Study design and recruitment
A single 12-week intervention was conducted among obese sedentary postmenopausal women recruited from two sources: research staff working at UCLA and the Love/Avon Army of Women. Eligibility criteria included not having had menses for two or more years, age less than 75 years old, being sedentary (undergoing less than 1 hour per week of moderate or vigorous intensity recreational exercise), and having a body-mass-index (BMI) between 25.0 kg/m2 and 35 kg/m2. We excluded women who could not physically undergo the exercise intervention, women currently on a weight loss program, women who ever had weight reduction surgery, or women consuming more than 3 servings of fruits and vegetables per day (to eliminate participants likely to consume healthy diets). We set an upper BMI limit of 35.0 kg/m2 to ensure women would not be too heavy to undergo the exercise intervention.
We utilized two sources of subjects: UCLA staff members and the Love/Avon Army of Women. We posted flyers around the UCLA campus inviting eligible women to participate, and a call was put out to the Army of Women (see reference #29) in early 2009
. A total of 50 women were interested in participating (25 from UCLA and 25 from the Love/Avon Army of Women), and all were screened over the telephone. Of the 50 women, 13 were ineligible, 3 changed their minds, and 26 were not able to fully participate. Five women from UCLA and 3 women from the Love/Avon Army of Women were enrolled. One participant later dropped midway into the study because of transportation difficulties. Seven women in total completed the study. The Institutional Review Board of the University of California at Los Angeles in accordance with assurances filed with and approved by the U.S. Department of Health and Human Services, approved the study, and participants provided written informed consent.
Baseline and post intervention assessments
We conducted blood sampling, breast fluid collection, exercise testing, and anthropometry measurements in the week prior to starting the study. Once the last baseline measurement was made, patients were started the next day. We scheduled measurements to occur over a consecutive three-day period, with fasting hormone blood levels and anthropometric measurements taken on one day, exercise testing on another day, and ductal lavage procedures on the third day. Order of measurement days was random for each subject. Once the order of measurements was established at baseline, the same order was repeated at the end of the intervention, thereby minimizing influence of measurement order on the results. At the 12th week, measurements taken at baseline were repeated.
Body composition was assessed using the body-mass index measure (BMI), DEXA (dual energy x-ray absorptiometry) scanning, and Bioelectrical Impedance Analysis (BIA). BMI measurement included height assessment with a stadiometer and weight (mass) with a calibrated beam scale. To provide accurate measures of total body fat, whole body scans were given to subjects positioned supine on the DEXA scanning table (DEXA, GE Lunar Prodigy ® Advance 2006). The same licensed technician performed all scans. Bioelectrical impedance analysis (310e Bioimpedance Analyzer; Biodynamics, Inc, Seattle, WA) assessed BMR (basal metabolic rate) used to estimate baseline caloric intake for the dietary intervention.
An aerobic fitness assessment determined maximum oxygen consumption at baseline and at the end of the intervention. The maximum oxygen consumption (VO2 max) value was used to provide an intensity starting point for the aerobic exercise 12-week program undertaken for each of the participants. A maximum strength assessment determined strength at baseline and was used similarly to determine starting level for the resistance training regimen.
Testing for aerobic fitness
The VMax Encore®, Palm Springs, CA, designed to measure O2 and CO2 volumes utilized at rest and during exercise, monitored a 12-lead EKG, heart rate, and gaseous inhalation and exhalation. The graded exercise tolerance test was conducted with electrocardiogram monitoring to exclude individuals with subjective or objective evidence of coronary artery disease. The participants were sedentary. Therefore we started fitness assessment at the lowest exercise intensity ramp, and calibrated response to the ramp, with all subjects tested at baseline using the lowest ramp. The speed of the bicycle increased at one-minute intervals and continued increasing until subjects could no longer continue. The level at which they could no longer continue was measured by O2 consumption (VO2Max). All participants improved their fitness level sufficiently to ‘graduate’ to a higher ramp, and therefore we increased the ramp at the end of the intervention.
Testing for muscular strength
The ‘gold standard’ of dynamic strength testing is the 1-repetition maximum (1-RM). The 1-RM is the heaviest weight that can be lifted only once, using good form
. We conducted 1-Rep Maximum strength testing using the Brzycki formula
 on calibrated air displacement weight equipment from Keiser®. The 1-Rep Maximum values were derived separately for the major muscle groups using a variety of weight machines that included Seated Leg Extension; Seated Chest Press; Seated Biceps Curl; Seated Triceps Pushdown; Seated Machine Row; Machine Squat; Seated Leg Press; and Hamstring Curl. Testing for 1-RM involved progressively increasing weight tension until participants reached their maximum exertion point.
Ductal lavage involves the infusion of small amounts of sterile saline into a breast duct through a microcatheter where saline is infused into the duct and then the ductal effluent including the saline is collected. The procedure is commonly used to collect breast epithelial cells but can also be used to measure the endocrine microenvironment of the breast
Ductal lavage was attempted immediately after nipple aspiration. Subjects were placed in the supine position. Skin in the nipple area was cleansed with 70% alcohol, and a fenestrated sterile drape was placed over the nipple. Ductal orifices were enlarged with dilators to facilitate cannulation. A separate microcatheter (available from Susan Love Foundation) was used for each duct cannulation to prevent cross-contamination between different individual ductal systems. After the microcatheter was inserted to a maximum depth of 1.5 cm, approximately 2 to 6 mL of normal saline was infused, and the breast compressed to facilitate recovery of ductal fluid into the collection chamber. This lavage procedure (infusion, compression and effluent collection) is repeated multiple times, instilling an average total volume of approximately 10 to 30 mL of normal saline and recovering approximately 5 to 20 mL of ductal effluent per duct. The exact infusion and effluent volumes was determined by weight to be used in the final calculations of the analytes. Location of each fluid-yielding duct and each cannulated duct was carefully marked on a 64-square nipple grid. The recovered ductal effluent was placed into tubes individually labeled for each cannulated duct. The effluent was spun at 3,000 rev for 10 minutes, the fluid transferred to aliquots, and frozen in a -80°C freezer.
We carefully recorded the breast we sampled and described the duct we cannulated at baseline, using the grid originally developed by Sartorius et al. (1977), so that we could return to the same duct at the 12th week of the intervention
. We limited our sampling to one duct per breast, and re-sampled the same duct from the same breast at the end of the intervention.
Blood specimen collection
Blood was drawn into tubes after a 12 hour fast. We collected serum in tubes without heparin, and plasma and buffy coat in tubes with heparin. Serum, plasma and buffy coat were transferred to aliquots and frozen in a −80°C freezer.
Laboratory measurement of serum and ductal lavage fluid
The study sample was small and all samples, both pre intervention and post intervention, were analyzed in the same batch. Serum leptin was determined by enzyme linked immuno assay (ELISA) from Diagnostic Systems Laboratory (Webster, TX). Serum interleukin 6 (IL
6) was determined using the Quantikine® human immunoassay kit (R&D Systems Inc., Minneapolis, MN). Serum estrone
sulfate was determined by radioimmunoassay using a RIA (radioimmunoassay) assay kit (DSL-5400) from Beckman Coulter (Fullerton, CA), with a manufacturer determined intra-assay precision range of 9.2% for 0.35 ng/mL; 4.6% for 8.89 ngmL; and 4.7% for 59.33 ng/mL, and, our laboratory determined intra-assay precision of 6.5% for a concentration of 15 ng/mL. Serum estradiol concentration was determined using the Cayman EIA kit (Cayman Chemical, Ann Arbor, MI) (intra-assay precision of 15.8% for 41.0 pg/mL; 13% for 102.4 pg/mL; and 7.1% for 640 pg/mL, and a limit of sensitivity of 6.6 pg/mL(manufacturer-determined)). Serum fatty acid analysis was performed in serum hexane extracts. Fatty acids were converted to methyl esters (FAME) according to the method by Bagga et al. (1997)
. FAME were separated and quantified by use of an Agilent Technologies (San Diego, CA) 5890A series II gas chromatography fitted with a model 7673 automatic split-injection system and flame ionization detector and SP2380 stabilized phase fused silica capillary column (30 m × 0.32 mm i.d., 0.25 um film thickness, Supelco, Inc, Bellefonte, PA). Quantification was based on the recovery of a known quantity of the internal standard (tridecanoic acid, NuChek Preparation Inc., Elysian, MN) and on the response ratio of fatty acid standards purchased from NuChek Preparation Inc. (Elysian, MN). For quality control a pooled serum sample was used.
Estradiol concentration in ductal lavage fluid (DLF) was determined by gaschromatography mass spectrometry (GC-MS/MS) analysis as described by Zhang et al. (2006)
. 17β-estradiol (Sigma, E-2758), dissolved in methanol, was used to establish the standard curves. Deuterated estradiol (17β-estradiol-d3, Sigma #491187) was added as internal standard to DLF samples prior to column concentration. Samples were concentrated using the Supelco Discovery solid phase extraction DPA-6S columns (Supelco, Bellefonte, PA), pre-bis(trimethylsilyl)trifluoro-acetamide (BSTFA) with 1% trimethylchlorosilane (TMCS) catalyst (Supelco, Bellefonte PA,#3-3148) for derivatization at 65°C for 30 min. The derivatized samples were dried under a nitrogen stream and reconstituted with 50 μl hexane. Analyses were carried out using a ThermoQuest TRACE™ 2000 gas chromatography coupled with a ThermoQuest TRACE™ MS. Sample injection (1 μL) was in splitless mode. The GC column used was a Restek RTx-5 column (15 m × 0.25 μm × 250 μm). Helium carrier gas was maintained at a constant flow rate of 1.0mLmin − 1.
Interleukin 6 (IL-6) concentration in breast ductal fluid was determined using the Quantikine® human immunoassay kit (R&D Systems Inc., Minneapolis, MN).
Description of the interventions
During the intervention period, patients were asked to consume only prepared meals distributed by the UCLA General Clinical Research Center (GCRC). Diets were comprised of 7 servings of fruits and vegetables, 20% fat, 30% protein, 50% carbohydrates, and 25–35 grams of fiber per day. Suggested caloric intake to achieve weight loss was based on a 500-calorie deficit of the participants’ basal metabolic rate (BMR) as determined by their BIA measurement conducted at the start of the study. BMR is derived from FFM (fat-free mass) using an equation by Grande et al.
. Base caloric level was established at 1200 calories per day.
Supplemental snacks, in addition to the three meal-per-day 1200 calorie regimen, tailored the intervention to each subjects’ BMR. All study subjects in the pilot diet and exercise intervention required snacks to supplement the 1200 calorie per day regimen, which suggested to us that the 1200 calorie baseline was a reasonable starting level to meet caloric requirements in the context of undergoing exercise training. Food was picked up from the GCRC by Center for Human Nutrition staff and distributed to study subjects twice per week. All participants received individual consultation on their respective diet plans with a registered dietitian throughout the study. Every four weeks (week 4 and week 8) subjects were weighed, and blood pressure and BMR measured.
Exercise training was custom-tailored to each woman’s strength and cardiovascular fitness. Participants began at an aerobic intensity that was 50% of their maximum oxygen uptake (VO2Max) determined from the VMax testing. We translated the VO2Max value into MET (metabolic equivalent of energy expenditure) units so that the corresponding intensity level for aerobic exercise could be determined
. Aerobic exercise occurred on treadmills and stationary bicycles. Weight training, beginning at 50% of the 1-repetition maximum (1-RM) recorded during muscular testing, was conducted using standard gym-based weight machines and free weights. Weight machines used in 1-RM testing were all available in the gym and were utilized along with supplemental free weights. At four week intervals, participants received updated exercises designed to increase their aerobic exercise by 10% of their VO2Max, and 10% of their 1-RM so that levels increased to 60%, 70% and 80% of their maximum values respectively.
Each participant met with the same exercise trainer for instruction using aerobic and weight resistance equipment at the designated exercise-training facility at UCLA, Fit Center South, for at least three times a week, for a minimum of one-hour sessions. Fit Center South was open 7 days a week during customary gym hours and was within close walking distance to the Center for Human Nutrition. On-site UCLA staff monitored subjects who use the exercise equipment. Notebooks contained each subject’s exercise routine. Subjects recorded their effort during their exercise intervention. Participants’ gym cards were scanned to validate attendance.
Data were summarized with means and standard deviations calculated for continuous variables and frequencies for categorical variables using SAS 9.2 (Cary, NC). We computed percentage change relative to baseline and 95% confidence intervals for all exercise performance, body composition, and biomarker variables. Change in strength was summarized by averaging the 1-Rep Maximum values derived from the various Kaiser weight machines used for testing, and computing percentage change from baseline. Means are expressed with ± standard deviation (S.D.).