Results at 10 Years of the RESOLVE Study (REverse Metabolic SyndrOme by Lifestyle and Various Exercise)

The RESOLVE + 10 protocol was design to provide a 10-year effect of a lifestyle intervention program (exercise, sedentary lifestyle and dietary diet) on abdominal visceral adipose tissue.

In the present protocol, parameters are measured just once, during 3 hours, to compared the results with those of the protocol RESOLVE (5 measurement times over a year 10 years ago: before the thermal modification program of the lifestyle ( exercise, sedentary lifestyle and diet) of 3 weeks (D0), after the thermal program (D21), at the 3rd month (M3), 6th month (M6) and 12th month of follow-up (M12), to which was matched a healthy control group).

Statistical analysis will be performed using Stata software (version 13; Stata-Corp, College Station, Tex., USA). All statistical tests will be two-sided and p inferior to 0.05 will be considered significant. Qualitative variables will be described in terms of numbers and proportions. Quantitative variables will be described in terms of numbers, mean standard deviation or median according to statistical distribution (normality studied using Shapiro-Wilk test). Graphic representations will be complete presentations of results. Comparisons between patients and controls, concerning loss of abdominal visceral adipose tissue at 10 years, will be performed using Student t-test or Mann-Whitney test if the assumptions of anova are not met (normality and homoscedasticity analyzed using Fisher-Snedecor test). The results will be expressed as effect-size and 95 percent confidence intervals. The primary analysis should be completed by multivariable analysis (linear regression with logarithmic transformation of dependent outcome if necessary) considering an adjustment on covariates fixed according to univariate results and clinical relevance. The results will be expressed as regression coefficients and 95percent confidence intervals. Furthermore, the abdominal visceral adipose tissue loss kinetics in the first year of follow-up will be considered an independent variable and treated as a functional variable. The intra-group comparisons (for patients and for controls) will be performed using paired statistical tests: paired t-test or Wilcoxon test. The results will be expressed as effect-size and 95 percent confidence intervals. Comparisons between independent groups will be done similarly as presented previously for other quantitative outcomes (body composition, quality of life, skin conductance, etc.). Multivariable analyses will be applied similarly as described previously. The comparisons concerning categorical variables will be realized using Chi-squared or when appropriate Fischer-exact test. The relations between quantitative outcomes will be analyzed using correlation coefficients (Pearson or Spearman according to statistical distribution). Considering the several multiple comparisons, a correction of the type I error will be applied (Sidak's correction).