD 10% 9% 8%  CS  GDM  PE/HT  Metabolic 5%  Hyper- syndrome cholesterolemia  LGA <20 20-24,9 25-29,9 30-35 >35,1  CS  PPWR BMI  IUFD

Figure 5. Distribution of maternal pre-pregnancy body mass index (BMI) in the study area at the start of recruitment to the VIGA trial in 2008. Figure 6. Consequences of excessive gestational weight gain (GWG) in women of reproductive age over the course of multiple pregnancies. Weight gain triggered by consecutive pregnancies contributes to the increased prevalence of overweight and obesity in reproductive age and postmenopausal women (27, 31, 32), and metabolic and/or behavioural changes that occur during the pregnancy

18 Ann-Kristin rönnberg Gestational Weight Gain 19 ANN-KrIStIN röNNBerg Gestational Weight Gain Median weight gains in this Swedish population were higher than IOM recommenda- Long-term maternal effects tions for many BMI categories (from overweight to obesity Grade II). Median GWG Excessive GWG can affect the woman’s subsequent weight-related health. There are among women with obesity Grade III (BMI >40) was however, within the IOM rec- a number of observational studies linking excessive GWG to excessive PPWR (26) ommendations, but 10% of women in this group gained essentially no weight (0.5 kg) and increased risk of subsequent obesity (27). An extensive meta-analysis by Neh- during pregnancy (Figure 4). Among women with normal BMI, the median GWG ring et al concluded that excess weight gained in pregnancy was still retained >15 was within the IOM recommendations, whereas in underweight women, the median years postpartum (28). The degree of weight retention is, however, highly variable was at the lower limit. The observed distributions of GWG in this Swedish population and women with pre-pregnancy overweight or obesity are considered more likely to were considerably broader than the range of the IOM recommendations. experience a high degree of PPWR (15, 29). The proportion of Swedish women with overweight or obesity (BMI>25) at ad- Entering a potential subsequent pregnancy with a pre-pregnancy BMI >30 has im- mission to maternal health care has increased from 25.4% in 1992 to 38.1% in 2014. plications for a number of maternal and fetal obstetric outcomes (30). Complications The study area for the VIGA trial, discussed in this thesis, had similar distribution during pregnancy, delivery and postpartum, associated with maternal obesity will not of prepregnancy BMI at admission to maternal health care during the study period be covered in this thesis other than to be mentioned here as a consequence of previous (Figure 5). excessive GWG. The long-term impact of PPWR on subsequent pregnancies is, how- 50% 48% ever, an important factor motivating intervention to reach GWG within recommended range among women entering first pregnancy with normal BMI.

40% Excessiv GWG Second Long during first pregnancy term pregnancy 30% 26%

18 Ann-Kristin rönnberg Gestational Weight Gain 19 ANN-KrIStIN röNNBerg Gestational Weight Gain are likely to play a role in the weight status of mothers postpartum and in future life. maternal GWG could therefore have a potential for positive long-term health effects The risk of becoming overweight or obese seems to be even greater when exces- for the offspring. sive GWG and/or PPWR occurs in successive pregnancies and when combined with short inter-pregnancy intervals (33, 34). In recent data presented by Bogaerts et al, Lifestyle intervention in relation to pregnancy weight retention of ≥2 BMI units between first and second pregnancy was associated Pregnancy has been proposed to be an optimal time for implementing positive chang- with an increased risk of gestational diabetes mellitus (GDM), pregnancy induced es in lifestyle behaviours. Multiple visits to maternal health care during pregnancy hypertension (PHT), caesarean section (CS) and large for gestational age (LGA) off- present many opportunities for different types of motivational intervention and infor- spring. These findings were not dependent on the woman’s previous BMI but applied mation about the importance of healthy lifestyle for both mother and child. Positive also to women still within normal BMI ranges. This is concurrent with earlier data results in achieving smoking cessation have spurred maternal health care staff to ad- by Villamor et al (2006) linking inter-pregnancy weight changes of >3 BMI units to dress other health-related factors such as diet and PA as well as screening for alcohol similar negative obstetric outcomes (35). abuse and domestic violence during pregnancy. Diet and PA are the main elements of The association between maternal inter-pregnancy weight gain and risk of still- weight control during pregnancy since the option of pharmaceutical or surgical meth- birth and infant mortality has been analysed in data from the Swedish Medical Birth ods is limited due to issues of safety. An exception is Metformin, which has been tried Register where inter-pregnancy weight gain >4 BMI units was associated to an in- among obese women without diabetes, with variable results: one study showed signif- reased risk of stillbirth regardless of BMI status at first pregnancy. An increased risk icant reduction of GWG compared with placebo among obese women, but no effect of infant mortality was seen among offspring to women with normal weight at first on FBW (39), and another concluded that there was no effect on GWG or FBW (40). pregnancy and inter-pregnancy gain >2 BMI units. Moreover, in overweight women, The optimal mode for effective intervention regarding GWG as well as the optimal weight loss between pregnancies reduced the risk of neonatal mortality among second timing is still unclear and this lack of evidence was also the rationale for this thesis. A offspring. These findings further support that inter-pregnancy weight gain should be review of antenatal interventions, published in 2009, will be presented as Paper I (41) prevented and that weight loss should be promoted before second pregnancy in over- in this thesis. A more recent meta-analysis has been published by the Cochrane Col- weight women (36). laboration in 2015 (42), concluding that high-quality evidence indicates that diet or exercise, or both combined, during pregnancy can reduce the risk of excessive GWG. Trans-generational effects Relevant outcomes such as risk of CS, macrosomia, preterm delivery, pre-eclampsia The increasing prevalence of childhood obesity in developed countries is considered and poor neonatal outcome were however, not significantly reduced by interventions a major threat to future public health. Effective early preventive efforts are important included in the Cohrane review. No significant difference in effect on GWG between as health consequences of obesity already begin during childhood and treatment of diet and PA intervention was detected. already established obesity has shown limited effect (37). Besides physical activity Considering that pre-pregnancy maternal BMI is considered the most important (PA), diet, parental BMI, environmental factors and genetics, the intrauterine envi- factor determining risk of adverse events during pregnancy, delivery and postpar- ronment is considered to influence the risk of developing childhood obesity. Exces- tum, an effective pre-conception intervention would be preferable. The importance sive maternal GWG has been associated not only with significantly higher FBW, but of entering pregnancy with a healthy weight cannot be overstated. No randomized also with childhood overweight/obesity (3, 4). Whether the association is explained controlled trials (RCTs) that assessed the effect of pre-conception health programmes by direct intra-uterine causal mechanisms or by environmental, lifestyle-related or and interventions in overweight and obese women with the aim of improving preg- genetic characteristics is still unclear. nancy outcomes was, however, found in a recently published Cochrane search (43) . Potential intra-uterine programming effects of excessive maternal GWG should Until the effectiveness of pre-conception health programmes and interventions can be be considered. The fetal overnutrition hypothesis suggests that increased placental established, no strong recommendations can be made. transfer of nutrients to the developing fetus in women with excessive GWG may subsequently affect fetal development, fetal fat deposition and the development of the hypothalamic–endocrine system, which controls appetite and energy metabolism (38). This may theoretically predispose offspring to a greater risk of adverse obesity-related health outcomes in later life. Effective strategies to avoid or limit excessive

20 Ann-Kristin rönnberg Gestational Weight Gain 21 ANN-KrIStIN röNNBerg Gestational Weight Gain are likely to play a role in the weight status of mothers postpartum and in future life. maternal GWG could therefore have a potential for positive long-term health effects The risk of becoming overweight or obese seems to be even greater when exces- for the offspring. sive GWG and/or PPWR occurs in successive pregnancies and when combined with short inter-pregnancy intervals (33, 34). In recent data presented by Bogaerts et al, Lifestyle intervention in relation to pregnancy weight retention of ≥2 BMI units between first and second pregnancy was associated Pregnancy has been proposed to be an optimal time for implementing positive chang- with an increased risk of gestational diabetes mellitus (GDM), pregnancy induced es in lifestyle behaviours. Multiple visits to maternal health care during pregnancy hypertension (PHT), caesarean section (CS) and large for gestational age (LGA) off- present many opportunities for different types of motivational intervention and infor- spring. These findings were not dependent on the woman’s previous BMI but applied mation about the importance of healthy lifestyle for both mother and child. Positive also to women still within normal BMI ranges. This is concurrent with earlier data results in achieving smoking cessation have spurred maternal health care staff to ad- by Villamor et al (2006) linking inter-pregnancy weight changes of >3 BMI units to dress other health-related factors such as diet and PA as well as screening for alcohol similar negative obstetric outcomes (35). abuse and domestic violence during pregnancy. Diet and PA are the main elements of The association between maternal inter-pregnancy weight gain and risk of still- weight control during pregnancy since the option of pharmaceutical or surgical meth- birth and infant mortality has been analysed in data from the Swedish Medical Birth ods is limited due to issues of safety. An exception is Metformin, which has been tried Register where inter-pregnancy weight gain >4 BMI units was associated to an in- among obese women without diabetes, with variable results: one study showed signif- reased risk of stillbirth regardless of BMI status at first pregnancy. An increased risk icant reduction of GWG compared with placebo among obese women, but no effect of infant mortality was seen among offspring to women with normal weight at first on FBW (39), and another concluded that there was no effect on GWG or FBW (40). pregnancy and inter-pregnancy gain >2 BMI units. Moreover, in overweight women, The optimal mode for effective intervention regarding GWG as well as the optimal weight loss between pregnancies reduced the risk of neonatal mortality among second timing is still unclear and this lack of evidence was also the rationale for this thesis. A offspring. These findings further support that inter-pregnancy weight gain should be review of antenatal interventions, published in 2009, will be presented as Paper I (41) prevented and that weight loss should be promoted before second pregnancy in over- in this thesis. A more recent meta-analysis has been published by the Cochrane Col- weight women (36). laboration in 2015 (42), concluding that high-quality evidence indicates that diet or exercise, or both combined, during pregnancy can reduce the risk of excessive GWG. Trans-generational effects Relevant outcomes such as risk of CS, macrosomia, preterm delivery, pre-eclampsia The increasing prevalence of childhood obesity in developed countries is considered and poor neonatal outcome were however, not significantly reduced by interventions a major threat to future public health. Effective early preventive efforts are important included in the Cohrane review. No significant difference in effect on GWG between as health consequences of obesity already begin during childhood and treatment of diet and PA intervention was detected. already established obesity has shown limited effect (37). Besides physical activity Considering that pre-pregnancy maternal BMI is considered the most important (PA), diet, parental BMI, environmental factors and genetics, the intrauterine envi- factor determining risk of adverse events during pregnancy, delivery and postpar- ronment is considered to influence the risk of developing childhood obesity. Exces- tum, an effective pre-conception intervention would be preferable. The importance sive maternal GWG has been associated not only with significantly higher FBW, but of entering pregnancy with a healthy weight cannot be overstated. No randomized also with childhood overweight/obesity (3, 4). Whether the association is explained controlled trials (RCTs) that assessed the effect of pre-conception health programmes by direct intra-uterine causal mechanisms or by environmental, lifestyle-related or and interventions in overweight and obese women with the aim of improving preg- genetic characteristics is still unclear. nancy outcomes was, however, found in a recently published Cochrane search (43) . Potential intra-uterine programming effects of excessive maternal GWG should Until the effectiveness of pre-conception health programmes and interventions can be be considered. The fetal overnutrition hypothesis suggests that increased placental established, no strong recommendations can be made. transfer of nutrients to the developing fetus in women with excessive GWG may subsequently affect fetal development, fetal fat deposition and the development of the hypothalamic–endocrine system, which controls appetite and energy metabolism (38). This may theoretically predispose offspring to a greater risk of adverse obesity-related health outcomes in later life. Effective strategies to avoid or limit excessive

20 Ann-Kristin rönnberg Gestational Weight Gain 21 ANN-KrIStIN röNNBerg Gestational Weight Gain collaboration of primarily methodologists, and introduced in 2004 (47). The GRADE working group has developed a common, sensible and transparent approach to grading quality (or certainty) of evidence and strength of recommendations. Many international organizations have provided input into the development of the GRADE system which is now considered an international standard in guideline development. The system applies a systematic approach to judgments about the quality of evi- Preconceptional Antenatal Postpartum dence and the strength of recommendations (47, 48) taking into account study design, intervention? intervention? intervention? quality, consistency and directness. Strength of recommendation is determined by the balance between desirable and undesirable consequences of alternative management strategies, quality of evidence, variability in values and preferences and resource use. The strength of recommendations reflects the extent to which we can be confident that the desirable effects of an intervention outweigh the undesirable effects across the range of patients for whom the recommendations are intended.

Quality of evidence Comment Figure 7. Alternative timing for lifestyle intervention in relation to gestation. High quality Further research is very unlikely to change our confi- ✚✚✚✚ dence in the estimate of the effect. An alternative time for intervention is post-delivery/inter-pregnancy. The effect of PA Moderate quality Further research is likely to have an important impact ✚✚✚ on our confidence in the estimate of the effect and may strategies on weight loss in women postpartum was reviewed by Nascimento et al in change the estimate. 2014 (44) and their meta-analysis as well as results presented by Van der Pligt et al in Low quality Further research is very likely to have an important im- 2013 (45) showed significant effects of lifestyle interventions initiated post-delivery ✚✚ pact on our confidence in the estimate of the effect and is on weight loss. Interventions during the postpartum period may have the additional likely to change the estimate. benefit of focusing on a period when many women report high motivation for weight Very low qualty ✚ Any estimate of the effect is very uncertain. loss and when the potential risk of negative effects on intra-uterine fetal growth is no Strength of recommendation longer an issue. One Cochrane review on postpartum intervention concludes that both diet and exercise combined and diet alone helped women to lose weight after child- Strong recommendation for using an intervention ⬆⬆ birth (46). However, improved maternal cardiorespiratory fitness and preservation of Weak recommendation for using an intervention ⬆? fat-free mass indicate that composite interventions including exercise are preferred. Weak recommendation against using an intervention The postpartum interventions seemed not to affect breastfeeding performance but ⬇? more evidence is needed to confirm whether diet and/or exercise, have negative ef- Strong recommendation against using an intervention ⬇⬇ fects on mother or child. Figure 8. Grading quality of evidence and strength of recommendation, according to the Reviewing scientific evidence – the GRADE system Grading of Recommendations, Assessment, Development and Evaluation (GRADE) system. Introducing new treatments, medical technology or interventions in healthcare is always characterized by some uncertainty. The base of evidence considered is often heterogenic in quality and provides results of varying coherence. The importance of methods for systematic evaluation of published results has been increasingly acknowledged in the research community and among decision makers during the last three decades. Several more or less complex systems for grading of evidence have been constructed and applied. The need for an internationally accepted system led to development of the GRADE system, originating from an informal international

22 Ann-Kristin rönnberg Gestational Weight Gain 23 ANN-KrIStIN röNNBerg Gestational Weight Gain collaboration of primarily methodologists, and introduced in 2004 (47). The GRADE working group has developed a common, sensible and transparent approach to grading quality (or certainty) of evidence and strength of recommendations. Many international organizations have provided input into the development of the GRADE system which is now considered an international standard in guideline development. The system applies a systematic approach to judgments about the quality of evi- Preconceptional Antenatal Postpartum dence and the strength of recommendations (47, 48) taking into account study design, intervention? intervention? intervention? quality, consistency and directness. Strength of recommendation is determined by the balance between desirable and undesirable consequences of alternative management strategies, quality of evidence, variability in values and preferences and resource use. The strength of recommendations reflects the extent to which we can be confident that the desirable effects of an intervention outweigh the undesirable effects across the range of patients for whom the recommendations are intended.

22 Ann-Kristin rönnberg Gestational Weight Gain 23 ANN-KrIStIN röNNBerg Gestational Weight Gain ETHICAL CONSIDERATIONS AIMS Before implementing an interventional programme potential unintended consequenc- The overall aim of this thesis was to systematically review and grade current scientific es on psychosocial wellbeing, equality, social and cultural values, informed choice, evidence regarding lifestyle interventions aimed at reducing excessive GWG and to privacy and the attributions of personal responsibilities and liberty should be evalu- perform and analyse effects of an original RCT of an antenatal intervention on out- ated. comes related to GWG, PPWR and offspring obesity at birth and at age 5. Ethical aspects involved in programmes to prevent excessive GWG largely over- lap with ethical aspects of prevention of other types of unhealthy behaviour such Paper I The primary objective was to determine whether published controlled as smoking. Attitudes as to whether, and to what extent, pregnant women have an trials of intervention aimed at reducing excessive GWG were of sufficient extended duty to behave in a healthy manner should be addressed among health quality and provided sufficient data to enable evidence-based recommen- care providers, as should possible harmful effects of projecting a message of guilt to dations to be developed for clinical practice in antenatal care. women not adherent to recommendations. That obesity is a social stigma is widely acknowledged in Sweden and an intervention specifically aimed at this population Paper II The primary objective was to evaluate whether a feasible, low-cost ante- would require additional consideration of the interaction between health care provid- natal lifestyle intervention could reduce the proportion of women gain- ers and the obese pregnant woman. ing weight above the Institute of Medicine (IOM) recommendations on The VIGA trial was approved by the Regional Board of Research Ethics in Up- GWG, compared with standard maternity care. Mean GWG (kg) was also psala, Sweden, under registration number 2007/031 on 21 February 2007. Written, compared between groups and a sub-analysis regarding differences in informed consent was obtained from all participants. outcome related to pre-pregnancy BMI and parity was performed as a secondary aim.

Paper III The primary objective was to evaluate whether an antenatal lifestyle intervention could reduce short- and/or long-term maternal PPWR compared with standard maternity care. Risk estimates for excessive weight retention (>5 kg) 1 year after delivery were analysed as a secondary aim.

Paper IV The primary objective was to evaluate whether an antenatal lifestyle intervention could affect offspring obesity at birth or age 5. Mean BMI z-scores and proportion (%) of over- and undernutrition (BMI z-score >+2 SD) in offspring was compared between groups at birth and age 5. Risk estimates for offspring obesity at age 5 were analysed with regard to maternal GWG and maternal pre-pregnancy BMI as a secondary aim.

24 Ann-Kristin rönnberg Gestational Weight Gain 25 ANN-KrIStIN röNNBerg Gestational Weight Gain ETHICAL CONSIDERATIONS AIMS Before implementing an interventional programme potential unintended consequenc- The overall aim of this thesis was to systematically review and grade current scientific es on psychosocial wellbeing, equality, social and cultural values, informed choice, evidence regarding lifestyle interventions aimed at reducing excessive GWG and to privacy and the attributions of personal responsibilities and liberty should be evalu- perform and analyse effects of an original RCT of an antenatal intervention on out- ated. comes related to GWG, PPWR and offspring obesity at birth and at age 5. Ethical aspects involved in programmes to prevent excessive GWG largely over- lap with ethical aspects of prevention of other types of unhealthy behaviour such Paper I The primary objective was to determine whether published controlled as smoking. Attitudes as to whether, and to what extent, pregnant women have an trials of intervention aimed at reducing excessive GWG were of sufficient extended duty to behave in a healthy manner should be addressed among health quality and provided sufficient data to enable evidence-based recommen- care providers, as should possible harmful effects of projecting a message of guilt to dations to be developed for clinical practice in antenatal care. women not adherent to recommendations. That obesity is a social stigma is widely acknowledged in Sweden and an intervention specifically aimed at this population Paper II The primary objective was to evaluate whether a feasible, low-cost ante- would require additional consideration of the interaction between health care provid- natal lifestyle intervention could reduce the proportion of women gain- ers and the obese pregnant woman. ing weight above the Institute of Medicine (IOM) recommendations on The VIGA trial was approved by the Regional Board of Research Ethics in Up- GWG, compared with standard maternity care. Mean GWG (kg) was also psala, Sweden, under registration number 2007/031 on 21 February 2007. Written, compared between groups and a sub-analysis regarding differences in informed consent was obtained from all participants. outcome related to pre-pregnancy BMI and parity was performed as a secondary aim.

24 Ann-Kristin rönnberg Gestational Weight Gain 25 ANN-KrIStIN röNNBerg Gestational Weight Gain MATERIALS AND METHODS Papers II-IV Women were included in the VIGA trial between 2007 and 2010. The research was Study Design Method Primary outcome performed as a RCT with no blinding to treatment. Consecutive randomization was I Systematic Literature study using the Grade of current quality of review GRADE method. evidence and strength of recom- applied after written, informed consent was obtained. Women, stratified by BMI cat- mendations. egory (normal, overweight, obese) were randomly assigned to the standard care con- II Randomized Prescribed PA, education on Proportion of women with exces- trol group or the lifestyle intervention. Eligible participants were all pregnant women controlled trial IOM recommendations, per- sive GWG (%) in relation to study of intervention sonalised weight-graph and group. Mean GWG (kg) in relation aged >18 who signed in for maternity care in Orebro Region in Sweden at <16 weeks increased body weight meas- to study group. pregnancy according to last menstrual period (LMP) or an early ultrasound dating. urements. Women with a previous medical history including treatment of an eating disorder III Secondary anal- Prescribed PA, education on Mean PPWR (kg) in relation to ysis of a RCT of IOM recommendations, per- study group and time after de- or with a history of having a growth-restricted infant were excluded from the study intervention sonalized weight-graph and livery. before randomization. We also excluded women with existing chronic illness, who increased body weight measurements. required primarily specialized maternity care during pregnancy. Additional criteria for exclusion were inadequate knowledge of Swedish, BMI <19 at first antenatal visit, IV Secondary anal- Antenatal maternal lifestyle Offspring mean BMI z-score at and multiple pregnancy. The first offspring was born in 2007 and the last offspring ysis of a RCT of intervention. Standardized birth and age 5 in relation to intervention measure of weight and height study group. Proportion of over- reached age 5 in 2015. All offspring to women analysed for GWG in the original in offspring within child health and undernutrition (%) at birth study (Paper II) were included in the follow-up study of offspring obesity (Paper IV). care. and age 5 in relation to study group. Intervention The intervention programme consisted of individual education/information about the BMI = body mass index; GRADE = Grading of Recommendations, Assessment, Development IOM guidelines for recommended GWG, according to pre-pregnancy BMI category and Evaluation; GWG = gestational weight gain; IOM = Institute of Medicine; PPWR = at first antenatal visit. Since the study period started in 2007 the IOM guidelines from postpartum weight retention; RCT = randomized controlled trial. 1990 were applied throughout the study (Table 2).

Paper I Pre-pregnancy BMI Recommended GWG (kg) Underweight: <19.8 12.5–18.0 A literature search was conducted in the scientific databases Pub-Med, Cochrane Li- Normal weight: 19.8-26.0 11.5–16.0 brary, CINAHL and PEDro, and the reference lists of relevant articles were reviewed. The literature search was concluded on 15 August 2009. All RCTs were considered Overweight: 26.1–29.0 7.0–11.5 for inclusion. As the number of published RCTs was limited, all non-randomized Obese: >29.0 >6.8–11.5* intervention studies that included a control group were also considered for inclusion. Table 2. Institute of Medicine (IOM) 1990 recommendations on gestational weight gain Systematic reviews were searched and examined in order to identify additional orig- (GWG) depending on pre-pregnancy body mass index (BMI) (8). inal studies. *No upper limit for obese women was established in the 1990 guidelines. In the study proto- Two reviewers independently assessed the quality of the design, methods and re- col the same upper limit as for overweight women was applied. sults of all included articles. Data were extracted from articles using a structured data collection form and classified using GRADE (47). The GRADE system offers four The education was supplemented by a personalized graph showing the recommended levels of evidence quality: high, moderate, low, and very low (Figure 8). Quality may interval of weight gain and allowing the woman, together with her midwife, to follow be downgraded as a result of limitations in study design or implementation, impreci- her own weight gain during pregnancy (Figure 9). Maternal weight was measured sion of estimates (wide confidence interval (CI)), variability in results, indirectness of without shoes and outdoor clothing and discussed at each antenatal visit. Measured evidence or publication bias. Quality may be upgraded due to a very large magnitude weights were recorded on the graph as well as in the study case record files. of effect, or presence of a dose-response gradient. It may also be upgraded if all plau- The midwife issued written formalized prescriptions of PA (swedish acronym = sible biases would reduce an apparent treatment effect. FaR). The general recommendation for PA during pregnancy was a moderate level

26 Ann-Kristin rönnberg Gestational Weight Gain 27 ANN-KrIStIN röNNBerg Gestational Weight Gain MATERIALS AND METHODS Papers II-IV Women were included in the VIGA trial between 2007 and 2010. The research was Study Design Method Primary outcome performed as a RCT with no blinding to treatment. Consecutive randomization was I Systematic Literature study using the Grade of current quality of review GRADE method. evidence and strength of recom- applied after written, informed consent was obtained. Women, stratified by BMI cat- mendations. egory (normal, overweight, obese) were randomly assigned to the standard care con- II Randomized Prescribed PA, education on Proportion of women with exces- trol group or the lifestyle intervention. Eligible participants were all pregnant women controlled trial IOM recommendations, per- sive GWG (%) in relation to study of intervention sonalised weight-graph and group. Mean GWG (kg) in relation aged >18 who signed in for maternity care in Orebro Region in Sweden at <16 weeks increased body weight meas- to study group. pregnancy according to last menstrual period (LMP) or an early ultrasound dating. urements. Women with a previous medical history including treatment of an eating disorder III Secondary anal- Prescribed PA, education on Mean PPWR (kg) in relation to ysis of a RCT of IOM recommendations, per- study group and time after de- or with a history of having a growth-restricted infant were excluded from the study intervention sonalized weight-graph and livery. before randomization. We also excluded women with existing chronic illness, who increased body weight measurements. required primarily specialized maternity care during pregnancy. Additional criteria for exclusion were inadequate knowledge of Swedish, BMI <19 at first antenatal visit, IV Secondary anal- Antenatal maternal lifestyle Offspring mean BMI z-score at and multiple pregnancy. The first offspring was born in 2007 and the last offspring ysis of a RCT of intervention. Standardized birth and age 5 in relation to intervention measure of weight and height study group. Proportion of over- reached age 5 in 2015. All offspring to women analysed for GWG in the original in offspring within child health and undernutrition (%) at birth study (Paper II) were included in the follow-up study of offspring obesity (Paper IV). care. and age 5 in relation to study group. Intervention The intervention programme consisted of individual education/information about the BMI = body mass index; GRADE = Grading of Recommendations, Assessment, Development IOM guidelines for recommended GWG, according to pre-pregnancy BMI category and Evaluation; GWG = gestational weight gain; IOM = Institute of Medicine; PPWR = at first antenatal visit. Since the study period started in 2007 the IOM guidelines from postpartum weight retention; RCT = randomized controlled trial. 1990 were applied throughout the study (Table 2).

26 Ann-Kristin rönnberg Gestational Weight Gain 27 ANN-KrIStIN röNNBerg Gestational Weight Gain of exertion during a total of approximately 30 minutes per day. Activities should be Maternal pre-pregnancy weight was defined as the first measured weight at admis- chosen as to entail a minimal risk of falling. The activity chosen should be based on sion to maternal health care before 16 weeks pregnancy and total GWG was defined individual abilities/interests/skills and as physiological changes during pregnancy can as the difference between pre-pregnancy weight and weight at term pregnancy (>37 influence and progressively alter conditions for the woman, the activity prescribed weeks). Data on breastfeeding status were collected at a visit to maternal health care might also be adjusted during the pregnancy. The midwife was instructed to follow up <16 weeks after delivery. and renew the prescription of exercise at every antenatal visit during the pregnancy. Women randomized to the control group were given standard maternity care. In No extra antenatal visits were scheduled for this group and no additional funding was accordance with national guidelines in Sweden, maternal weight was registered three allocated to the care of women receiving intervention. times during pregnancy: at the first visit, at 33 weeks of pregnancy, and when admitted to the delivery unit. Standard maternity care in Sweden includes recommendations on dietary intake during pregnancy according to guidelines from the Swedish National Food Administration (49). This advice was given at admission in early pregnancy, all study patients were given the same standardized written and oral information.

Randomization Randomization was stratified for BMI group (normal weight, overweight, obese) in accordance with WHO standards and the sequence was computer-generated, in blocks of variable sizes between 4 and 8. The allocation sequence was concealed in opaque, sealed, sequentially numbered envelopes, and kept by administrative person- nel not related to the study. When including a pregnant woman in the study, the recruiting midwife contacted the study administrator by telephone and she then opened the lowest numbered envelope in the relevant BMI category and gave information on the assigned study group to the midwife.

Statistical methods, Papers II-IV All data analyses were carried out according to a modified intention to treat approach and the pre-established analysis plan. Statistical analysis was performed in IBM SPSS Statistics for Windows, version 21 and 22 (IBM Corp., Armonk, NY, USA). Chi-square test and independent samples t-test with 95% CIs were used to compare Figure 9. Example of personalized weight graph applied in the trial. differences in proportions and means between groups. A two-sided significance test No recalled or self-reported measures of body weight were used in any of the studies was used and a p-value of <0.05 was regarded as statistically significant. The cor- in this thesis. Information on participating women’s height was, however, self-report- relation between GWG and PPWR in Paper III was analysed using Pearson product ed, which is routine in Swedish maternity care. The children were measured using the moment correlation. Swedish standards for child health care services at birth, 1, 4, 6, 8, 10, 12, 18 months, In Paper IV, BMI z-scores, also called “BMI standard deviation (SD) scores”, and 2 1/2, 3, 4 and 5 years of age. Scales were calibrated according to child health were calculated using the World Health Organization (WHO) Child Growth Stan- care routines. An infant scale was used up until age 2 and thereafter a standing paedi- dards as external reference (50). Risk estimates were calculated by binary logistic atric scale was used. Small children were weighed without clothes and diapers while regression and linear logistic regression, with adjustment for relevant confounding underwear was accepted when weighing the older children. Weight was registered in factors. Mixed model analysis of variance was used to analyse BMI z-scores in multi- grams up to 9,999 g and thereafter in kilograms with one decimal. Length was mea- ple measures during the study period (eleven measures, 0-5 years). A heterogeneous sured in a standardized manner, with the child laying down until age 2 and thereafter first-order autoregressive covariance structure was used for the mixed model analysis. standing up. Length was reported in centimetres with one decimal.

28 Ann-Kristin rönnberg Gestational Weight Gain 29 ANN-KrIStIN röNNBerg Gestational Weight Gain of exertion during a total of approximately 30 minutes per day. Activities should be Maternal pre-pregnancy weight was defined as the first measured weight at admis- chosen as to entail a minimal risk of falling. The activity chosen should be based on sion to maternal health care before 16 weeks pregnancy and total GWG was defined individual abilities/interests/skills and as physiological changes during pregnancy can as the difference between pre-pregnancy weight and weight at term pregnancy (>37 influence and progressively alter conditions for the woman, the activity prescribed weeks). Data on breastfeeding status were collected at a visit to maternal health care might also be adjusted during the pregnancy. The midwife was instructed to follow up <16 weeks after delivery. and renew the prescription of exercise at every antenatal visit during the pregnancy. Women randomized to the control group were given standard maternity care. In No extra antenatal visits were scheduled for this group and no additional funding was accordance with national guidelines in Sweden, maternal weight was registered three allocated to the care of women receiving intervention. times during pregnancy: at the first visit, at 33 weeks of pregnancy, and when admitted to the delivery unit. Standard maternity care in Sweden includes recommendations on dietary intake during pregnancy according to guidelines from the Swedish National Food Administration (49). This advice was given at admission in early pregnancy, all study patients were given the same standardized written and oral information.

28 Ann-Kristin rönnberg Gestational Weight Gain 29 ANN-KrIStIN röNNBerg Gestational Weight Gain Sample size RESULTS The sample size for the VIGA trial was based on the anticipated effects of the intervention on GWG. An improvement where 60% of women, compared with 45% of Paper I women previously, had GWG within IOM recommendations was considered clinically relevant. With a two-sided 5% significance level, a power of 80%, and an antic- Eleven trials of intervention were initially identified and considered for inclusion. ipated dropout rate of 20%, a sample size of 210 women per group was estimated as After full text assessment, two studies were excluded because of a lack of a control adequate in the original study on GWG (Paper II). Sample size was not recalculated group. One further study was excluded because of its primary focus on effects on in the follow-up study of mean PPWR (Paper III). blood pressure of a low-sodium diet in pregnancy. Three intervention studies with a To evaluate the statistical power to detect differences in our secondary analysis randomized controlled design and one study with a quasi-randomized design quali- of offspring obesity, we performed an additional power calculation for the primary fied for inclusion in this review (306 patients). Three were conducted in the USA and outcome, mean BMI z-scores. Given an α of 0.05, a β of 0.80, and a BMI z-score SD one in Denmark (Table 3). Furthermore, four intervention studies with a non-random- of 1.0, a difference between the intervention and the control group in offspring mean ized controlled design met the inclusion criteria and were therefore assessed (1,232 BMI z-score of 0.29 could be detected with a sample size of 187 children in each patients). The non-randomized studies included were conducted in Finland, Sweden, group. All offspring (374 children) to women included and analysed for GWG in the Canada and the USA (Table 4). original study were included in the follow-up study.

Child Growth Standards applied in Paper IV The WHO Child Growth Standards were used as an external reference in the assessment of BMI z-score in the offspring analysed in Paper IV (50). The WHO standards, issued in 2006, are based on the Multi-center Growth Reference Study (MGRS), which collected primary growth data and related information on approximately 8,500 children from widely different ethnic backgrounds and cultural settings (Brazil, Gha- na, India, Norway, Oman and the USA) between 1997 and 2003. The gender-specific BMI for age standard is based on a total of 26,985 records. The data from MGRS provide a single international standard that represents the best description of physiological growth for all children from birth to age 5 and to establish the breastfed infant as the normative model for growth and development. The WHO Child (0-5 years) Growth Standards are recommended by the Interna- tional Paediatric Association as well as the European Childhood Obesity Group and were adapted, already in 2011, as an official standard in more than 125 countries (51), representing 75% of the world´s 0-5 years children. The scrutiny that the WHO standards have undergone is without precedent in the history of developing and applying growth assessment tools, whether national or international.

30 Ann-Kristin rönnberg Gestational Weight Gain 31 ANN-KrIStIN röNNBerg Gestational Weight Gain Sample size RESULTS The sample size for the VIGA trial was based on the anticipated effects of the intervention on GWG. An improvement where 60% of women, compared with 45% of Paper I women previously, had GWG within IOM recommendations was considered clinically relevant. With a two-sided 5% significance level, a power of 80%, and an antic- Eleven trials of intervention were initially identified and considered for inclusion. ipated dropout rate of 20%, a sample size of 210 women per group was estimated as After full text assessment, two studies were excluded because of a lack of a control adequate in the original study on GWG (Paper II). Sample size was not recalculated group. One further study was excluded because of its primary focus on effects on in the follow-up study of mean PPWR (Paper III). blood pressure of a low-sodium diet in pregnancy. Three intervention studies with a To evaluate the statistical power to detect differences in our secondary analysis randomized controlled design and one study with a quasi-randomized design quali- of offspring obesity, we performed an additional power calculation for the primary fied for inclusion in this review (306 patients). Three were conducted in the USA and outcome, mean BMI z-scores. Given an α of 0.05, a β of 0.80, and a BMI z-score SD one in Denmark (Table 3). Furthermore, four intervention studies with a non-random- of 1.0, a difference between the intervention and the control group in offspring mean ized controlled design met the inclusion criteria and were therefore assessed (1,232 BMI z-score of 0.29 could be detected with a sample size of 187 children in each patients). The non-randomized studies included were conducted in Finland, Sweden, group. All offspring (374 children) to women included and analysed for GWG in the Canada and the USA (Table 4). original study were included in the follow-up study.

30 Ann-Kristin rönnberg Gestational Weight Gain 31 ANN-KrIStIN röNNBerg Gestational Weight Gain Results in mean No difference GWG (p=0.77). in pro- No difference IOM portion exceeding (p=0.053). in GWG No difference (p=0.3). Significant difference in ’low-income‘ sub- (p=0.01) group GWG in the Reduced group intervention (p<0.001). in rate No difference of GWG or total mean GWG (p=0.29). trimester; no difference; trimester; trimester; trimester; trimester; trimester; st nd rd Results in Significant decrease with normal BMI women IOM recommen- exceeding (58 vs 33%, p<0.05). dations among over- Opposite trend 59%, (32 vs. women weight p=0.09). GWG in the inter- Reduced (p=0.002). group vention 1 Significantly higher GWG (p<0.006) with intervention No significant in difference to IOM recom- adherence (p=0.21) mendations. 3 significantly less GWG with (p<0.001) intervention 2 Outcome Mean GWG (kg). ofProportion exceed- women ing IOM recommendations. ofProportion exceed- women ing IOM recommendations. Mean GWG (kg). ofRate GWG (kg)/week. Outcome ofProportion exceeding women IOM recommendations. Gestational (kg). gain weight (lb) gain Weight per trimester. retention Weight postpar- 6 weeks tum (lb). ofProportion exceeding women IOM recommendations. Control Standard care Standard care Standard care Standard Dietary care. assessment. Control Standard maternity care. Standard maternity care. Standard maternity Repeat- care. ed nutritional assessment. Standard maternity care. Intervention Individual counselling concerning diet and PA. ofEducation healthcare GWG-grid. providers. by- education Patient mail. and education Patient in- Frequent motivation. Aqua dividual sessions. exercise. aerobic Indi- education. Patient Dietary vidual sessions. assessment. Commu- nity activities focused advice. on nutritional groups. Exercise Intervention behavioural Stepped-care intervention. by-mail. education Patien of graph Personalized gain. weight Individual dietary consul- (10) during preg- tations nancy Group education. Patient nutri- Repeated sessions. tional assessment. pregnan- Dietician in early IOM about Information cy. and recommendations grid. weight to exercise. Encouraged Population in Finland Primipara >18 yrs Age = 49 n (Intervention) = 56 n (Control) USA. BMI 19.8-29.0 >18 yrs Age = 179 n (Intervention) = 381 n (Control) Sweden BMI > 30 = 155 n (Intervention) = 193 n (Control) Indians Cree in Canada. = 112 n (Intervention) = 107 n (Control) Population Low-income population in the USA BMI > 19.8 > 18 yrs Age = 57 n (Intervention) = 53 n (Control) Caucasian, non-smoking in Denmark. BMI >30 >18 yrs Age = 23 n (Intervention) = 27 n (Control) African-American in the USA. women 13-18 yrs Age = 22 n (Intervention) = 24 n (Control) USA BMI < 40.5 18-49 yrs Age = 57 n (Intervention) = 43 n (Control) Design Non-rand- con- omized trial. trolled Non-randomized. Historical con- group. trol Non-rand- con- omized trial. trolled Non-randomized. Historical con- group. trol Design RCT RCT Quasi-RCT RCT Study et al Kinnunen 2007 Olson et al 2004 Claesson et al 2007 Gray-Donald et al 2000 Study et al Polley 2002 Wolff et al 2008 Bechtel- Blackwell 2002 Asbee et al 2009 Table 4. Details of non-randomized controlled trials of antenatal intervention. Table 3. Details of randomized controlled trials (RCTs) of antenatal intervention.

32 Ann-Kristin rönnberg Gestational Weight Gain 33 ANN-KrIStIN röNNBerg Gestational Weight Gain Results in mean No difference GWG (p=0.77). in pro- No difference IOM portion exceeding (p=0.053). in GWG No difference (p=0.3). Significant difference in ’low-income‘ sub- (p=0.01) group GWG in the Reduced group intervention (p<0.001). in rate No difference of GWG or total mean GWG (p=0.29). trimester; no difference; trimester; trimester; trimester; trimester; trimester; st nd rd Results in Significant decrease with normal BMI women IOM recommen- exceeding (58 vs 33%, p<0.05). dations among over- Opposite trend 59%, (32 vs. women weight p=0.09). GWG in the inter- Reduced (p=0.002). group vention 1 Significantly higher GWG (p<0.006) with intervention No significant in difference to IOM recom- adherence (p=0.21) mendations. 3 significantly less GWG with (p<0.001) intervention 2 Outcome Mean GWG (kg). ofProportion exceed- women ing IOM recommendations. ofProportion exceed- women ing IOM recommendations. Mean GWG (kg). ofRate GWG (kg)/week. Outcome ofProportion exceeding women IOM recommendations. Gestational (kg). gain weight (lb) gain Weight per trimester. retention Weight postpar- 6 weeks tum (lb). ofProportion exceeding women IOM recommendations. Control Standard care Standard care Standard care Standard Dietary care. assessment. Control Standard maternity care. Standard maternity care. Standard maternity Repeat- care. ed nutritional assessment. Standard maternity care. Intervention Individual counselling concerning diet and PA. ofEducation healthcare GWG-grid. providers. by- education Patient mail. and education Patient in- Frequent motivation. Aqua dividual sessions. exercise. aerobic Indi- education. Patient Dietary vidual sessions. assessment. Commu- nity activities focused advice. on nutritional groups. Exercise Intervention behavioural Stepped-care intervention. by-mail. education Patien of graph Personalized gain. weight Individual dietary consul- (10) during preg- tations nancy Group education. Patient nutri- Repeated sessions. tional assessment. pregnan- Dietician in early IOM about Information cy. and recommendations grid. weight to exercise. Encouraged Population in Finland Primipara >18 yrs Age = 49 n (Intervention) = 56 n (Control) USA. BMI 19.8-29.0 >18 yrs Age = 179 n (Intervention) = 381 n (Control) Sweden BMI > 30 = 155 n (Intervention) = 193 n (Control) Indians Cree in Canada. = 112 n (Intervention) = 107 n (Control) Population Low-income population in the USA BMI > 19.8 > 18 yrs Age = 57 n (Intervention) = 53 n (Control) Caucasian, non-smoking in Denmark. BMI >30 >18 yrs Age = 23 n (Intervention) = 27 n (Control) African-American in the USA. women 13-18 yrs Age = 22 n (Intervention) = 24 n (Control) USA BMI < 40.5 18-49 yrs Age = 57 n (Intervention) = 43 n (Control) Design Non-rand- con- omized trial. trolled Non-randomized. Historical con- group. trol Non-rand- con- omized trial. trolled Non-randomized. Historical con- group. trol Design RCT RCT Quasi-RCT RCT Study et al Kinnunen 2007 Olson et al 2004 Claesson et al 2007 Gray-Donald et al 2000 Study et al Polley 2002 Wolff et al 2008 Bechtel- Blackwell 2002 Asbee et al 2009 Table 4. Details of non-randomized controlled trials of antenatal intervention. Table 3. Details of randomized controlled trials (RCTs) of antenatal intervention.

32 Ann-Kristin rönnberg Gestational Weight Gain 33 ANN-KrIStIN röNNBerg Gestational Weight Gain Randomized Studies Limita- Consist- Direct- Precision Publi- Quality of (partici- tions ency ness cation evidence n= 445 pants) bias 8 (1538)* Serious Important Indirect- No Unlikely Very low limitations inconsistency ness important imprecision -2 -1 –1 0 0 Allocated to intervention n=221 Allocated to standard care n= 221 Table 5. Quality of evidence according to the Grading of recommendations, Assessment, De- Recieved intervention n= 221 Recieved standard care n=224 velopment and Evaluation (GRADE) system. *Errata in Paper II. In this table the correct number of participants is reported.

Conclusion Paper I Lost to follow up Lost to follow up As a consequence of important limitations in study design as well as inconsistencies During pregnancy n=8 During pregnancy n=21 Dropout n=9 Dropout n=2 and lack of directness, the overall quality of evidence was judged to be very low (+) Misscarrige=7 Misscarrige=11 (Table 5). The results of published interventional trials were of insufficient quality to enable evidence-based recommendations to be developed for clinical practice in antenatal care.

Excluded in analysis n=5 Excluded in analysis n=8 Paper II Analysed for GWG n=192 Analysed for GWG n=182 A total of 445 women were randomized (221 women were allocated to intervention and 224 to standard care) to participate in the VIGA trial, 374 women remained for analysis after completing their pregnancy (Figure 10). A total of 14 maternal health care clinics in the region participated in the study. Analysis was limited to term de- Offspring analysed Women analysed Offspring analysed Women analysed liveries (≥37 weeks of gestation). There was an 8% loss to follow-up in the group at birth n=192 PPWR(1) n=137 at birth n=182 PPWR(1) n=130 at age five n=156 at age five n=144 receiving intervention and an 11% loss to follow-up in the group receiving standard PPWR(2) n=87 PPWR(2) n=81 care. Data were analysed according to a modified intention to treat approach.

Figure 10. CONSORT flow diagram of study participants in the VIGA trial. *Errata in Paper III. The diagram lost the number of miscarriages in each group during editing but is now complete. No significant baseline difference in maternal age, parity or mean pre-pregnancy BMI was detected between groups (Table 6). The majority of the women included were normal weight (72%). Length of gestation at inclusion and at delivery (Errata in Paper II, where the difference in gestational length at inclusion was overlooked) significantly dif- fered (p<0.001 at inclusion and p=0.034 at delivery) between groups.

34 Ann-Kristin rönnberg Gestational Weight Gain 35 ANN-KrIStIN röNNBerg Gestational Weight Gain Randomized Studies Limita- Consist- Direct- Precision Publi- Quality of (partici- tions ency ness cation evidence n= 445 pants) bias 8 (1538)* Serious Important Indirect- No Unlikely Very low limitations inconsistency ness important imprecision -2 -1 –1 0 0 Allocated to intervention n=221 Allocated to standard care n= 221 Table 5. Quality of evidence according to the Grading of recommendations, Assessment, De- Recieved intervention n= 221 Recieved standard care n=224 velopment and Evaluation (GRADE) system. *Errata in Paper II. In this table the correct number of participants is reported.

34 Ann-Kristin rönnberg Gestational Weight Gain 35 ANN-KrIStIN röNNBerg Gestational Weight Gain Variable Intervention group Control group Characteristics Intervention Control p-value n=192 n=182 group group Age, mean (SD), yrs 29.9 (4.5) 29.8 (4.8) GWG>IOM GWG>IOM n=79 n=91 Pre-pregnancy BMI, mean (SD) 25.2 (4.9) 25.3 (4.8) Normal pre-pregnancy BMI n (%) 48 (35) 55 (42) 0.449 Pre-pregnancy BMI group, n (%) 19.0-26.0 138 (72) 130 (71) Pre-pregnancy overweight n (%) 13 (56) 17 (81) 0.151

26.1-29.0 23 (12) 21 (12) Pre-pregnancy obesity n (%) 18 (58) 19 (61) 0.822 >29.0 31 (16) 31 (17) Length of gestation 10.7 (1.7) 9.7 (2.1) Table 8. Effects of intervention in relation to pre-pregnancy body mass index (BMI) among at inclusion, mean (SD), wks women with gestational weight gain (GWG) above Institute of Medicine (IOM) recommenda- Length of gestation 39.7 (1.2) 39.4 (1.2) tions. at delivery, mean (SD) , wks Ancillary analysis in relation to parity showed that the intervention significantly in- Table 6. Baseline maternal characteristics of the study population. creased the proportion of multiparous women who gained weight below IOM recom- The proportion of women with excessive GWG was 9% smaller among women who mendations (22% for the intervention group vs. 12% for controls, p=0.027). received the intervention (41.1% vs. 50.0%) but the difference was not statistically significant (p = 0.080). There was, however, a significant difference in mean GWG Conclusion Paper II (kg) between study groups in favour of intervention (14.19 kg vs. 15.31 kg, p= 0.028) A composite, low-cost, antenatal lifestyle intervention significantly reduced mean (Table 7). GWG, however, the proportion of women with excessive GWG was not significantly lower in the intervention group compared with standard maternity care. Outcome Intervention Control p-value Adjusteda group group p-value N=192 N=182 Paper III GWG >IOM, n (%) 79 (41.1) 91 (50.0) 0.086 0.080 Postpartum weight retention was analysed in Paper III, with the first weight measure- GWG, mean (SD), kg 14.19 (4.45) 15.31 (5.38) 0.029 0.028 ment at the postpartum visit approximately16 weeks after delivery and the second measurement 1 year post-delivery (Table 9). Of 445 women randomized to participate Table 7. Effect of intervention on proportion of women with excessive gestational weight gain in the VIGA trial, 267 remained for analysis at first follow-up postpartum and 168 at (GWG) and on mean GWG in relation to study group. second follow-up at 1 year postpartum (Figure 10). There were no significant differ- a Adjusted for differences in gestational length at inclusion and at birth. ences in baseline characteristics between study groups, neither among participants No significant difference in the proportion of women gaining weight below (19% vs remaining or lost to follow-up during the first year postpartum. 16%) or within (40% vs. 34%) IOM recommendations was seen in relation to study group. The effect of the intervention was analysed within sub-groups based on pre-pregnancy BMI and parity. A tendency towards lower weight gain among the intervention group was seen in all categories of pre-pregnancy BMI. The differences were, however, not statistically significant (Table 8).

36 Ann-Kristin rönnberg Gestational Weight Gain 37 ANN-KrIStIN röNNBerg Gestational Weight Gain Variable Intervention group Control group Characteristics Intervention Control p-value n=192 n=182 group group Age, mean (SD), yrs 29.9 (4.5) 29.8 (4.8) GWG>IOM GWG>IOM n=79 n=91 Pre-pregnancy BMI, mean (SD) 25.2 (4.9) 25.3 (4.8) Normal pre-pregnancy BMI n (%) 48 (35) 55 (42) 0.449 Pre-pregnancy BMI group, n (%) 19.0-26.0 138 (72) 130 (71) Pre-pregnancy overweight n (%) 13 (56) 17 (81) 0.151

36 Ann-Kristin rönnberg Gestational Weight Gain 37 ANN-KrIStIN röNNBerg Gestational Weight Gain Variable Interven- Control Missing Missing Missing Outcome Intervention Control Mean p-value CI (95%) tion group group Intervention Control analysis group group difference (kg) group group p-value Mean PPWR 1.81 (4.52) 3.19 (4.77) 1.38 0.016 0.26–2.50 <16 wks PP, n (%) kg (SD) <16 wks PP 137 130 55 (28.6) 52 (28.6) NS Mean PPWR 0.30 (5.52) 1.00 (5.46) 0.69 0.414 -0.98-2.37 1 y PP 87 81 105 (64.7) 101 (55.5) NS 1 yr PP, kg (SD) Mean age, yrs (SD) 30.1 (4.4) 29.7 (4.7) 29.3 ( 4.5) 29.9 (5.2) NS <16 wks PP, n (%) Table 10. Effect of intervention on mean postpartum weight retention (PPWR) (kg) at two Mean pre-pregnancy 25.6 (5.2) 25.2 (4.7) 24.3 (3.9) 25.6 (5.1) NS time points of maternal weight estimation. BMI (SD) Normal weight 93 (68) 95 (73) 45 (82) 35 (67) NS CI = confidence interval; PP = postpartum; SD = standard deviation. Overweight 17 (12) 15 (12) 6 (11) 6 (12) Obese 27 (20) 20 (15) 4 (7) 11 (21) Postpartum weight retention was also analysed in sub-groups of women to explore 1 yr PP, n (%) potential differences in effects of the intervention in relation to factors such as GWG Mean pre-pregnancy 24.5 (4.0) 25.1 (4.7) 25.8 (5.5) 25.6 (4.8) NS and pre-pregnancy BMI. No difference in the effect of the intervention in relation to BMI (SD) the woman’s adherence to IOM recommendations was seen even though a significant- Normal weight 69 (79) 58 (72) 69 (66) 72 (71) NS ly higher risk for high short-term PPWR was demonstrated in the category with high Overweight 5 (6) 9 (11) 18 (17) 12 (12) Obese 13 (15) 14 (17) 18 (17) 17 (17) GWG in the whole study population. Nulliparas, n (%) 64 (47) 69 (53) 21 (38) 24 (46) NS Mean GWG, kg (SD) 5 < IOM IOM > IOM <16 wks PP 13.9 (4.4) 15.5 (5.4) 14.8 (4.5) 14.9 (5.4) NS 1 yr PP 14.0 (4.2) 15.4 (5.1) 14.4 (4.7) 15.2 (5.6) NS 4 Breastfeeding at <16 wks PP, n (%) Yes 115 (88) 110 (85) 6 0 NS No 16 (12) 20 (15) NS 3 Mean time PP at weight measurements <16 w PP wks 10.2 (3.0) 10.0 (3.2) NS 2 (SD) 12.2 (0.8) 12.3 (0.7) NS 1 year PP mo (SD) 1

Table 9. Baseline characteristics of the study population after delivery and at 1 year postpar- mean PPWR (kg) tum. 0 <16w 1 year The intervention group had significantly lower mean PPWR at ≤16 weeks (1.81 kg vs. 3.19 kg). At 1 year postpartum, mean retention was still 0.69 kg lower in the inter- -1 vention group (0.30 kg vs. 1.00 kg),but the difference was not statistically significant (Table 10). -2

Figure 11. Mean postpartum weight retention (PPWR) (kg) in relation to adherence to the Institute of Medicine (IOM) recommendations, independent of study group. Women with normal pre-pregnancy BMI had a significant short-term effect of the intervention on PPWR but other BMI categories did not. At the long-term follow-up there were no significant differences regarding PPWR in any of the analysed sub-

38 Ann-Kristin rönnberg Gestational Weight Gain 39 ANN-KrIStIN röNNBerg Gestational Weight Gain Variable Interven- Control Missing Missing Missing Outcome Intervention Control Mean p-value CI (95%) tion group group Intervention Control analysis group group difference (kg) group group p-value Mean PPWR 1.81 (4.52) 3.19 (4.77) 1.38 0.016 0.26–2.50 <16 wks PP, n (%) kg (SD) <16 wks PP 137 130 55 (28.6) 52 (28.6) NS Mean PPWR 0.30 (5.52) 1.00 (5.46) 0.69 0.414 -0.98-2.37 1 y PP 87 81 105 (64.7) 101 (55.5) NS 1 yr PP, kg (SD) Mean age, yrs (SD) 30.1 (4.4) 29.7 (4.7) 29.3 ( 4.5) 29.9 (5.2) NS <16 wks PP, n (%) Table 10. Effect of intervention on mean postpartum weight retention (PPWR) (kg) at two Mean pre-pregnancy 25.6 (5.2) 25.2 (4.7) 24.3 (3.9) 25.6 (5.1) NS time points of maternal weight estimation. BMI (SD) Normal weight 93 (68) 95 (73) 45 (82) 35 (67) NS CI = confidence interval; PP = postpartum; SD = standard deviation. Overweight 17 (12) 15 (12) 6 (11) 6 (12) Obese 27 (20) 20 (15) 4 (7) 11 (21) Postpartum weight retention was also analysed in sub-groups of women to explore 1 yr PP, n (%) potential differences in effects of the intervention in relation to factors such as GWG Mean pre-pregnancy 24.5 (4.0) 25.1 (4.7) 25.8 (5.5) 25.6 (4.8) NS and pre-pregnancy BMI. No difference in the effect of the intervention in relation to BMI (SD) the woman’s adherence to IOM recommendations was seen even though a significant- Normal weight 69 (79) 58 (72) 69 (66) 72 (71) NS ly higher risk for high short-term PPWR was demonstrated in the category with high Overweight 5 (6) 9 (11) 18 (17) 12 (12) Obese 13 (15) 14 (17) 18 (17) 17 (17) GWG in the whole study population. Nulliparas, n (%) 64 (47) 69 (53) 21 (38) 24 (46) NS Mean GWG, kg (SD) 5 < IOM IOM > IOM <16 wks PP 13.9 (4.4) 15.5 (5.4) 14.8 (4.5) 14.9 (5.4) NS 1 yr PP 14.0 (4.2) 15.4 (5.1) 14.4 (4.7) 15.2 (5.6) NS 4 Breastfeeding at <16 wks PP, n (%) Yes 115 (88) 110 (85) 6 0 NS No 16 (12) 20 (15) NS 3 Mean time PP at weight measurements <16 w PP wks 10.2 (3.0) 10.0 (3.2) NS 2 (SD) 12.2 (0.8) 12.3 (0.7) NS 1 year PP mo (SD) 1

38 Ann-Kristin rönnberg Gestational Weight Gain 39 ANN-KrIStIN röNNBerg Gestational Weight Gain groups. Gestational weight gain above the IOM recommendations was a significant Paper IV risk factor for excessive weight retention (>5 kg) 1 year after delivery (odds ratio In all, 374 children were included at birth and 300 remained for analysis at age 5. The (OR) 2.44; 95% CI: 1.08–5.52, p=0.029), independent of study group (Table 11). flow of study participants is shown in Figure 10. There was no record of women not Variable Weight reten- Weight reten- p- OR CI (95%) receiving the intended treatment in either study group and all offspring were analysed tion < 5 kg tion > 5 kg value according to the group their mother was allocated to. No significant difference was detected in baseline characteristics, such as sex of N=139 N=29 offspring, maternal parity, age, pre-pregnancy BMI or prevalence of breast-feeding Mean age, yrs (SD) 30.2 (4.8) 29.5 (4.2) 0.503 0.97 0.89-1.06 between study groups. Mean gestational age at birth was, however, significantly higher among offspring in the intervention group (39.7 weeks vs. 39.4 weeks; 95% CI: Mean pre-pregnan- 24.6 ( 4.5) 25.8 (3.8) 0.161 1.06 0.98-1.16 0.02-0.50, p=0.034) (Table 12). cy BMI (SD) Parity, n (%) Variable Intervention Control p-value group group 0 61 (44) 13 (45) 0.926 0.98 0.63-1.63 >1 78 (56) 16 (55) 1.02 0.71-1.46 N=192 N=182 Study population, n (%) Male offspring, n (%) 101 (53) 89 (49) 0.535 Intervention 74 (53) 13 (45) 0.410 0.71 0.32-1.59 Gestational age at birth, wks (SD) 39.7 (1.2) 39.4 (1.2) 0.034 Control 65 (47) 16(55) 1.19 0.77-1.83 Mean maternal age, yrs (SD) 29.9 (4.5) 29.8 (4.8) 0.870 Mean GWG, kg 14.1 (4.3) 17.2 (5.7) 0.002 1.16 1.05-1.27 Mean maternal GWG, kg (SD) 14.19 (4.45) 15.31 (5.38) 0.029 (SD) Nulliparous mothers, n (%) 85 (44) 93 (51) 0.214 Non-adherence to IOM recommenda- Mean maternal pre-pregnancy BMI, 25.2 (4.9) 25.3 (4.8) 0.822 tions n(%) kg/m2 (SD) GWG >IOM 51 (37) 17 (59) 0.029 2.44 1.08-5.52 Maternal pre-pregnancy BMI group, n (%) Normal 138 (72) 130 (71) 0.969 Mean FBW, g (SD) 3,544 (506) 3,631 (395) 0.426 1.00 0.99-1.00 Overweight 23 (12) 21 (12) Breastfeeding, Obese 31 (16) 31 (17) n (%) Breast-feeding, n (%) Yes 104 (90) 19 (79) 0.359 0.52 0.13-2.16 No 12(10) 5 (21) 0.44 0.14-1.39 Yes 115 (60) 112 (62) 0.520 Missing data 23/139 (16) 5/29 (17) No 16 (8) 20 (11) Missing data 61 (32) 50 (27) Table 11. Risk estimates for excessive postpartum weight retention (PPWR) (>5 kg) 1 year after delivery. Table 12. Offspring and maternal baseline characteristics. BMI = body mass index; GWG = gestational weight gain; SD = standard deviation. BMI = body mass index; FBW = fetal birth weight; IOM = Institute of Medicine; OR = odds ratio; SD = standard deviation. No significant difference in mean BMI z-scores was seen at birth (intervention group 0.68 vs. controls 0.56; p=0.242) or at age 5 (intervention group 0.34 vs. controls 0.26; Conclusion Paper III p=0.510) and no significant difference in the proportion (%) of over- or undernutrition (BMI z-scores >+2, at birth or age 5, was seen between study groups. Adjustment Women who received antenatal lifestyle intervention had significantly less short-term for potential confounding by differences in gestational age at birth was performed PPWR. One year after delivery there was no significant difference in weight retention without significant changes in the primary outcome results (Table 13). between the study groups. Excessive GWG was a significant risk factor for excessive Alternative measures of fetal size at birth were also analysed and no significant weight retention (>5 kg) 1 year after delivery, independent of study group. difference in fetal size, except for fetal length, was seen after adjusting for differences in gestational age at birth (Table 13).

40 Ann-Kristin rönnberg Gestational Weight Gain 41 ANN-KrIStIN röNNBerg Gestational Weight Gain groups. Gestational weight gain above the IOM recommendations was a significant Paper IV risk factor for excessive weight retention (>5 kg) 1 year after delivery (odds ratio In all, 374 children were included at birth and 300 remained for analysis at age 5. The (OR) 2.44; 95% CI: 1.08–5.52, p=0.029), independent of study group (Table 11). flow of study participants is shown in Figure 10. There was no record of women not Variable Weight reten- Weight reten- p- OR CI (95%) receiving the intended treatment in either study group and all offspring were analysed tion < 5 kg tion > 5 kg value according to the group their mother was allocated to. No significant difference was detected in baseline characteristics, such as sex of N=139 N=29 offspring, maternal parity, age, pre-pregnancy BMI or prevalence of breast-feeding Mean age, yrs (SD) 30.2 (4.8) 29.5 (4.2) 0.503 0.97 0.89-1.06 between study groups. Mean gestational age at birth was, however, significantly higher among offspring in the intervention group (39.7 weeks vs. 39.4 weeks; 95% CI: Mean pre-pregnan- 24.6 ( 4.5) 25.8 (3.8) 0.161 1.06 0.98-1.16 0.02-0.50, p=0.034) (Table 12). cy BMI (SD) Parity, n (%) Variable Intervention Control p-value group group 0 61 (44) 13 (45) 0.926 0.98 0.63-1.63 >1 78 (56) 16 (55) 1.02 0.71-1.46 N=192 N=182 Study population, n (%) Male offspring, n (%) 101 (53) 89 (49) 0.535 Intervention 74 (53) 13 (45) 0.410 0.71 0.32-1.59 Gestational age at birth, wks (SD) 39.7 (1.2) 39.4 (1.2) 0.034 Control 65 (47) 16(55) 1.19 0.77-1.83 Mean maternal age, yrs (SD) 29.9 (4.5) 29.8 (4.8) 0.870 Mean GWG, kg 14.1 (4.3) 17.2 (5.7) 0.002 1.16 1.05-1.27 Mean maternal GWG, kg (SD) 14.19 (4.45) 15.31 (5.38) 0.029 (SD) Nulliparous mothers, n (%) 85 (44) 93 (51) 0.214 Non-adherence to IOM recommenda- Mean maternal pre-pregnancy BMI, 25.2 (4.9) 25.3 (4.8) 0.822 tions n(%) kg/m2 (SD) GWG >IOM 51 (37) 17 (59) 0.029 2.44 1.08-5.52 Maternal pre-pregnancy BMI group, n (%) Normal 138 (72) 130 (71) 0.969 Mean FBW, g (SD) 3,544 (506) 3,631 (395) 0.426 1.00 0.99-1.00 Overweight 23 (12) 21 (12) Breastfeeding, Obese 31 (16) 31 (17) n (%) Breast-feeding, n (%) Yes 104 (90) 19 (79) 0.359 0.52 0.13-2.16 No 12(10) 5 (21) 0.44 0.14-1.39 Yes 115 (60) 112 (62) 0.520 Missing data 23/139 (16) 5/29 (17) No 16 (8) 20 (11) Missing data 61 (32) 50 (27) Table 11. Risk estimates for excessive postpartum weight retention (PPWR) (>5 kg) 1 year after delivery. Table 12. Offspring and maternal baseline characteristics. BMI = body mass index; GWG = gestational weight gain; SD = standard deviation. BMI = body mass index; FBW = fetal birth weight; IOM = Institute of Medicine; OR = odds ratio; SD = standard deviation. No significant difference in mean BMI z-scores was seen at birth (intervention group 0.68 vs. controls 0.56; p=0.242) or at age 5 (intervention group 0.34 vs. controls 0.26; Conclusion Paper III p=0.510) and no significant difference in the proportion (%) of over- or undernutrition (BMI z-scores >+2, at birth or age 5, was seen between study groups. Adjustment Women who received antenatal lifestyle intervention had significantly less short-term for potential confounding by differences in gestational age at birth was performed PPWR. One year after delivery there was no significant difference in weight retention without significant changes in the primary outcome results (Table 13). between the study groups. Excessive GWG was a significant risk factor for excessive Alternative measures of fetal size at birth were also analysed and no significant weight retention (>5 kg) 1 year after delivery, independent of study group. difference in fetal size, except for fetal length, was seen after adjusting for differences in gestational age at birth (Table 13).

40 Ann-Kristin rönnberg Gestational Weight Gain 41 ANN-KrIStIN röNNBerg Gestational Weight Gain Outcome Intervention Control Crude Adjusted Mean BMI z-score was consistently slightly higher among offspring to women who group group p-value p-valuea had received the lifestyle intervention (Figure 12). Point-estimate comparison of n= 192 n=182 mean BMI z-scores showed significantly differences at age 2,5 years only; however, Mean offspring BMI, in a mixed model analysis of multiple measurements, adjusted for differences in ges- kg/cm2 (SD) Birth 14.3 (1.3) 14.2 (1.3) 0.237 0.473 tational age at birth, the difference in z-score was 0.16 (p=0.034; 95% CI 0.01-0.30). Age 5 yrs 15.8 (1.4) 15.7 (1.4) 0.488 n=156 n=144 1,0 Intervention Control Mean offspring BMI z-score, (SD) 0,8 Birth 0.68 (0.99) 0.56 (0.99) 0.242 0.479 Age 5 yrs 0.34 (0.96) 0.26 (0.96) 0.510 0,6 n=156 n=144 0,4 Mean FBW, kg (SD) 3.661 (0.461) 3.548 (0.489) 0.022 0.109 0,2 LGA (>2 SD), n (%) 15 (8) 11 (6) 0.501 0.452 AGA, n (%) 174 (91) 168 (92) 0.561 0.507 0,0 mean BMI z-score SGA (<2 SD), n (%) 3 (2) 3 (2) 0.947 0.962 -0,2 Mean fetal birth length, cm (SD) 50.5 (1.8) 49.9 (1.8) 0.005 0.032 -0,4 Birth 2 mo 4 mo 6 mo 8 mo 10 mo 12 mo 18 mo 2.5 yrs 4 yrs 5 yrs FBW >4,000 g, n (%) 47 (24) 28 (15) 0.029 0.091 FBW >4,500 g, n (%) 8 (4) 8 (4) 0.913 0.580 Figure 12. Trajectory of mean body mass index (BMI) z-scores in offspring from birth to age Ponderal Index, weight (kg) x100/length 2.84 2.84 0.861 0.989 5, in relation to study group. (cm)3 Independent of study group, GWG above IOM recommendations, was a signifi- Table 13. Effect of the intervention on offspring mean body mass index (BMI) and mean BMI cant risk factor for offspring overnutrition (BMI z-score >2 SD) at birth (OR=4.51, z-score, by study group and age. The Table also shows the effect of intervention on alternative p<0.001; 95% CI 1.95-10.44) but not at age 5. Gestational weight gain below IOM measures of fetal size at birth. recommendations had a significant negative association with overnutrition at birth a Adjusted for differences in gestational age at birth. (OR 0.89, p=0.005; 95% CI 0.86-0.93) but not at age 5 (OR 0.76, p=0.734; 95% CI AGA = average for gestational age; FBW = fetal birth weight; LGA = large for gestational 0.17-3.54) (Table 14). age; SD = standard deviation; SGA = small for gestational age. Pre-pregnancy maternal BMI had no significant association with offspring over- Outcome Intervention Control Crude Adjusted nutrition at birth (OR=1.32, p= 0.570; 95% CI 0.51-3.45), however, a strong positive group group p-value p-valuea association with childhood obesity at age 5 was seen, independent of maternal GWG n=192 n=182 (Figure 13). Offspring to women with normal pre-pregnancy weight had a reduced Overnutrition in offspring, risk of obesity at age 5 and offspring to obese women had an increased risk of obesity (BMI z-score ≥ 2 SD) Birth, n (%) 18 (9) 15 (8) 0.699 0.932 at age 5 (Table 15). Age 5 yrs, n (%) 5 (3) 9 (6) 0.221 0.228 n=156 n=144 Undernutrition in offspring, (BMI z-score ≤ -2 SD) Birth, n (%) 2 (1) 0 0.167 0.995 Age 5 yrs, n (%) 3 (2) 1 (1) 0.347 0.437 n=156 n=144

Table 14. Effects of the intervention on the precentage of over/undernutrition (BMI z-score > +2 standard deviations (SDs)) in offspring, by study group and age. a Adjusted for differences in gestational age at birth.

42 Ann-Kristin rönnberg Gestational Weight Gain 43 ANN-KrIStIN röNNBerg Gestational Weight Gain Outcome Intervention Control Crude Adjusted Mean BMI z-score was consistently slightly higher among offspring to women who group group p-value p-valuea had received the lifestyle intervention (Figure 12). Point-estimate comparison of n= 192 n=182 mean BMI z-scores showed significantly differences at age 2,5 years only; however, Mean offspring BMI, in a mixed model analysis of multiple measurements, adjusted for differences in ges- kg/cm2 (SD) Birth 14.3 (1.3) 14.2 (1.3) 0.237 0.473 tational age at birth, the difference in z-score was 0.16 (p=0.034; 95% CI 0.01-0.30). Age 5 yrs 15.8 (1.4) 15.7 (1.4) 0.488 n=156 n=144 1,0 Intervention Control Mean offspring BMI z-score, (SD) 0,8 Birth 0.68 (0.99) 0.56 (0.99) 0.242 0.479 Age 5 yrs 0.34 (0.96) 0.26 (0.96) 0.510 0,6 n=156 n=144 0,4 Mean FBW, kg (SD) 3.661 (0.461) 3.548 (0.489) 0.022 0.109 0,2 LGA (>2 SD), n (%) 15 (8) 11 (6) 0.501 0.452 AGA, n (%) 174 (91) 168 (92) 0.561 0.507 0,0 mean BMI z-score SGA (<2 SD), n (%) 3 (2) 3 (2) 0.947 0.962 -0,2 Mean fetal birth length, cm (SD) 50.5 (1.8) 49.9 (1.8) 0.005 0.032 -0,4 Birth 2 mo 4 mo 6 mo 8 mo 10 mo 12 mo 18 mo 2.5 yrs 4 yrs 5 yrs FBW >4,000 g, n (%) 47 (24) 28 (15) 0.029 0.091 FBW >4,500 g, n (%) 8 (4) 8 (4) 0.913 0.580 Figure 12. Trajectory of mean body mass index (BMI) z-scores in offspring from birth to age Ponderal Index, weight (kg) x100/length 2.84 2.84 0.861 0.989 5, in relation to study group. (cm)3 Independent of study group, GWG above IOM recommendations, was a signifi- Table 13. Effect of the intervention on offspring mean body mass index (BMI) and mean BMI cant risk factor for offspring overnutrition (BMI z-score >2 SD) at birth (OR=4.51, z-score, by study group and age. The Table also shows the effect of intervention on alternative p<0.001; 95% CI 1.95-10.44) but not at age 5. Gestational weight gain below IOM measures of fetal size at birth. recommendations had a significant negative association with overnutrition at birth a Adjusted for differences in gestational age at birth. (OR 0.89, p=0.005; 95% CI 0.86-0.93) but not at age 5 (OR 0.76, p=0.734; 95% CI AGA = average for gestational age; FBW = fetal birth weight; LGA = large for gestational 0.17-3.54) (Table 14). age; SD = standard deviation; SGA = small for gestational age. Pre-pregnancy maternal BMI had no significant association with offspring over- Outcome Intervention Control Crude Adjusted nutrition at birth (OR=1.32, p= 0.570; 95% CI 0.51-3.45), however, a strong positive group group p-value p-valuea association with childhood obesity at age 5 was seen, independent of maternal GWG n=192 n=182 (Figure 13). Offspring to women with normal pre-pregnancy weight had a reduced Overnutrition in offspring, risk of obesity at age 5 and offspring to obese women had an increased risk of obesity (BMI z-score ≥ 2 SD) Birth, n (%) 18 (9) 15 (8) 0.699 0.932 at age 5 (Table 15). Age 5 yrs, n (%) 5 (3) 9 (6) 0.221 0.228 n=156 n=144 Undernutrition in offspring, (BMI z-score ≤ -2 SD) Birth, n (%) 2 (1) 0 0.167 0.995 Age 5 yrs, n (%) 3 (2) 1 (1) 0.347 0.437 n=156 n=144

42 Ann-Kristin rönnberg Gestational Weight Gain 43 ANN-KrIStIN röNNBerg Gestational Weight Gain Crude OR Crude Adjusted Adjusted Conclusion Paper IV (95% CI) p-value OR (95% CI) p-value The composite antenatal lifestyle intervention did not significantly reduce offspring Pre-pregnancy maternal obesity mean BMI z-scores or the proportions of childhood obesity at birth or age 5. Maternal n=54 (18%) 5.08 0.004 4.81 0.006a (1.70–15.18) (1.56-14.83)a excessive GWG was not significantly associated with an increased risk of offspring Normal maternal pre-pregnancy obesity at age 5 but maternal pre-pregnancy obesity had a strong positive association weight with childhood obesity at age 5, independent of maternal GWG or study group. n=207 (69%) 0.16 0.003 0.17 0.004a (0.05-0.53) (0.05-0.56)a Maternal GWG above IOM recommendations n=139 (46%) 1.17 0.778 1.01 0.991b (0.40-3.41) (0.33-3.05)b

Table 15. Risk analysis of maternal factors associated with offspring obesity (body mass index (BMI) z-score ≥2 standard deviations (SDs)) at age 5, independent of study group. a Adjusted for maternal gestational weight gain (GWG) (kg); b Adjusted for maternal prepregnancy BMI (kg/cm2).

CI = confidence interval; IOM = Institute of Medicine; OR = odds ratio.

0,8

0,7 * 0,6 0,5 * 0,4

0,3

0,2 Mean BMI z-score at age ve at Mean BMI z-score 0,1

0,0 Normal weight Overweight Obese Figure 13. Mean body mass index (BMI) z-scores in offspring at age 5, in relation to maternal pre-pregnancy BMI category. *p<0.01, significant difference between groups.

44 Ann-Kristin rönnberg Gestational Weight Gain 45 ANN-KrIStIN röNNBerg Gestational Weight Gain Crude OR Crude Adjusted Adjusted Conclusion Paper IV (95% CI) p-value OR (95% CI) p-value The composite antenatal lifestyle intervention did not significantly reduce offspring Pre-pregnancy maternal obesity mean BMI z-scores or the proportions of childhood obesity at birth or age 5. Maternal n=54 (18%) 5.08 0.004 4.81 0.006a (1.70–15.18) (1.56-14.83)a excessive GWG was not significantly associated with an increased risk of offspring Normal maternal pre-pregnancy obesity at age 5 but maternal pre-pregnancy obesity had a strong positive association weight with childhood obesity at age 5, independent of maternal GWG or study group. n=207 (69%) 0.16 0.003 0.17 0.004a (0.05-0.53) (0.05-0.56)a Maternal GWG above IOM recommendations n=139 (46%) 1.17 0.778 1.01 0.991b (0.40-3.41) (0.33-3.05)b

CI = confidence interval; IOM = Institute of Medicine; OR = odds ratio.

0,8

0,7 * 0,6 0,5 * 0,4

0,3

0,2 Mean BMI z-score at age ve at Mean BMI z-score 0,1

44 Ann-Kristin rönnberg Gestational Weight Gain 45 ANN-KrIStIN röNNBerg Gestational Weight Gain DISCUSSION charts of GWG for gestational age in relation to maternal BMI (25) could however have improved the depth in our analysis by providing a more exact measure of weight gain in relation to gestational age. The charts were however not available during the Summary of main findings analysis of our data. The general rationale for this thesis was the need to expand knowledge of the impact According to the original VIGA protocol, the women were educated and moti- of an antenatal lifestyle intervention aimed at reducing excessive GWG. Current sci- vated to reach GWG as per IOM recommendations of 1990 (8). At the end of the entific evidence was reviewed and a composite antenatal lifestyle intervention was study period the IOM published revised recommendations on GWG (20). Body mass designed and performed as a RCT. index categories were adapted to WHO standards and the upper limit for women with The body of evidence from published intervention trials in 2009 was found to be pre-gestational obesity was set to 9 kg. A small number of the participating women insufficient to enable development of evidence-based recommendations for clinical would have been classified to a different BMI category based on the new guidelines. practice in maternal health care. Our antenatal lifestyle intervention, tested within They would therefore have received a different education and a different personal- the VIGA trial, was safely performed according to study protocol and resulted in ized graph. The upper limit for women with pre-pregnancy obesity would have been significantly reduced mean maternal GWG but no significant reduction in the propor- lower. Since it was not possible to foresee how the participating women would have tion of women gaining excessively according to IOM recommendations. Women who adhered to the intervention with these new recommendations, our data on efficiency received the intervention had significantly less short-term weight retention after de- of the intervention were analysed in accordance with our original study protocol us- livery, however, no significant difference in weight retention, compared with standard ing the IOM 1990 guidelines. care, was seen 1 year post-delivery. Offspring to women receiving the intervention Our intervention was feasible and could be readily applied in most antenatal care did not have significantly lower mean BMI z-scores at birth or at age 5 and no effect settings with minimal additional funding. Our follow-up period was longer compared of the intervention was seen in the proportion of children with over- or undernutrition with the follow-up in many other, similar published trials, providing valuable infor- at birth or age 5. mation on how the efficacy of our intervention changed over time for both the women and their offspring. Strength and limitations of included studies There was no sign of systematic errors in the original recruitment to the different The choice of scientific methods in this thesis strengthens the grade of the evidence study groups, which could have confounded our data. The high rate of loss to fol- presented. The GRADE system used in Paper I, is internationally recognized and con- low-up regarding long-term PPWR limits the power to detect differences between sidered transparent and systematic. The randomized controlled study design, and the the groups in this aspect. We considered the possibility that a differential bias had fact that the study was performed in a routine clinical setting strengthens our findings been introduced in our results during the follow-up period, and performed an analysis in Papers II-IV. of missing data on mean PPWR and offspring weight and height. Participants lost The original study population was relatively large and representative of all BMI to follow-up did not differ significantly in number, age, parity, pre-pregnancy BMI, categories except underweight women, who were not included. No self-reported body offspring gender or incidence of breastfeeding between the groups. Data lost to fol- weights were used and in accordance with international guidelines, first measurement low-up in Papers II-IV were thereby considered missing at random. of maternal weight occurred at the time of the first antenatal visit (52). Information The quality of data on breastfeeding status was low and limited any ancillary on participating women’s height was self-reported, as is routine in Swedish maternity analysis of associations between breastfeeding status, PPWR and childhood obesi- care. Self-reported height is usually slightly overestimated (53), however, studies of ty. Baseline incidence of breastfeeding did, however, not differ between groups and the use of self-reported values to derive BMI have not found self-reported height to potential confounding of our results by this variable was considered unlikely. So- impact overall conclusions (53). The results are considered to be generalizable to cio-economic status and smoking habits among participants were not included in the other women and offspring sharing the same characteristics. analysis, which must be considered a limitation even though BMI can be considered Total GWG is highly correlated with gestational duration as longer pregnancy du- a valid surrogate variable for socio-economic status in a Swedish population (54). ration provides more opportunity to gain weight. The significant difference in gesta- Measures collected to estimate maternal pre-pregnancy metabolic status and off- tional length between study groups in Papers II-IV was a potential bias and statistical spring obesity were limited to height and weight. The addition of circumferential adjustment was made accordingly in the analysis. Using recently published z-score measures may have increased the quality of our data and limited the major weakness

46 Ann-Kristin rönnberg Gestational Weight Gain 47 ANN-KrIStIN röNNBerg Gestational Weight Gain DISCUSSION charts of GWG for gestational age in relation to maternal BMI (25) could however have improved the depth in our analysis by providing a more exact measure of weight gain in relation to gestational age. The charts were however not available during the Summary of main findings analysis of our data. The general rationale for this thesis was the need to expand knowledge of the impact According to the original VIGA protocol, the women were educated and moti- of an antenatal lifestyle intervention aimed at reducing excessive GWG. Current sci- vated to reach GWG as per IOM recommendations of 1990 (8). At the end of the entific evidence was reviewed and a composite antenatal lifestyle intervention was study period the IOM published revised recommendations on GWG (20). Body mass designed and performed as a RCT. index categories were adapted to WHO standards and the upper limit for women with The body of evidence from published intervention trials in 2009 was found to be pre-gestational obesity was set to 9 kg. A small number of the participating women insufficient to enable development of evidence-based recommendations for clinical would have been classified to a different BMI category based on the new guidelines. practice in maternal health care. Our antenatal lifestyle intervention, tested within They would therefore have received a different education and a different personal- the VIGA trial, was safely performed according to study protocol and resulted in ized graph. The upper limit for women with pre-pregnancy obesity would have been significantly reduced mean maternal GWG but no significant reduction in the propor- lower. Since it was not possible to foresee how the participating women would have tion of women gaining excessively according to IOM recommendations. Women who adhered to the intervention with these new recommendations, our data on efficiency received the intervention had significantly less short-term weight retention after de- of the intervention were analysed in accordance with our original study protocol us- livery, however, no significant difference in weight retention, compared with standard ing the IOM 1990 guidelines. care, was seen 1 year post-delivery. Offspring to women receiving the intervention Our intervention was feasible and could be readily applied in most antenatal care did not have significantly lower mean BMI z-scores at birth or at age 5 and no effect settings with minimal additional funding. Our follow-up period was longer compared of the intervention was seen in the proportion of children with over- or undernutrition with the follow-up in many other, similar published trials, providing valuable infor- at birth or age 5. mation on how the efficacy of our intervention changed over time for both the women and their offspring. Strength and limitations of included studies There was no sign of systematic errors in the original recruitment to the different The choice of scientific methods in this thesis strengthens the grade of the evidence study groups, which could have confounded our data. The high rate of loss to fol- presented. The GRADE system used in Paper I, is internationally recognized and con- low-up regarding long-term PPWR limits the power to detect differences between sidered transparent and systematic. The randomized controlled study design, and the the groups in this aspect. We considered the possibility that a differential bias had fact that the study was performed in a routine clinical setting strengthens our findings been introduced in our results during the follow-up period, and performed an analysis in Papers II-IV. of missing data on mean PPWR and offspring weight and height. Participants lost The original study population was relatively large and representative of all BMI to follow-up did not differ significantly in number, age, parity, pre-pregnancy BMI, categories except underweight women, who were not included. No self-reported body offspring gender or incidence of breastfeeding between the groups. Data lost to fol- weights were used and in accordance with international guidelines, first measurement low-up in Papers II-IV were thereby considered missing at random. of maternal weight occurred at the time of the first antenatal visit (52). Information The quality of data on breastfeeding status was low and limited any ancillary on participating women’s height was self-reported, as is routine in Swedish maternity analysis of associations between breastfeeding status, PPWR and childhood obesi- care. Self-reported height is usually slightly overestimated (53), however, studies of ty. Baseline incidence of breastfeeding did, however, not differ between groups and the use of self-reported values to derive BMI have not found self-reported height to potential confounding of our results by this variable was considered unlikely. So- impact overall conclusions (53). The results are considered to be generalizable to cio-economic status and smoking habits among participants were not included in the other women and offspring sharing the same characteristics. analysis, which must be considered a limitation even though BMI can be considered Total GWG is highly correlated with gestational duration as longer pregnancy du- a valid surrogate variable for socio-economic status in a Swedish population (54). ration provides more opportunity to gain weight. The significant difference in gesta- Measures collected to estimate maternal pre-pregnancy metabolic status and off- tional length between study groups in Papers II-IV was a potential bias and statistical spring obesity were limited to height and weight. The addition of circumferential adjustment was made accordingly in the analysis. Using recently published z-score measures may have increased the quality of our data and limited the major weakness

46 Ann-Kristin rönnberg Gestational Weight Gain 47 ANN-KrIStIN röNNBerg Gestational Weight Gain of BMI, which is its inability to distinguish between elevated adiposity and elevated was seen in the VIGA trial or in other published trials, however, meta-analyses of lean mass. Direct measures of body composition, such as dual X-ray absorptiometry negative maternal and fetal outcome in subgroups with increased risk are still lack- (DXA), would also have added information to our data but were not within the scope ing. Our ancillary analysis of offspring differences, using multiple measures recorded of this study. A high correlation has, however, been reported between the percentage during childhood, demonstrated significantly higher BMI z-scores among offspring of fat mass measured by DXA and the WHO z-scores in children (55). Weight and after antenatal intervention. This finding will, however, need to be confirmed in future height are simple, inexpensive, non-invasive measurements that are routinely record- studies before further speculation on its validity is made. ed in maternal and child health care settings. Research using a weight/height index According to reports of inadequate mean GWG among underweight women in 9 to define obesity has advantages that outweigh its limitations. Despite the potential out of 21 health care regions (24) in Sweden and also among 10% of Swedish women for misclassification of the small percentage of individuals whose high BMI is due with pre-gestational obesity Grade III (25), the issue of safety is highly relevant when to lean muscle mass, the great majority of individuals with high BMI do have excess initiating lifestyle interventions aimed at reducing GWG in a Swedish population. body fat. The overall modest effect of the VIGA trial intervention on GWG and PPWR and The actual “reach” of our intervention was not analysed (i.e. reasons for the de- the null effect on childhood obesity concurs with several similar intervention studies cline, number of actual visits to the midwife, number of weight graphs distributed, in the literature (58-63). Our primary objective was to improve adherence to IOM rec- number of prescriptions of PA, performance of PA). Neither was behavioural change ommendations in an overall healthy pregnant population consisting of women with objectively measured (level of PA, motivational level). This was, however, part of a predominantly normal pre-pregnancy BMI. This was, however, not achieved other conscious, pragmatic approach to our intervention. The cornerstones of our model than in a sub-group of parous women where significant effect was found. The reduc- were a motivational toolbox meant to strengthen and empower the participants’ own tion in mean GWG (1.12 kg) is relatively small and the clinical significance and ex- motivation to avoid excessive GWG and thereby increase the likelihood of more last- pected relevance for future weigh-related outcome can be questioned. To significantly ing behavioural changes and long-term beneficial health effects. Extensive analysis of improve the proportion of women who achieve GWG within the IOM recommenda- “reach” could also influence results by enhancing effects of intervention in a manner tions, a more elaborate and costly mode of intervention than tested in the VIGA trial not reproducible outside of the study setting. is likely to be required. A large body of evidence based on retrospective observational studies associates excessive GWG with adverse maternal and fetal outcome in terms Interpretation and implications of study results of high FBW and PPWR and increased risk of childhood obesity (1, 29, 64). The The quality of evidence of antenatal intervention targeting GWG has improved since lack of prospectively reported positive effects on long-term maternal and offspring 2009, when we concluded our systematic review of current evidence (41). A review outcome does however pose the question as to the extent to which limiting excessive published in 2012 found moderate evidence for successful intervention to limit GWG GWG is beneficial. and pointed at dietary interventions as being most effective (56). A further updated Avoidance of excessive GWG among severly obese women as a mean to improve meta-analysis, published by the Cochrane Collaboration in 2015, conclude that high maternal metabolic function and perinatal outcome has been specifically questioned quality evidence indicate that diet and/or PA, during pregnancy can reduce the risk of by Catalano et al, who suggests that metabolic alterations affect maternal/placental excessive GWG by an avarage of 20% (RR 0.80, 95% CI: 0.73-0.87). This updated function already in the first trimester of pregnancy among these women, and there- review included 65 RCTs in total, with 49 RCTs involving approximately 11,444 par- fore occur before most interventions are initiated (65). According to their hypothe- ticipants, contributing data (42). The risk of GWG below recommendations was how- sis, obese women would have to improve their metabolic condition already before ever increased with 14% (RR 1.14, CI 95%: 1.02-1.27). Data for reduction in mean pregnancy to decrease complications related to fetal size and GDM. The increased GWG were considered too heterogenous to pool and weight-related morbidity, such expression of lipogenic and inflammatory genes in maternal adipose tissue and pla- as risk of CS, macrosomia, preterm delivery and pre-eclampsia were not significantly centa in the first trimester of pregnancy demonstrated, before any phenotypic changes reduced by interventions targeting GWG. A recent review of dietary interventions in weight become apparent, implicate that severly obese women may have limited alone, revealed large methodological variability in studies performed among over- benefit from lifestyle interventions during pregnancy but this remains to be further weight and obese pregnant women (57). The lack of consensus still limits the ability investigated in future studies. to develop clinical guidelines and apply the evidence in clinical practice. The highest risk of excessive GWG is however, among women with pre-pregnancy No evidence of harm as a result of the dietary and PA-based antenatal interventions BMI between 25-35 and in a public health perspective, multiple visits to maternal

48 Ann-Kristin rönnberg Gestational Weight Gain 49 ANN-KrIStIN röNNBerg Gestational Weight Gain of BMI, which is its inability to distinguish between elevated adiposity and elevated was seen in the VIGA trial or in other published trials, however, meta-analyses of lean mass. Direct measures of body composition, such as dual X-ray absorptiometry negative maternal and fetal outcome in subgroups with increased risk are still lack- (DXA), would also have added information to our data but were not within the scope ing. Our ancillary analysis of offspring differences, using multiple measures recorded of this study. A high correlation has, however, been reported between the percentage during childhood, demonstrated significantly higher BMI z-scores among offspring of fat mass measured by DXA and the WHO z-scores in children (55). Weight and after antenatal intervention. This finding will, however, need to be confirmed in future height are simple, inexpensive, non-invasive measurements that are routinely record- studies before further speculation on its validity is made. ed in maternal and child health care settings. Research using a weight/height index According to reports of inadequate mean GWG among underweight women in 9 to define obesity has advantages that outweigh its limitations. Despite the potential out of 21 health care regions (24) in Sweden and also among 10% of Swedish women for misclassification of the small percentage of individuals whose high BMI is due with pre-gestational obesity Grade III (25), the issue of safety is highly relevant when to lean muscle mass, the great majority of individuals with high BMI do have excess initiating lifestyle interventions aimed at reducing GWG in a Swedish population. body fat. The overall modest effect of the VIGA trial intervention on GWG and PPWR and The actual “reach” of our intervention was not analysed (i.e. reasons for the de- the null effect on childhood obesity concurs with several similar intervention studies cline, number of actual visits to the midwife, number of weight graphs distributed, in the literature (58-63). Our primary objective was to improve adherence to IOM rec- number of prescriptions of PA, performance of PA). Neither was behavioural change ommendations in an overall healthy pregnant population consisting of women with objectively measured (level of PA, motivational level). This was, however, part of a predominantly normal pre-pregnancy BMI. This was, however, not achieved other conscious, pragmatic approach to our intervention. The cornerstones of our model than in a sub-group of parous women where significant effect was found. The reduc- were a motivational toolbox meant to strengthen and empower the participants’ own tion in mean GWG (1.12 kg) is relatively small and the clinical significance and ex- motivation to avoid excessive GWG and thereby increase the likelihood of more last- pected relevance for future weigh-related outcome can be questioned. To significantly ing behavioural changes and long-term beneficial health effects. Extensive analysis of improve the proportion of women who achieve GWG within the IOM recommenda- “reach” could also influence results by enhancing effects of intervention in a manner tions, a more elaborate and costly mode of intervention than tested in the VIGA trial not reproducible outside of the study setting. is likely to be required. A large body of evidence based on retrospective observational studies associates excessive GWG with adverse maternal and fetal outcome in terms Interpretation and implications of study results of high FBW and PPWR and increased risk of childhood obesity (1, 29, 64). The The quality of evidence of antenatal intervention targeting GWG has improved since lack of prospectively reported positive effects on long-term maternal and offspring 2009, when we concluded our systematic review of current evidence (41). A review outcome does however pose the question as to the extent to which limiting excessive published in 2012 found moderate evidence for successful intervention to limit GWG GWG is beneficial. and pointed at dietary interventions as being most effective (56). A further updated Avoidance of excessive GWG among severly obese women as a mean to improve meta-analysis, published by the Cochrane Collaboration in 2015, conclude that high maternal metabolic function and perinatal outcome has been specifically questioned quality evidence indicate that diet and/or PA, during pregnancy can reduce the risk of by Catalano et al, who suggests that metabolic alterations affect maternal/placental excessive GWG by an avarage of 20% (RR 0.80, 95% CI: 0.73-0.87). This updated function already in the first trimester of pregnancy among these women, and there- review included 65 RCTs in total, with 49 RCTs involving approximately 11,444 par- fore occur before most interventions are initiated (65). According to their hypothe- ticipants, contributing data (42). The risk of GWG below recommendations was how- sis, obese women would have to improve their metabolic condition already before ever increased with 14% (RR 1.14, CI 95%: 1.02-1.27). Data for reduction in mean pregnancy to decrease complications related to fetal size and GDM. The increased GWG were considered too heterogenous to pool and weight-related morbidity, such expression of lipogenic and inflammatory genes in maternal adipose tissue and pla- as risk of CS, macrosomia, preterm delivery and pre-eclampsia were not significantly centa in the first trimester of pregnancy demonstrated, before any phenotypic changes reduced by interventions targeting GWG. A recent review of dietary interventions in weight become apparent, implicate that severly obese women may have limited alone, revealed large methodological variability in studies performed among over- benefit from lifestyle interventions during pregnancy but this remains to be further weight and obese pregnant women (57). The lack of consensus still limits the ability investigated in future studies. to develop clinical guidelines and apply the evidence in clinical practice. The highest risk of excessive GWG is however, among women with pre-pregnancy No evidence of harm as a result of the dietary and PA-based antenatal interventions BMI between 25-35 and in a public health perspective, multiple visits to maternal

48 Ann-Kristin rönnberg Gestational Weight Gain 49 ANN-KrIStIN röNNBerg Gestational Weight Gain health care during pregnancy present a timely opportunity for motivation and infor- CONCLUSION mation about the general importance of healthy lifestyle regarding diet and PA as well as tobacco and alcohol abuse. Evidence-based advice on GWG should be systemat- Lifestyle behaviours are influenced by an accumulation of experiences during life ically implemented in maternal health care in order to facilitate health weight goals and a life-long approach is needed on multiple levels in society to promote PA and pre-conception, during pregnancy and postpartum. healthy diet in order to reduce the risk of obesity-related disease. The reproductive cycle begins before conception and continues through the first year postpartum and Future research recommendations maternal weight status throughout this entire cycle has a potential effect on mother and offspring both in the short- and in the long-term. Research encompassing implications of gestational weight changes on fetal and ma- In a healthy, predominantly normal weight, Swedish population we found a sig- ternal outcome is still in its cradle. nificant but moderately sized reduction of mean GWG with our antenatal lifestyle Further understanding of the importance of timing and composition of excess or intervention. The proportion of women exceeding recommended GWG was however inadequate weight gained during pregnancy is needed as the potential metabolic im- not significantly reduced. The reduction in mean GWG did not affect maternal long- pact and contribution to long-term risk is likely to vary not just in relation to maternal term PPWR or offspring adiposity at birth or age 5. pre-pregnancy BMI but also depending on timing and fat mass accrual as well as the The relationship between maternal metabolic status and pre-conception, gestation- site of fat depot. al and postpartum outcome is complex and the relevance of maternal weight gain as a Prospective controlled studies of safety regarding GWG below IOM recommen- modifier of risk needs further research. Studies included in this thesis add to the base dations among obese women, stratified for grade of obesity, are needed. An ethical of evidence on the topic but do not support the implementation of the tested interven- framework for trials involving lifestyle modification among pregnant women in a tion in maternal health care in Sweden. health care setting should also be developed in order to allow analysis of possible unintended consequences. Further studies are also needed on pre-conception and intra-uterine programming of maternal adipose tissue, placenta and offspring metabolism. Mechanisms for epigenetic modification in offspring, elicited by (either excessive or inadequate) maternal weight changes, could give us further understanding of long- term trans-generational effects. Adequately powered and performed RCTs of pre-conception and inter-pregnancy lifestyle interventions are of particular interest to answer questions as to what intervention works for whom, why and at what cost.

50 Ann-Kristin rönnberg Gestational Weight Gain 51 ANN-KrIStIN röNNBerg Gestational Weight Gain health care during pregnancy present a timely opportunity for motivation and infor- CONCLUSION mation about the general importance of healthy lifestyle regarding diet and PA as well as tobacco and alcohol abuse. Evidence-based advice on GWG should be systemat- Lifestyle behaviours are influenced by an accumulation of experiences during life ically implemented in maternal health care in order to facilitate health weight goals and a life-long approach is needed on multiple levels in society to promote PA and pre-conception, during pregnancy and postpartum. healthy diet in order to reduce the risk of obesity-related disease. The reproductive cycle begins before conception and continues through the first year postpartum and Future research recommendations maternal weight status throughout this entire cycle has a potential effect on mother and offspring both in the short- and in the long-term. Research encompassing implications of gestational weight changes on fetal and ma- In a healthy, predominantly normal weight, Swedish population we found a sig- ternal outcome is still in its cradle. nificant but moderately sized reduction of mean GWG with our antenatal lifestyle Further understanding of the importance of timing and composition of excess or intervention. The proportion of women exceeding recommended GWG was however inadequate weight gained during pregnancy is needed as the potential metabolic im- not significantly reduced. The reduction in mean GWG did not affect maternal long- pact and contribution to long-term risk is likely to vary not just in relation to maternal term PPWR or offspring adiposity at birth or age 5. pre-pregnancy BMI but also depending on timing and fat mass accrual as well as the The relationship between maternal metabolic status and pre-conception, gestation- site of fat depot. al and postpartum outcome is complex and the relevance of maternal weight gain as a Prospective controlled studies of safety regarding GWG below IOM recommen- modifier of risk needs further research. Studies included in this thesis add to the base dations among obese women, stratified for grade of obesity, are needed. An ethical of evidence on the topic but do not support the implementation of the tested interven- framework for trials involving lifestyle modification among pregnant women in a tion in maternal health care in Sweden. health care setting should also be developed in order to allow analysis of possible unintended consequences. Further studies are also needed on pre-conception and intra-uterine programming of maternal adipose tissue, placenta and offspring metabolism. Mechanisms for epigenetic modification in offspring, elicited by (either excessive or inadequate) maternal weight changes, could give us further understanding of long- term trans-generational effects. Adequately powered and performed RCTs of pre-conception and inter-pregnancy lifestyle interventions are of particular interest to answer questions as to what intervention works for whom, why and at what cost.

50 Ann-Kristin rönnberg Gestational Weight Gain 51 ANN-KrIStIN röNNBerg Gestational Weight Gain ACKNOWLEDGEMENTS siasm and exceptional professional stamina.

I want to express my deepest gratitude to everyone who has helped and supported me All nurses in primary child- and school health care in Örebro Region for assistance during the process of working with this thesis. First of all I would like to thank all the in following offspring growth development and copying charts. participating women for finding the time and courage to contribute to research during a period in life filled with other concerns regarding pregnancy and parenthood.Y our Study administrator Annika Haag for expert handling of randomization and CRF contribution has taken us further. files and research midwife Anette Trygg for diligent work in collecting offspring growth charts. I would also especially like to thank; All midwifes and assistant nurses at the Specialized maternal health care (SMVC) My main supervisor Professor Kerstin Nilsson for guiding me through my PhD edu- unit and at the Delivery- and Postpartum-ward at Örebro University Hospital for cation with constructive advice and an everlasting positive approach to research. Your making our department such an enjoyable workplace. knowledge, integrity and fearless attitude towards new challenges in life are a well of inspiration from which I hope to continue tapping in the future. All my past and present fellow Obstetricians during my fifteen years at the department; Francis K, Lars D, Anna L, Gabriella F, Anna-Lena B, Torbjörn S, Emil- My co-supervisor Associate Professor Ulf Hanson for sharing your vast knowledge ian U, Hanna Ö, Karin H and Ann-Christine N, for being inspirational, educational in the field of maternal health and bringing your calm scientific mind to our meetings and a pleasure to work with in many different ways. You have all added to my bank and manuscripts. Your contribution was inspirational and invaluable Ulf. of experience in the exciting world of obstetrics.

Co-author and former Head of the Department of Gynecology and Obstetrics at Öre- All my other colleagues and friends at the department of obstetrics and gynecology bro University Hospital Ingrid Östlund for sharing your experience of research in at Örebro University Hospital for widening my horizon beyond obstetrical issues. primary maternal health care in the design of the VIGA trial and for creating a positive research environment during your time as Head of the department. My family Marit, Jan, Kerstin and Martin for happy memories and for providing me with a connection between past, present and future. Co-authors Tomas Gottvall and Helena Fadl for inspiration, constructive advice and input on the VIGA trial and for contributing to the budding scientific spirit at our My father Bengt, in living memory, for teaching me to appreciate the beauty of nature department. and for being a constant reminder of the simple necessities of a happy life, far from academic achievements. Statistician Anders Magnusson for assisting in data analysis and discussing statistical methodology with me. You made a significant difference Anders (p<0.001). My friend through life Anna-Rosa, for showing me numerous doors to open and for always pointing out possibilities rather than limits. Former and present Head of department of Gynecology and Obstetrics at Örebro Uni- versity Hospital; René Bangshöj and Ingrid Strandman for past and present focus Isak, Adam and Felicia and your partners and friends for being an inspirational and on facilitating scientific achievements in our department. caring extended family.

Head of midwifes in primary maternal health care in Region Örebro Yvonne Skogs- My sons Cornelis and Hampus for filling my heart with love, for the joy of watching dahl for promoting research in primary maternal health care in our region and facili- you grow and for being spectacular arrows into the future. tating the VIGA trial in every possible way. My Thomas, for exceptional emotional, editorial and graphical support, for your in- All midwifes in primary health care units in Örebro Region for informing, recruiting, quisitive nature and because life with you is just beautiful. I love you. performing intervention and following up patients during the VIGA trial with enthu-

52 Ann-Kristin rönnberg Gestational Weight Gain 53 ANN-KrIStIN röNNBerg Gestational Weight Gain ACKNOWLEDGEMENTS siasm and exceptional professional stamina.

52 Ann-Kristin rönnberg Gestational Weight Gain 53 ANN-KrIStIN röNNBerg Gestational Weight Gain REFERENCES 13. Blomberg M. Maternal and neonatal outcomes among obese women with weight gain below the new Institute of Medicine recommendations. Obstet Gynecol. 1. Ludwig DS, Currie J. The association between pregnancy weight gain and 2011;117(5):1065-70. birthweight: a within-family comparison. Lancet. 2010;376(9745):984-90. 14. Bogaerts A, Ameye L, Martens E, Devlieger R. Weight loss in obese pregnant 2. Rasmussen KM, Abrams B, Bodnar LM, Butte NF, Catalano PM, Maria women and risk for adverse perinatal outcomes. Obstet Gynecol. 2015;125(3):566- Siega-Riz A. Recommendations for weight gain during pregnancy in the context of 75. the obesity epidemic. Obstet Gynecol. 2010;116(5):1191-5. 15. Siega-Riz AM, Viswanathan M, Moos MK, Deierlein A, Mumford S, 3. Tie HT, Xia YY, Zeng YS, Zhang Y, Dai CL, Guo JJ, et al. Risk of childhood Knaack J, et al. A systematic review of outcomes of maternal weight gain according overweight or obesity associated with excessive weight gain during pregnancy: a to the Institute of Medicine recommendations: birthweight, fetal growth, and postpar- meta-analysis. Arch Gynecol Obstet. 2014;289(2):247-57. tum weight retention. Am J Obstet Gynecol. 2009;201(4):339 e1-14. 4. Nehring I, Lehmann S, von Kries R. Gestational weight gain in accordance 16. Pitkin RM. Nutritional support in obstetrics and gynecology. Clin Obstet to the IOM/NRC criteria and the risk for childhood overweight: a meta-analysis. Pe- Gynecol. 1976;19(3):489-513. diatr Obes. 2013;8(3):218-24. 17. Lederman SA, Paxton A, Heymsfield SB, Wang J, Thornton J, Pierson 5. Kapadia MZ, Park CK, Beyene J, Giglia L, Maxwell C, McDonald SD. RN, Jr. Body fat and water changes during pregnancy in women with different body Weight Loss Instead of Weight Gain within the Guidelines in Obese Women during weight and weight gain. Obstet Gynecol. 1997;90(4 Pt 1):483-8. Pregnancy: A Systematic Review and Meta-Analyses of Maternal and Infant Out- 18. Berggren EK, Groh-Wargo S, Presley L, Hauguel-de Mouzon S, Catalano comes. PLoS One. 2015;10(7):e0132650. PM. Maternal fat, but not lean, mass is increased among overweight/obese women 6. Beyerlein A, Schiessl B, Lack N, von Kries R. Associations of gestation- with excess gestational weight gain. Am J Obstet Gynecol. 2015. al weight loss with birth-related outcome: a retrospective cohort study. BJOG. 19. Despres JP, Lemieux I. Abdominal obesity and metabolic syndrome. Nature. 2011;118(1):55-61. 2006;444(7121):881-7. 7. Catalano PM, Mele L, Landon MB, Ramin SM, Reddy UM, Casey B, et al. 20. In: Rasmussen KM, Yaktine AL, editors. Weight Gain During Pregnancy: Inadequate weight gain in overweight and obese pregnant women: what is the effect Reexamining the Guidelines. The National Academies Collection: Reports funded by on fetal growth? Am J Obstet Gynecol. 2014;211(2):137 e1-7. National Institutes of Health. Washington (DC)2009. 8. IOM (Institute of Medicine). Nutrition During Pregnancy. Part I, Weight 21. Tomedi LE, Simhan HN, Chang CC, McTigue KM, Bodnar LM. Gestational Gain; Part II, Nutrient Supplements Washington, DC: National Academy Press; 1990 weight gain, early pregnancy maternal adiposity distribution, and maternal hypergly- [cited 2014 11 Feb]. 468 pp]. Available from: http://www.nap.edu/openbook.php?re- cemia. Maternal and child health journal. 2014;18(5):1265-70. cord_id=1451. 22. Knabl J, Riedel C, Gmach J, Ensenauer R, Brandlhuber L, Rasmussen KM, 9. National Research Council. Weight gain during pregnancy: Reexamining et al. Prediction of excessive gestational weight gain from week-specific cutoff val- the guidelines Washington, DC: The National Academies Press; 2009 [cited 2014 11 ues: a cohort study. J Perinatol. 2014;34(5):351-6. Feb]. Available from: http://www.nap.edu/catalog.php?record_id=12584. 23. Catov JM, Abatemarco D, Althouse A, Davis EM, Hubel C. Patterns of ges- 10. Nohr EA, Vaeth M, Baker JL, Sorensen TIA, Olsen J, Rasmussen KM. Com- tational weight gain related to fetal growth among women with overweight and obe- bined associations of prepregnancy body mass index and gestational weight gain with sity. Obesity. 2015;23(5):1071-8. the outcome of pregnancy (vol 87, pg 1750, 2008). Am J Clin Nutr. 2008;88(6):1705-. 24. The Swedish Pregnancy Register. 2016. 11. Beyerlein A, Schiessl B, Lack N, von Kries R. Optimal gestational weight 25. Johansson K, Hutcheon JA, Stephansson O, Cnattingius S. Pregnancy gain ranges for the avoidance of adverse birth weight outcomes: a novel approach. weight gain by gestational age and BMI in Sweden: a population-based cohort study. Am J Clin Nutr. 2009;90(6):1552-8. Am J Clin Nutr. 2016;103(5):1278-84. 12. Bodnar LM, Siega-Riz AM, Simhan HN, Himes KP, Abrams B. Se- 26. Rong K, Yu K, Han X, Szeto IM, Qin X, Wang J, et al. Pre-pregnancy BMI, vere obesity, gestational weight gain, and adverse birth outcomes. Am J Clin Nutr. gestational weight gain and postpartum weight retention: a meta-analysis of observa- 2010;91(6):1642-8. tional studies. Public Health Nutr. 2015;18(12):2172-82. 27. Mamun AA, Kinarivala M, O’Callaghan MJ, Williams GM, Najman JM,

54 Ann-Kristin rönnberg Gestational Weight Gain 55 ANN-KrIStIN röNNBerg Gestational Weight Gain REFERENCES 13. Blomberg M. Maternal and neonatal outcomes among obese women with weight gain below the new Institute of Medicine recommendations. Obstet Gynecol. 1. Ludwig DS, Currie J. The association between pregnancy weight gain and 2011;117(5):1065-70. birthweight: a within-family comparison. Lancet. 2010;376(9745):984-90. 14. Bogaerts A, Ameye L, Martens E, Devlieger R. Weight loss in obese pregnant 2. Rasmussen KM, Abrams B, Bodnar LM, Butte NF, Catalano PM, Maria women and risk for adverse perinatal outcomes. Obstet Gynecol. 2015;125(3):566- Siega-Riz A. Recommendations for weight gain during pregnancy in the context of 75. the obesity epidemic. Obstet Gynecol. 2010;116(5):1191-5. 15. Siega-Riz AM, Viswanathan M, Moos MK, Deierlein A, Mumford S, 3. Tie HT, Xia YY, Zeng YS, Zhang Y, Dai CL, Guo JJ, et al. Risk of childhood Knaack J, et al. A systematic review of outcomes of maternal weight gain according overweight or obesity associated with excessive weight gain during pregnancy: a to the Institute of Medicine recommendations: birthweight, fetal growth, and postpar- meta-analysis. Arch Gynecol Obstet. 2014;289(2):247-57. tum weight retention. Am J Obstet Gynecol. 2009;201(4):339 e1-14. 4. Nehring I, Lehmann S, von Kries R. Gestational weight gain in accordance 16. Pitkin RM. Nutritional support in obstetrics and gynecology. Clin Obstet to the IOM/NRC criteria and the risk for childhood overweight: a meta-analysis. Pe- Gynecol. 1976;19(3):489-513. diatr Obes. 2013;8(3):218-24. 17. Lederman SA, Paxton A, Heymsfield SB, Wang J, Thornton J, Pierson 5. Kapadia MZ, Park CK, Beyene J, Giglia L, Maxwell C, McDonald SD. RN, Jr. Body fat and water changes during pregnancy in women with different body Weight Loss Instead of Weight Gain within the Guidelines in Obese Women during weight and weight gain. Obstet Gynecol. 1997;90(4 Pt 1):483-8. Pregnancy: A Systematic Review and Meta-Analyses of Maternal and Infant Out- 18. Berggren EK, Groh-Wargo S, Presley L, Hauguel-de Mouzon S, Catalano comes. PLoS One. 2015;10(7):e0132650. PM. Maternal fat, but not lean, mass is increased among overweight/obese women 6. Beyerlein A, Schiessl B, Lack N, von Kries R. Associations of gestation- with excess gestational weight gain. Am J Obstet Gynecol. 2015. al weight loss with birth-related outcome: a retrospective cohort study. BJOG. 19. Despres JP, Lemieux I. Abdominal obesity and metabolic syndrome. Nature. 2011;118(1):55-61. 2006;444(7121):881-7. 7. Catalano PM, Mele L, Landon MB, Ramin SM, Reddy UM, Casey B, et al. 20. In: Rasmussen KM, Yaktine AL, editors. Weight Gain During Pregnancy: Inadequate weight gain in overweight and obese pregnant women: what is the effect Reexamining the Guidelines. The National Academies Collection: Reports funded by on fetal growth? Am J Obstet Gynecol. 2014;211(2):137 e1-7. National Institutes of Health. Washington (DC)2009. 8. IOM (Institute of Medicine). Nutrition During Pregnancy. Part I, Weight 21. Tomedi LE, Simhan HN, Chang CC, McTigue KM, Bodnar LM. Gestational Gain; Part II, Nutrient Supplements Washington, DC: National Academy Press; 1990 weight gain, early pregnancy maternal adiposity distribution, and maternal hypergly- [cited 2014 11 Feb]. 468 pp]. Available from: http://www.nap.edu/openbook.php?re- cemia. Maternal and child health journal. 2014;18(5):1265-70. cord_id=1451. 22. Knabl J, Riedel C, Gmach J, Ensenauer R, Brandlhuber L, Rasmussen KM, 9. National Research Council. Weight gain during pregnancy: Reexamining et al. Prediction of excessive gestational weight gain from week-specific cutoff val- the guidelines Washington, DC: The National Academies Press; 2009 [cited 2014 11 ues: a cohort study. J Perinatol. 2014;34(5):351-6. Feb]. Available from: http://www.nap.edu/catalog.php?record_id=12584. 23. Catov JM, Abatemarco D, Althouse A, Davis EM, Hubel C. Patterns of ges- 10. Nohr EA, Vaeth M, Baker JL, Sorensen TIA, Olsen J, Rasmussen KM. Com- tational weight gain related to fetal growth among women with overweight and obe- bined associations of prepregnancy body mass index and gestational weight gain with sity. Obesity. 2015;23(5):1071-8. the outcome of pregnancy (vol 87, pg 1750, 2008). Am J Clin Nutr. 2008;88(6):1705-. 24. The Swedish Pregnancy Register. 2016. 11. Beyerlein A, Schiessl B, Lack N, von Kries R. Optimal gestational weight 25. Johansson K, Hutcheon JA, Stephansson O, Cnattingius S. Pregnancy gain ranges for the avoidance of adverse birth weight outcomes: a novel approach. weight gain by gestational age and BMI in Sweden: a population-based cohort study. Am J Clin Nutr. 2009;90(6):1552-8. Am J Clin Nutr. 2016;103(5):1278-84. 12. Bodnar LM, Siega-Riz AM, Simhan HN, Himes KP, Abrams B. Se- 26. Rong K, Yu K, Han X, Szeto IM, Qin X, Wang J, et al. Pre-pregnancy BMI, vere obesity, gestational weight gain, and adverse birth outcomes. Am J Clin Nutr. gestational weight gain and postpartum weight retention: a meta-analysis of observa- 2010;91(6):1642-8. tional studies. Public Health Nutr. 2015;18(12):2172-82. 27. Mamun AA, Kinarivala M, O’Callaghan MJ, Williams GM, Najman JM,

54 Ann-Kristin rönnberg Gestational Weight Gain 55 ANN-KrIStIN röNNBerg Gestational Weight Gain Callaway LK. Associations of excess weight gain during pregnancy with long-term (EMPOWaR): a randomised, double-blind, placebo-controlled trial. The lancet Dia- maternal overweight and obesity: evidence from 21 y postpartum follow-up. Am J betes & endocrinology. 2015;3(10):778-86. Clin Nutr. 2010;91(5):1336-41. 41. Ronnberg AK, Nilsson K. Interventions during pregnancy to reduce exces- 28. Nehring I, Schmoll S, Beyerlein A, Hauner H, von Kries R. Gestational sive gestational weight gain: a systematic review assessing current clinical evidence weight gain and long-term postpartum weight retention: a meta-analysis. Am J Clin using the Grading of Recommendations, Assessment, Development and Evaluation Nutr. 2011;94(5):1225-31. (GRADE) system. BJOG. 2010;117(11):1327-34. 29. Mannan M, Doi SA, Mamun AA. Association between weight gain during 42. Muktabhant B, Lawrie TA, Lumbiganon P, Laopaiboon M. Diet or exercise, pregnancy and postpartum weight retention and obesity: a bias-adjusted meta-analy- or both, for preventing excessive weight gain in pregnancy. The Cochrane database sis. Nutr Rev. 2013;71(6):343-52. of systematic reviews. 2015;6:CD007145. 30. Devlieger R, Benhalima K, Damm P, Van Assche A, Mathieu C, Mahmood 43. Opray N, Grivell RM, Deussen AR, Dodd JM. Directed preconception T, et al. Maternal obesity in Europe: where do we stand and how to move forward?: A health programs and interventions for improving pregnancy outcomes for wom- scientific paper commissioned by the European Board and College of Obstetrics and en who are overweight or obese. The Cochrane database of systematic reviews. Gynaecology (EBCOG). Eur J Obstet Gynecol Reprod Biol. 2016. 2015;7:CD010932. 31. Cohen AK, Chaffee BW, Rehkopf DH, Coyle JR, Abrams B. Excessive ges- 44. Nascimento SL, Pudwell J, Surita FG, Adamo KB, Smith GN. The effect of tational weight gain over multiple pregnancies and the prevalence of obesity at age physical exercise strategies on weight loss in postpartum women: a systematic review 40. Int J Obes (Lond). 2014;38(5):714-8. and meta-analysis. Int J Obes (Lond). 2014;38(5):626-35. 32. Amorim AR, Rossner S, Neovius M, Lourenco PM, Linne Y. Does excess 45. van der Pligt P, Willcox J, Hesketh KD, Ball K, Wilkinson S, Crawford D, et pregnancy weight gain constitute a major risk for increasing long-term BMI? Obesity. al. Systematic review of lifestyle interventions to limit postpartum weight retention: 2007;15(5):1278-86. implications for future opportunities to prevent maternal overweight and obesity fol- 33. Davis EM, Babineau DC, Wang X, Zyzanski S, Abrams B, Bodnar LM, et lowing childbirth. Obes Rev. 2013;14(10):792-805. al. Short inter-pregnancy intervals, parity, excessive pregnancy weight gain and risk 46. Amorim Adegboye AR, Linne YM. Diet or exercise, or both, for weight of maternal obesity. Maternal and child health journal. 2014;18(3):554-62. reduction in women after childbirth. The Cochrane database of systematic reviews. 34. Schmitt NM, Nicholson WK, Schmitt J. The association of pregnancy and 2013;7:CD005627. the development of obesity - results of a systematic review and meta-analysis on the 47. Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S, et al. Grading natural history of postpartum weight retention. Int J Obes (Lond). 2007;31(11):1642- quality of evidence and strength of recommendations. BMJ. 2004;328(7454):1490. 51. 48. Guyatt GH, Oxman AD, Kunz R, Falck-Ytter Y, Vist GE, Liberati A, et al. 35. Villamor E, Cnattingius S. Interpregnancy weight change and risk of adverse Going from evidence to recommendations. BMJ. 2008;336(7652):1049-51. pregnancy outcomes: a population-based study. Lancet. 2006;368(9542):1164-70. 49. Advice About Food for You Who are Pregnant. The Swedish National Food 36. Cnattingius S, Villamor E. Weight change between successive pregnan- Administration Stockholm Sweden2011 [updated 2015]. Available from: http:// cies and risks of stillbirth and infant mortality: a nationwide cohort study. Lancet. www.livsmedelsverket.se/globalassets/english/food-habits-health-environment/di- 2016;387(10018):558-65. etary-guidelines/advice-about-food-for-you-who-are-pregnant.pdf). 37. Pandita A, Sharma D, Pandita D, Pawar S, Tariq M, Kaul A. Childhood obe- 50. Group WHOMGRS. WHO Child Growth Standards based on length/height, sity: prevention is better than cure. Diabetes Metab Syndr Obes. 2016;9:83-9. weight and age. Acta Paediatr Suppl. 2006;450:76-85. 38. Poston L. Gestational weight gain: influences on the long-term health of the 51. de Onis M, Onyango A, Borghi E, Siyam A, Blossner M, Lutter C, et al. child. Curr Opin Clin Nutr Metab Care. 2012;15(3):252-7. Worldwide implementation of the WHO Child Growth Standards. Public Health Nutr. 39. Syngelaki A, Nicolaides KH, Balani J, Hyer S, Akolekar R, Kotecha R, et al. 2012;15(9):1603-10. Metformin versus Placebo in Obese Pregnant Women without Diabetes Mellitus. N 52. (NICE). NIoHaCE. Weight management before, during and after pregnancy. Engl J Med. 2016;374(5):434-43. . Manchester, UK: National Institute for Health and Care Excellence. ; 2010. 40. Chiswick C, Reynolds RM, Denison F, Drake AJ, Forbes S, Newby DE, et 53. Brunner Huber LR. Validity of self-reported height and weight in women of al. Effect of metformin on maternal and fetal outcomes in obese pregnant women reproductive age. Maternal and child health journal. 2007;11(2):137-44.

56 Ann-Kristin rönnberg Gestational Weight Gain 57 ANN-KrIStIN röNNBerg Gestational Weight Gain Callaway LK. Associations of excess weight gain during pregnancy with long-term (EMPOWaR): a randomised, double-blind, placebo-controlled trial. The lancet Dia- maternal overweight and obesity: evidence from 21 y postpartum follow-up. Am J betes & endocrinology. 2015;3(10):778-86. Clin Nutr. 2010;91(5):1336-41. 41. Ronnberg AK, Nilsson K. Interventions during pregnancy to reduce exces- 28. Nehring I, Schmoll S, Beyerlein A, Hauner H, von Kries R. Gestational sive gestational weight gain: a systematic review assessing current clinical evidence weight gain and long-term postpartum weight retention: a meta-analysis. Am J Clin using the Grading of Recommendations, Assessment, Development and Evaluation Nutr. 2011;94(5):1225-31. (GRADE) system. BJOG. 2010;117(11):1327-34. 29. Mannan M, Doi SA, Mamun AA. Association between weight gain during 42. Muktabhant B, Lawrie TA, Lumbiganon P, Laopaiboon M. Diet or exercise, pregnancy and postpartum weight retention and obesity: a bias-adjusted meta-analy- or both, for preventing excessive weight gain in pregnancy. The Cochrane database sis. Nutr Rev. 2013;71(6):343-52. of systematic reviews. 2015;6:CD007145. 30. Devlieger R, Benhalima K, Damm P, Van Assche A, Mathieu C, Mahmood 43. Opray N, Grivell RM, Deussen AR, Dodd JM. Directed preconception T, et al. Maternal obesity in Europe: where do we stand and how to move forward?: A health programs and interventions for improving pregnancy outcomes for wom- scientific paper commissioned by the European Board and College of Obstetrics and en who are overweight or obese. The Cochrane database of systematic reviews. Gynaecology (EBCOG). Eur J Obstet Gynecol Reprod Biol. 2016. 2015;7:CD010932. 31. Cohen AK, Chaffee BW, Rehkopf DH, Coyle JR, Abrams B. Excessive ges- 44. Nascimento SL, Pudwell J, Surita FG, Adamo KB, Smith GN. The effect of tational weight gain over multiple pregnancies and the prevalence of obesity at age physical exercise strategies on weight loss in postpartum women: a systematic review 40. Int J Obes (Lond). 2014;38(5):714-8. and meta-analysis. Int J Obes (Lond). 2014;38(5):626-35. 32. Amorim AR, Rossner S, Neovius M, Lourenco PM, Linne Y. Does excess 45. van der Pligt P, Willcox J, Hesketh KD, Ball K, Wilkinson S, Crawford D, et pregnancy weight gain constitute a major risk for increasing long-term BMI? Obesity. al. Systematic review of lifestyle interventions to limit postpartum weight retention: 2007;15(5):1278-86. implications for future opportunities to prevent maternal overweight and obesity fol- 33. Davis EM, Babineau DC, Wang X, Zyzanski S, Abrams B, Bodnar LM, et lowing childbirth. Obes Rev. 2013;14(10):792-805. al. Short inter-pregnancy intervals, parity, excessive pregnancy weight gain and risk 46. Amorim Adegboye AR, Linne YM. Diet or exercise, or both, for weight of maternal obesity. Maternal and child health journal. 2014;18(3):554-62. reduction in women after childbirth. The Cochrane database of systematic reviews. 34. Schmitt NM, Nicholson WK, Schmitt J. The association of pregnancy and 2013;7:CD005627. the development of obesity - results of a systematic review and meta-analysis on the 47. Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S, et al. Grading natural history of postpartum weight retention. Int J Obes (Lond). 2007;31(11):1642- quality of evidence and strength of recommendations. BMJ. 2004;328(7454):1490. 51. 48. Guyatt GH, Oxman AD, Kunz R, Falck-Ytter Y, Vist GE, Liberati A, et al. 35. Villamor E, Cnattingius S. Interpregnancy weight change and risk of adverse Going from evidence to recommendations. BMJ. 2008;336(7652):1049-51. pregnancy outcomes: a population-based study. Lancet. 2006;368(9542):1164-70. 49. Advice About Food for You Who are Pregnant. The Swedish National Food 36. Cnattingius S, Villamor E. Weight change between successive pregnan- Administration Stockholm Sweden2011 [updated 2015]. Available from: http:// cies and risks of stillbirth and infant mortality: a nationwide cohort study. Lancet. www.livsmedelsverket.se/globalassets/english/food-habits-health-environment/di- 2016;387(10018):558-65. etary-guidelines/advice-about-food-for-you-who-are-pregnant.pdf). 37. Pandita A, Sharma D, Pandita D, Pawar S, Tariq M, Kaul A. Childhood obe- 50. Group WHOMGRS. WHO Child Growth Standards based on length/height, sity: prevention is better than cure. Diabetes Metab Syndr Obes. 2016;9:83-9. weight and age. Acta Paediatr Suppl. 2006;450:76-85. 38. Poston L. Gestational weight gain: influences on the long-term health of the 51. de Onis M, Onyango A, Borghi E, Siyam A, Blossner M, Lutter C, et al. child. Curr Opin Clin Nutr Metab Care. 2012;15(3):252-7. Worldwide implementation of the WHO Child Growth Standards. Public Health Nutr. 39. Syngelaki A, Nicolaides KH, Balani J, Hyer S, Akolekar R, Kotecha R, et al. 2012;15(9):1603-10. Metformin versus Placebo in Obese Pregnant Women without Diabetes Mellitus. N 52. (NICE). NIoHaCE. Weight management before, during and after pregnancy. Engl J Med. 2016;374(5):434-43. . Manchester, UK: National Institute for Health and Care Excellence. ; 2010. 40. Chiswick C, Reynolds RM, Denison F, Drake AJ, Forbes S, Newby DE, et 53. Brunner Huber LR. Validity of self-reported height and weight in women of al. Effect of metformin on maternal and fetal outcomes in obese pregnant women reproductive age. Maternal and child health journal. 2007;11(2):137-44.

56 Ann-Kristin rönnberg Gestational Weight Gain 57 ANN-KrIStIN röNNBerg Gestational Weight Gain 54. Holowko N, Chaparro MP, Nilsson K, Ivarsson A, Mishra G, Koupil I, et 65. Catalano P, deMouzon SH. Maternal obesity and metabolic risk to the off- al. Social inequality in pre-pregnancy BMI and gestational weight gain in the first spring: why lifestyle interventions may have not achieved the desired outcomes. Int J and second pregnancy among women in Sweden. J Epidemiol Community Health. Obes (Lond). 2015;39(4):642-9. 2015;69(12):1154-61. 55. Gomez-Campos R, David Langer R, de Fatima Guimaraes R, Contiero San Martini M, Cossio-Bolanos M, de Arruda M, et al. Accuracy of Body Mass Index Cutoffs for Classifying Obesity in Chilean Children and Adolescents. International journal of environmental research and public health. 2016;13(5). 56. Thangaratinam S, Rogozinska E, Jolly K, Glinkowski S, Duda W. Interven- tions to reduce or prevent obesity in pregnant women: a systematic review. Health Technol Assess. 2012;16(31):191. 57. Flynn AC, Dalrymple K, Barr S, Poston L, Goff LM, Rogozinska E, et al. Dietary interventions in overweight and obese pregnant women: a systematic review of the content, delivery, and outcomes of randomized controlled trials. Nutr Rev. 2016;74(5):312-28. 58. Dodd JM, Turnbull D, McPhee AJ, Deussen AR, Grivell RM, Yelland LN, et al. Antenatal lifestyle advice for women who are overweight or obese: LIMIT randomised trial. BMJ. 2014;348:g1285. 59. Dodd JM, Deussen AR, Mohamad I, Rifas-Shiman SL, Yelland LN, Louise J, et al. The effect of antenatal lifestyle advice for women who are overweight or obese on secondary measures of neonatal body composition: the LIMIT randomised trial. BJOG. 2016;123(2):244-53. 60. Poston L, Bell R, Croker H, Flynn AC, Godfrey KM, Goff L, et al. Ef- fect of a behavioural intervention in obese pregnant women (the UPBEAT study): a multicentre, randomised controlled trial. The lancet Diabetes & endocrinology. 2015;3(10):767-77. 61. 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(12):2502-7. 62. Simmons D, Jelsma JG, Galjaard S, Devlieger R, van Assche A, Jans 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(9):1650-6. 63. Phelan S, Phipps MG, Abrams B, Darroch F, Grantham K, Schaffner A, et al. Does behavioral intervention in pregnancy reduce postpartum weight retention? Twelve-month outcomes of the Fit for Delivery randomized trial. Am J Clin Nutr. 2014;99(2):302-11. 64. Woo Baidal JA, Locks LM, Cheng ER, Blake-Lamb TL, Perkins ME, Tav- eras EM. Risk Factors for Childhood Obesity in the First 1,000 Days: A Systematic Review. Am J Prev Med. 2016.

58 Ann-Kristin rönnberg Gestational Weight Gain 59 ANN-KrIStIN röNNBerg Gestational Weight Gain 54. Holowko N, Chaparro MP, Nilsson K, Ivarsson A, Mishra G, Koupil I, et 65. Catalano P, deMouzon SH. Maternal obesity and metabolic risk to the off- al. Social inequality in pre-pregnancy BMI and gestational weight gain in the first spring: why lifestyle interventions may have not achieved the desired outcomes. Int J and second pregnancy among women in Sweden. J Epidemiol Community Health. Obes (Lond). 2015;39(4):642-9. 2015;69(12):1154-61. 55. Gomez-Campos R, David Langer R, de Fatima Guimaraes R, Contiero San Martini M, Cossio-Bolanos M, de Arruda M, et al. Accuracy of Body Mass Index Cutoffs for Classifying Obesity in Chilean Children and Adolescents. International journal of environmental research and public health. 2016;13(5). 56. Thangaratinam S, Rogozinska E, Jolly K, Glinkowski S, Duda W. Interven- tions to reduce or prevent obesity in pregnant women: a systematic review. Health Technol Assess. 2012;16(31):191. 57. Flynn AC, Dalrymple K, Barr S, Poston L, Goff LM, Rogozinska E, et al. Dietary interventions in overweight and obese pregnant women: a systematic review of the content, delivery, and outcomes of randomized controlled trials. Nutr Rev. 2016;74(5):312-28. 58. Dodd JM, Turnbull D, McPhee AJ, Deussen AR, Grivell RM, Yelland LN, et al. Antenatal lifestyle advice for women who are overweight or obese: LIMIT randomised trial. BMJ. 2014;348:g1285. 59. Dodd JM, Deussen AR, Mohamad I, Rifas-Shiman SL, Yelland LN, Louise J, et al. The effect of antenatal lifestyle advice for women who are overweight or obese on secondary measures of neonatal body composition: the LIMIT randomised trial. BJOG. 2016;123(2):244-53. 60. Poston L, Bell R, Croker H, Flynn AC, Godfrey KM, Goff L, et al. Ef- fect of a behavioural intervention in obese pregnant women (the UPBEAT study): a multicentre, randomised controlled trial. The lancet Diabetes & endocrinology. 2015;3(10):767-77. 61. 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(12):2502-7. 62. Simmons D, Jelsma JG, Galjaard S, Devlieger R, van Assche A, Jans 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(9):1650-6. 63. Phelan S, Phipps MG, Abrams B, Darroch F, Grantham K, Schaffner A, et al. Does behavioral intervention in pregnancy reduce postpartum weight retention? Twelve-month outcomes of the Fit for Delivery randomized trial. Am J Clin Nutr. 2014;99(2):302-11. 64. Woo Baidal JA, Locks LM, Cheng ER, Blake-Lamb TL, Perkins ME, Tav- eras EM. Risk Factors for Childhood Obesity in the First 1,000 Days: A Systematic Review. Am J Prev Med. 2016.

58 Ann-Kristin rönnberg Gestational Weight Gain 59 ANN-KrIStIN röNNBerg Gestational Weight Gain SUMMARY IN SWEDISH naden var inte statistiskt signifikant (41% versus 50%; p = 0.086). I studien av kvinnornas återgång till pregravid vikt efter förlossningen visade det SAMMANFATTNING PÅ SVENSKA sig att kvinnor i interventionsgrupp hade en lägre viktretention vid korttidsuppföl- jning efter förlossningen (1.81 kg mot 3.19 kg, p=0.016). Ett år efter förlossningen Viktökning under graviditet (VIGA) hade interventionsgruppen fortfarande mindre viktretention men skillnaden var då – effekter av antenatal livsstilsintervention inte statistiskt signifikant (0.3 kg mot 1.0 kg, p=0.414). Överflödig viktuppgång under graviditet är vanligt förekommande i västvärlden och Överflödig viktuppgång under graviditet var en stark riskfaktor för överflödig vik- har kopplats till en ökad risk för komplikationer under graviditet, förlossning och tretention (>5 kg) ett år efter förlossningen oberoende av vilken studiegrupp kvinnan postpartum för både mor och barn. Framförallt ökar risken för komplikationer som är tillhört (OR 2.44; 95% CI; 1.08–5.52, p=0.029). relaterade till barnets födelsevikt men överflödig viktuppgång har även kopplats till Alla 374 barn till kvinnor i VIGA-studien analyserades i den uppföljande studien en ökad risk för kvinnan att inte återgå till sin pregravida vikt efter förlossningen och om barnfetma. Medelvärdet för barnens BMI z-score eller andelen barn med fetma att barnet får en framtida fetma. var inte lägre hos barnen i interventionsgrupp. Andelen barn med undervikt skiljde En effektiv intervention som förhindrar eller begränsar överflödig viktuppgång sig inte heller signifikant mellan grupperna. Överflödig viktuppgång under graviditet under graviditet skulle därmed teoretiskt, kunna förbättra förlossningsutfall och även gav en ökad risk för barnfetma (BMI z-score >2) vid födelsen (OR=4.51; 95% CI ge positiva långtidseffekter för mor och barn genom att minska risk för framtida fet- 1.95-10.44, p=<0.001) men inte vid fem års ålder. Den starkaste prediktiva riskfak- ma-relaterad sjuklighet. En ökande medvetenhet inom svensk mödrahälsovård om torn för barnfetma vid fem års ålder var maternell pregravid obesitas (OR 4.81; 95% den höga förekomsten av överflödig viktuppgång och dess negativa konsekvenser var CI 1.56-14.83, p=0.006). utgångspunkten för denna avhandling, som omfattar studier av livsstilsintervention Sammanfattningsvis gav VIGA studiens livsstils-intervention en minskad total med syfte att begränsa överflödig viktuppgång under graviditet. viktuppgång under graviditet och en lägre maternell viktretention på kort sikt. Ande- Aktuell forskning inom området eftersöktes, granskades och sammanställdes sys- len kvinnor med överflödig viktuppgång minskades dock inte. Positiva långtidseffek- tematiskt enligt GRADE-metoden 2009. Evidensgraden i då publicerade interven- ter med minskad maternell viktretention och minskad andel barnfetma hos avkomma tionsstudier bedömdes sammantaget vara mycket låg och några evidensbaserade rik- kunde inte heller påvisas. Alternativa metoder och timing för livsstils-intervention i tlinjer för mödrahälsovården avseende intervention kunde inte ges. relation till reproduktion, bör prövas i framtida studier. Minskad förekomst av pre- En randomiserad kontrollerad interventionsstudie (VIGA) inkluderande 445 friska gravid obesitas måste även fortsättningsvis anses vara det primära medlet för att för- kvinnor med BMI ≥19, ålder ≥18 år och ≤16 veckors graviditet genomfördes mellan bättra maternell och fetalt utfall på kort och lång sikt. 2007 och 2010 inom Mödrahälsovården i Region Örebro. Kvinnorna följdes under graviditet och under ett år postpartum. Barnen till inkluderade kvinnor i originalstudi- en följdes sedan inom Barnhälsovården under fem år efter födseln och VIGA-studien avslutades 2015. VIGA studien jämförde standard mödrahälsovård med en kombinerad intervention bestående av utbildning om rekommenderad viktuppgång enligt internationella rekommendationer, utdelning av personlig viktgraf, fysisk aktivitet på recept (FaR) och tätare mätning av kvinnans vikt. Barnen följdes med standardiserade mätningar av vikt och längd och deras tillväxt analyserades i relation till WHO´s internationella standard för tillväxt upp till fem års ålder. Totalt 374 kvinnor kunde analyseras efter fullgången graviditet (Intervention=192, Kontroll=182). Interventionen som testades inom ramen för VIGA studien gav en signifikant lägre total viktuppgång under graviditet (medelvärde intervention=14.2 kg vs. medelvärde kontroll=15.3 kg, p=0.028). Andelen kvinnor som överskred internationella rekommendationer för viktuppgång var lägre efter intervention men skill-

60 Ann-Kristin rönnberg Gestational Weight Gain 61 ANN-KrIStIN röNNBerg Gestational Weight Gain SUMMARY IN SWEDISH naden var inte statistiskt signifikant (41% versus 50%; p = 0.086). I studien av kvinnornas återgång till pregravid vikt efter förlossningen visade det SAMMANFATTNING PÅ SVENSKA sig att kvinnor i interventionsgrupp hade en lägre viktretention vid korttidsuppföl- jning efter förlossningen (1.81 kg mot 3.19 kg, p=0.016). Ett år efter förlossningen Viktökning under graviditet (VIGA) hade interventionsgruppen fortfarande mindre viktretention men skillnaden var då – effekter av antenatal livsstilsintervention inte statistiskt signifikant (0.3 kg mot 1.0 kg, p=0.414). Överflödig viktuppgång under graviditet är vanligt förekommande i västvärlden och Överflödig viktuppgång under graviditet var en stark riskfaktor för överflödig vik- har kopplats till en ökad risk för komplikationer under graviditet, förlossning och tretention (>5 kg) ett år efter förlossningen oberoende av vilken studiegrupp kvinnan postpartum för både mor och barn. Framförallt ökar risken för komplikationer som är tillhört (OR 2.44; 95% CI; 1.08–5.52, p=0.029). relaterade till barnets födelsevikt men överflödig viktuppgång har även kopplats till Alla 374 barn till kvinnor i VIGA-studien analyserades i den uppföljande studien en ökad risk för kvinnan att inte återgå till sin pregravida vikt efter förlossningen och om barnfetma. Medelvärdet för barnens BMI z-score eller andelen barn med fetma att barnet får en framtida fetma. var inte lägre hos barnen i interventionsgrupp. Andelen barn med undervikt skiljde En effektiv intervention som förhindrar eller begränsar överflödig viktuppgång sig inte heller signifikant mellan grupperna. Överflödig viktuppgång under graviditet under graviditet skulle därmed teoretiskt, kunna förbättra förlossningsutfall och även gav en ökad risk för barnfetma (BMI z-score >2) vid födelsen (OR=4.51; 95% CI ge positiva långtidseffekter för mor och barn genom att minska risk för framtida fet- 1.95-10.44, p=<0.001) men inte vid fem års ålder. Den starkaste prediktiva riskfak- ma-relaterad sjuklighet. En ökande medvetenhet inom svensk mödrahälsovård om torn för barnfetma vid fem års ålder var maternell pregravid obesitas (OR 4.81; 95% den höga förekomsten av överflödig viktuppgång och dess negativa konsekvenser var CI 1.56-14.83, p=0.006). utgångspunkten för denna avhandling, som omfattar studier av livsstilsintervention Sammanfattningsvis gav VIGA studiens livsstils-intervention en minskad total med syfte att begränsa överflödig viktuppgång under graviditet. viktuppgång under graviditet och en lägre maternell viktretention på kort sikt. Ande- Aktuell forskning inom området eftersöktes, granskades och sammanställdes sys- len kvinnor med överflödig viktuppgång minskades dock inte. Positiva långtidseffek- tematiskt enligt GRADE-metoden 2009. Evidensgraden i då publicerade interven- ter med minskad maternell viktretention och minskad andel barnfetma hos avkomma tionsstudier bedömdes sammantaget vara mycket låg och några evidensbaserade rik- kunde inte heller påvisas. Alternativa metoder och timing för livsstils-intervention i tlinjer för mödrahälsovården avseende intervention kunde inte ges. relation till reproduktion, bör prövas i framtida studier. Minskad förekomst av pre- En randomiserad kontrollerad interventionsstudie (VIGA) inkluderande 445 friska gravid obesitas måste även fortsättningsvis anses vara det primära medlet för att för- kvinnor med BMI ≥19, ålder ≥18 år och ≤16 veckors graviditet genomfördes mellan bättra maternell och fetalt utfall på kort och lång sikt. 2007 och 2010 inom Mödrahälsovården i Region Örebro. Kvinnorna följdes under graviditet och under ett år postpartum. Barnen till inkluderade kvinnor i originalstudi- en följdes sedan inom Barnhälsovården under fem år efter födseln och VIGA-studien avslutades 2015. VIGA studien jämförde standard mödrahälsovård med en kombinerad intervention bestående av utbildning om rekommenderad viktuppgång enligt internationella rekommendationer, utdelning av personlig viktgraf, fysisk aktivitet på recept (FaR) och tätare mätning av kvinnans vikt. Barnen följdes med standardiserade mätningar av vikt och längd och deras tillväxt analyserades i relation till WHO´s internationella standard för tillväxt upp till fem års ålder. Totalt 374 kvinnor kunde analyseras efter fullgången graviditet (Intervention=192, Kontroll=182). Interventionen som testades inom ramen för VIGA studien gav en signifikant lägre total viktuppgång under graviditet (medelvärde intervention=14.2 kg vs. medelvärde kontroll=15.3 kg, p=0.028). Andelen kvinnor som överskred internationella rekommendationer för viktuppgång var lägre efter intervention men skill-

60 Ann-Kristin rönnberg Gestational Weight Gain 61 ANN-KrIStIN röNNBerg Gestational Weight Gain

Publications in the series Örebro Studies in Medicine

1. Bergemalm, Per-Olof (2004). Audiologic and cognitive long-term sequelae from closed head injury. 2. Jansson, Kjell (2004). Intraperitoneal Microdialysis. Technique and Results. 3. Windahl, Torgny (2004). Clinical aspects of laser treatment of lichen sclerosus and squamous cell carcinoma of the penis. 4. Carlsson, Per-Inge (2004). Hearing impairment and deafness. Genetic and environmental factors – interactions – consequences. A clinical audiological approach. 5. Wågsäter, Dick (2005). CXCL16 and CD137 in Atherosclerosis. 6. Jatta, Ken (2006). Inflammation in Atherosclerosis. 7. Dreifaldt, Ann Charlotte (2006). Epidemiological Aspects on Malignant Diseases in Childhood. 8. Jurstrand, Margaretha (2006). Detection of Chlamydia trachomatis and Mycoplasma genitalium by genetic and serological methods. 9. Norén, Torbjörn (2006). Clostridium difficile, epidemiology and antibiotic resistance. 10. Anderzén Carlsson, Agneta (2007). Children with Cancer – Focusing on their Fear and on how their Fear is Handled. 11. Ocaya, Pauline (2007). Retinoid metabolism and signalling in vascular smooth muscle cells. 12. Nilsson, Andreas (2008). Physical activity assessed by accelerometry in children. 13. Eliasson, Henrik (2008). Tularemia – epidemiological, clinical and diagnostic aspects. 14. Walldén, Jakob (2008). The influence of opioids on gastric function: experimental and clinical studies. 15. Andrén, Ove (2008). Natural history and prognostic factors in localized prostate cancer. 16. Svantesson, Mia (2008). Postpone death? Nurse-physician perspectives and ethics rounds. 17. Björk, Tabita (2008). Measuring Eating Disorder Outcome – Definitions, dropouts and patients’ perspectives. 18. Ahlsson, Anders (2008). Atrial Fibrillation in Cardiac Surgery. 19. Parihar, Vishal Singh (2008). Human Listeriosis – Sources and Routes. 20. Berglund, Carolina (2008). Molecular Epidemiology of Methicillin- Resistant Staphylococcus aureus. Epidemiological aspects of MRSA and the dissemination in the community and in hospitals. 21. Nilsagård, Ylva (2008). Walking ability, balance and accidental falls in persons with Multiple Sclerosis. 22. Johansson, Ann-Christin (2008). Psychosocial factors in patients with lumbar disc herniation: Enhancing postoperative outcome by the identification of predictive factors and optimised physiotherapy. 23. Larsson, Matz (2008). Secondary exposure to inhaled tobacco products. 24. Hahn-Strömberg, Victoria (2008). Cell adhesion proteins in different invasive patterns of colon carcinoma: A morphometric and molecular genetic study. 25. Böttiger, Anna (2008). Genetic Variation in the Folate Receptor-α and Methylenetetrahydrofolate Reductase Genes as Determinants of Plasma Homocysteine Concentrations. 26. Andersson, Gunnel (2009). Urinary incontinence. Prevalence, treatment seeking behaviour, experiences and perceptions among persons with and without urinary leakage. 27. Elfström, Peter (2009). Associated disorders in celiac disease. 28. Skårberg, Kurt (2009). Anabolic-androgenic steroid users in treatment: Social background, drug use patterns and criminality. 29. de Man Lapidoth, Joakim (2009). Binge Eating and Obesity Treatment – Prevalence, Measurement and Long-term Outcome. 30. Vumma, Ravi (2009). Functional Characterization of Tyrosine and Tryptophan Transport in Fibroblasts from Healthy Controls, Patients with Schizophrenia and Bipolar Disorder. 31. Jacobsson, Susanne (2009). Characterisation of Neisseria meningitidis from a virulence and immunogenic perspective that includes variations in novel vaccine antigens. 32. Allvin, Renée (2009). Postoperative Recovery. Development of a Multi-Dimensional Questionnaire for Assessment of Recovery. 33. Hagnelius, Nils-Olof (2009). Vascular Mechanisms in Dementia with Special Reference to Folate and Fibrinolysis. 34. Duberg, Ann-Sofi (2009). Hepatitis C virus infection. A nationwide study of assiciated morbidity and mortality. 35. Söderqvist, Fredrik (2009). Health symptoms and potential effects on the blood-brain and blood-cerebrospinal fluid barriers associated with use of wireless telephones. 36. Neander, Kerstin (2009). Indispensable Interaction. Parents’ perspectives on parent–child interaction interventions and beneficial meetings. 37. Ekwall, Eva (2009). Women’s Experiences of Gynecological Cancer and Interaction with the Health Care System through Different Phases of the Disease. 38. Thulin Hedberg, Sara (2009). Antibiotic susceptibility and resistance in Neisseria meningitidis – phenotypic and genotypic characteristics. 39. Hammer, Ann (2010). Forced use on arm function after stroke. Clinically rated and self-reported outcome and measurement during the sub-acute phase. 40. Westman, Anders (2010). Musculoskeletal pain in primary health care: A biopsychosocial perspective for assessment and treatment. 41. Gustafsson, Sanna Aila (2010). The importance of being thin – Perceived expectations from self and others and the effect on self-evaluation in girls with disordered eating. 42. Johansson, Bengt (2010). Long-term outcome research on PDR brachytherapy with focus on breast, base of tongue and lip cancer. 43. Tina, Elisabet (2010). Biological markers in breast cancer and acute leukaemia with focus on drug resistance. 44. Overmeer, Thomas (2010). Implementing psychosocial factors in physical therapy treatment for patients with musculoskeletal pain in primary care. 45. Prenkert, Malin (2010). On mechanisms of drug resistance in acute myloid leukemia. 46. de Leon, Alex (2010). Effects of Anesthesia on Esophageal Sphincters in Obese Patients. 47. Josefson, Anna (2010). Nickel allergy and hand eczema – epidemiological aspects. 48. Almon, Ricardo (2010). Lactase Persistence and Lactase Non- Persistence. Prevalence, influence on body fat, body height, and relation to the metabolic syndrome. 49. Ohlin, Andreas (2010). Aspects on early diagnosis of neonatal sepsis. 50. Oliynyk, Igor (2010). Advances in Pharmacological Treatment of Cystic Fibrosis. 51. Franzén, Karin (2011). Interventions for Urinary Incontinence in Women. Survey and effects on population and patient level. 52. Loiske, Karin (2011). Echocardiographic measurements of the heart. With focus on the right ventricle. 53. Hellmark, Bengt (2011). Genotypic and phenotypic characterisation of Staphylococcus epidermidis isolated from prosthetic joint infections. 54. Eriksson Crommert, Martin (2011). On the role of transversus abdominis in trunk motor control. 55. Ahlstrand, Rebecca (2011). Effects of Anesthesia on Esophageal Sphincters. 56. Holländare, Fredrik (2011). Managing Depression via the Internet – self-report measures, treatment & relapse prevention. 57. Johansson, Jessica (2011). Amino Acid Transport and Receptor Binding Properties in Neuropsychiatric Disorders using the Fibroblast Cell Model. 58. Vidlund, Mårten (2011). Glutamate for Metabolic Intervention in Coronary Surgery with special reference to the GLUTAMICS-trial. 59. Zakrisson, Ann-Britt (2011). Management of patients with Chronic Obstructive Pulmonary Disease in Primary Health Care. A study of a nurse-led multidisciplinary programme of pulmonary rehabilitation. 60. Lindgren, Rickard (2011). Aspects of anastomotic leakage, anorectal function and defunctioning stoma in Low Anterior Resection of the rectum for cancer. 61. Karlsson, Christina (2011). Biomarkers in non-small cell lung carcinoma. Methodological aspects and influence of gender, histology and smoking habits on estrogen receptor and epidermal growth factor family receptor signalling. 62. Varelogianni, Georgia (2011). Chloride Transport and Inflammation in Cystic Fibrosis Airways. 63. Makdoumi, Karim (2011). Ultraviolet Light A (UVA) Photoactivation of Riboflavin as a Potential Therapy for Infectious Keratitis. 64. Nordin Olsson, Inger (2012). Rational drug treatment in the elderly: ”To treat or not to treat”. 65. Fadl, Helena (2012). Gestational diabetes mellitus in Sweden: screening, outcomes, and consequences. 66. Essving, Per (2012). Local Infiltration Analgesia in Knee Arthroplasty. 67. Thuresson, Marie (2012). The Initial Phase of an Acute Coronary Syndrome. Symptoms, patients’ response to symptoms and opportunity to reduce time to seek care and to increase ambulance use. 68. Mårild, Karl (2012). Risk Factors and Associated Disorders of Celiac Disease. 69. Fant, Federica (2012). Optimization of the Perioperative Anaesthetic Care for Prostate Cancer Surgery. Clinical studies on Pain, Stress Response and Immunomodulation. 70. Almroth, Henrik (2012). Atrial Fibrillation: Inflammatory and pharmacological studies. 71. Elmabsout, Ali Ateia (2012). CYP26B1 as regulator of retinoic acid in vascular cells and atherosclerotic lesions. 72. Stenberg, Reidun (2012). Dietary antibodies and gluten related seromarkers in children and young adults with cerebral palsy. 73. Skeppner, Elisabeth (2012). Penile Carcinoma: From First Symptom to Sexual Function and Life Satisfaction. Following Organ-Sparing Laser Treatment. 74. Carlsson, Jessica (2012). Identification of miRNA expression profiles for diagnosis and prognosis of prostate cancer. 75. Gustavsson, Anders (2012): Therapy in Inflammatory Bowel Disease. 76. Paulson Karlsson, Gunilla (2012): Anorexia nervosa – treatment expectations, outcome and satisfaction. 77. Larzon, Thomas (2012): Aspects of endovascular treatment of abdominal aortic aneurysms. 78. Magnusson, Niklas (2012): Postoperative aspects of inguinal hernia surgery – pain and recurrences. 79. Khalili, Payam (2012): Risk factors for cardiovascular events and incident hospital-treated diabetes in the population. 80. Gabrielson, Marike (2013): The mitochondrial protein SLC25A43 and its possible role in HER2-positive breast cancer. 81. Falck, Eva (2013): Genomic and genetic alterations in endometrial adenocarcinoma. 82. Svensson, Maria A (2013): Assessing the ERG rearrangement for clinical use in patients with prostate cancer. 83. Lönn, Johanna (2013): The role of periodontitis and hepatocyte growth factor in systemic inflammation. 84. Kumawat, Ashok Kumar (2013): Adaptive Immune Responses in the Intestinal Mucosa of Microscopic Colitis Patients. 85. Nordenskjöld, Axel (2013): Electroconvulsive therapy for depression. 86. Davidsson, Sabina (2013): Infection induced chronic inflammation and its association with prostate cancer initiation and progression. 87. Johansson, Benny (2013): No touch vein harvesting technique in coronary by-pass surgery. Impact on patency rate, development of atherosclerosis, left ventricular function and clinical outcome during 16 years follow-up. 88. Sahdo, Berolla (2013): Inflammasomes: defense guardians in host-microbe interactions. 89. Hörer, Tal (2013): Early detection of major surgical postoperative complications evaluated by microdialysis. 90. Malakkaran Lindqvist, Breezy (2013): Biological signature of HER2- positive breast cancer. 91. Lidén, Mats (2013): The stack mode review of volumetric datasets – applications for urinary stone disease. 92. Emilsson, Louise (2013): Cardiac Complications in Celiac Disease. 93. Dreifaldt, Mats (2013): Conduits in coronary artery bypass grafting surgery: Saphenous vein, radial and internal thoracic arteries. 94. Perniola, Andrea (2013): A new technique for postoperative pain management with local anaesthetic after abdominal hysterectomy. 95. Ahlstrand, Erik (2013): Coagulase-negative Staphylococci in Hematological Malignancy. 96. Sundh, Josefin (2013): Quality of life, mortality and exacerbations in COPD. 97. Skoog, Per (2013): On the metabolic consequences of abdominal compartment syndrome. 98. Palmetun Ekbäck, Maria (2013): Hirsutism and Quality of Life with Aspects on Social Support, Anxiety and Depression. 99. Hussain, Rashida (2013): Cell Responses in Infected and Cystic Fibrosis Respiratory Epithelium. 100. Farkas, Sanja (2014): DNA methylation in the placenta and in cancer with special reference to folate transporting genes. 101. Jildenstål, Pether (2014): Influence of depth of anaesthesia on postoperative cognitive dysfunction (POCD) and inflammatory marker. 102. Söderström, Ulf (2014): Type 1 diabetes in children with non-Swedish background – epidemiology and clinical outcome 103. Wilhelmsson Göstas, Mona (2014): Psychotherapy patients in mental health care: Attachment styles, interpersonal problems and therapy experiences 104. Jarl, Gustav (2014): The Orthotics and Prosthetics Users´ Survey: Translation and validity evidence for the Swedish version 105. Demirel, Isak (2014): Uropathogenic Escherichia coli, multidrug- resistance and induction of host defense mechanisms 106. Mohseni, Shahin (2014): The role of ß-blockade and anticoagula- tion therapy in traumatic brain injury 107. Bašić, Vladimir T. (2014): Molecular mechanisms mediating development of pulmonary cachexia in COPD 108. Kirrander, Peter (2014): Penile Cancer: Studies on Prognostic Factors 109. Törös, Bianca (2014): Genome-based characterization of Neisseria meningitidis with focus on the emergent serogroup Y disease 110. von Beckerath, Mathias (2014): Photodynamic therapy in the Head and Neck 111. Waldenborg, Micael (2014): Echocardiographic measurements at Takotsubo cardiomyopathy - transient left ventricular dysfunction. 112. Lillsunde Larsson, Gabriella (2014): Characterization of HPV-induced vaginal and vulvar carcinoma. 113. Palm, Eleonor (2015): Inflammatory responses of gingival fibroblasts in the interaction with the periodontal pathogen Porphyromonas gingivlis. 114. Sundin, Johanna (2015): Microbe-Host Interactions in Post-infectious Irritable Bowel Syndrome. 115. Olsson, Lovisa (2015): Subjective well-being in old age and its association with biochemical and genetic biomarkers and with physical activity. 116. Klarström Engström, Kristin (2015): Platelets as immune cells in sensing bacterial infection. 117. Landström, Fredrik (2015): Curative Electrochemotherapy in the Head and Neck Area. 118. Jurcevic, Sanja (2015): MicroRNA expression profiling in endometrial adenocarcinoma. 119. Savilampi, Johanna (2015): Effects of Remifentanil on Esophageal Sphincters and Swallowing Function. 120. Pelto-Piri, Veikko (2015): Ethical considerations in psychiatric inpatient care. The ethical landscape in everyday practice as described by staff. 121. Athlin, Simon (2015): Detection of Polysaccharides and Polysaccharide Antibodies in Pneumococcal Pneumonia. 122. Evert, Jasmine (2015): Molecular Studies of Radiotheray and Chemotherapy in Colorectal Cancer. 123. Göthlin-Eremo, Anna (2015): Biological profiles of endocrine breast cancer. 124. Malm, Kerstin (2015): Diagnostic strategies for blood borne infections in Sweden. 125. Kumakech, Edward (2015): Human Immunodeficiency Virus (HIV), Human Papillomavirus (HPV) and Cervical Cancer Prevention in Uganda: Prevalence, Risk factors, Benefits and Challenges of Post- Exposure Prophylaxis, Screening Integration and Vaccination. 126. Thunborg, Charlotta (2015): Exploring dementia care dyads’ person transfer situations from a behavioral medicine perspective in physiotherapy. Development of an assessmement scale. 127. Zhang, Boxi (2015): Modulaton of gene expression in human aortic smooth muscle cells by Porphyromonas gingivalis - a possible association between periodontitis and atherosclerosis. 128. Nyberg, Jan (2015): On implant integration in irradiated bone: - clinical and experimental studies. 129. Brocki, Barbara C. (2015): Physiotherapy interventions and outcomes following lung cancer surgery. 130. Ulfenborg, Benjamin (2016): Bioinformatics tools for discovery and evaluation of biomarkers. Applications in clinical assessment of cancer. 131. Lindström, Caisa (2016): Burnout in parents of chronically ill children. 132. Günaltay, Sezin (2016): Dysregulated Mucosal Immune Responses in Microscopic Colitis Patients. 133. Koskela von Sydow, Anita (2016): Regulation of fibroblast activity by kera- tinocytes, TGF-β and IL-1α –studies in two- and three dimensional in vitro models. 134. Kozlowski, Piotr (2016): Prognostic factors, treatment and outcome in adult acute lymphoblastic leukemia. Population-based studies in Sweden. 135. Darvish, Bijan (2016): Post-Dural Puncture Headache in Obstetrics. Audiological, Clinical and Epidemiological studies. 136. Sahlberg Bang, Charlotte (2016): Carbon monoxide and nitric oxide as antimicrobial agents – focus on ESBL-producing uropathogenic E. coli. 137. Alshamari, Muhammed (2016): Low-dose computed tomography of the abdomen and lumbar spine. 138. Jayaprakash, Kartheyaene (2016): Monocyte and Neutrophil Inflammatory Responses to the Periodontopathogen Porphyromonas gingivalis. 139. Elwin Marie (2016): Description and measurement of sensory symptoms in autism spectrum. 140. Östlund Lagerström, Lina (2016): ”The gut matters” - an interdisciplinary approach to health and gut function in older adults. 141. Zhulina, Yaroslava (2016): Crohn’s disease; aspects of epidemiology, clinical course, and fecal calprotectin. 142. Nordenskjöld, Anna (2016): Unrecognized myocardial infarction and cardiac biochemical markers in patients with stable coronary artery disease. 143. Floodeen, Hannah (2016): Defunctioning stoma in low anterior resection of the rectum for cancer: Aspects of stoma reversal, anastomotic leakage, anorectal function, and cost-effectiveness. 144. Duberg, Anna (2016): Dance Intervention for Adolescent Girls with Inter- nalizing Problems. Effects and Experiences. 145. Samano, Ninos (2016): No-Touch Saphenous Veins in Coronary Artery Bypass Grafting. Long-term Angiographic, Surgical, and Clinical Aspects. 146. Rönnberg, Ann-Kristin (2016): Gestational Weight Gain. Implications of an Antenatal Lifestyle Intervention.