Is surgery better

1
Is surgery better?

• Bariatric surgery versus medical management in
  type 2 diabetes mellitus

Jordan Kautz, MS IV Ambulatory Care Conference,
Jan 08 UNC-CH School of Medicine
2
For the soon-to-be medical resident

• LUKE SKYWALKER Is the dark side stronger?
• YODA No, no, no. Quicker, easier, more
  seductive.
• Empire Strikes Back (1980)

3
Quasi-hypothetical case

• Mr. A, 48-year-old male, pmhx significant for
  htn, hyperlipidemia, dm, and obesity (BMI 35)
  presents for diabetes f/u. DM dxd 03. Tried
  diet, exercise. Metformin. Glipizide. Most
  recent A1c 8.6 despite optimal mgmt.
• Add third agent? Insulin? Other ideas?
• Doc, what about that weight loss surgery?
• In addition to weight loss (and perhaps as a
  result), what health benefits accrue secondary to
  bariatric surgery for such a patient?

4
Another quasi-hypothetical case

• Mrs. B, 35-year-old female, pmhx significant only
  for obesity (BMI 39), presents c new dx type 2
  dm. Battling wt since college. Not started any
  meds. Baseline A1c 7.8. Father c dm, on
  insulin, good control. Sister underwent gastric
  banding, good result.
• Recommendations?
• I heard on the news gastric banding can cure
  diabetes if done earlyis that right?
• For new diagnosis type 2 dm patients (preserved
  pancreatic beta-cell function), is there
  recovery from diabetes and might it be
  clinically significant?

5
Med Student Reasoning

• Twin public health epidemics
• Not all diabetics are obese and not all obese
  persons have diabetes, butDIABESITY?!?!
• Early and intensive treatment of diabetes
  improves health outcomes, quality of life
• Weight loss improves blood glucose control,
  morbidity and mortality
• Bariatric surgery leads to greater weight loss
  than therapeutic lifestyle change (in most cases
  but not without assuming greater risks)
• Observational studies suggest that surgically
  induced weight loss may be an effective treatment
  for type 2 diabetes mellitus, especially
  proximally

6
What How it (might) work (we think)

• Not talking about one procedure
• Laparoscopic adjustable gastric banding (LAGB),
  roux-en-Y, biliary-pancreatic diversion
• Each procedure involves different risks,
  benefits, and metabolic consequences
• Purely a secondary outcome of surgically-induced
  wt loss
• Foregut hypothesis Improved diabetic control by
  excluding the duodenum and proximal jejunum from
  nutrient flow
• Hindgut hypothesis More rapid delivery of
  nutrients to distal intestine results in improved
  glucose metabolism (via GLP-1, other peptides)
• Questions re safety, invasiveness,
  cost-effectiveness
• Underrepresented as treatment option in ADA
  guidelines, recommendations by similar bodies

7
First RCT hot off the press

• Dixon JB, OBrien PE, Playfair J, et al.
  Adjustable gastric banding and conventional
  therapy for type 2 diabetes A randomized
  controlled trial. JAMA 2008 299(3)316-23
• PICO For adults, age 20-60, BMI 30-40, and T2DM
  dxd w/i two yrs, does LAGB compared to wt loss
  by lifestyle change lead to increased remission
  of type 2 diabetes (FPGlt126, A1c lt6.2, no
  glycemic rx)?

8
Study Population (n60)

• Inclusion Age 20-60, BMI 30-40, Dxd c T2DM w/i
  2yrs and NO evidence of renal impairment,
  diabetic retinopathy
• Exclusion T1DM, DM secondary to specific dz,
  previous bariatric surgery, h/o mental
  impairment, drug or EtOH addiction, recent major
  vascular event, internal malignancy, portal HTN

9
Study Design

• Run-in to maximize current mgmt and assess
  compliance
• NON-blinded computer randomization
• Table 1 appears successful
• Conventional treatment GP, dietitian, nurse,
  diabetes educator visit c at least 1 team member
  q6wks during 2yrs diet physical activity
  requirements
• Not treated according to standardized
  algorithmoptimal (?) management
• Surgical program ALL aspects of conventional rx
  PLUS LAGB progress reviewed by bariatric surgery
  team q4-6wks adjustments to band volume using
  standard clinical criteria
• Highly experienced, specialized surgical
  groupreproducible (?) success

10
Study Metrics

• Primary Proportion of patients achieving
  remission (as previously defined)
• Secondary change in A1c, wt, bp, waist
  circumference, lipids, change in medication use,
  proportion of pts c metabolic syndrome, change in
  direct measures of insulin resistance

11
Results

• Diabetes remission 73 (surgical) versus 13
  (medical), plt0.001
• Greater percentage of wt loss at two yrs and
  lower baseline A1c values were independently
  associated with remission
• Percentage of weight loss explained most of the
  variance
• 20 (surgical) versus 1.4 (medical), plt0.001

12
Primary and Secondary Outcomes at 2 Years
STATISTICALLY significant improvements in
secondary end points though the study was NOT
powered to assess multiple outcome
measures Adverse events (surgical group) included
superficial wound infection, gastric pouch
enlargement requiring nonurgent revision, and
band removal Minor events (surgical group)
included postop fever, hypoglycemic episode, and
GI intolerance to metformin
13
Limitations

• Restricted to recent diagnosis diabetes
• Results may not apply to those with longer hx of
  dz due to deterioration of beta-cell fxn
• Sample size and duration of follow-up
• More diverse population and longer f/u to see if
  benefits persist, evaluate hard end points

14
Ladies first

• Based on this study, its strengths and
  limitations, what would you say to Mrs. B?
• What in the literature or about the patient would
  be persuasive in recommending LAGB?
• Should the bariatric surgeon be on speed dial for
  all newly diagnosed diabetics when diet and
  exercise fail?

15
Swedish Obese Subjects (SOS)

• Study design Prospective, nonrandomized
  intervention trial (2y n4,047 10y n1,703)
• Intervention Fixed or variable banding, vertical
  banded gastroplasty, gastric bypass compared to
  customary treatment (non-standardized including
  no treatment)
• Subjects Obese subjects who underwent gastric
  surgery and contemporaneously matched,
  conventionally treated obese control subjects
• Inclusion BMIgt34 (M), gt38 (F), Age 37-60
• Exclusion Not well detailed, available
  elsewhere

16
And the winners are
TWO YEARS -23.4 (surgical) 0.1 (control) TEN
YEARS-16.1 (surgical) 1.6 (control)
17
Effect on incidence of and recovery from risk
conditions (1)
18
Effect on incidence of and recovery from risk
conditions (2)
19
Questions

• What biologically plausible mechanism might
  account for more successful recovery from than
  incidence of hypertension, dyslipidemia?
• Does dichotomizing variables (or interventions
  for that matter) obscure information that may be
  meaningful for clinical practice? Patient health
  outcomes?

20
Getting back to the other guy

• Based on this study, its strengths and
  limitations, what would you say to Mr. A?
• Is a disease-free interval likely to pay
  dividends for this gentleman years later (when we
  know micro- and macro-vascular complications
  operate on such a time scale)?
• Yeah, well, but what if he just looks and feels
  great?

21
References

• Dixon JB, OBrien PE, Playfair J, et al.
  Adjustable gastric banding and conventional
  therapy for type 2 diabetes A randomized
  controlled trial. JAMA 2008 299(3)316-23
• Sjostrom L, Lindroos AK, Peltonen M, et al.
  Lifestyle, diabetes, and cardiovascular risk
  factors 10 years after bariatric surgery. NEJM
  2004 351(26)2683-93
• Buchwald H, Avidor Y, Braunwald E, et al.
  Bariatric surgery a systematic review and
  meta-analysis. JAMA 2004 292(14) 1724-37
• Ferchak CV and Meneghini LF. Obesity, bariatric
  surgery, and type 2 diabetesa systematic review.
  Diabetes Metab Res Rev 2004 20(6)438-45
• Dixon JB, Pories WJ, OBrien PE, et al. Surgery
  as an effective early intervention for diabesity
  why the reluctance? Diabetes Care 2005
  28(2)472-4
• Chapman AE, Kiroff G, Game P. LAGB in the
  treatment of obesity a systematic literature
  review. Surgery 2004 135326-51
• PubMed search bariatric surgery MeSH AND
  diabetes mellitus, type 2 MeSH, limits
  English (134 articles)