Effects of Obesity and Physical Inactivity on Pregnancy Outcomes

Exercise during pregnancy is associated with the delivery of leaner, lighter, and healthier infants (Clapp 1990, Clapp 2000). Subsequently, high infant adiposity and birth weight are strong predictors of childhood obesity and adult adiposity (Catalano 2006, Danielzk 2002). Therefore, maternal physical inactivity during pregnancy may have significant ramifications for the child, the effects of which may extend well into adulthood. Exercise during pregnancy also plays an important role in the health of the mother. Active pregnant women tend to gain less weight during pregnancy (Clapp 1995) and retain less weight following pregnancy (O'Toole, 2003). With excessive gestational weight gain being the strongest risk factor for maternal overweight and obesity postpartum, as well as being associated with many adverse maternal and neonatal metabolic outcomes such as adiposity and insulin resistance (Heerwagen 2010), the impact of exercise on maternal and neonatal outcomes could be substantial. The mechanisms underlying these changes are poorly understood and studies which strive to expose them are critical.

Habitual physical activity in non-gravid individuals has been shown to positively alter lipid metabolism by increasing fatty acid oxidation (Martin 1996), but the effect of physical activity on maternal lipid metabolism during pregnancy has not been studied. Due to previous research suggesting that an altered intrauterine metabolic environment may play a significant role in fetal programming (Heerwagen 2010), it is reasonable to believe improvements in maternal lipid metabolism may contribute to improved neonatal metabolic outcomes in exercising pregnant women. Preliminary data from our group found that in obese and lean pregnant women, lipid oxidation rate was significantly correlated to offspring birth weight; suggesting maternal lipid metabolism may contribute to neonatal metabolic outcomes. In inactive pregnant women, impaired lipid oxidative capacity in conjunction with known increased physiologic adipose tissue lipolysis that occurs during pregnancy and obesity would result in excess un-oxidized plasma fatty acids that are likely to be re-esterified in adipose tissue and/or delivered to the fetus. This series of events may contribute to increased maternal and neonatal adiposity.

In addition, generation of excess reactive oxygen species, known byproducts of lipid metabolism, may contribute to altered/abnormal oxidative stress profiles in obese pregnant women. Reactive oxygen species are up-regulated during physiologic pregnancy as well as non-gravid obesity, and research suggests oxidative stress may be related to poorer neonatal outcomes (Heerwagen 2010) . In non-gravid individuals, long-term physical activity has been shown to improve oxidative stress profiles (Fisher-Wellman 2009). Therefore, women who exercise during pregnancy may also have higher antioxidant capacity and lower markers of oxidative stress; both of which may contribute to favorable neonatal outcomes. However, this has not yet been studied.

Obesity is also believed to adversely influence lipid metabolism, oxidative stress, and neonatal outcomes in pregnancy. Therefore, we plan to compare these parameters in obese active and obese inactive pregnant women. This study design will allow us to compare groups in order to determine if unfavorable maternal lipid metabolism and oxidative stress profiles, and neonatal metabolic outcomes (adiposity and insulin resistance) are more attributable to physical inactivity or obesity. Previous research with non-gravid adults suggests that the presence of comorbidity is more correlated with physical activity levels than with body weight (Blair 1989, 1999, Sui 2007). This finding is contrary to much of the previous literature on pregnancy which suggests "obesity may be the most common health risk for the developing fetus" (Heerwagen 2010). Knowledge about maternal lipid metabolism and oxidative stress profiles and their relationships in neonatal outcomes in active and inactive pregnant women can guide lifestyle and medical interventions designed to target factors that may be contributing to poor outcomes in obese pregnancy. Currently, we have collected data on lean, inactive pregnant women that can be used for comparison at the end of all data collection.

This is the first study to examine the relationship between physical activity, lipid metabolism and oxidative stress in obese pregnancy. We anticipate that results from the proposed study will demonstrate the importance of a physically active lifestyle during pregnancy (irrespective of body weight) in order to maximize the short and long-term health of the neonate. In addition, we hope that these results will encourage obese and overweight women of childbearing age to remain or become physically active by demonstrating that physical inactivity has a greater effect on poor neonatal outcomes than obesity. These findings are unique as much of the current literature focuses on the negative impact of maternal obesity on neonatal outcomes. Blair et al. has consistently demonstrated in non-gravid populations that physical inactivity is a stronger predictor of all-cause mortality than obesity (Blair 1989, 1999). Similarly, we believe physical activity in pregnancy is more important than simply maintaining a healthy body weight in improving neonatal outcomes. This idea is novel and innovative as it is previously unexplored in pregnancy and pregnancy outcomes.

In addition to determining the effect of regular physical activity on neonatal outcomes, measuring maternal lipid metabolism and oxidative stress profiles will provide valuable knowledge about mechanisms responsible for improved outcomes in physically active pregnant women. Also, measuring maternal lipid metabolism and oxidative stress profiles during exercise is novel and clinically insightful. This paradigm holds the potential to reveal alterations in maternal metabolism and/or oxidative stress profiles that may not be detected when measuring these factors at rest. Additionally, measuring metabolism and oxidative stress during exercise is clinically useful by providing information about maternal metabolism during activities that will mimic daily lifestyle tasks such as childcare, household chores, etc. (~3-5 METS). Thus, this study design will provide us with valuable insight and enhanced understanding of maternal lipid metabolism and oxidative stress profiles during everyday lifestyle activities.

METHODS

Subjects:

All women who seek pre-natal care at the Women's Health Clinic at Barnes Jewish Hospital/Washington University will be screened for inclusion BMI by history at the clinic. Subjects will be recruited late in their 2nd trimester at the women's health clinic after asking about their exercise habits. All patients who meet criterion with on-going pregnancies will be approached for enrollment in the study. This study will compare 2 groups of pregnant women between 30 and 35 weeks gestation. The first group will inactive obese women and the other will be active obese women. We will recruit ~15 subjects per group (N=30). Groups will be race-matched.

Sample Size Calculation Data from Pomeroy et al.in 2012 stated that when using an accelerometer to measure physical activity and air displacement plethysmography to measure neonatal body composition (the same measurements we are proposing to use), the Spearman correlation coefficient showing the association between maternal physical activity level and neonatal fat free mass is r=0.5226. Using this R-value and an alpha of 0.05, 30 total participants (15 per group) are needed to adequately power our study at .85 (beta=.15).

Study Procedures:

All study procedures will be performed at the Washington University School of Medicine (WUSM) Institute for Clinical and Translational Sciences Clinical Research Unit (CRU).

CRU Visit #1 of 2: Body composition and Fitness Assessment (32-37 weeks gestation):

Maternal Body Composition:

A skin fold measurement will be performed to determine maternal body composition (% body fat). This will be done by pressing folds of the skin at 7 sites with a caliper and recording its thickness as previously described (Jackson and Pollock 1980).

Maternal Physical Fitness Levels:

Maternal fitness levels will be assessed using a submaximal cycle test on a recumbent bicycle. Subjects will sit comfortably on the bicycle while the pedals are properly adjusted so there is a slight bend in the knee when legs are extended. They will then complete the YMCA submaximal multistage cycle ergometer test according ACSM's guidelines for exercise testing and prescription (Thompson 2010). Sady and colleagues concluded that the VO2-Heart Rate extrapolation method is the most precise way to predict VO2max in pregnancy (1985)., and the YMCA test utilizes this method. A 3-lead ECG will be applied to monitor heart rate during the exercise test.

Maternal Physical Activity Levels:

Daily maternal physical activity will be assessed in the week following these tests using the ActiGraph GT3X+ accelerometer (ActiGraph LLC, Pensacola, FL) in order to objectively measure daily physical activity levels. ActiGraph data will be collected for seven consecutive days on the non-dominant wrist at 30 Hertz. Time spent in sedentary and active (light, lifestyle, moderate, vigorous, and very vigorous) activities will be calculated using algorithms from Freedson and colleagues using ActiGraph software (Freedson 1998). We will also measurement maternal physical activity levels subjectively using the Pregnancy Physical Activity Questionnaire (PPAQ). The PPAQ is a valid and reliable instrument to measure physical activity levels during pregnancy (Chasan-Taber 2011). Not only will the PPAQ provide us with additional details about their activity levels, but it will allow us to account for activities that the actigraph may be unable to detect (i.e. riding a stationary bicycle).

Dietary Intake and Composition:

In order to account for differences in diet, subjects will complete the National Institutes of Health's Dietary History Questionnaire II. This dietary assessment has been rigorously validated (Subar 2001) and is widely used among many different populations. Previous literature also demonstrates that dietary history questionnaires are valid and reproducible among pregnant populations (Vioque 2013).

CRU Visit #2 of 2: Lipid metabolism during exercise study (32-37 weeks gestation):

After obtaining height, weight, and vitals, a catheter (IV) will be placed in a hand vein and heated in a warming box prior to each blood draw. Participants will rest for approximately 30 minutes prior to measuring lipid oxidation rate using indirect calorimetry (True One 2400, Parvo Medics, Sandy, UT). Participants will lay supine while a hood device is placed over their head for 15 minutes to measure oxygen consumption and carbon dioxide production in order to determine lipid oxidation rate34. After the initial indirect calorimetry measurement is taken, basal blood collection will be obtained. Basal insulin and glucose levels will be used to calculate maternal insulin resistance via a homeostatic model assessment-insulin resistance (HOMA-IR). After this blood draw, participants will exercise at approximately 50% of their predicted VO2max (based on the YMCA submaximal cycle test) for 30 minutes on the recumbent cycle ergometer (Lode Corvial, InMed, New South Wales, Australia). Blood will be collected at various time points during exercise. Indirect calorimetry (using a mouthpiece, nose clip, and exercise version of the software) will also be performed for 2 minutes at a time to measure lipid oxidation and total body oxygen consumption during low-level exercise. After exercise termination, participants will return to a supine position. Recovery blood draws will be taken and indirect calorimetry will be performed.

Blood drawn at different time points will be used to measure glucose, insulin, free fatty acids, reactive oxygen species (F2- isoprostanes by mass spectrometry (also referred to as 8-iso-PGF2α) (Milne 2007)), and total antioxidant capacity (Total Antioxidant Capacity Assay (TAC), Cell Biolabs, Inc., San Diego, CA). All of these measurements will help us to better understand insulin resistance, oxidative stress, and mechanisms that could be contributing to either condition.

Parturition:

At parturition, maternal weight will be measured and gestational weight gain will be determined. Neonatal weight, length, and head circumference will also be obtained. Infant HOMA-IR and fatty acid delivery to the fetus will be determined by measuring umbilical cord plasma glucose, insulin, and fatty acid concentrations at parturition. Within 48 hours of delivery, neonatal body composition (fat and lean mass) will be measured by skin fold thickness measurement and by air displacement plethysmography (Pea Pod, Life Measurement, Inc., Concord, CA) in the CRU at WUSM.

Statistical Analysis: Repeated measures ANOVA (group x time) will be used to compare lipid oxidation rates and oxidative stress profiles between the 2 groups during pregnancy before, during, and after exercise. Pearson product moment correlation coefficients for normally distributed variables and Spearmen's rank order coefficients for non-normally distributed variables will be used to examine the relationships between maternal lipid oxidation rate, plasma oxidative stress markers, and neonatal metabolic outcomes. We may also use a regression analysis to examine the relationship between maternal physical activity levels in obese women and neonatal body composition and/or insulin resistance (similar to what has been done in normal weight pregnant women by Pomeroy et al. 2012).