Transcript A Retrospective Study of Bariatric Surgery Patients with Sub

Shah Faisal MBBS
Mentor
Joseph A. Caruana MD
Aim and Objective
Aim and Objective
 Some patients after Roux-en-Y Gastric Bypass undergo
“Sub-optimal” Weight Loss (SWL).
Aim and Objective
 Some patients after Roux-en-Y Gastric Bypass undergo
“Sub-optimal” Weight Loss (SWL).
 Age, Gender and BMI are the major contributors to weight
loss pattern.
Aim and Objective
 Some patients after Roux-en-Y Gastric Bypass undergo
“Sub-optimal” Weight Loss (SWL).
 Age, Gender and BMI are the major contributors to weight
loss pattern.
 Are there any more subtle and/or modifiable variables
which play a role?
Problem of Obesity
Problem of Obesity
 Two thirds of individuals living in the US are overweight,
and of those, almost half are obese.
Problem of Obesity
 Two thirds of individuals living in the US are overweight,
and of those, almost half are obese.
 DM-2, dyslipidemia, CAD, HTN, OSA are a few of the
myriad complications that obesity predisposes to.
Problem of Obesity
 Two thirds of individuals living in the US are overweight,
and of those, almost half are obese.
 DM-2, dyslipidemia, CAD, HTN, OSA are a few of the
myriad complications that obesity predisposes to.
 Known association with some common cancers:
Problem of Obesity
 Two thirds of individuals living in the US are overweight,
and of those, almost half are obese.
 DM-2, dyslipidemia, CAD, HTN, OSA are a few of the
myriad complications that obesity predisposes to.
 Known association with some common cancers:
 Liver, kidney, breast, endometrial, prostate, colon…
Problem of Obesity
 Two thirds of individuals living in the US are overweight,
and of those, almost half are obese.
 DM-2, dyslipidemia, CAD, HTN, OSA are a few of the
myriad complications that obesity predisposes to.
 Known association with some common cancers:
 Liver, kidney, breast, endometrial, prostate, colon…
 Huge burden on healthcare cost:
Problem of Obesity
 Two thirds of individuals living in the US are overweight,
and of those, almost half are obese.
 DM-2, dyslipidemia, CAD, HTN, OSA are a few of the
myriad complications that obesity predisposes to.
 Known association with some common cancers:
 Liver, kidney, breast, endometrial, prostate, colon…
 Huge burden on healthcare cost:
 25 % greater among subjects with a BMI 30 to 34.9
Problem of Obesity
 Two thirds of individuals living in the US are overweight,
and of those, almost half are obese.
 DM-2, dyslipidemia, CAD, HTN, OSA are a few of the
myriad complications that obesity predisposes to.
 Known association with some common cancers.
 Liver, kidney, breast, endometrial, prostate, colon…
 Huge burden on healthcare cost:
 25 % greater among subjects with a BMI 30 to 34.9
 44 % greater among those with a BMI > 35
Treatment
Treatment
 Medical treatment:
Treatment
 Medical treatment:
 Diet, exercise and life style modification.
Treatment
 Medical treatment:
 Diet, exercise and life style modification.
 Medications.
Treatment
 Medical treatment:
 Diet, exercise and life style modification.
 Medications.
 Surgical treatment:
Treatment
 Medical treatment:
 Diet, exercise and life style modification.
 Medications.
 Surgical treatment:
 Roux-en-Y Gastric By-pass.
Treatment
 Medical treatment:
 Diet, exercise and life style modification.
 Medications.
 Surgical treatment:
 Roux-en-Y Gastric By-pass.
 Adjustable Gastric Banding.
Treatment
 Medical treatment:
 Diet, exercise and life style modification.
 Medications.
 Surgical treatment:
 Roux-en-Y Gastric By-pass.
 Adjustable Gastric Banding.
 Diet, exercise and life style modification are part of surgical
treatment.
Dietary and Exercise recommendation after
Gastric Bypass
Dietary and Exercise recommendation after
Gastric Bypass
 Diet:
 1200-1400 Cal per day.
 60-80 g protein per day.
 4-5 small meals (especially breakfast).
 32 oz water per day.
Dietary and Exercise recommendation after
Gastric Bypass
 Diet:
 1200-1400 Cal per day.
 60-80 g protein per day.
 4-5 small meals (especially breakfast).
 32 oz water per day.
 Exercise:
 Walking 30 min/day, five to seven days per week.
Roux en Y Gastric By-pass
Adjustable Gastric band
Weight Loss after Gastric Bypass
Weight Loss after Gastric Bypass
 How do Age, Gender and BMI influence weight loss?
Weight Loss after Gastric Bypass
 How do Age, Gender and BMI influence weight loss?
 Recent work done by Dr Caruana:
Weight Loss after Gastric Bypass
 How do Age, Gender and BMI influence weight loss?
 Recent work done by Dr Caruana:
 Weight Loss Curves.
Weight Loss after Gastric Bypass
 How do Age, Gender and BMI influence weight loss?
 Recent work done by Dr Caruana:
 Weight Loss Curves.
 Effect of Age, Gender and BMI on weight loss.
Weight-Loss Curve-Percentiles
95
90
75
50
25
10
5
Effect of Age on Weight Loss
Red Lines >41 yr
Black Lines <41 yr
95
90
75
50
25
10
5
Effect of Gender on Weight Loss
Red Lines Females
Black Lines Males
95
90
75
50
25
10
5
Effect of BMI on Weight Loss
Red Lines BMI>55
Black Lines BMI<55
95
90
75
50
25
10
5
Sub-Optimal Weight Loss (SWL)
Sub-Optimal Weight Loss (SWL)
 SWL < 40% (to 50%) Excess Body Weight (EBW) lost.
Sub-Optimal Weight Loss (SWL)
 SWL < 40% (to 50%) Excess Body Weight (EBW) lost.
 EBW is actual weight minus Ideal Body Weight (IBW).
Sub-Optimal Weight Loss (SWL)
 SWL < 40% (to 50%) Excess Body Weight (EBW) lost.
 EBW is actual weight minus Ideal Body Weight (IBW).
 IBW from 1983 Metropolitan Height and Weight Tables.
Sub-Optimal Weight Loss (SWL)
 SWL < 40% (to 50%) Excess Body Weight (EBW) lost.
 EBW is actual weight minus Ideal Body Weight (IBW).
 IBW from 1983 Metropolitan Height and Weight Tables.
 Around 10% (to 20%) of Gastric Bypass patients have
SWL:
Sub-Optimal Weight Loss (SWL)
 SWL < 40% (to 50%) Excess Body Weight (EBW) lost.
 EBW is actual weight minus Ideal Body Weight (IBW).
 IBW from 1983 Metropolitan Height and Weight Tables.
 Around 10% (to 20%) of Gastric Bypass patients have
SWL:
 Most of the patients actually regain the weight.
Methods
Methods
 Matched Case-Control Study.
Methods
 Matched Case-Control Study.
 Controls matched to cases for Age, Gender & BMI.
Methods
 Matched Case-Control Study.
 Controls matched to cases for Age, Gender & BMI.
 Four matched controls for every case.
Inclusion/Exclusion Criteria
Inclusion/Exclusion Criteria
 Gastric bypass between Jan ’00 and Dec ‘07.
Inclusion/Exclusion Criteria
 Gastric bypass between Jan ’00 and Dec ‘07.
 Followed up at 18 (±2) months after surgery.
Inclusion/Exclusion Criteria
 Gastric bypass between Jan ’00 and Dec ‘07.
 Followed up at 18 (±2) months after surgery.
 At 18 (± 2) months:
Inclusion/Exclusion Criteria
 Gastric bypass between Jan ’00 and Dec ‘07.
 Followed up at 18 (±2) months after surgery.
 At 18 (± 2) months:
 If lost < 40% EBW—Case.
Inclusion/Exclusion Criteria
 Gastric bypass between Jan ’00 and Dec ‘07.
 Followed up at 18 (±2) months after surgery.
 At 18 (± 2) months:
 If lost < 40% EBW—Case.
 If lost >40% EBW—Potential Control.
Break Down of Cases & Controls
Cases
Controls
Break Down of Cases & Controls
Number
Cases
Controls
21
84
Break Down of Cases & Controls
Number
Gender
Females
Males
Cases
Controls
21
84
20
1
80
4
Break Down of Cases & Controls
Number
Gender
Age (Years)
Females
Males
Mean
Range
Cases
Controls
21
20
1
45
35
84
80
4
45
38
Break Down of Cases & Controls
Number
Gender
Age (Years)
BMI
Females
Males
Mean
Range
Mean
Range
Cases
Controls
21
20
1
45
35
55
37
84
80
4
45
38
55
40
Break Down of Cases & Controls
Cases
Controls
21
84
Females
Males
Mean
Range
Mean
Range
20
1
45
35
55
37
80
4
45
38
55
40
Yes
No
60
40
64
36
Number
Gender
Age (Years)
BMI
Working (%)
Break Down of Cases & Controls
Number
Gender
Age (Years)
BMI
Working (%)
Smoking (%)
Cases
Controls
Females
Males
Mean
Range
Mean
Range
Yes
No
21
20
1
45
35
55
37
60
40
84
80
4
45
38
55
40
64
36
Yes
No
10
90
17
83
Break Down of Cases & Controls
Number
Cases
Controls
21
84
Gender
Females
Males
20
1
80
4
Age (Years)
Mean
Range
Mean
Range
Yes
No
45
35
55
37
60
40
45
38
55
40
64
36
Smoking (%)
Yes
No
10
90
17
83
Race (%)
White
Black
90
10
83
17
BMI
Working (%)
Data Collection
 Retrospective Chart Review.
Data Collection
 Retrospective Chart Review.
 Variables:
Data Collection
 Retrospective Chart Review.
 Variables:
 Demographic
 Anthropometric
 Social
 Compliance
 Disease Characteristics
 Co-morbidities
 Related Diseases
 Peri-operative
Type of Variable Variables
Demographic
Race, (Age and Gender Matched)
Type of Variable Variables
Demographic
Race, (Age and Gender Matched)
Anthropometric
(BMI Matched)
Type of Variable Variables
Demographic
Race, (Age and Gender Matched)
Anthropometric
(BMI Matched)
Social
Work Status
Smoking Status
Type of Variable Variables
Demographic
Race, (Age and Gender Matched)
Anthropometric
(BMI Matched)
Social
Work Status
Smoking Status
Compliance
Total Number of Follow-up Visits
Dietary Compliance at Follow-up
Exercise Compliance at Follow-up
Type of Variable Variables
Demographic
Race, (Age and Gender Matched)
Anthropometric
(BMI Matched)
Social
Work Status
Smoking Status
Compliance
Total Number of Follow-up Visits
Dietary Compliance at Follow-up
Exercise Compliance at Follow-up
Disease
Characteristics
Duration of Obesity
Family H/O Obesity
Type of Variable Variables
Demographic
Race, (Age and Gender Matched)
Anthropometric
(BMI Matched)
Social
Work Status
Smoking Status
Compliance
Total Number of Follow-up Visits
Dietary Compliance at Follow-up
Exercise Compliance at Follow-up
Disease
Characteristics
Duration of Obesity
Family H/O Obesity
Co-morbidities
No. of Co-morbidities (DM, HTN, Dyslipidemia, OSA, Arthritis, and GERD)
Hepatic Steatosis
No. of Prescription Medications
Severity of DM, HTN and Dyslipidemia
Type of Variable Variables
Demographic
Race, (Age and Gender Matched)
Anthropometric
(BMI Matched)
Social
Work Status
Smoking Status
Compliance
Total Number of Follow-up Visits
Dietary Compliance at Follow-up
Exercise Compliance at Follow-up
Disease
Characteristics
Duration of Obesity
Family H/O Obesity
Co-morbidities
No. of Co-morbidities (DM, HTN, Dyslipidemia, OSA, Arthritis, and GERD)
Hepatic Steatosis
No. of Prescription Medications
Severity of DM, HTN and Dyslipidemia
Related Diseases
Depression
Thyroid Dysfunction
H/O Major Surgeries
Type of Variable Variables
Demographic
Race, (Age and Gender Matched)
Anthropometric
(BMI Matched)
Social
Work Status
Smoking Status
Compliance
Total Number of Follow-up Visits
Dietary Compliance at Follow-up
Exercise Compliance at Follow-up
Disease
Characteristics
Duration of Obesity
Family H/O Obesity
Co-morbidities
No. of Co-morbidities (DM, HTN, Dyslipidemia, OSA, Arthritis, and GERD)
Hepatic Steatosis
No. of Prescription Medications
Severity of DM, HTN and Dyslipidemia
Related Diseases
Depression
Thyroid Dysfunction
H/O Major Surgeries
Peri-operative
Cholecystectomy done at OR
Post-op Leak
Post-op PE
Return to OR in 30 days
Statistical Analysis
Statistical Analysis
 Binary predictors between the case and control group—
the exact Cochran-Mantel-Haenszel test:
Statistical Analysis
 Binary predictors between the case and control group—
the exact Cochran-Mantel-Haenszel test:
 Common odds ratio with a corresponding exact 95%
confidence interval was estimated.
Statistical Analysis
 Binary predictors between the case and control group—
the exact Cochran-Mantel-Haenszel test:
 Common odds ratio with a corresponding exact 95%
confidence interval was estimated.
 Mixed linear—Between-group Mean Difference.
Statistical Analysis
 Binary predictors between the case and control group—
the exact Cochran-Mantel-Haenszel test:
 Common odds ratio with a corresponding exact 95%
confidence interval was estimated.
 Mixed linear—Between-group Mean Difference.
 A nominal significance level of 0.05 was used.
Statistical Analysis
 Binary predictors between the case and control group—
the exact Cochran-Mantel-Haenszel test:
 Common odds ratio with a corresponding exact 95%
confidence interval was estimated.
 Mixed linear—Between-group Mean Difference.
 A nominal significance level of 0.05 was used.
 A multivariate model was constructed using a stepwise
conditional logistic regression procedure.
Statistical Analysis
 Binary predictors between the case and control group—
the exact Cochran-Mantel-Haenszel test:
 Common odds ratio with a corresponding exact 95%
confidence interval was estimated.
 Mixed linear—Between-group Mean Difference.
 A nominal significance level of 0.05 was used.
 A multivariate model was constructed using a stepwise
conditional logistic regression procedure.
 SAS version 9.1.3 statistical software (Cary, NC).
Binary Predictors—the exact
Cochran-Mantel-Haenszel test
Variable
Diabetes
Y vs. N
OR
95% CI
p- Value
0.6
(0.1931, 2.4302)
0.567
Binary Predictors—the exact
Cochran-Mantel-Haenszel test
Variable
Diabetes
Y vs. N
Binary Predictor—Yes or No
OR
95% CI
p- Value
0.6
(0.1931, 2.4302)
0.567
Binary Predictors—the exact
Cochran-Mantel-Haenszel test
Variable
Diabetes
Y vs. N
OR
95% CI
p- Value
0.6
(0.1931, 2.4302)
0.567
Binary Predictor—Yes or No
Odds of normal weight loss if DM
=0.6
Odds of normal weight loss if not DM
Binary Predictors—the exact
Cochran-Mantel-Haenszel test
Variable
Diabetes
Y vs. N
OR
95% CI
p- Value
0.6
(0.1931, 2.4302)
0.567
Binary Predictor—Yes or No
Odds of normal weight loss if DM
=0.6
Odds of normal weight loss if not DM
95% Confidence that OR lies between (0.1931, 2.4302)
Binary Predictors—the exact
Cochran-Mantel-Haenszel test
Variable
Diabetes
Y vs. N
OR
95% CI
p- Value
0.6
(0.1931, 2.4302)
0.567
Binary Predictor—Yes or No
Odds of normal weight loss if DM
=0.6
Odds of normal weight loss if not DM
95% Confidence that OR lies between (0.1931, 2.4302)
P-Value < 0.05 for a
relationship to be
statistically
significant,
otherwise could be
just by chance
Mixed Linear Models—
Between-group mean difference
Variable
Estimated Mean Difference 95% CI
(Control—Case)
p- Value
No. of Medications @ follow-up
-1.4
0.0031
(0.5079, 2.4206)
Mixed Linear Models—
Between-group mean difference
Variable
Estimated Mean Difference 95% CI
(Control—Case)
p- Value
No. of Medications @ follow-up
-1.4
0.0031
Linear Variable—Number
(0.5079, 2.4206)
Mixed Linear Models—
Between-group mean difference
Variable
Estimated Mean Difference 95% CI
(Control—Case)
p- Value
No. of Medications @ follow-up
-1.4
0.0031
Linear Variable—Number
Average number of extra medications
taken by Controls
(0.5079, 2.4206)
Results—Univariate
Variable
Comment
Results—Univariate
Demographic
Variable
Comment
Race
Age
Gender
Insignificant
Matched
Matched
Results—Univariate
Variable
Comment
Demographic
Race
Age
Gender
Insignificant
Matched
Matched
Anthropometric
BMI
Matched
Results—Univariate
Variable
Comment
Demographic
Race
Age
Gender
Insignificant
Matched
Matched
Anthropometric
BMI
Matched
Social
Work Status
Smoking Status
Insignificant
Insignificant
Results—Univariate
Variable
Comment
Demographic
Race
Age
Gender
Insignificant
Matched
Matched
Anthropometric
BMI
Matched
Social
Work Status
Smoking Status
Insignificant
Insignificant
Compliance
Total Number of Follow-up Visits
Dietary Compliance at Follow-up
Exercise Compliance at Follow-up
Insignificant
Significant
Insignificant
Results—Univariate
Variable
Comment
Demographic
Race
Age
Gender
Insignificant
Matched
Matched
Anthropometric
BMI
Matched
Social
Work Status
Smoking Status
Insignificant
Insignificant
Compliance
Total Number of Follow-up Visits
Dietary Compliance at Follow-up
Exercise Compliance at Follow-up
Insignificant
Significant
Insignificant
Disease
Characteristics
Duration of Obesity
Family H/O Obesity
Insignificant
Insignificant
Results—Univariate
Variables
Comment
Results—Univariate
Co-morbidities
Variables
Comment
No. of Co-morbidities
Hepatic Steatosis
DM
Insignificant
Insignificant
Insignificant
Results—Univariate
Co-morbidities
Variables
Comment
No. of Co-morbidities
Hepatic Steatosis
DM
Insignificant
Insignificant
Insignificant
Dyslipidemia
On Anti-dyslipidemic Medications
@ Baseline
@Follow-up
Insignificant
Insignificant
Insignificant
Results—Univariate
Co-morbidities
Variables
Comment
No. of Co-morbidities
Hepatic Steatosis
DM
Insignificant
Insignificant
Insignificant
Dyslipidemia
On Anti-dyslipidemic Medications
@ Baseline
@Follow-up
Insignificant
HTN
On Anti-HTN Medications
@ Baseline
@Follow-up
Insignificant
Insignificant
Insignificant
Significant
Insignificant
Results—Univariate
Co-morbidities
Variables
Comment
No. of Co-morbidities
Hepatic Steatosis
DM
Insignificant
Insignificant
Insignificant
Dyslipidemia
On Anti-dyslipidemic Medications
@ Baseline
@Follow-up
Insignificant
HTN
On Anti-HTN Medications
@ Baseline
@Follow-up
Insignificant
Prescription Medications
@ Baseline
@Follow-up
Insignificant
Insignificant
Significant
Insignificant
Insignificant
Significant
Results—Univariate
Variable
Comment
Results—Univariate
Related Diseases
Variable
Comment
H/O Depression
H/O Thyroid dysfunction
H/O Major Surgeries
Insignificant
Insignificant
Insignificant
Results—Univariate
Variable
Comment
Related Diseases
H/O Depression
H/O Thyroid dysfunction
H/O Major Surgeries
Insignificant
Insignificant
Insignificant
Peri-operative
Cholecystectomy done at OR
Post-op Leak
Post-op PE
Return to OR in 30 days
Insignificant
Insignificant
Insignificant
Insignificant
Results—Multivariate
Variable
Compliance
Dietary Compliance at Follow-up
Significant
Co-morbidities
Prescriptions Medications at Follow-up
Significant
Conclusion
 After negating effects of Age, Gender and BMI:
 Dietary non-compliance (reported by patients) predisposes
to SWL.
Conclusion
 After negating effects of Age, Gender and BMI:
 Dietary non-compliance (reported by patients) predisposes
to SWL.
 This should be stressed to patients before and after gastric
bypass surgery.
Conclusion
 After negating effects of Age, Gender and BMI:
 Dietary non-compliance (reported by patients) predisposes
to SWL.
 This should be stressed to patients before and after gastric
bypass surgery.
 “Self-reported loss of control over eating was related to
weight regain after Gastric Bypass surgery and may be an
important target for clinical intervention” (Kalarchian, Obes Surg
2002 Apr;12(2):270-5).
Conclusion
 After negating effects of Age, Gender and BMI:
 Dietary non-compliance (reported by patients) predisposes
to SWL.
 This should be stressed to patients before and after gastric
bypass surgery.
 “Self-reported loss of control over eating was related to
weight regain after Gastric Bypass surgery and may be an
important target for clinical intervention” (Kalarchian, Obes Surg
2002 Apr;12(2):270-5).
 Gradual enlargement of the gastric pouch.
Dietary and Exercise recommendation after
Gastric Bypass
 Diet:
 1200-1400 Cal per day.
 60-80 g protein per day.
 4-5 small meals (especially breakfast).
 32 oz water per day.
 Exercise:
 Walking 30 min/day, five to seven days per week.
Roux en Y Gastric By-pass
Conclusion
 Anti-HTN at baseline:
 Probably a direct effect of a more serious disease and hence
more SWL.
Conclusion
 Anti-HTN at baseline:
 Probably a direct effect of a more serious disease and hence
more SWL.
 Prescription medications at follow-up:
 An indirect clue that patients who (were non-compliant and)
did not loose enough weight continued to need more
medications.
Conclusion
 Bigger and more powerful studies could potentially show
other significant associations.
Conclusion
 Bigger and more powerful studies could potentially show
other significant associations.
 Being a retrospective study, cause-effect relationship
cannot be established—results show only associations.
Conclusion
 Bigger and more powerful studies could potentially show
other significant associations.
 Being a retrospective study, cause-effect relationship
cannot be established—results show only associations.
 Data was partly patient reported/subjective- potential
recall bias.
References
1. WHO Fact sheet N°311, September 2006
2. Walter J. Pories; Who Would Have Thought It? ANNALS OF SURGERY
Vol. 222, No. 3, 339-352
3. Sjöström L; Lifestyle, Diabetes, and Cardiovascular Risk Factors 10 Years
after Bariatric Surgery. N Engl J Med 2004;3512683-93
4. Walter J. Pories; Bariatric Surgery: Risks and Rewards, J Clin Endocrinol
Metab 93: S89–S96, 2008)
5. Sjöström L; Effects of Bariatric Surgery on Mortality in Swedish Obese
Subjects. N Eng J Med 2007; 357: 741-52
6. Buchwald H; Yoav Avidor; Eugene Braunwald; et al. JAMA.
2004;292(14):1724-37
7. National Center for Health Statistics NHANES IV Report. Available at:
http://www.cdc.gov/nchs /product/pubs/pubd/hestats/obes/obese99
.htm2002. Accessed September 9, 2002.
References
8. Halverson JK, Koehler RE. Gastric bypass: analysis of weight loss
and factors determining success. Surgery. 1981;90:446-455.
9. Brolin RE, Kenler HA, Gorman RC, Cody RP. The dilemma of
outcome assessment after operations for morbid obesity. Surgery.
1989;105:337-346.
10. Carbonell AM, Wolfe LG, Meador JG, et al. Does diabetes affect
weight loss after gastric bypass? Surg Obes Relat Dis 2008;4:441-4.
11. Caruana et al; The Weight Loss Curve: A Nomogram for
Monitoring the Response to Gastric Bypass
12. 12 Branson R; Impact of age, sex and body mass index on
outcomes at four years after gastric banding: Obesity Surgery.
15(6):834-42, 2005 Jun-Jul
References
13. Melton GB; Suboptimal weight loss after gastric bypass surgery:
correlation of demographics, comorbidities, and insurance status
with outcomes: Journal of Gastrointestinal Surgery. 12(2):250-5,
2008 Feb
14. . Wittgrove AC, Clark GW. Laparoscopic gastric bypass, RouxenY—500 patients: technique and results, with 3–60 month follow-up.
Obes Surg 2000;10(3):233–239, Jun
15. Prachand VN, Davee RT, Alverdy JC. Duodenal switch provides
superior weight loss in the super-obese (BMI > or =50 kg/m2)
compared with gastric bypass. Ann Surg 2006;244(4):611–619,Oct
16. Choban PS, Flancbaum L. The effect of Roux limb lengths on
outcome after Roux-en-Y gastric bypass: a prospective, randomized
clinical trial. Obes Surg 2002;12(4):540–545, Aug.
17. Metropolitan Life Insurance Company