Relationship between insulin sensitivity and long-term weight change in adults

Objective: To determine whether insulin resistance is associated with reduced risk of weight gain. Methods: In this retrospective study, we reviewed medical records of volunteers who had undergone a euglycemic, hyperinsulinemic clamp in research studies at Mayo Clinic between October 20, 1986, and January 30, 2002. Data from volunteers who had at least 1 year of follow-up were analyzed, which included height, weight, date of birth, ethnicity, body composition (percentage body fat and kilograms of fat-free mass), waist-to-hip ratio, date of insulin clamp, fasting plasma glucose and insulin, average plasma glucose and insulin concentrations during the final 30 minutes of each insulin clamp step, and average glucose infusion rate (μmol·kg fat free mass^-1·min^-1) during the final 30 minutes of each insulin clamp step. We abstracted the following for each medical encounter: date of visit, height, weight, diagnoses, procedures, and medication use. For the purposes of statistical analysis, the diagnoses, procedures, and medications were then converted into Current Procedural Terminology codes and National Drug Codes. Results: One hundred sixteen patients had at least 1 year of follow-up. The average baseline body mass index was 29.4 ± 5.3 kg/m², and the follow-up time averaged 8.1 ± 5.9 years. We found no significant correlation between baseline insulin action and annual weight change or overall weight change (P = .60 and P = .11, respectively) or between log-transformed insulin action and annual weight gain (P = .61). Conclusion: These results suggest that in free-living, healthy, white adults, there is not a clinically meaningful relationship between insulin action and subsequent weight change.