Anthropometric measurements, blood and urinary examinations were assessed at a checkup which we held before and after intervention for six months. Height, weight, body mass index (BMI), systolic and diastolic blood pressures, pulse rate and waist circumference were measured as anthropometry. Brachial-ankle pulse wave velocity (baPWV) and bone mineral density (BMD) at the calcaneus were determined using a pulse pressure analyzer (Form PWV/ABI; Nihon Colin Co., Tokyo, Japan) and an ultrasound bone densitometer (CM-200; Erk Co., Tokyo, Japan), respectively. Blood pressure, BMD and baPWV were measured twice and the mean values were adopted.
All blood samples were drawn between 0700 and 0800 h a.m. after an overnight fast. A 75-g oral glucose tolerance test (OGTT) for measuring glucose and insulin drawn before and at 30 and 120 min after oral glucose ingestion was performed. Additional assessments included as follows: screening blood tests, HbA1c, plasma homocystein and 3-methoxy-4-hydroxy-phenylglycol (MHPG), serum cystatin C, high sensitive C-reactive protein (hsCRP), leptin, high-molecular-weight adiponectin (hmw-adiponectin), prolactin, E2, T, DHEA-S, undercarboxylated osteocalcin (ucOC), N-telopeptides of type I collagen (NTx) and ratio of urinary 8-hydroxy-2’-deoxyguanosine (8-OHdG) to creatinine. Plasma and serum was obtained by centrifuging the blood at 1,000 x g for 15 min at 4C immediately after drawing blood. We commissioned SRL Inc. (Tachikawa, Japan) to measure all of the hematological variables, serum and urinary biochemical and hormonal concentrations. Homocystein and MHPG were measured by HPLC, cystatin C by colloidal gold enhanced immunoassay, leptin, E2 and ucOC by RIA, hmw-adiponectin, prolactin, T and DHEA-S by CLEIA, and hsCRP, NTx and 8-OHdG by ELISA.
The homeostatic model assessment-insulin resistance index (HOMA-IR), which is calculated as fasting immunoreactive insulin (FIRI) (μU/mL) x fasting plasma glucose (FPG) (mg/dL) divided by 405, was used to assess insulin resistance. HOMA of β-cell function index (HOMA-B), computed as the product of 360 x FIRI (μU/mL) divided by the value of FPG (mg/dL) minus 63, has been proposed to be a good measure of steady state β-cell function. Insulinogenic index (IGI), which is the ratio of the 30-min increment in insulin level to the 30-min increment in glucose level in the OGTT, was used as an index of early-phase insulin secretion from β-cells. Estimated glomerular filtration rate (eGFR) and non-high density lipoprotein cholesterol (non-HDL-C) were calculated as 194 x age−0.287 x serum creatinine (CRE) −1.094 (x 0.739 if female) and TC - HDL-C, respectively.
General health status was assessed with the use of the Japanese version of the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36v2TM
] before and after intervention. The SF-36, a standardized and written questionnaire, evaluates eight dimensions of health: physical functioning (PF), role limitations due to physical problems (RP), bodily pain (BP), general health perception (GH), vitality (VT), social functioning (SF), role of limitation due to emotional problems (RE) and mental health (MH). Scores for each domain range from 0 to 100. A higher score indicates better health status. The scores were calculated by a Japanese version of the scoring program. The SF-36v2TM
and the scoring program were obtained from the Institute for Health Outcomes & Process Evaluation Research (Kyoto, Japan).
A target sample size of 58 participants (29 subjects per group) was estimated to provide > =80% power at a 5% level of significance (2-sided) to detect a 0.9 difference in the change in hematocrit (Ht) from baseline to the end of intervention period between the 2 groups, with an assumption of an SD of 5.0 and allowance for a 10% loss to follow up.
Analyses were done according to intention to treat. Data were expressed as mean ± SEM. Baseline characteristics were compared between the RJ group and the control group using the Fisher exact test for categorical variables and Student’s t-test for continuous ones. Non-normally distributed variables were log transformed for further analysis if they were applicable. An analysis of covariance (ANCOVA), adjusted for age, sex, smoking and drinking habits, hypertension, diabetes mellitus, dyslipidemia, arrhythmia, history of ischemic heart disease and apoplexy and the baseline value as covariates, was used to compare change from baseline to 6 months after intervention in anthropometric and biochemical variables and the SF-36 subscale scores between the RJ and control groups. Regarding DHEA-S, T and E2, analyses by gender were also done because the normal range of these hormones was quite different between the sexes. P values less than 0.05 were considered statistically significant. SAS 9.1.3 Service Pack 4 for Windows (SAS Institute Inc., Cary, NC) was used for all statistical analyses.