Weight cycling increases mortality risk
High body weight variability is associated with an increased risk for death independent of traditional cardiovascular risk factors, mean body weight and body weight change, according to an analysis of Korean adults published in The Journal of Clinical Endocrinology & Metabolism.
In a population-based analysis of adults participating in the ongoing Korean Genome and Epidemiology Study, the researchers also found that the effect of body weight fluctuation on incident diabetes was protective in adults with obesity, but negative in lean adults, whereas there was no association observed between weight cycling and CVD.
“It is important to maintain a heathy body weight and avoid body weight cycling,” Hak C. Jang, MD, PhD, professor in the department of internal medicine at Seoul National University (SNU) Bundang Hospital and SNU College of Medicine in Seongnam, South Korea, told Endocrine Today. “Weight cycling, or fluctuation, is an important predictor for poor health outcomes.”
Jang and colleagues analyzed data from 3,678 adults aged 40 to 69 years in 2001-2002, following the cohort biannually (last data obtained in November 2016; 1,616 men). At each visit, body weight and BMI were measured. Intraindividual body weight variability, defined as average successive variability, was calculated as the mean of each body weight change. Mortality was assessed via death statistics from the Korean National Statistics Office; baseline and follow-up events of CVD (acute myocardial infarction, coronary artery disease, cerebrovascular disease and peripheral artery disease) were recorded via questionnaire.
Mean baseline body weight was 61.5 kg, and mean follow-up body weight was 61 kg for the cohort. The delta change of body weight from baseline to last observation was –1.3 kg; median average successive variability of body weight was 1.75, with this cutoff value used to stratify the cohort into two groups, according to the researchers.
The researchers found that adults with high body weight variability (above the cutoff) were more likely to have obesity and lost more body weight vs. those with low body weight variability (below the cutoff). At baseline, more adults in the high body weight variability group vs. the low variability group had hypertension (28.6% vs. 25.6%; P = .049), diabetes (13.8% vs. 9.4%; P < .001) or CVD (3.7% vs. 2.1%; P = .004).
During a median follow-up of 14 years, 90 adults (4.9%) in the low body weight variability group and 173 adults (9.4%) in the high body weight variability group died (P < .001), with 43 of 263 total deaths attributed to CV causes, according to the researchers. They did not observe between-group differences with respect to CV event rates in the high vs. low body weight variability groups (4.4% vs. 3.6%). Additionally, those in the high body weight variability group developed diabetes at a similar rate (8.4%) when compared with adults in the low body weight variability group (8.5%).
The researchers also found that, when comparing event rates between quintiles, the rate of mortality increased with each higher quintile of body weight variability; however, a similar trend was not observed for diabetes or CVD.
In Cox proportional hazard models with average successive variability of body weight as a continuous variable, researchers found that for each 1-U increase of body weight average successive variability, HR for mortality was 1.46 (95% CI, 1.32-1.62), whereas HR for diabetes was 1.05 (95% CI, 0.91-1.21) and HR for CVD was 1.11 (95% CI, 0.9-1.35). Results persisted after adjustment for age, sex, HbA1c, systolic blood pressure and other factors.
“In the view of pathophysiology, mechanistic study is needed such as the biological change of adipose tissue in people with higher body weight fluctuation,” Jang said. “In the clinical perspective, it is necessary to develop new interventions to maintain a heathier body weight after weight reduction.” – by Regina Schaffer
For more information:
Hak C. Jang, MD, PhD, can be reached at Seoul National University College of Medicine, 166 Gumi-ro, Bundang-Gu, Seongnam-Si, Gyeonggi-Do, 463-707, South Korea; email: firstname.lastname@example.org.
Disclosures: The Korea Centers for Disease Control and Prevention supported this study. The authors report no relevant financial disclosures.