Laparoscopic Vs Open Gastric Bypass Surgery Differences in Patient Demographics, Safety, and Outcomes

ORIGINAL ARTICLE Laparoscopic vs Open Gastric Bypass Surgery Differences in Patient Demographics, Safety, and Outcomes

Gaurav Banka, MD; Gavitt Woodard, MD; Tina Hernandez-Boussard, PhD, MPH; John M. Morton, MD, MPH

Objective: To determine national outcome differences pitals (69% vs 61%; PϽ.001). A higher percentage of between laparoscopic Roux-en-Y gastric bypass (LRYGB) ORYGB compared with LRYGB patients were Medicare and open Roux-en-Y gastric bypass (ORYGB). (9.3% vs 7.1%) and Medicaid (10.4% vs 5.9%; PϽ.01) beneficiaries. More ORYGB patients compared with Design: Retrospective cohort study. LRYGB patients were discharged with nonroutine dispositions (7.7% vs 2.4%; P=.005), died (0.2% vs 0.1%; Setting: The Nationwide Inpatient Sample. PϽ.001), and had 1 or more complications (18.7% vs 12.3%; PϽ.001). All Patient Safety Indicator rates were Patients: Patients undergoing ORYGB and LRYGB. higher for ORYGB. Patients who had ORYGB compared with LRYGB also had longer median lengths of stay (3.5 Main Outcome Measures: Outcome measures were vs 2.4 days; PϽ.001) and higher total charges ($35 018 number of procedures performed, patient and hospital vs $32 671; PϽ.001). Patients who had LRYGB had a characteristics, patient complications, mortality, length lower odds ratio than patients who had ORYGB for both of stay, resource use, and Agency for Healthcare Re- mortality (odds ratio, 0.54; PϽ.001) and having 1 or more search and Quality Patient Safety Indicators. Both demo- complications (odds ratio, 0.66; PϽ.001) even after ad- graphic and outcomes variables were compared by either justing for confounding variables. t test or ␹2 analysis, with regression analysis adjusting for confounding variables. Conclusion: In this population-based study, LRYGB pro- vided greater safety than ORYGB even after adjusting for Results: The ORYGB and LRYGB cohorts consisted of patient-level socioeconomic and comorbidity differ- 41 094 and 115 177 cases, respectively. From 2005 to ences. 2007, LRYGB was more commonly performed than ORYGB (72% vs 28%; PϽ.001) and at high-volume hos- Arch Surg. 2012;147(6):550-556

BESITY IS A MAJOR PUBLIC to the United States, as obesity has be- health concern as rates of come an international issue, even in less obesity have signifi- developed countries.7-9 cantly increased. In the United States, preva- CME available online at lenceO of obesity in adults increased from 15% www.jamaarchivescme.com in 1980 to 32% in 2004 and 6% to 19% in and questions on page 495 children during the same period.1-3 Accord- ing to Centers for Disease Control and Pre- vention reports, 34% of US adults older than Obesity has been associated with nu- 20 years are obese. In addition, the Cen- merous adverse health conditions includ- ters for Disease Control and Prevention re- ing diabetes mellitus, cardiovascular dis- ports 66% of US adults are overweight or ease, nonalcoholic liver disease, increased obese and 5 million are extremely obese.4 risk of disability, hypertension, dyslipid- The 2008 FasinFatreport indicates that emia, some forms of cancers, gallstones, 25% of adults are obese in 28 states and that and musculoskeletal disorders.10-12 These no state has had a decrease in the numbers adverse health conditions lead to a sig- of obese individuals.5 nificant increase in early mortality and re- Author Affiliations: Stanford Studies have also shown troubling so- duction in life years in obese individu- Center for Outcomes Research ciodemographic disparities in the United als.13,14 In addition, obese individuals report and Evaluation, Section of States because higher proportions of Afri- lower quality of life and greater discrimi- Minimally Invasive and 15,16 Bariatric Surgery, Department of can American individuals, Hispanic indi- nation, bias, and stigma. Treating obe- Surgery, Stanford University viduals, Medicaid patients, and those of sity and its complications has been esti- School of Medicine, Stanford, low education and income status are mated to cost approximately $92 billion California. obese.6 Obesity concerns are not limited a year.17

ARCH SURG/ VOL 147 (NO. 6), JUNE 2012 WWW.ARCHSURG.COM 550

©2012 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/29/2021 Bariatric surgery has proven to be the most effective urban vs rural location, and teaching vs nonteaching status. and enduring option in treating the morbidly obese. In Teaching status of the hospital was determined by hospital af- the 10-year Swedish Obesity Study, gastric bypass sur- filiation with either a medical school or an Accreditation gery patients showed greater long-term weight loss, health- Council for Graduate Medical Education residency program. related quality-of-life improvements, and reduction in vari- Hospital size (small, medium, and large) has differing defini- tions depending on urban or rural and teaching status. The ous risk factors compared with controls receiving NIS includes sample weights, designated by the sampling de- conventional treatment. The Swedish Obesity Study trial sign, to make nationally representative estimates of inpatient found that the decrease in weight and risk factors was use. The NIS is based on a stratified probability sample of all greater in those treated by gastric bypass than those treated nonfederal US hospitals, which provides a national estimate by medical intervention, banding, or vertical banded gas- of inpatient health services. The data set contains information troplasty.18,19 on primary and secondary diagnoses, demographic character- However, mortality and other complications are se- istics, procedure use, length of stay, payer, total charges, and rious risks associated with bariatric surgery and are in- admission and discharge status. The NIS does not have versely correlated with the volume of cases performed unique patient identifiers, and as a result, patients cannot be by the surgeon.20 The Centers for Medicare & Medicaid followed up longitudinally. Additional information about the NIS is available from the Agency for Healthcare Research and Services determined that bariatric procedures should only Quality.30 be performed at high-volume facilities that are either cer- tified by the American College of Surgeons or the Ameri- STUDY SAMPLE can Society for Metabolic and Bariatric Surgery/Surgery Review Corporation as a Bariatric Surgery Center of Ex- 21,22 In this retrospective cohort study, patients were identified by cellence. The Centers for Medicare & Medicaid Ser- the ICD-9-CM procedure codes for LRYGB (44.38) and vices defines high-volume practices as performing 125 ORYGB (44.31 and 44.39).29 To ensure homogeneity, cases or more weight-loss operations annually.23 were restricted to adult nontransfer elective patients with a di- The efficacy of gastric bypass surgery has made it the agnosis-related group code 288, obesity. Patients were also most commonly performed bariatric surgery in the United excluded if they had malignant neoplasm of digestive organs States. In 2002, 88% of all bariatric surgeries were gas- or peritoneum (ICD-9-CM codes 150-159), inflammatory tric bypass procedures.24 Laparoscopic Roux-en-Y gas- bowel disease (ICD-9-CM codes 555.0-556.9), or noninfec- tric bypass (LRYGB) was first introduced by Wittgrove tious colitis (ICD-9-CM codes 557.0-558.9). The period stud- et al25 in 1994 and since then has become more com- ied was from January 1, 2005, through December 31, 2007. This period was chosen so that the results would be suffi- monly performed than open Roux-en-Y gastric bypass 26 ciently recent to include the new ICD-9 LRYGB code (44.38), (ORYGB). Weller and Rosati report that 75% of gastric which was first introduced in 2004 and not readily used in bypass procedures in 2005 were performed laparoscopi- practice until 2005. cally. The safety of LRYGB vs ORYGB has been addressed PATIENT CHARACTERISTICS in single-center trials, but to our knowledge, no national data exist comparing the 2 approaches. Two ran- Patient age, sex, accompanying diagnoses, and comorbidities domized controlled trials have shown better outcomes were examined. The Deyo modification of the Charlson Co- for LRYGB compared with ORYGB.27,28 Prospective ran- morbidity Index31 (0-3, with 3 indicating greatest comorbid- domized controlled trials are used to establish the effi- ity) was calculated for each patient based on ICD-9-CM diag- cacy of interventions, but population-based studies are nosis codes. In addition, Elixhauser comorbidity measures were 32 critical to determine the effectiveness and adoption of in- calculated. To adjust for potential differences in delivery of terventions. However, few population-based studies have care that might bias outcomes, structure of care was assessed considered the effectiveness of LRYGB compared with by examining payer status, hospital size, hospital volume, and teaching status. ORYGB in terms of patient safety because International Classification of Diseases, Ninth Revision, Clinical Modi- fication (ICD-9-CM) laparoscopic codes for gastric by- OUTCOMES pass surgery were not available until 2004 and not fully 29 Our outcomes of interests were number of procedures per- adopted until 2005. formed, patient and hospital characteristics, patient complications, mortality, length of stay, resource use, and Agency for METHODS Healthcare Research and Quality Patient Safety Indicators (PSIs). The ICD-9-CM codes for complications were gathered by ref- erencing Santry et al24 and performing our own review of ICD- DATA SOURCE 9-CM codes (see codes in the eTable, http://www.archsurg .com). The PSIs were adjusted for age, sex, diagnosis-related Data were derived from the 2005-2007 Nationwide Inpatient group, and Elixhauser comorbidity index.32 High-volume hos- Sample (NIS), a hospital discharge database created as part of pitals were defined as being hospitals that performed 125 or the Agency for Healthcare Research and Quality Healthcare more bariatric surgery cases per year. Cost and Utilization Project. The NIS is the largest publicly The PSIs were developed by the Agency for Healthcare Re- available, all-payer, inpatient database in the United States. search and Quality Evidence-Based Practice Center at the Uni- For each year, the NIS contains information on 5 to 8 million versity of California, San Francisco and Stanford University. inpatient stays from about 1000 hospitals from 37 different The PSIs are a set of indicators providing information on po- states (approximately 20% of all community hospitals in the tential inpatient adverse events following surgeries, medical pro- United States), stratified by geographic region, hospital size, cedures, and childbirth. The PSIs were developed through a

ARCH SURG/ VOL 147 (NO. 6), JUNE 2012 WWW.ARCHSURG.COM 551

©2012 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/29/2021 2008) is available as module programs for use with SAS soft- Table 1. Patient Demographics: Elective Gastric Bypass, ware version 9.1 (SAS Institute Inc). 2005 to 2007

% STATISTICAL ANALYSIS P Open Laparoscopic Total Value We applied the sampling weights designated by the NIS sam- Procedures, No. (%) 41 094 (26) 115 177 (74) 156 271 pling design to make nationally representative estimates of the Age, y number of procedures performed between 2005 and 2007. We 18-29 12.4 11.8 11.9 .51 established statistically significant differences in the categori- 30-39 27.7 28.5 28.5 .94 cal data using the Rao-Scott ␹2 and t test for continuous vari- 40-49 30.1 30.0 30.0 .97 ables.34 All point estimates, standard error estimates, and tests 50-59 24.2 24.0 24 .84 of significance accounted for the weighted nature of the data. Ն 60 5.6 5.8 5.8 .47 Regression analyses were applied to complication and mortal- Female 80.5 82.9 82.5 .008 ity outcomes to correct for potential confounders. All analyses White 73.9 75 75 .70 were performed with SAS software version 9.1. Insurance Medicare 9.3 7.1 7.8 .26 Medicaid 10.4 5.9 7.2 .001 RESULTS Private insurance 69.9 75.1 73.8 .35 Other 10.4 12.1 11.2 .40 Characteristics of patients undergoing elective gastric bypass procedures from 2005 to 2007 appear in Table 1. The final cohorts of ORYGB and LRYGB were 41 094 and 115 177 weighted discharges, respectively (PϽ.001). The median age of the population was 42.7 years. The ma- Open Laparoscopic jority of patients were white (75%) and female (82.5%). 80 We found that a higher percentage of ORYGB than LRYGB P < .001 70 P = .73 patients were covered by Medicare (9.3% vs 7.1%; P=.26) 60 P = .71 and Medicaid (10.4% vs 5.9%; P=.001). 50 The Figure describes characteristics of hospitals per-

% 40 forming elective bariatric surgery procedures from 2005 30 to 2007. A majority of the gastric bypass procedures were 20 performed at nonprofit (57%), teaching (61.3%), and 10 high-volume hospitals (60.7%). There were no signifi- 0 cant differences between LRYGB vs ORYGB for non- Nonprofit Teaching HVH profit or teaching status, but there was a significant dif- Type of Hospital ference with regard to high-volume status. A higher proportion of LRYGBs (69%) were performed at high- Figure. Characteristics of hospitals performing gastric bypass in 2005 to volume hospitals than ORYGB (61%) cases (PϽ.001). 2007. HVH indicates high-volume hospital. A comparison of patient severity of illness is presented in Table 2. A higher percentage of ORYGB patients had Charlson Comorbidity Index scores of 2 or Table 2. Severity of Illness: Gastric Bypass Surgery greater than LRYGB patients (7.8% vs 7.1%; P=.24). A Patients, 2005 to 2007 greater proportion of ORYGB compared with LRYGB patients had comorbidities of diabetes mellitus (31.4% vs % P 29.6%; P=.03) and obstructive sleep apnea (28.6% vs Open Laparoscopic Value 26.2%; PϽ.001), but a greater proportion of LRYGB com- Charlson Comorbidity Index score pared with ORYGB patients had venous stasis disease 0 55.4 57.1 .34 (1.3% vs 0.7%; PϽ.001). There were no significant dif- 1 36.7 35.8 .24 ferences with hypertension or lysis of adhesion (surro- 2 7.1 6.6 .56 gate marker for previous surgery) between the 2 co- 3 0.7 0.5 .43 horts. In terms of Elixhauser comorbidity measures, we Diabetes mellitus 31.4 29.6 .03 Hypertension 52.7 53.4 .44 found slightly higher rates in ORYGB than LRYGB cases Obstructive sleep apnea 28.6 26.2 Ͻ.001 for the following: cardiac arrhythmia (5.76% vs 4.26%; Venous stasis disease 0.7 1.3 Ͻ.001 P=.008), congestive heart failure (2.03% vs 1.43%; Lysis of adhesion 8.9 9.2 .62 P=.002), diabetes complicated (1.74% vs 1.20%; P=.01), pulmonary circulatory disorders (0.92% vs 0.47%; P=.001), and peripheral vascular disorders (0.46% vs 0.30%; P=.07). There were no significant differences be- comprehensive literature review, ICD-9-CM code study, review by a physician panel, risk adjustment measures, and em- tween the 2 cohorts with the 25 other Elixhauser comor- pirical analyses. The PSIs screen for problems that patients ex- bidity measures (Table 3). perience as a result of exposure to the health care system and In Table 4, in-hospital outcomes are presented: total that are likely amenable to prevention by changes at the sys- charges, length of stay, and rates of complication, mor- tem or provider level.33 The PSI software version 3.2 (March tality, and nonroutine disposition. A significantly greater

ARCH SURG/ VOL 147 (NO. 6), JUNE 2012 WWW.ARCHSURG.COM 552

©2012 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/29/2021 proportion of ORYGB compared with LRYGB patients cations (0.8% vs 0.4%; PϽ.001), splenic complications were discharged with nonroutine dispositions (7.7% vs (0.5% vs 0.04%; PϽ.001), genitourinary tract complica- 2.4%), died (0.2% vs 0.1%), and had 1 or more compli- tions (2.3% vs 1.3%; PϽ.001), cardiac arrhythmia (4.3% cations (18.7% vs 12.3%). Patients who had ORYGB com- vs 2.8%; PϽ.001), cardiac complication not otherwise pared with LRYGB also had longer median lengths of stay specified (1.2% vs 0.7%; PϽ.001), other and unspecified (3.5 vs 2.4 days; PϽ.001) and higher total charges effects of external causes (9.2% vs 6.1%; PϽ.001), gas- ($35 018 vs $32 671; PϽ.001). trointestinal leak (1.9% vs 1.2%; PϽ.001), sepsis (0.9% Rates and odds ratios (ORs) of specific complications vs 0.4%; PϽ.001), and unexpected reoperations for sur- for LRYGB vs ORYGB are presented in Table 5. A higher gical complications (1.1% vs 0.3%; PϽ.001). The OR for percentage of ORYGB than LRYGB patients had blood LRYGB vs ORYGB was 0.76 for blood transfusion Ͻ transfusion (2.7% vs 1.9%; P .001), abscess (0.7% vs 0.4%; (PϽ.001), 0.57 for abscess (PϽ.001), 0.24 for pulmo- Ͻ Ͻ P .001), pulmonary embolism (0.3% vs 0.1%; P .001), nary embolism (PϽ.001), 0.50 for pneumonia (PϽ.001), pneumonia (1.1% vs 0.5%; PϽ.001), other pulmonary Ͻ Ͻ 0.55 for other pulmonary complications (P .001), 0.53 complications (2.6% vs 1.2%; P .001), wound compli- for wound complications (PϽ.001), 0.07 for spleen complications (PϽ.001), 0.64 for genitourinary tract compli- Table 3. Selected Elixhauser Comorbidity Measures: cations (PϽ.001), 0.69 for cardiac arrhythmia (PϽ.001), Elective Gastric Bypass Patients, 2005 to 2007 0.56 for cardiac complications not otherwise specified (PϽ.001), 0.65 for other and unspecified effects of exter- % nal causes (PϽ.001), 0.66 for gastrointestinal leak P Ͻ Comorbidity Open Laparoscopic Value (P .001), 0.29 for unexpected reoperations for surgical Ͻ Ͼ complications (P .001), 0.70 for small-bowel obstruc- Hypertension 53.4 53.4 .99 Ͻ Ͻ Diabetes, uncomplicated 29.3 28.4 .66 tion (P .001), and 0.51 for sepsis (P .001). COPD 19.9 18.6 .44 Selected PSI rates appear in Table 6. All PSI rates were Depression 19.4 24.2 Ͻ.001 significantly higher for ORYGB than LRYGB. Rates were Liver disease 10.5 9.42 .55 more than twice as high for ORYGB vs LRYGB for fail- Hypothyroidism 10.2 10.8 .74 ure to rescue, selected infections due to medical care, post- Cardiac arrhythmia 5.76 4.26 .008 Fluid/electrolyte disorders 5.80 3.05 Ͻ.001 operative hemorrhage or hematoma, postoperative re- Congestive heart failure 2.03 1.43 .002 spiratory failure, and postoperative pulmonary embolism Valvular disease 1.71 1.42 .63 or deep vein thrombosis. Diabetes, complicated 1.74 1.20 .01 Data were entered into a multivariate logistic regres- Deficiency anemia 1.05 1.05 Ͼ.99 sion analysis to determine adjusted ORs for LRYGB vs Pulmonary circulatory 0.92 0.47 .001 disorders ORYGB (Table 7). Laparoscopic gastric bypass was RA/collagen vascular 0.89 1.05 .33 highly protective for mortality (OR, 0.54; PϽ.001), 1 or disorders more complications (OR, 0.66; PϽ.001), and nonrou- Other neurological 0.86 0.86 Ͼ.99 tine disposition (OR, 0.43; PϽ.001). The model for 1 or disorders more complications is presented in detail. Using pa- Coagulation disorders 0.54 0.35 .35 Peripheral vascular 0.46 0.30 .07 tients who underwent ORYGB as a referent group, uni- disorders variate analysis indicated that LRYGB had a reduced risk Renal failure 0.45 0.47 .82 for in-hospital complications (OR, 0.657; PϽ.001) as did Drug abuse 0.27 0.20 .73 high-volume status (OR, 0.785; P=.003). The OR for com- PUD, excluding bleeding 0.24 0.21 .35 plications was higher for age 60 years and older (OR, Blood loss, anemia 0.23 0.33 .44 Ͻ Alcohol abuse 0.22 0.16 .65 2.174; P .001), age between 50 and 59 years (OR, 1.818; Psychoses 0.16 0.14 .82 PϽ.001), age between 40 and 49 years (OR, 1.430; Paralysis 0.05 0.05 Ͼ.99 PϽ.001), presence of lysis of adhesion (OR, 1.576; Lymphoma 0.04 0.03 .62 PϽ.001), and Medicare insurance status (OR, 1.422; Solid tumor without 0.03 0.05 .56 PϽ.001). Calculation of incremental ORs indicated that metastasis the laparoscopic approach had the largest protective ef- Abbreviations: COPD, chronic obstructive pulmonary disease; PUD, peptic fect for complications, and there was little additive in- ulcer disease; RA, rheumatoid arthritis. teraction between other known risk factors (Table 8).

Table 4. In-Hospital Outcomes: Gastric Bypass Patients, 2005 to 2007

Adjusted Odds P Open Laparoscopic Ratio (95% CI) Value LOS, d, median 3.5 2.4 0.43 (0.32-0.56) Ͻ.001 Total charges, $, median 35 018 32 671 0.79 (0.68-0.92) Ͻ.001 Ն1 Complications, % 18.7 12.3 0.66 (0.56-0.77) Ͻ.001 Mortality, % 0.2 0.1 0.54 (0.40-0.73) Ͻ.001 Nonroutine disposition, % 7.7 2.4 0.29 (0.12-0.69) .005

Abbrevation: LOS, length of stay.

ARCH SURG/ VOL 147 (NO. 6), JUNE 2012 WWW.ARCHSURG.COM 553

% Adjusted Odds P Complication Open Laparoscopic Ratio (95% CI) Value Other 9.2 6.1 0.65 (0.62-0.68) Ͻ.001 Cardiac arrhythmia 4.3 2.8 0.69 (0.64-0.74) Ͻ.001 Blood transfusion 2.7 1.9 0.76 (0.70-0.82) Ͻ.001 Other pulmonary 2.6 1.2 0.55 (0.50-0.60) Ͻ.001 GU 2.3 1.3 0.64 (0.58-0.70) Ͻ.001 GI leak 1.9 1.2 0.66 (0.60-0.73) Ͻ.001 Hemorrhagic 1.7 1.8 1.05 (0.95-1.14) .21 DVT 1.4 1.4 1.04 (0.93-1.15) .19 SBO 1.3 0.9 0.70 (0.62-0.78) Ͻ.001 Cardiac, NOS 1.2 0.7 0.56 (0.49-0.63) Ͻ.001 Pneumonia 1.1 0.5 0.50 (0.43-0.57) Ͻ.001 Reoperations 1.1 0.3 0.29 (0.25-0.34) Ͻ.001 Sepsis 0.9 0.4 0.51 (0.44-0.60) Ͻ.001 Wound 0.8 0.4 0.53 (0.45-0.62) Ͻ.001 Abscess 0.7 0.4 0.57 (0.48-0.67) Ͻ.001 Spleen 0.5 0.04 0.07 (0.04-0.10) Ͻ.001 PE 0.3 0.1 0.24 (0.17-0.34) Ͻ.001 MI 0.1 0.1 0.67 (0.43-1.02) .76

Abbrevations: DVT, deep vein thrombosis; GI, gastrointestinal; GU, genitourinary tract; LOS, length of stay; MI, myocardial infarction; NOS, not otherwise specified; PE, pulmonary embolism; SBO, small-bowel obstruction.

Table 6. Incidence Rates for Selected Agency for Health Table 7. Adjusted Odds Ratios for In-Hospital Care Research and Quality Patient Safety Indicators: Complications: Gastric Bypass Patients, 2005 to 2007 Gastric Bypass Surgeries, 2005 to 2007a Laparoscopic vs Open Odds P % Regression Variable Ratio (95% CI) Value P Ͻ Open Laparoscopic Value Mortality 0.54 (0.40-0.73) .001 Ն1 Complications 0.66 (0.56-0.77) Ͻ.001 Ͻ Failure to rescue 136.31 75.71 .001 Nonroutine 0.43 (0.34-0.62) Ͻ.001 Postoperative pulmonary 17.22 6.19 Ͻ.001 disposition embolism or deep vein thrombosis Specific Independent Odds Ratio (95% CI) P Value Ն Postoperative respiratory 10.87 3.94 Ͻ.001 Variables for 1 failure Complications Ͻ Accidental puncture or 5.61 3.27 Ͻ.001 Laparoscopic vs open 0.657 (0.563-0.767) .001 laceration Age, y Selected infections due to 5.42 1.17 Ͻ.001 Ͻ30 1 [Reference] medical care 30-39 1.161 (1.009-1.335) .04 Postoperative hemorrhage 4.34 1.94 Ͻ.001 40-49 1.430 (1.244-1.644) Ͻ.001 or hematoma 50-59 1.818 (1.566-2.109) Ͻ.001 Postoperative physiologic 0.96 0.63 Ͻ.001 Ն60 2.174 (1.768-2.672) Ͻ.001 and metabolic Lysis of adhesion 1.576 (1.362-1.824) Ͻ.001 derangements Male 1.351 (1.207-1.512) Ͻ.001 Charlson Comorbidity a Smoothed risk-adjusted incidence rates per 1000. Index score 0 1 [Reference] 1 1.049 (0.970-1.135) .25 2 1.219 (1.034-1.437) .02 COMMENT 3 1.202 (0.809-1.785) .33 Insurance To our knowledge, this is the first national, all-payer, all- Private insurance 1 [Reference] Medicare 1.422 (1.248-1.620) Ͻ.001 hospital, and largest comparison of LRYBG and ORYGB. Medicaid 1.165 (0.982-1.382) .19 We found that the majority of patients in our study sample Other 0.964 (0.703-1.226) .34 underwent LRYGB (73.7%). This is consistent with 2 re- High-volume hospital 0.785 (0.668-0.923) .003 cent national estimates26,35 but much higher than estimates done prior to the introduction of ICD-9-CM codes specific to LRYGB.36 A greater proportion of LRYGB cases sistent with results from a population-based study looking were performed at high-volume hospitals and this may at data from academic centers across the United States be reflective of the additional experience required to per- from 2004 to 2006.35 form LRYGB. A few significant differences in comorbidi- In addition, Livingston and Ko6 reported that of those ties between ORYGB and LRYGB patients existed con- eligible for bariatric surgery, 16% were covered by Medi-

ARCH SURG/ VOL 147 (NO. 6), JUNE 2012 WWW.ARCHSURG.COM 554

©2012 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/29/2021 The minimally invasive approach of LRYGB appears Table 8. Incremental Odds Ratio for Development to allow greater safety and lower resource use than of In-Hospital Complications: Gastric Bypass Patients, ORYGB. This large, nationally representative compari- 2005 to 2007 son confirms and replicates prior randomized trial evidence supporting the laparoscopic approach, indicating Model Odds Ratio (95% CI)a safe dissemination of this technology. For bariatric sur- LRYGB vs ORYGB 0.641 (0.54-0.76) gery, patient safety may be further enhanced by appro- LRYGB vs ORYGB, age 0.631 (0.54-0.75) LRYGB vs ORYGB, age, lysis of adhesion 0.630 (0.53-0.74) priate application of the laparoscopic approach. LRYGB vs ORYGB, age, lysis 0.634 (0.54-0.75) of adhesion, sex Accepted for Publication: January 6, 2012. LRYGB vs ORYGB, age, lysis of adhesion, 0.636 (0.54-0.75) sex, Charlson Comorbidity Index Correspondence: John M. Morton, MD, MPH, Depart- LRYGB vs ORYGB, age, lysis of adhesion, 0.645 (0.55-0.76) ment of Surgery, Stanford University Medical Center, 300 sex, Charlson Comorbidity Index, insurance Pasteur Dr, H3680, Stanford, CA 94305-5655 (morton LRYGB vs ORYGB, age, lysis of adhesion, 0.657 (0.56-0.77) @stanford.edu). sex, Charlson Comorbidity Index, Author Contributions: Study concept and design: Banka, insurance, hospital volume Woodard, and Morton. Acquisition of data: Hernandez- Abbrevations: LRYGB, laparoscopic Roux-en-Y gastric bypass; ORYGB, Boussard. Analysis and interpretation of data: Banka, open Roux-en-Y gastric bypass. Woodard, Hernandez-Boussard, and Morton. Drafting of a Significance: P Ͻ .001. the manuscript: Banka, Woodard, and Morton. Critical revision of the manuscrpt for important intellectual content: Banka, Woodard, Hernandez-Boussard, and Morton. Sta- care and 12%, by Medicaid. Among those who received tistical analysis: Banka, Woodard, Hernandez-Boussard, bariatric surgery, only 5% were Medicare patients and 8%, and Morton. Obtained funding: Morton. Administrative, Medicaid patients. This study shows that Medicaid (7.2%) technical, and material support: Morton. Study supervi- or Medicare (7.8%) coverage remains low among those sion: Morton. receiving bariatric surgery. In addition, a higher percent- Financial Disclosure: None reported. age of ORYGB (10.4%) compared with LRYGB (5.9%) Funding/Support: This research was funded by the Stan- patients were covered by Medicaid (PϽ.01). These dis- ford Center for Outcomes Research and Evaluation, Stan- parities among Medicaid and nonwhite patients may be ford University Medical School and Stanford University caused by the low socioeconomic status associated with Hospitals and Clinics, Stanford, California. these 2 groups. Greater efforts should be made to edu- Online-Only Material: The eTable is available at http: cate these groups about the benefits of gastric bypass sur- //www.archsurg.com. gery and ensure they have access to appropriate bariatric procedures, especially by improving coverage and reimbursement for bariatric surgery in these groups. REFERENCES Livingston and Ko6 also reported that males composed 36% of the population eligible for gastric bypass surgery 1. Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adults, 1999-2000. JAMA. 2002;288(14):1723-1727. but only 14% of the population who received gastric by- 2. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Preva- pass surgery. Consistent with studies by Livingston and Ko lence of overweight and obesity in the United States, 1999-2004. JAMA. 2006; and Santry et al,24 we also observed that only 17.5% of gas- 295(13):1549-1555. tric bypass procedures were performed on men. This dis- 3. Ogden CL, Flegal KM, Carroll MD, Johnson CL. Prevalence and trends in overweight among US children and adolescents, 1999-2000. JAMA. 2002;288(14): parity between males and females may be due to greater 1728-1732. societal pressures on females to be thin and/or reluctance 4. Ogden CL, Carroll MD, McDowell MA, Flegal KM. Obesity among adults in the from males to seek medical help for morbid obesity. United States: no statistically significant change since 2003-2004. http://www Most importantly and consistent with previous pro- .medpagetoday.com/upload/2007/11/28/Nobesedb01.pdf. Published Novem- spective randomized trials and population-based stud- ber 2007. Accessed April 3, 2012. 5. Levi J, Vinter S, Richardson L, St Laurent R, Segal LM. F as in Fat: How Obesity ies, LRYGB vs ORYGB had significantly lower rates of Policies Are Failing in America. Princeton, NJ: The Robert Wood Johnson Foun- complications, length of stay, charges, and mortal- dation; 2008. ity.26-28,35-37 To our knowledge, this is the first study to 6. Livingston EH, Ko CY. Socioeconomic characteristics of the population eligible compare PSI rates between LRYGB and ORYGB proce- for obesity surgery. Surgery. 2004;135(3):288-296. 7. Rennie KL, Jebb SA. Prevalence of obesity in Great Britain. Obes Rev. 2005;6(1): dures and we found that rates for LRYGB were lower for 11-12. all PSIs compared with ORYGB. 8. Padez C, Fernandes T, Moura˜o I, Moreira P, Rosado V. Prevalence of overweight Limitations of this study are reflective of the data and obesity in 7-9-year-old Portuguese children: trends in body mass index from source. Data are collected administratively, and as a 1970-2002. Am J Hum Biol. 2004;16(6):670-678. result, important clinical information such as body 9. Luo J, Hu FB. Time trends of obesity in pre-school children in China from 1989 to 1997. Int J Obes Relat Metab Disord. 2002;26(4):553-558. mass index is not present. Furthermore, the data set 10. Bray GA. Risks of obesity. Endocrinol Metab Clin North Am. 2003;32(4):787-804. only addresses inpatient complications and longer- 11. Kral JG. Morbidity of severe obesity. Surg Clin North Am. 2001;81(5):1039-1061. term complications are not available. Furthermore, the 12. Field AE, Coakley EH, Must A, et al. Impact of overweight on the risk of devel- generally longer length of stay associated with ORYGB oping common chronic diseases during a 10-year period. Arch Intern Med. 2001; 161(13):1581-1586. may lead to increased complication rates, given the 13. Dorn JM, Schisterman EF, Winkelstein W Jr, Trevisan M. Body mass index and greater opportunity for iatrogenic harm while in the mortality in a general population sample of men and women: the Buffalo Health hospital. Study. Am J Epidemiol. 1997;146(11):919-931.

ARCH SURG/ VOL 147 (NO. 6), JUNE 2012 WWW.ARCHSURG.COM 555

©2012 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/29/2021 14. Calle EE, Thun MJ, Petrelli JM, Rodriguez C, Heath CW Jr. Body-mass index and 27. Luja´n JA, Frutos MD, Herna´ndez Q, et al. Laparoscopic versus open gastric by- mortality in a prospective cohort of U.S. adults. N Engl J Med. 1999;341(15): pass in the treatment of morbid obesity: a randomized prospective study. Ann 1097-1105. Surg. 2004;239(4):433-437. 15. Fontaine KR, Barofsky I. Obesity and health-related quality of life. Obes Rev. 2001; 28. Nguyen NT, Goldman C, Rosenquist CJ, et al. Laparoscopic versus open gastric 2(3):173-182. bypass: a randomized study of outcomes, quality of life, and costs. Ann Surg. 16. Puhl R, Brownell KD. Bias, discrimination, and obesity. Obes Res. 2001;9(12):788- 2001;234(3):279-289, discussion 289-291. 805. 29. US Department of Health and Human Services. International Classification of Dis- 17. Finkelstein EA, Fiebelkorn IC, Wang G. National medical spending attributable to eases, Ninth Revision, Clinical Modification (ICD-9-CM). Hyattsville, MD: Na- overweight and obesity: how much, and who’s paying? Health Aff (Millwood). tional Center for Health Statistics; 2007. 2003;W3(suppl web exclusives):W3-219-26. 30. Overview of the Nationwide Inpatient Sample (NIS). Healthcare Cost and Utili- 18. Sjo¨stro¨m L, Lindroos AK, Peltonen M, et al; Swedish Obese Subjects Study Sci- zation Project (HCUP) website. www.hcup-us.ahrq.gov/nisoverview.jsp. Pub- entific Group. Lifestyle, diabetes, and cardiovascular risk factors 10 years after lished 2007. Accessed August 2011. bariatric surgery. N Engl J Med. 2004;351(26):2683-2693. 31. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prog- 19. Sjo¨stro¨m L, Narbro K, Sjo¨stro¨m CD, et al; Swedish Obese Subjects Study. Effects nostic comorbidity in longitudinal studies: development and validation. J Chronic of bariatric surgery on mortality in Swedish obese subjects. N Engl J Med. 2007; Dis. 1987;40(5):373-383. 357(8):741-752. 32. Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use 20. Flum DR, Salem L, Elrod JA, Dellinger EP, Cheadle A, Chan L. Early mortality with administrative data. Med Care. 1998;36(1):8-27. among Medicare beneficiaries undergoing bariatric surgical procedures. JAMA. 33. Patient Safety Indicators Overview. Agency for Healthcare Research and Quality 2005;294(15):1903-1908. website. http://qualityindicators.ahrq.gov/Modules/psi_overview.aspx. Pub- 21. Decision memo for bariatric surgery for the treatment of morbid obesity lished 2006. Accessed February 2011. (CAG-00250R). Centers for Medicare & Medicaid Services website. http://www 34. Rao JNK, Scott AJ. The analysis of categorical data from complex sample surveys: .cms.gov. Published 2008. Accessed August 19, 2011. chi-squared tests for good and fit and independence in two-way tables. JAmStat 22. American Society for Metabolic and Bariatric Surgery website. www.asmbs.org. Assoc. 1981;76(374):221-230. doi:10.2307/2287815. Published 2008. Accessed August 18, 2011. 35. Nguyen NT, Hinojosa M, Fayad C, Varela E, Wilson SE. Use and outcomes of lapa- 23. American College of Surgeons website. http://www.facs.org/. Published 2008. roscopic versus open gastric bypass at academic medical centers. J Am Coll Surg. Accessed August 20, 2011. 2007;205(2):248-255. 24. Santry HP, Gillen DL, Lauderdale DS. Trends in bariatric surgical procedures. 36. Ricciardi R, Town RJ, Kellogg TA, Ikramuddin S, Baxter NN. Outcomes after open JAMA. 2005;294(15):1909-1917. versus laparoscopic gastric bypass. Surg Laparosc Endosc Percutan Tech. 2006; 25. Wittgrove AC, Clark GW, Tremblay LJ. Laparoscopic gastric bypass, Roux- 16(5):317-320. en-Y: preliminary report of five cases. Obes Surg. 1994;4(4):353-357. 37. Flum DR, Belle SH, King WC, et al; Longitudinal Assessment of Bariatric Sur- 26. Weller WE, Rosati C. Comparing outcomes of laparoscopic versus open bariat- gery (LABS) Consortium. Perioperative safety in the longitudinal assessment of ric surgery. Ann Surg. 2008;248(1):10-15. bariatric surgery. N Engl J Med. 2009;361(5):445-454.

ARCH SURG/ VOL 147 (NO. 6), JUNE 2012 WWW.ARCHSURG.COM 556