Exploratory analysis of eating- and physical activity-related outcomes from a randomized controlled trial for weight loss maintenance with exercise and liraglutide single or combination treatment

Weight regain following weight loss is a significant challenge in obesity treatment. While initial weight loss can be achieved through calorie restriction, individuals often regain weight, typically returning to their original weight within five years. This weight regain involves biological factors like reduced energy expenditure and changes in appetite-regulating hormones promoting increased food intake. However, alterations in eating and sedentary behaviors also play a crucial role in weight management. Obesity is characterized by behaviors such as low cognitive restraint, high emotional eating, and increased reward responses to energy-dense foods, making weight management difficult. Sedentary behavior and physical activity significantly impact daily energy expenditure; reduced physical activity doesn't always lead to proportional energy intake reduction, and a sedentary lifestyle increases obesity risk. The relationship between weight loss and physical activity is debated, with some suggesting decreased non-exercise physical activity after weight loss hindering maintenance, while others propose weight loss reducing physical burden and facilitating activity. GLP-1 receptor agonists (GLP-1RAs) are known to induce weight loss by enhancing satiety and reducing hunger in short-term studies. However, their long-term effects on appetite during weight loss maintenance remain less understood. Exercise is recommended for weight loss maintenance to increase energy expenditure and counteract post-weight loss energy expenditure reduction. While it's proposed to improve satiety and reduce cravings, the effects of exercise on appetite after weight loss require further investigation. This study aimed to investigate the effects of exercise, liraglutide, or both on appetite, eating habits, sedentary behavior, and physical activity during a one-year weight loss maintenance period following an initial diet-induced weight loss.

Existing literature highlights the complexities of weight loss maintenance. Studies indicate that adaptive thermogenesis, a decrease in resting and non-resting energy expenditure beyond what's expected based on body mass loss, contributes to weight regain. Changes in appetite-regulating hormones further complicate weight maintenance by favoring increased food intake. Eating behaviors in obese individuals, such as low cognitive restraint, high emotional and uncontrolled eating, and increased reward responses to energy-dense foods, present challenges in weight management. Sedentary behavior and physical activity significantly influence total daily energy expenditure. Studies suggest a lack of proportional decrease in energy intake with decreased physical activity, and a sedentary lifestyle is linked to increased obesity risk. While the impact of weight loss on physical activity is debated, an increase in sedentary behavior could hinder weight maintenance. Short-term studies show GLP-1RAs like liraglutide promote satiety and reduce hunger, but their effects during long-term weight loss maintenance are less clear. Exercise, while beneficial for weight management, has a complex relationship with appetite; it might improve satiety but also increase appetite to match energy demands. Existing studies on these interventions, particularly concerning their combined effect on eating and sedentary behavior during weight maintenance, are limited, highlighting the need for this study.

This study was a randomized, placebo-controlled, 2-by-2 factorial trial. 215 adults with obesity (BMI 32-43 kg/m²) without diabetes were recruited. Participants underwent an initial 8-week low-calorie diet (-800 kcal/day). Those losing ≥5% body weight (n=195) were then randomized (1:1:1:1 ratio, stratified by sex and age) to one of four groups for a one-year weight loss maintenance phase: 1. Exercise + placebo 2. Liraglutide 3.0 mg/day 3. Exercise + liraglutide 4. Placebo Liraglutide (or placebo) dosage started at 0.6 mg/day and increased weekly to 3.0 mg/day (or highest tolerated dose). Exercise involved a 7-week ramp-up, followed by supervised group sessions (twice weekly) and individual moderate-to-vigorous intensity exercise (twice weekly). Placebo and liraglutide groups maintained their usual physical activity. Eating behavior was assessed using a Danish version of the Three-Factor Eating Questionnaire-R18, measuring cognitive restraint, emotional eating, and uncontrolled eating. Appetite was evaluated using a visual analog scale (VAS) for prospective food consumption, hunger, fullness, and satiety, both in fasting and postprandial states after a standardized mixed meal (600 kcal). An overall appetite suppression score was calculated. Food preferences were assessed using the Leeds Food Preference Questionnaire (LFPQ), measuring explicit liking and implicit wanting for four food categories. Exercise energy expenditure was measured using sports watches with heart rate monitors. Sedentary behavior and physical activity were objectively measured using wrist-worn accelerometers over seven consecutive days at five time points (before and after diet, weeks 13, 26, and 52). The International Physical Activity Questionnaire-Short Form (IPAQ-SF) provided subjective measures of physical activity and sedentary time. Linear mixed models analyzed outcome changes, accounting for time, group, sex, age, and time-group interactions. Multiple linear regression examined associations between behavioral changes and weight change. Both per-protocol and intention-to-treat analyses were conducted.

After one year, the placebo group showed a 14% decrease in postprandial appetite suppression score, a 31-minute increase in daily sedentary time, and weight regain (6.1 kg). Liraglutide prevented the appetite suppression decrease (0% vs. -14%, P=0.023) and maintained weight loss. The exercise group did not differ from placebo in appetite or sedentary behavior despite maintaining weight loss. The combination group showed a 13% increase in cognitive restraint score (13% vs. -9%, P=0.042) and a 41-minute decrease in daily sedentary time (-10 vs. 31 min/day, P=0.049) compared to placebo, leading to additional weight loss. Regression analysis revealed associations between increased cognitive restraint (P=0.003) and increased moderate-to-vigorous physical activity (P=0.016) with less weight regain. Increased uncontrolled eating score showed a trend toward weight regain (P=0.07). Liraglutide prevented the increase in postprandial hunger and prospective food consumption observed in the placebo group. Exercise, despite increased energy expenditure, did not increase appetite more than placebo. The combination group had improved cognitive restraint and reduced sedentary time. Intention-to-treat analyses generally supported the per-protocol findings, although the accelerometer-measured sedentary time in the combination group was not significantly different from the placebo group in this analysis.

This study demonstrates that weight loss is associated with increased appetite and sedentary time, potentially contributing to weight regain. Liraglutide effectively countered this increased appetite, maintaining weight loss. Exercise, while not increasing appetite further than placebo, helped maintain weight loss when combined with increased energy expenditure. The combination of liraglutide and exercise resulted in improved cognitive restraint and reduced sedentary time, contributing to substantial weight loss. These findings highlight the multifaceted nature of weight management, where targeting both eating behaviors and physical activity is crucial for long-term success. The beneficial effects of liraglutide on appetite align with its known mechanisms of action. Exercise, despite its known benefits, did not significantly impact appetite in this context. The combination strategy proved most effective by addressing both appetite and sedentary behavior. These results underscore the need for comprehensive weight maintenance strategies that address both biological and behavioral aspects of weight control.

This study demonstrates that weight regain after weight loss is associated with increased appetite and sedentary time. Liraglutide effectively prevented the increase in appetite, while exercise, despite increased energy expenditure, did not further increase appetite compared to the placebo group. The combination of liraglutide and exercise was the most effective strategy, improving cognitive restraint and reducing sedentary time, leading to greater weight loss. These results emphasize the importance of comprehensive weight management strategies that target both eating behaviors and physical activity. Future research should investigate the long-term effects of these interventions and explore other strategies for managing appetite and sedentary behavior after weight loss.

While the study's strengths include its randomized, placebo-controlled design and one-year intervention period, certain limitations exist. Causal inferences between the measured outcomes and weight changes cannot be definitively established, although they appear probable. Subset selection bias may affect the per-protocol analysis. However, intention-to-treat analyses generally support the per-protocol findings, minimizing this concern. Lastly, it's important to note that this study's exploratory analyses were unadjusted for multiple comparisons, limiting the strength of some statistical inferences.