TY - JOUR
T1 - Association between weight fluctuation and fasting insulin concentration in Japanese men
AU - Yatsuya, H.
AU - Tamakoshi, K.
AU - Yoshida, T.
AU - Hori, Y.
AU - Zhang, H.
AU - Ishikawa, M.
AU - Zhu, S.
AU - Kondo, T.
AU - Toyoshima, H.
N1 - Funding Information:
This work was supported by Grant 13770192, 13470087, 12670352 from the Ministry of Education, Culture, Sports, Science and Technology of Japan.
PY - 2003/4/1
Y1 - 2003/4/1
N2 - OBJECTIVE: To investigate whether long-term weight fluctuation is associated with the fasting serum insulin concentration. DESIGN AND SUBJECTS: Weight histories of 1932 male Japanese workers aged 40-59y were analyzed in relation to their current fasting serum insulin concentration. MEASUREMENTS: Individual weight fluctuation was calculated by root mean square error (RMSF) along the linear regression line of weight measured at five to six different ages. RESULTS: The mean RMSF and fasting insulin concentration were 1.22 kg and 4.5 μU/ml, respectively. The multivariate adjusted insulin level became higher with the increase in weight fluctuation. Subanalysis stratified by current body mass index (BMI) showed that the multivariate adjusted insulin level in individuals in the top quartile of fluctuation was 4.3 μU/ml, against 3.9 μU/ml in those in the bottom quartile (P = 0.018, analysis of covariance (ANCOVA)) in the normal weight subgroup with current BMI below 25 kg/m2. In the overweight subgroup with BMI 25 kg/m2 or above, the level was 6.9 μU/ml in individuals in the top quartile and 6.2 μU/ml in those in the bottom quartile (P = 0.054, ANCOVA). CONCLUSION: The results suggest that weight fluctuation increases the risk of developing hyperinsulinemia. Prospective observations together with measurement of changes in adiposity are needed for confirmation.
AB - OBJECTIVE: To investigate whether long-term weight fluctuation is associated with the fasting serum insulin concentration. DESIGN AND SUBJECTS: Weight histories of 1932 male Japanese workers aged 40-59y were analyzed in relation to their current fasting serum insulin concentration. MEASUREMENTS: Individual weight fluctuation was calculated by root mean square error (RMSF) along the linear regression line of weight measured at five to six different ages. RESULTS: The mean RMSF and fasting insulin concentration were 1.22 kg and 4.5 μU/ml, respectively. The multivariate adjusted insulin level became higher with the increase in weight fluctuation. Subanalysis stratified by current body mass index (BMI) showed that the multivariate adjusted insulin level in individuals in the top quartile of fluctuation was 4.3 μU/ml, against 3.9 μU/ml in those in the bottom quartile (P = 0.018, analysis of covariance (ANCOVA)) in the normal weight subgroup with current BMI below 25 kg/m2. In the overweight subgroup with BMI 25 kg/m2 or above, the level was 6.9 μU/ml in individuals in the top quartile and 6.2 μU/ml in those in the bottom quartile (P = 0.054, ANCOVA). CONCLUSION: The results suggest that weight fluctuation increases the risk of developing hyperinsulinemia. Prospective observations together with measurement of changes in adiposity are needed for confirmation.
UR - http://www.scopus.com/inward/record.url?scp=0037398106&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0037398106&partnerID=8YFLogxK
U2 - 10.1038/sj.ijo.0802221
DO - 10.1038/sj.ijo.0802221
M3 - Article
C2 - 12664081
AN - SCOPUS:0037398106
SN - 0307-0565
VL - 27
SP - 478
EP - 483
JO - International Journal of Obesity
JF - International Journal of Obesity
IS - 4
ER -
