LAUSR

Background: The characteristics and prognosis of healthy obesity (HO) still remain unclear. We aimed to examine the characteristics of healthy severe obesity (HSO), defined by a novel approach, with a focus on self-reported physical activity (PA) and a genetic risk score for type 2 diabetes (GRS DM2). Methods: A cross-sectional analysis was carried out in a subsample of severly obese subjects (BMI≥35 kg/m2; n = 809) selected from the population-based Malmo Preventive Project (MPP). Subjects with HSO (n = 57) were defined by having no records of hospitalisation in the Swedish Hospital Discharge Register during a time period of 33.4 ± 3.9 years between baseline (1974–1992) and the end of follow-up (31st of December 2014). They were compared to subjects with unhealthy severe obesity (USO; n = 752), as well as age- and sex-matched non-obese controls (n = 1618). Results: Subjects with HSO had a significantly lower GRS DM2 (HSO 40.4 ± 3.7 vs. USO 41.8 ± 3.8, p = 0.007). Compared to subjects with USO, the HSO subjects reported significantly more leisure-time physical activity, PA (p = 0.016). There were no significant differences between HSO and USO subjects in the distribution of fat mass or obesity-associated gene phenotypes (FTO gene; variant rs9939609; p = 0.8). Conclusion: Higher PA and lower GRS DM2 might be protective factors against all-cause hospitalisation in subjects with severe obesity. These findings support the concept of HO being fat but fit. Still, it remains unclear whether higher PA is causally related to HSO, and which role environmental factors such as PA play in the interaction with genetic factors such as GRS DM2. (Less)