Meta-analysis of the Resolution rate of Type 2 diabetes mellitus in patients undergoing Roux-en-Y gastric bypass surgery

(拉杰斯)，(葛海燕)

1Tongji University School of Medicine，Shanghai 200072，China2Department of Gastro-Intestinal Surgery，Shanghai Tenth People’s Hospital，Shanghai 200072，China

Meta-analysis of the Resolution rate of Type 2 diabetes mellitus in patients undergoing Roux-en-Y gastric bypass surgery

Ramsohok Lajeswar1(拉杰斯)，Ge Haiyan2(葛海燕)

1Tongji University School of Medicine，Shanghai200072，China2Department of Gastro-Intestinal Surgery，Shanghai Tenth People’s Hospital，Shanghai200072，China

Background One of the major co-morbidities in obese and morbidly obese patients is Type 2 diabetes(T2DM)with approximately 20%of them suffering from this disease.The objective of this review was to determine the impact of Roux-en-Y gastric bypass(RYGB)surgery on the resolution rate of T2DM in obese diabetic patients.

MethodsA systematic review was made into the literature using MEDLINE database to search for relevant studies reporting the resolution or improvement of T2DM post RYGB.Data from the case series and systematic review were collected for different age groups and different degrees of obesity(obese，morbidly obese，super obese)and meta-analysed.The various other aspects of the surgery regarding the complications，mortality rate，cost analysis，improvement in quality of life were also reviewed.

ResultsData from 56 relevant studies were retrieved containing a total of 10289 patients.Meta-analysis of the data from case series and other systematic reviews showed that RYGB is an effective method for obtaining weight reduction in obese，morbidly obese patients and in elderly obese patients.The mean age of patients was 42.8 years，mean baseline BMI was 47.4 kg/m2.The mean percentage excess weight loss(%EWL)was 65.3%and mean change in BMI was 17.1 kg/m2.The mean resolution rate of diabetes was 76.4%while the mean improved/resolved rate of T2DM was 87.4%.

ConclusionRYGB is a highly effective method for patients of various degrees of obesity and different age groups for treatment of obesity and T2DM.It also appears to be an interesting option economically due to savings from diabetic medications which outweigh the costs of the surgery and follow-up coupled with increased quality of life.

1 Introduction

The global epidemic of overweight and obesity－“globesity”is rapidly becoming a major public health problem in many parts of the world.Paradoxically coexisting with undernutrition in developing countries，the increasing prevalence of overweight and obesity is associated with many diet-related chronic diseases including diabetes mellitus，cardiovasculardisease，stroke，hypertension and certain cancers.Because this disease is strongly linked to obesity，the term“diabesity”has been coined to describe the confluence of the two disease processes.World Health Orgnisation’s(WHO)latest projections indicate that globally in 2005 approximately 1.6 billion adults(age 15+)were overweight;at least 400 million adults were obese.WHO further projects that by 2015，approximately 2.3 billion adults will be overweight and more than 700 million will be obese.At least 20 million children under the age of 5 years were overweight globally in 2005.Diabetes deaths are likely to increase by more than 50%in the next 10 years without urgent action［1］.

Diabetes and its complications have a significant economic impact on individuals，families，health systems and countries.For example，WHO estimates that in the period 2006 －2015，China will lose$558 billion in foregone national income due to heart disease，stroke and diabetes alone［1］.Recently，evidence for reduction of complications of type 2 diabetes with tight control of hyperglycemia has been reported［2］but current therapies，including diet，exercise，behavior modification，oral hypoglycemic agents，and insulin，rarely return patients to euglycemia［3］.There are currently no truly effective pharmaceutical agents to treat obesity，especially morbid obesity.Bariatric surgery has increased 10-fold within the past decade，from 14 000 procedures in 1993 to an estimated 140 000 in 2004.Gastric bypass now accounts for 93%of bariatric operations［4］.There are currently no truly effective phar-maceutical agents to treat obesity，especially morbid obesity.The application of laparoscopic techniques to bariatric surgery in the past decade has reduced perioperative morbidity and has contributed to a remarkable increase in interest in the surgical treatment of morbid obesity.

2 Definitions

Body Mass Index(BMI)is a simple index of weight-for-height that is commonly used to classify underweight，overweight and obesity in adults.It is defined as the weight in kilograms divided by the square of the height in metres(kg/m2).Morbid obesity，is defined in most articles as patients who exceed their ideal weight by at least 100 lb or are more than 200%of ideal body weight or having a BMI＞40 kg/m2.

During the search of the literature other terms that are found to be in use to describe overweight patients using BMI are morbidly obese BMI＞40，super obese BMI＞ 50，super super obese BMI＞60.Very often obese diabetic patients have a constellation of other metabolic abnormalities and known as Metabolic Syndrome.

3 Methods

The main objective of the study was to analyze the impact ofRoux-en-Y gastric bypass surgery(RYGB)on different groups of obese diabetic patients based on the BMI and on age with regard to resolution of Type 2 diabetes mellitus(T2DM).

By established systematic review methods，MEDLINE database was searched from the period January 1980 to October 2010 for articles in English.The search terms were“Obesity/surgery”，“Bariatric Surgery”，“Type 2 diabetes”，“Diabetes mellitus”，“weight loss surgery”，“gastric bypass”，“Roux-en-Y”.After a broad search，the abstracts of the articles were reviewed and relevant articles were selected.Final selection of articles was based on the inclusion criteria which were case series reporting diabetes relevant statistics in their RYGB cohort.

The next level of inclusion involved selecting those studies that dealt with surgical outcomes(including efficacy)，guidelines，and disease impact.Exclusion criteria were follow up less than 3months，number of diabetic patients in the RYGB cohort less than 5 languages other than English.

For data extraction the screened studies had to report outcomes for diabetes in the RYGB cohort.Extracted studies could be of any design and between January 1980 to October 2010 and having at least 5 diabetic patients in the RYGB cohort.All outcomes were preferentially extracted at the time points for which the comorbidity outcomes were available.Samples of which the follow-up time was not clear were not included.Some studies reported on mixed treatment groups where outcomes were not separable by procedure，these studies were not included.

Kin studies defined as multiple publications describing the same or overlapping series of patients were identified and entered into our catalog only once to avoid the double counting of patients.

Resolved and Resolved or Improved.

Outcomesofthe selected comorbidity were grouped into categories of resolved and resolved or improved.For the calculation of the percentage resolved，we included those studies reporting the number of patients in which comorbid conditions disappeared or no longer required therapy.We preferentially extracted the number of patients evaluated as the denominator wherever possible.For the calculation of the percentage resolved or improved，we included studies reporting numbers of patients in both of these 2 categories(in which case，the 2 were summed)，as well as studies that only used the term improved.Consequently，the percentage resolved or improved may be lower than the percentage deemed resolved due to different study cohorts and，therefore，different denomi-nators for the percentage calculations.Resolution of diabetes was often regarded as stoppage of all diabetic medications(oral hypoglycaemics and insulin)，and fasting blood glucose or glycosylated haemoglobin(HbA1c)back to normal values.

Indications for diabetes surgery

In 1991，the National Institutes of Health Consensus set guidelines for surgical treatment of obesity recommending patients with either a BMI of＞40 kg/m2or a BMI of＞35 kg/m2with associated comorbidities such as diabetes to be candidates for surgery［2］.But this also means that 45%of type 2 patients with diabetes worldwide demonstrate a BMI＞30 and would not qualify for surgical treatment based on these criteria.Numerous studies have demonstrated the high levels of metabolic risk factors at relatively low levels of BMI among Asian population because of more proneness to have central obesity.A study in China found a 9.8%prevalence of T2DM in a population with only 4.3%obesity［5］.

In comparison with the western population，Asians had a higher prevalence of diabetes，higher blood pressures，higher plasma triglyceride，and lower highdensity lipoprotein cholesterol concentrations at body mass index(BMI)that are much lower than WHO-recommended cut-off values for obesity.The 2005 A-sia Pacific Bariatric Surgery Group(APBSG)meeting modified the indications for bariatric surgery for Asian people:(1)obese patients with BMI＞37 kg/m2，(2)obese patients with BMI＞ 32 kg/m2in the presence of diabetes or two significant obesity-related comorbidities，(3)have been unable to lose or maintain weight loss by dietary or medical measures，(4)age of patient＞18 years and ＜65 years and no alcohol abuse and concurrent psychiatric illness［5］.

The bariatric surgeries being performed can be divided into two broad categories.One is restriction of oral intake and the other is malabsorption of ingested food.Some operations involve the use of only one mechanism for weight loss，whereas others combine the two.Restrictive operations include Vertical banded gastroplasty(VBG)，Laparoscopic adjustable gastric banding(LAGB);largely restrictive mildly malabsorptiveRoux-en-Y gastricbypass(RYGB);largely malabsorptive/mildly restrictive Biliopancreatic Diversion(BPD)and Duodenal switch(DS).Roux-en-Y gastric bypass is the predominant approach in the United States，whereas LAGB is most common in Europe and Australia.

Mechanism of RYGB

The Roux-en-Y gastric bypass has proven to be a far more successful procedure than other popular bariatric surgeries like adjustable gastric banding(AGB).It consists of creating a small gastric pouch along with forming a Roux-en-Y limb with the proximal bowel.The surgery works both by restricting food intake as well as causing malabsorption.This procedure also leads to significant changes in postprandial gut hormone levels［6］.Numerous individual studies have detailed the effects these surgeries have on the different markers of diabetes［2］.Two principal theories explain the possible mechanism of resolution of diabetes after bariatric procedure:the foregut and hindgut theory.

Results

Systematic review

A total of 399 citations were initially screened.After reviewing the abstracts，kin studies(separate reports of the same study groups)were eliminated.

The patients characteristics were varied from adolescents to＞60years of age and of different BMI subgroups.Some were paired studies comparing RYGB to other bariatric surgeries or to non surgical cohorts.Studies involving open，laparoscopic and hand assisted laparoscopic RYGB were all included.There were minor variations in the procedures depending on institutional preferences or the study cohort characteristics like the length of Roux limb.Study cohorts for revisional RYGB surgeries were not included.After elimination of non English case series and those having less than diabetic patients in their study cohorts we were left with 56 study series.

Statistical analysis

The data from these series were extracted and compiled into a comprehensive spss database and analyzed for relationships between the resolution rates of T2DM and other study characteristics using multiple regression and SPSS 14.0 software.

Descriptive statistics of the data

The total number of studies with relevant data that were retrieved was 56 and the total number of patients was 10289.The mean age of the subjects was 42.8 years(Standard Deviation［SD］8.5).The mean BMI was 47.4 kg/m2(SD 6.69).The mean change in BMI after RYGB was 17.1 kg/m2(SD 6.04).The mean percentage of excess weight loss(EWL)was 65.3% .The mean resolution rate of diabetes of all the studies was 76.4%and the mean percentage of improved/resolved T2DM rate was 87.4%.

Analysis according to age of patients

The outcomes of RYGB in different age groups varied because of the different characteristics of the patients.Patients ＞60 years old are usually sicker with more comorbidities.This puts them at greater risk for complications or death after bariatric surgery.But some studies in the veteran population have shown that despite the unique qualities RYGB can achieve long-term improvements in weight and obesity related comorbidities.The excess weight loss(EWL)achieved by the veteran(VA)subjects peaks between post operative year 1 and 2 similar to outcomes in the non veteran population.However，the peak EWL of 52% is not near the 65%to 80%EWL commonly achieved by subjects in the non VA studies［7］.

In our analysis we found that there is a significant regression relationship between the age of the patients in the study cohorts and the resolution of T2DM.The younger the age group the higher the resolution rate(P ＜0.007).Comparing older patients with a younger cohort，one study found that the weight loss was greater in younger patients but greater reduction of medication use in the older patients with similar complication rates and mortality rates in both groups［8］.

Analysis according to the different levels of BMI

On division of the studies into 3 groups(BMI＜34.9，35 ＜ BMI＜50，BMI＞50)and comparing the mean resolution rates of T2DM among them there is no significant difference in the rates.Similar results were obtained in another study which compared the resolution rates between two groups(BMI＜35 AND BMI＞35)and there was no significant difference found［3］.

RYGB in super obese who stay morbidly obese after surgery

In the super-obese patient，even when the surgical procedure is successful，the amount of weight loss achieved might still leave these patients morbidly obese.Despite this，the dramatic resolution of many of the key co-morbidities can still be obtained.RYGB thus offers to these patients significant benefits that medical therapy cannot achieve［8］.Some patients after RYGB have an initial resolution of T2DM and then in the long term follow-up there is re-emergence of diabetes.In one study 26%of the patients with initial resolution experienced recurrence of their T2DM and 20% worsened over time［9］.Significant differences were found in weight loss profiles between these two groups.The peak%EWL was lower and the failure rate and percentage of lost weight regained were greater for those who experienced a subsequent recurrence or worsening of their T2DM despite their lower initial BMI.

Effect of RYGB on the further evolution of diabetes

It is interesting to note that bariatric surgery does not only improve glycemic values but also reduces or delays the onset of diabetes in patients with morbid obesity.The SOS-study(Swedish Obese Subjects)compared two groups of patients with class III obesity and demonstrated a 32-fold reduction in the incidence of DM2 after 2 years of postoperative follow-up and a 5-fold reduction after 8 years for patients submitted to bariatric surgery［10］.Gastric bypass is effective in reducing the 5 year measured and the 10-year estimated Framingham risk of CHD events in morbidly obese individuals.The major risk reduction was observed in patients with T2DM［11］.

Resolution rates based of length of time the patients were suffering from diabetes

Some studies have found that the duration of diabetes in the patients had an effect on the resolution rates of T2DM.Patients with the shortest duration(＜5 years)，the mildest form of T2DM(diet controlled)，and the greatest weight loss after surgery were most likely to achieve complete resolution of T2DM［12］.The data showed resolution of T2DM in 91%of the patients if they carried the diagnosis for less than 10 years vs.38%if greater than 10 years［21］.Analysis of the cases of unresolved diabetes in another trial observed that the patients usually had T2DM for ＞5 years，they needed higher doses of insulin，and they were older than the rest of the patients.It is reasonable to conclude that the resolution of T2DM is influenced by the duration and the severity of the disease［12］.Sugerman et al showed that younger age patients are more likely to resolve their T2DM［13］.

Predictors of success after RYGB:

Among the preoperative variables considered(age，sex，race，marital status，parental status，employment status，depression，smoking，binge eating，and BMI)［14］，only preoperative BMI and marital status had a statistically significant effect on postoperative weight loss.Having said that，when looking at the number of pounds each group lost，we found that super obese patients lost significantly less weight than morbidly obese individuals.Marital status had a significant effect on the amount of weight loss after LRYGB，with single or divorced patients achieving better results［15］.

Outcome studies to date have identified that patient factors，surgeon experience，and variations in surgical technique may affect complication rates and/or the magnitude and duration of weight loss.Male sex is associated with central obesity，which has also been associated with increased perioperative risk.Surgeon experience is inversely related to operative complication rates and mortality.Centers and surgeons that perform more than100 gastric bypass operations per year report 2-to 3-fold lower complication rates and mortality rates［14］.The%EWL for the patients with T2DM remission has been found to be significantly greater than that for those in whom T2DM did not undergo remission.Younger age and a greater%EWL were independent predictors of T2DM remission.Each additional 12 years of age reduced the chance of T2DM remission by 20%，and each additional 10%EWL increased the chance of T2DM remission by an additional 8%.Multivariate analysis identified LRYGB，longer follow-up duration，and female gender as independent predictors of a greater%EWL［14］.

Obesity in adolescents

Extreme obesity(＞99th percentile of BMI for age)may affect 2%and 6%of all children and adolescents.Youth are increasingly developing health complications of obesity and also increased morbidity and later mortality compared with nonobese youth［16，17］.Given the increasing prevalence of childhood obesity，some have suggested that T2DM could develop in as many as 33%to 50%of all Americans born in the year 2000［18］.The lack of any major medical or surgical complications suggests that the risk/benefit ratio for RYGB in adolescents with T2DM is favorable.This is not unexpected，because these adolescents have likely not yet developed the cardiovascular pathology that is seen in obese diabetic adults.Indeed，recent analyses suggest that both perioperative morbidity and mortality risks may be lower for adolescents compared with adults undergoing bariatric surgery［19，20］

Indications for surgery in adolescence are failure of at least 6 months of organized，medically supervised weight loss attempts，ages 13 to 18 for girls，and 14 to 18 for boys，BMI＞40 with presence of severe obesityrelated comorbidity，BMI ＞50 with less severe obesityrelated comorbidities.Contraindications for bariatric surgery are substance abuse problem within the preceding year，psychiatric diagnosis that would impair ability to adhere to postoperative dietary or medication regimen(e.g.psychosis)，medically correctable cause of obesity，inability or unwillingness of patient or parent to fully comprehend the surgical procedure and its medical consequences，inability or refusal to participate in lifelong medical surveillance［21］.

Complications after RYGB

Complications after RYGB are varied;rates vary depending on the centre’s yearly volume of bariatric surgeries，the surgeons experience and the selection of patients groups among others.Short-term complications represent any symptom that occurs within 30 days of surgery，whereas long-term complications are those occurring 1 month post-surgery［2］.

Short-term complications—Stomal stenosis can result in persistent vomiting.Marginal ulceration along the anastomotic sites occurs in 1%－16% of patients and haematemesis，melena，and orthostatic hypotension can be early signs of a bleeding ulcer.Constipation is a common short term complication.Pulmonary embolism(PE)and pulmonary failure are two common causes of death in early post operative period［22］.

Long-term complications—special attention should be given to symptomatic cholelithiasis，dumping syndrome，persistent vomiting，and nutritional deficiencies

A fairly common side effect is postgastrectomy dumping syndrome and a rare complication is hypoglycemia from nesidioblastosis characterized by an endogenous hyperinsulinemic hypoglycemia［22］.

It has also been suggested that the identification of risk factors for adverse events after bariatric surgery could help to define high-risk groups and improve patient safety.They found that male gender was predictive of both accidental puncture or laceration and respiratory failure and advanced age was a strong risk factor for accidental puncture or laceration，pulmonary embolus or deep venous thrombosis，and respiratory failure.Advanced age，high body mass index(BMI)，and male sex have been associated with an increased risk of perioperative complications as well as mortality［23］.

Nutritional deficiencies after RYGB

Nutritional complications of RYGBP are associated with restricted intake of macro-and micro-nutrients and the bypass of absorptive and secretory areas of stomach and small intestine.Decreased intake and absorption of protein can occasionally lead to protein-calorie malnutrition，especially after malabsorptive procedures［24］.

Marinari et al have reported hypoalbuminemia that was mild in 11%and severe in 2.4%.Hypoalbuminemia is rare after restrictive procedures，including proximal gastric bypass.However，longer Roux limbs result in increased risk of protein-calorie malnutrition［15］.Micro-nutrient deficiencies after bariatric surgery can result in anemia，neurological deficits and osteopenia，as a consequence of deficiency of iron，folic acid，vitamin B12，vitamin B1，calcium，vitamin D and others［25］.Postoperative iron deficiency is more common after the combined malabsorptive-restrictive procedures，particularly in menstruating women.

Intensive care admission in postbariatric surgery

As many as one-fifth of the surgical ICU patients are obese and up to 7% are morbidly obese.Both chronic inflammation and diminished physiologic reserve place obese patients at risk for postoperative complications，as well as worse outcomes once these complications occur［26］.Men older than 50 year and with a BMI of more than 60 kg/m2，or who had immediate perioperative complications，would likely require critical care intervention.ICU care was required for more than 24 hours in 24%of patients［26］.Pulmonary，anastomotic，and thrombo-embolic complications were the most common reasons for ICU admission.One interesting finding was that approximately 40%of the cohort included men.

Mortality rates from diabetes

A recently published study of retrospective cohorts comparing obese patients who underwent surgical treatment to obese patients who had no surgical intervention showed a 40%decrease in overall mortality，56%decrease in specific mortality due to coronary heart disease，and 92%decrease in mortality due to diabetes.Flum et al.noted a 4.8%vs.1.7%mortality rate for patients over 65 years of age vs.younger patients.Mortality rates for ＜30days for open RYGB is around 0.44(0.25 －0.64)and for laparoscopic RYGB 0.16(.09 －.23).Mortality rates for ＞30 days for open RYGB 0.69(.03 －1.35)and for laparoscopic RYGB 0.09(0.00 －0.18)［27］.

Failure of RYGB

Despite these excellent results，5% －15%of patients will not achieve successful weight loss after gastric bypass，with“success”being defined as reaching and maintaining a BMI＜30 kg/m2or the resolution of co-morbidities［24］.Numerous factors can contribute to failed weight loss，including:the preoperative BMI(i.e.super-obese);anatomical problems such as a gastro-gastric fistula，large gastric pouch，or dilated gastrojejunal anastomosis;and non-compliance with exercise and dietary guidelines.Some patients seem to be predisposed to persistent obesity despite precise surgical technique and close follow-up.

Cost benefit ratio of RYGB in patients with comorbidities.

Higher BMIs are associated with an increased number and cost of medications，primarily for treatment of hypertension，diabetes and congestive heart disease［28］.One study demonstrated a significant reduction in number of medications and cost of medications for the treatment of obesity-related diabetes and hypertension after RYGBP［28］.Patients studied represent those with optimal management of the obesity and obesity-related co-morbidities preoperatively，and all patients were included in the study end-points.Even at a relatively short follow-up，this retrospective pilot study found that the pharmaceutical cost-savings are significant when compared favorably to the cost of RYGBP［28］.

Conclusion

In light of the above meta-analysis of the resolution of diabetes we are of the opinion that RYGB surgery is an effective means to deal with the problem and can successfully cure T2DM in select group of patients.LRYGB which is the more commonly practiced procedure nowadays has a more steeper learning curve but provides faster recovery for the patients in short term.More research studies are needed in non obese T2DM patients to fully evaluate the efficacy of RYGB in this group.

［1］ Wild S，Gojka R，Green A，et al.Global prevalence of diabetes-Estimates for the year 2000 and projections for 2030 ［J］.Diabetes Care，2004，27(5):1047 －1053.

［2］ Buchwald H，Avidor Y，Braunwald E，et al.Bariatric Surgery:A Systematic Review and meta-analysis［J］.JAMA，2004，292(14):1724 －1737.

［3］ Lee WJ，Wang W，Lee YC.Effect of Laparoscopic Mini-Gastric Bypass for Type 2 Diabetes Mellitus:Comparison of BMI ＞35 and ＜35 kg/m2［J］.Journal of Gastrointestinal Surgery，2008，12(5):945 －952.

［4］ Mokdad AH，Serdula MK The spread of the obesity epidemic in the United States，1991 － 1998［J］.JAMA，1999，282(16):1519 －1522.

［5］ Lee WJ，Wang W.Bariatric Surgery:Asia-Pacific Perspective［J］.Obesity Surgery，2005，15(6):751 －757.

［6］ Samuel Coffin，Chandana Konduru，Schwarz M Surgical Approaches for the Prevention and Treatment of Type 2 Diabetes Mellitus［J］.Cardiology in Review，2009，17(6):275－279

［7］ Yan E，Ko E，Luong V，et al.Long-term changes in weight lossand obesity-related comorbiditiesafter Roux-en-Y gastric bypass:a primary care experience［J］.The American Journal of Surgery，2008，195(1):94－98.

［8］ Bennett JC，Wang H，Schirmer BD.Quality of life and resolution of co-morbidities in super-obese patients remaining morbidly obese after Roux-en-Y gastric bypass［J］.Surgery for Obesity and Related Diseases，2007，3(3):387－391.

［9］ DiGiorgi M，Rosen DJ，Choi JJ，et al.Re-emergence of diabetes after gastric bypass in patients with mid-to long-term follow-up［J］.Surgery for Obesity and Related Diseases，2010，6(3):249 －253.

［10］ Dunkle-Blatter SE，St Jean MR，Whitehead C，et al.Outcomes among elderly bariatric patients at a high-volume center［J］.Surgery for Obesity and Related Diseases，2007，3(2):163 －170.

［11］ Rubino F.Is type 2 diabetes an operable intestinal disease?A provocative yet reasonable hypothesis［J］.Diabetes Care，2008，31(2):S290 － S296.

［12］ Schauer PR，Burguera B，Ikramuddin S，et al.Effect of Laparoscopic Roux-En Y Gastric Bypass on Type 2 Diabetes Mellitus［J］.Annals of Surgery，2003，238(4):467－485.

［13］ Sugerman HJ，Wolfe LG，Sica DA，et al.Diabetes and hypertension in severe obesity and effects of gastric bypass-induced weight loss［J］.Annals of Surgery，2003，237(6):751 －758.

［14］ Nguyen NT，Paya M，Stevens M，et al.The relationship between hospital volume and outcomes in bariatric surgery at academic medical centers［J］.Annals of Surgery，2004，240(4):586 －594.

［15］ Marinari GM，Murelli F，Cmerini G，et al.A 15-year evaluation of biliopancreatic diversion according to the Bariatric Analysis Reporting Outcome System(BAROS)［J］.Obesity Surgery，2004，14(3):325 －328.

［16］ Sonne-Holm S，Sorensen TIA，Christensen U.Risk of early death in extremely overweight young men［J］.British Medical Journal，1983，287(6395):795 －797.

［17］ Fontaine KR，Redden DT.Years of life lost due to obesity［J］.JAMA，2003，289(2):187 －193.

［18］ Narayan KMV，Boyle JP，Thompson TJ，et al.Lifetime risk for diabetes mellitus in the United States［J］.JAMA，2003，290(14):1884 －1890.

［19］ Varela JE，Hinojosa MW，Nguyen NT.Perioperative outcomes of bariatric surgery in adolescents compared with adults at academic medical centers［J］.Surgical Obesity Related Disorder，2007，3(5):537 －540.

［20］ Tsai WS，Inge TH，Burd RS.Bariatric surgery in adolescents:recent national trends in use and in-hospital outcome［J］.Archives of Pediatrics ＆ Adolescent Medicine，2007，161(3):217 －221.

［21］ Inge TH，Garcia V，Daniels S，et al.A Multidisciplinary Approach to the Adolescent Bariatric Surgical Patient［J］.J Pediatr Surg，2004，39(3):442 －447.

［22］ Jamal MK，DeMaria EJ，Johnson JM，et al.Impact of major co-morbidities on mortality and complications after gastric bypass［J］.Surgery for Obesity and Related Diseases，2005，1(6):511 －516.

［23］ Livingston EH，Huerta S.Male gender is a predictor of morbidity and age a predictor of mortality for patients undergoing gastric bypass surgery ［J］.Ann Surg 2002，236(5):576 －582.

［24］ Inabnet BW，Quinn T，Gagner M，et al.Laparoscopic Roux-en-Y Gastric Bypass in Patients with BMI＜50:A Prospective Randomized Trial Comparing Short and Long Limb Lengths［J］.Obesity Surgery，2005，15(1):51－57.

［25］ Bloomberg RD，Fleishman A，Nale JE，et al.Nutritional deficiencies following bariatric surgery:what have we learned?［J］.Obesity Surgery，2005，15(2):145 －154.

［26］ Cannon-Diehl R.Emerging Issues for the Postbariatric Surgical Patient［J］.Critical Care Nursing Quarterly 2010，33(4):361 －370.

［27］ Buchwald H，Estok R，Farbach K，et al.Trends in mortality in bariatric surgery:A systematic review and metaanalysis［J］.Surgery，2007，142(4):621 －635.

［28］ Potteiger CE，Paragi PR，Inverso NA，et al.Bariatric Surgery:Shedding the Monetary Weight of Prescription Costs in the Managed Care Arena［J］.Obesity Surgery，2004，14(6):725 －730.