Update in Diabetes Care Exercise and Bariatric Surgery Ted Adams, PhD, MPH Intermountain LiVe Well Center Salt Lake October 6, 2017
There is no drug in current or perspective use that holds as much promise for sustained health as a lifetime program of physical activity. American Medical Association
Physical Activity/Exercise and Diabetes: A Position Statement of the American Diabetes Association Benefits of exercise/physical activity in diabetic patients Improve blood glucose control in type 2 diabetes Reduced cardiovascular risk factors Contributes to weight loss Improves well-being May prevent or delay type 2 diabetes Improve cardiovascular fitness, muscle strength, insulin sensitivity, etc. Diabetes Care 2016;39:2065-2079
R x Exercise/physical activity should be tailored to meet specific needs of each individual patient Refer to Table 3, page 2070, for specific recommendations related to types, intensity, duration and progression of exercise prescription. Refer to Table 5, page 2073, for physical activity considerations. Precautions and recommended activities for exercising with health-related complications. Be mindful of pregnancy, hypo- and hyper-glycemia, medication effects and heat-related illness Diabetes Care 2016;39:2065-2079
ADA Abridged Standards of Care, 2017 R x Exercise/physical prescription - children and adolescents Type 1, type 2 or prediabetes 60 min/day of moderate or vigorous intensity activity, with vigorous, muscle strengthening and bonestrengthening activities included at least 3 days/week Clinical Diabetes published online 12/15/2016
ADA Abridged Standards of Care, 2017 R x Exercise/physical prescription - most adults, Type 1 and 2 Type 1 or type 2 150 min or more of moderate-tovigorous intensity activity per week (spread out over at least 3 days/wk, not missing more than 2 days in a row 2-3 sessions/wk or resistance exercise (not in a row) Decrease time spent in sedentary behavior; interrupt prolonged sitting every 30 min Flexibility and balance training 2-3 times/wk for older adults (yoga and tai chi - flexibility, strength, balance Clinical Diabetes published online 12/15/2016
Prospective and retrospective observational studies in bariatric surgery Can contribute (in a unique way) to the understanding of long-term outcomes of bariatric surgery. Electronic medical record Health plans claims data Combined health plans claims data RCTs in bariatric surgery have some limitations Difficult to recruit participants May be under-powered for adequate analyses
Reduction of Micro- and Macrovascular Complications (Johnson BJ et al.; 2013) Pre-bariatric surgical patients (BAR) with T2DM (n=2580); surgery, 1996-2009 Moderate and obese non-bariatric surgical patients (No-BAR) with T2DM (n=13,371) Both groups without MI, angina, CHF, stroke, or advanced microvascular disease Study outcome: First major macrovascular or microvascular event BAR to No-BAR Disease HR Macrovascular 0.39; 95% CI, 0.29-0.51 Microvacular 0.22; 95% CI, 0.09-0.49 Johnson BJ et al. J Am Coll Surg 2013;216:545-58
Multisite Study of Long-term Remission and Relapse of T2DM (Arterburn; 2013) Pre-bariatric surgical patients with T2DM (n=4,434); 1995-2008 Initial T2DM remission within 5 years post-surgery: 68.2% (95% CI, 66-70%) Remitters who redeveloped T2DM remission within 5 years 35.1% (95% CI, 32-38%) Arterburn DE et al. Obes Surg 2013;23(1):93
Retrospective Pre-surgery T2DM and CVD Incidence and Mortality (Eliasson; 2015) Two large Swedish registries merged National Diabetes Registry & Scandinavian Obesity Surgery Registry 6132 RYGB patients and 6132 non-surgery control patients all patients with T2DM Matched on sex, age, BMI, calendar time 3.5 years follow-up Eliasson, B, Lancet Diabetes Endocrinol, 2015;3:847
Eliasson cont. all patients with T2DM (n=6132; each group, ) Events RYGB Patients (N) Non-operated (N) HR (95% CI) All-cause mortality 82 288 0.42 (0.3-0.6) CVD-caused mortality 13 67 0.41 (0.2-0.9) Fatal or non-fatal MI 24 67 0.51 (0.3-0.91) Eliasson, B, Lancet Diabetes Endocrinol, 2015;3:847
PROMISE study NIH Funded (Arterburn; 2016) Joint collaboration between 4 U.S. health care systems: Kaiser Permanente Northern California (KPNC) Kaiser Permanente Southern California (KPSC) HealthPartners Research Foundation in Minnesota (HPRF) Group Health Cooperative (GHC) Approximately 10,000 bariatric surgical patients who were diabetic prior to surgery (surgery 2001 through 2011; 40% racial/ethnic minority)
PROMISE study (continued) First manuscript Long-term microvascular disease (first occurrence of retinopathy, neuropathy, and/or nephropathy) Those who remitted T2DM = 29% lower risk of microvascular disease compared to never-remitted (HR 0.71 [95% CI 0.60, 0.85]) For those who remitted T2DM and later relapsed, the longer the remission time inversely related to risk for incident microvascular disease Every additional year of remittance = reduced risk of 19% (HR 0.81 [95% CI 0.67, 0.99]) Coleman KJ et al. Diab Care 2016 (online pub)
Geisinger Obesity Institute CVD Risk Factors and CHF (Benotti; 2017) RYGB group and matched controls, n=1724 Matched on: age, BMI, sex, Framingham Risk Score, smoking Hx, HTN meds, T2DM and date of surgery Primary end points: MI, stroke and CHF Secondary end points: CVD risk factors (risk score, lipids, SBP, and T2DM) Follow-up: 12 years post-surgery (mean = 6.3 years) Benotti PN et al. J Am Heart Assoc 2017;6:e005126
Geisinger Obesity Institute results Reduction in major composite CV events (63 in RYGB group and 110 in control group) HR 0.58 (95% CI: 0.42, 0.82) Groups MI (N) Stroke (N) CHF RYGB 12 31 24 Matched Controls 17 49 55 Reduction in CHF in RYGB compared to controls (p=0.0077) but not significantly different for MI and stroke alone Improvements in CV risk factors (p<0.0001) Benotti PN et al. J Am Heart Assoc 2017;6:e005126
Thereaux J et al. Retrospective Nationwide Matched Cohort Study French health care system database, SNIRAM, linked to French medical discharge database, PMSI 1633 bariatric surgery patients and 1633 matched controls Follow-up of 6 years; greatest benefit in RYGB subgroup Events Bariatric Surgery Patients Non-operated (N) Remitted T2DM 50% 9% T2DM Incidence 1% 12% Thereaux, J et al, data presented at IFSO London, 2017.
SOS Long-term Micro- and Macro-vascular Complications (Sjöström L et al.) Macrovascular complications of BAR vs. No-BAR BAR 20.6 per 1000 person years No-BAR 41.8 per 1000 person-years HR 0.44; 95% CI, 0.34-0.5; p<0.001 Microvascular complications of BAR vs. No-BAR BAR 31.7 per 1000 person years No-BAR 44.2 per 1000 person-years HR 0.68; 95% CI, 0.54-0.85; p=0.001 Sjöström L et al. JAMA 2014;311:2297
One RCT - Micorvascular & Macrovascular Outcomes - STAMPEDE (Schauer, P. et al.) Reduction in urinary albumin-to-creatinine ratio (UACR) of SURG vs. No-SURG (p<0.04) for both GBP and sleeve groups Patients with baseline albuminuria who returned to normal values: 62% (8 of 13; p=0.04 within group comparison) 80% (8 of 10; p=0.11 within group comparison) 25% (1 of 4; p=1.00 within group comparison) Schauer, P. et al. NEJM 2014;370:2002
Utah Study - Hypothesis Does Roux-en-Y gastric bypass (RYGB) surgery result in improved morbidity when severely obese non-surgical groups (population-based and denied-surgery) are compared to RYGB patients? Note: Gastric banding was not approved in US at time of study initiation (2001).
Recruitment Scheme: RYGB Group and Non-surgery Group 1 Patients come to surgeon s clinic to learn about RYGB and get health insurance advice. Patients informed of study. Patient agrees to participate Patient undergoes baseline testing Health plan approves RYGB or patient provides self-pay and RYGB surgery performed Health plan denies coverage of RYGB or patient chooses not to have RYGB performed Patient assigned to RYGB Group Patient assigned to Non-surgical group 1
Recruitment Scheme: Non-surgery Group 2 Severely obese participants not seeking bariatric surgery. Randomly selected from the Utah Health Family Tree program database: 150,000+ families (>1 million family members) Health data, including reported weight status.
Recruitment Hypothesis Denied surgery patients (Non-surgical Group 1) represent characteristics associated with seeking surgery (potentially self-selected for significant comorbidities and other conditions). Population-based severely obese participants (Non-surgical Group 2) represent characteristics associated with severely obese subjects in the general population.
Long-term Prospective Utah Study (NIH-funded 2001-2015) Exam 1 Exam 2 Exam 3 Exam 4 Surgical Center RYGB 418 RYGB Surgery 410 387 388 Utah Health Family Tree Program No RYGB 417 321 Non-Surgery 1 400 Non-Surgery 2 319 388 315 364 301 Baseline 2 Years 6 Years Adams et al. Obesity 2010;18:121-30 Adams et al. JAMA 2012;308:1122-31 12 Years
12-year Results Follow-up End Point RYGB Surgery Group Non-surgery Group 1 Non-surgery Group 2 (minus deaths) (minus deaths) (minus deaths) Potential Participation, n 392 378 303 Any data source, n (%) 388 (98.7%) 364 (96.3%) 301 (99.3%) Minimum of Weight, SBP, and either glucose or HbA1c, n (%) Later had bariatric surgery, n (%) 353 (90.1%) 342 (90.5%) 285 (94.1%) --- 147 (35%) 39 (12%) Adams, T. et al. NEJM 2017;377:1143-55
12-year Results Mortality End Point RYGB Surgery Group Non-surgery Group 1 Non-surgery Group 2 (minus deaths) (minus deaths) (minus deaths) All-cause, n (%) 26 (6.2%) 39 (9.4%) 18 (5.6%) CVD, (n) (%) 3 17 8 Cancer, n (%) 4 5 7 Suicide, n (%) 5 (1.2%) 2 (0.5%) 0 Poisoning, n (%) 4 (1.0%) 2 (0.5%) 0 = Death subsequent to bariatric surgery Adams, T. et al. NEJM 2017;377:1143-55
% Change % Change in Baseline Body Weight (mean adjusted) 10 0-10 -20-30 -40 Control 2 Control 1 Surgery -50 BL 2 Years 6 Years 12 Years = Surgery group differs significantly from non-surgery groups; p<0.001 Adams, T. et al. NEJM 2017;377:1143-55
Percent weight change from baseline 20 % Change in Baseline Body Weight RYGB Group 10 0-10 -20-30 -40-50 Baseline 2-yr 6-yr 12-yr % weight lost -60-70 2001 2004 2006 2009 2012 2014 2017 # of participants Baseline 2 Years 6 Years 12 Years RYGB patients 418 409 379 387 Deaths --- 3 9 14 Total 418 412 388 401
Percent weight change from baseline % Change in Baseline Body Weight Non-Surgery Group 1 70 60 50 40 30 20 10 0-10 -20-30 -40-50 -60-70 2001 2004 2006 2009 2012 2014 2017 # of participants Baseline 2 Years 6 Years 12 Years Non-surgery group 1 417 373 294 217 Had surgery later --- 28 89 146 Deaths --- 3 11 25 Total 417 404 394 388 Baseline 2-yr 6-yr 12-yr 2-yr surgery 6-yr surgery 12-yr surgery Non-surgical controls All controls including surgeries
Percent weight change from baseline % Change in Baseline Body Weight Non-Surgery Group 2 70 60 50 40 30 20 2-yr 10 6-yr 0-10 -20-30 -40-50 -60-70 2001 2004 2006 2009 2012 2014 2017 # of participants Baseline 2 Years 6 Years 12 Years Non-surgery group 2 321 312 294 262 Had surgery later --- 8 19 39 Deaths --- --- 3 15 Total 321 320 316 316 Baseline 12-yr 2-yr surgery 6-yr surgery 12-yr surgery Population controls All controls including surgeries
Weight Loss Maintenance at 12 Years: RYGB Group 360 of 387 RYGB patients (93%) maintained at least a 10% weight loss from baseline to year 12 271 (70%) maintained at least 20% weight loss 155 (40%) maintained at least 30% weight loss Only 4 of 387 (1%) had regained all postsurgical weight loss Adams, T. et al. NEJM 2017;377:1143-55
Diabetes remission after RYGB Group 2 years 6 years 12 years Surgery, % 74% 62% 51% Surgery, n 66/88 54/87 43/84 Adams, T. et al. NEJM 2017;377:1143-55
Diabetes and Hypertension: 12-year Remission End Point Surgery % NS 1, % NS 2, % Adj. OR S vs NS 1 Adj. OR S vs NS 2 T2D 51% 10% 5% HTN 36% 10% 14% 8.9 (2.0, 40.0) 5.1 (1.7, 15.6) 14.8 (2.9, 75.5) 2.4 (0.9, 5.9) = P<0.001 Adams, T. et al. NEJM 2017;377:1143-55
Diabetes and Hypertension: 12-year Incidence End Point Surgery % NS 1, % NS 2, % Adj. OR S vs NS 1 Adj. OR S vs NS 2 T2D 3% 26% 26% 0.08 (0.03, 0.24) HTN 16% 41% 47% 0.23 (0.11, 0.49) 0.09 (0.03, 0.24) 0.09 (0.11, 0.51) = P<0.001 Adams, T. et al. NEJM 2017;377:1143-55
mg/dl Unadjusted Mean HDL-Cholesterol 65 60 55 50 45 40 35 30 BL 2 Years 6 Years 12 Years Surgery Group Non-S Group 1 Non-S Group 2 Surgery group mean (adjusted for medication use) differs significantly from non-surgery groups; p<0.05 Adams et al. JAMA 2012;308:1122-31 Adams et al. Obesity 2010;18:121-30 Adams, et al. NEJM 2017;377:1143-55
grams 250 Left Ventricular Mass 200 150 100 50 0 BL 2 Years 6 Years 12 Years Surgery Group Non-S Group 1 Non-S Group 2 Adams et al. JAMA 2012;308:1122-31 Adams et al. Obesity 2010;18:121-30 Adams, et al. NEJM 2017;377:1143-55
Conclusions Long-term durability of weight loss after RYGB with minimal weight increase between the 6- and 12-year follow-up Very minimal incidence of T2DM Remission rate 51% at the 12-year follow-up Marked improvement in systolic hypertension and lipid levels
Acknowledgement: Funding from NIH-NIDDK
Thank You
There is no drug in current or perspective use that holds as much promise for sustained health as a lifetime program of physical activity. Journal of the American Medical Association
Benefits of Physical Activity Lower death rates Prevents or reduces risk of: High blood pressure Osteoporosis Overweight and obesity Depression and anxiety Diabetes Some cancers (i.e. colon)
Benefits of Physical Activity Improves structure and function of ligaments, tendons and joints Improves: Muscular strength Sleep patterns Health-related quality of life Independent living with increasing age
R x Structured Exercise Program Frequency: 3-5 days/week Intensity: 55/65% - 90% max heart rate
R x Structured Exercise Program continued Duration: 20-60 min (continuous or intermittent - minimum 10 min. bouts accumulated over the day) Mode: Large muscle groups, continuous, aerobic in nature
Modes for Structured Exercise Walking-hiking Running-jogging Cycling-bicycling Aerobic dance Stair climbing Swimming Endurance games
R x Structured Exercise Program continued Resistance training: One set, 8-10 exercises, 8-12 reps, 2-3 days/wk
Examples of Using Your Own Weight
R x Structured Exercise Program continued Flexibility training: 2-3 days/wk of stretching major muscle groups
Flexibility Exercises
R x Lifestyle Physical Activity Accumulate 30 minutes or more of moderate-intensity physical activity on most, preferably all days of the week. Resistance training 2 times per week
Examples: Moderate Intensity Activities Walking Yard Work Stair Climbing Housework Dancing
Resting Metabolic Rate (RMR) Indirect calorimetry using a Parvomedics TrueOne 2400 metabolic cart. Measurements taken in the morning after a 12-hour fast, minimal exercise in prior 24 hours, and participants having rested in a supine position for at least 30 minutes. 24-hour estimates based on at least 10 minutes of stable resting values. 986 participants had at least one assessment Mixed model regression was used to determine group-wise differences at each time point.
kcals/day Resting Metabolic Rate (RMR) 2500 2400 2300 2200 2100 2000 1900 1800 1700 Non-Surgery 2 Non-Surgery 1 Surgery 1600 1500 BL 2 Years 6 Years 12 Years RMR has an expected drop with weight loss in Surgery group Long-term rate of RMR decrease mirrors age-related drop in controls // Within-group change from previous assessment is significant (p<0.05) Surgery value differs significantly from Non-Surgery 1 group (p<0.05) Surgery value differs significantly from Non-Surgery 2 group (p<0.05)
kcals/day Resting Metabolic Rate (RMR) per kg body weight 22 21 20 19 18 17 16 15 Non-Surgery 2 Non-Surgery 1 Surgery 14 13 12 BL 2 Years 6 Years 12 Years Per kilogram RMR remains improved long-term, perhaps explaining maintenance of post-op weight loss Difference begins to wane by 12 years // Within-group change from previous assessment is significant (p<0.05) Surgery value differs significantly from Non-Surgery 1 group (p<0.05) Surgery value differs significantly from Non-Surgery 2 group (p<0.05)
kilograms RYGB: influence of Fat and FFM on RMR 160 140 120 100 2500 2000 1500 FFM FM Weight RMR 80 60 40 20 1000 500 0 BL 2 yrs 6 yrs 12 yrs 0 RMR does not decrease as much as expected with weight loss FFM constitutes 1/3 rd of weight loss, but is only 11% of regain /// Change is significant (p<0.05) compared with previous assessment
kilograms Kcals/kg/day RYGB: influence of Fat and FFM on RMR/kg 160 140 120 100 20 15 FFM FM Weight RMR/kg 80 60 40 10 5 20 0 BL 2 yrs 6 yrs 12 yrs 0 RMR/kg changes reflect increased FFM relative to FM /// Change is significant (p<0.05) compared with previous assessment
Cardiorespiratory fitness: time on treadmill Graded exercise test with 12-lead ECG Submaximal tests at baseline and 12 years (maximal tests at 2 and 6 years) Means are seconds to test completion Time Test type RYGB Baseline Submax (80%) (seconds) NS 1 (seconds) NS 2 (seconds) 602±13 593±14 582±13 2 years Maximal 873±13 623±16 586±14 6 years Maximal 828±14 641±17 679±13 12 years Submax (80%) 643±14 598±16 556±17 After surgical weight loss, RYGB patients are able to last significantly longer on the treadmill 10-12 years after surgery The effect is lost when controlling for weight = (p<0.05) = (p<0.0001)