Bariatric and metabolic surgery has become the most effective treatment for achieving stable weight loss and metabolic improvements in morbidly obese patients. This reviews summarizes the most commonly performed bariatric procedures used today, their indications and contraindications and provides recently published data on the metabolic outcome in type 2 diabetes mellitus patients.
Indications for bariatric surgery
Today, health care systems all over the world are faced with the rapid increase in the prevalence of obesity and morbid obesity. One out of every five American dollars spent on health care in the USA is for the treatment of obesity;1 the steady increase in average body mass index (BMI) over the last decade, observed globally, is contributing to a health economics problem of immense dimensions.
As lifestyle modifications or medical treatment fail to achieve stable weight loss success,2,3 the demand for bariatric surgery is rising. Thus, the number of bariatric procedures performed annually worldwide increased from 146 301 in 2003 to 344 221 in 2008.4 From 2008 to 2011 the total number of procedures slightly decreased to 340 768, indicating that, at that time, the numeric limit for bariatric surgery may have been reached. Bariatric surgery can never be carried out on all morbidly obese patients who fail to achieve weight loss success; therefore, a more appropriate criterion to select patients for surgery should be related to metabolic improvements or prevention of microvascular and macrovascular type 2 diabetes mellitus complications, more than for weight loss itself.
The guidelines for the application of bariatric surgery, published back in 1991 by the US National Institutes of Health (NIH), recommend surgery for patients with a BMI of 40 kg/m2 or greater, as well as surgery for patients with a BMI of 35 kg/m2 or greater suffering from obesity-associated comorbidities, such as hypertension and type 2 diabetes mellitus.5 Despite the fact that new insights into the metabolic effects of subcutaneous and visceral fat distribution have been discovered, these dated recommendations remain the basis for all patient selection criteria in bariatric surgery.
While a selection criterion based on BMI has the advantage of being generally accepted and easy to calculate, a selection criterion based on the more sophisticated waist-to-height ratio would enable patients with particular visceral obesity to undergo bariatric surgery, even at a lower BMI. Using long-term data from the Swedish Obese Subjects (SOS) study, Sjöström et al.6 demonstrated that bariatric surgery performed early in patients with type 2 diabetes mellitus significantly improved prevention of microvascular and macrovascular complications in the long-term follow-up.
Visceral obesity significantly increases the risk of type 2 diabetes mellitus, as demonstrated by the Nurses’ Health Study,7 which followed a cohort of 43 581 US women between 1986 and 1994. The risk of developing type 2 diabetes mellitus correlates to waist circumference, defining abdominal or visceral obesity by a waist circumference of 102 cm or greater in men and 88 cm or greater in women.8 However, today, the NIH guidelines as well as the European guidelines for bariatric surgery9 are still strictly based on BMI alone.
Despite circumstances and comorbidities excluding patients from all kinds of bariatric interventions, such as pregnancy, severe psychiatric disorders or the simple refusal to participate in a life-long follow-up, some special factors might exclude bariatric candidates from a specific procedure. For example, a bus driver should not undergo Roux-en-Y gastric bypass (RYGB) because of the high incidence of post-operative hyperinsulinaemic hypoglycaemia.10 As some changes in taste are observed after RYGB,11 if the patient has a profession that depends on this sense, such as chefs or sommeliers, this might act as a contraindication for this kind of procedure. These procedure-specific contraindications also exist for sleeve gastrectomy as patients diagnosed with Barrett’s disease should not be considered for sleeve gastrectomy.
At present, worldwide, RYGB and sleeve gastrectomy are the most commonly performed bariatric procedures,12 in addition to laparoscopic adjustable gastric banding (LAGB) and biliopancreatic diversion (BPD). Omega loop bypass (OLB), also known as ‘mini-gastric bypass’ or ‘one-anastomosis gastric bypass’, is a relatively new procedure, which already makes up 4.3% of all bariatric procedures performed in Europe in 2011.12
For the last decade, the weight loss mechanisms of bariatric surgery were seen to act via two simple pathways. ‘Restriction’, achieved by the surgical reduction of the gastric volume to a small gastric pouch, limits food uptake and therefore also the amount of food entering the intestine. The reduced number of calories taken in contributes to a negative calorie balance. ‘Malabsorption’, achieved by the surgical limitation of the functional absorptive intestinal length, also leads to the absorption of fewer calories from the intestine. LAGB is based strictly on restriction, while BPD predominantly acts by malabsorption. Some bariatric procedures, like the gastric bypass and the duodenal switch, combine these two major principles to achieve weight loss.
In addition to the simplified dualism of restriction and malabsorption, and the already known effects of numerous gut hormones13,14 interfering in the regulation of satiety and hunger, a variety of new weight loss effectors have been found by recent bariatric research focusing on bile salts,15–17 the alteration of the gut flora18,19 and changes in taste or food preference.20
Different bariatric procedures such as gastric bypass, gastric banding and BPD result in a wide variability in excessive weight loss or type 2 diabetes mellitus remission rates. For all kinds of procedures, the extent of type 2 diabetes mellitus remission is also influenced by the duration of type 2 diabetes mellitus prior to surgery and the type of medication (insulin vs. medical treatment) and patients’ age.21
Remission rates of type 2 diabetes mellitus for gastric bypass, gastric banding and BPD are 80.3%, 56.7% and 95.1%, respectively.22 Furthermore, bariatric surgery can even prevent the development of type 2 diabetes mellitus in the long-term follow-up. Analysing the 15-year follow-up of the SOS study, Carlsson et al.23 observed a reduction of the long-term incidence of type 2 diabetes mellitus by 78% in the surgery group, of which the procedures predominantly consisted of the, nowadays abandoned, vertical banded gastroplasty, in addition to LAGB and RYGB.
Metabolic surgery can lead to substantial improvements in obesity-associated comorbidities, such as hypertension, hypercholesterolaemia, hypertriglyceridaemia or hyperuricaemia, summarized within the metabolic syndrome. In their 2004 meta-analysis, which included a total of 22 094 patients, Buchwald et al.24 found that hyperlipidaemia improved in 76% of patients, hypertension resolved in 62% and obstructive sleep apnoea resolved in 85% of patients.
Roux-en-Y gastric bypass
With more than 150 000 procedures performed worldwide annually, RYGB, which was first performed laparoscopically in 1993 by Wittgrove et al.,25 still represents the most commonly performed bariatric procedure, based on data from 2011 by Buchwald et al.12
Roux-en-Y gastric bypass combines two simplified mechanisms of weight loss surgery – restriction and mild malabsorption. Restriction is achieved by forming a small gastric pouch, which is connected to an alimentary limb of 100–150 cm length. The biliopancreatic limb is anastomosed to the alimentary limb in a Roux-en-Y fashion. Mild malabsorption is based on the exclusion of the biliopancreatic limb from the intestinal nutrient absorption and the limitation of fat uptake to the common limb.
In addition to restriction and mild malabsorption, a variety of gut hormones also play a major role in weight loss. Metabolic improvements observed after RYGB are caused by altered secretion patterns of gut hormones, such as peptide tyrosine tyrosine,13,26 glucagon-like peptide-113,27 and ghrelin.28,29
A well-known side-effect of RYGB is post-prandial dumping syndrome. In a recent study published by Kefurt et al.,10 post-operative hyperinsulinaemic hypoglycaemia following RYGB was diagnosed in up to 75% of the patients by continuous glucose monitoring; a frequency higher than expected. While the uptake of simple carbohydrates, which can be seen as non-adherence to dietary recommendations, can trigger hyperinsulinaemic hypoglycaemia, nocturnal hypoglycaemic episodes were also found in 38% of patients. Hypoglycaemia might also play a role in weight regain in the longer follow-up after RYGB, as patients change their eating behaviour to counteract blood glucose fluctuations.30 Severe neuroglucopenia might also lead to non-specific symptoms such as confusion, syncope and seizures and might therefore contribute to the number of unexplained accidental deaths following RYGB.31 Therefore, bus drivers, in particular, should not be offered RYGB.
In 2009, a type 2 diabetes mellitus remission rate of 80.3% after gastric bypass was presented in a meta-analysis carried out by Buchwald et al.22 In a recent randomized prospective study comparing gastric bypass, BPD and medical type 2 diabetes mellitus treatment, Mingrone et al.3 observed type 2 diabetes mellitus remission, defined as a fasting glucose level of less than 100 mg per decilitre (5.6 mmol per litre) and glycated haemoglobin (HbA1c) levels of < 6.5%, in 74% of type 2 diabetes mellitus gastric bypass patients in the 2 years following surgery. All 20 RYGB patients were able to discontinue their type 2 diabetes mellitus medication within 15 days of surgery. Another randomized controlled study published recently by Schauer et al.2 compared medical type 2 diabetes mellitus therapy with gastric bypass or sleeve gastrectomy. In this study, 42% of the 50 gastric bypass patients reached the goal of having HbA1c of 6.0% or less within 12 months, while mean HbA1c decreased from 9.3% to 6.4%. After surgery, only 4% of the patients were still on insulin compared with 44% before surgery. A mean body weight loss of 27% was observed in the gastric bypass group. The 3-year update32 of the Systemic Therapy in Advancing or Metastatic Prostate Cancer: Evaluation of Drug Efficacy (STAMPEDE) trial confirmed the superiority of gastric bypass over sleeve gastrectomy in type 2 diabetes mellitus patients, as gastric bypass patients required only the half of the glucose-lowering medications per day, compared with sleeve gastrectomy patients, combined with a significantly higher reduction of body weight in the gastric bypass group.
Patients with type 2 diabetes mellitus should not wait to undergo gastric bypass surgery, because of two recent findings: (1) Sjöström et al.6 demonstrated that bariatric surgery performed within 1 year of type 2 diabetes mellitus diagnosis significantly improved prevention of microvascular and macrovascular complications in the long-term follow-up than bariatric surgery performed later after type 2 diabetes mellitus diagnosis; (2) Arterburn et al.21 found complete remission of type 2 diabetes mellitus within 5 years of surgery in 68% of gastric bypass patients. However, with a median duration of type 2 diabetes mellitus remission of 8.3 years, 35% of these patients redeveloped type 2 diabetes mellitus within 5 years.21 Predictors of type 2 diabetes mellitus relapse were longer duration of type 2 diabetes mellitus prior to surgery, insulin use and poor glycaemic control prior to surgery.21
Omega loop gastric bypass
In OLB, an antecolic gastrojejunal anastomosis is established between a long and narrow gastric pouch and a jejunal limb at approximately 200 cm from the ligament of Treitz. This procedure was initially promoted by Robert Rutledge, who presented promising short-33 and mid-term34 follow-up for weight loss success, combined with a short operative time of 37 minutes (mean) in a group of 1274 patients.33 To prevent biliary reflux, Garcia-Caballero et al.35 further introduced antireflux stitches from the biliopancreatic loop to the left part of the gastric pouch.
The long and narrow pouch reduces possible tension on the gastrojejunostomy, which might explain the low leakage rates at the level of the anastomosis published so far.36,37 The long and narrow pouch also prevents pouch enlargement, biliary reflux to the gastro-oesophageal junction and slows down the food passage to the small bowel, which might contribute to an improved feeling of satiety and less hypoglycaemia. With only one anastomosis to be established, the operative time is shorter than RYGB.38 While an incidence of up to 5%39,40 is reported for internal herniation following RYGB, this risk might even be lower in OLB.
Several studies have been published recently, each more than 1000 patients, that include a follow-up of 7 years or more after OLB.36,37,41,42 In these studies, leakage rates at the gastrojejunal anastomosis were found to range from 0.3% to 0.6%, marginal ulcers occurred in approximately 1.7% and biliary reflux was observed in only 0.4% to 0.9% of the patients.
Disse et al.43 recently published a comparative study on short-time weight loss following OLB versus RYGB. With 89% versus 71%, they found higher excess BMI loss (EBL%) in the OLB group at 1 year after surgery in this matched non-randomized study of 20 OLB patients and 60 RYGB patients, but further prospective randomized studies are needed to assess any superiority of one procedure over the other.
In addition to primary OLB surgery, this bariatric procedure can be applied to revisional surgery in case of weight loss failure following LAGB36 or sleeve gastrectomy.44 In these cases, pouch or sleeve enlargement are to be expected, so a careful narrow pouch formation has to be achieved to prevent biliary reflux.
Laparoscopic adjustable gastric banding
In LAGB a silicone band is wrapped around the upper part of the stomach, creating a small gastric pouch above the band. LAGB, based strictly on restriction, represents the least invasive bariatric procedure, as no transection or stapling of the stomach or gut exclusion from the nutrient absorption is made. Furthermore, LAGB demonstrates the lowest morbidity and mortality rates of all bariatric procedures performed today.
In addition to this, other advantages of gastric banding are its individual adjustability to the demands of the patient and the reversibility of the procedure. Complications of LAGB consist of band slipping, band erosion, oesophageal dilatation or frequent vomiting as a result of food intolerance or tightly adjusted bands.45 Compared with other bariatric procedures, weight loss24 and alteration of gut hormones46 contributing to metabolic effects are less frequent after LAGB.
While LAGB provided promising short-term weight loss success at the beginning of laparoscopic bariatric surgery,47,48 poor long-term results, for both the incidence of band-related complications45 and weight regain,49 explain the rapid decrease in the number of LAGBs performed in both the USA and Europe, as published by Buchwald and Oien in 2013.12 In cases of weight loss failure, surgical treatment consists of the conversion from band to RYGB45,50 or OLB.36
Sleeve gastrectomy has gained enormous popularity among bariatric surgeons within the last few years. Initially developed as part of the biliopancreatic diversion with duodenal switch, sleeve gastrectomy was introduced as a sole bariatric procedure as the first part of a two-step laparoscopic gastric bypass in ‘super-super obese’ (i.e. BMI > 60 kg/m2) patients.51 Owing to promising short-time weight loss success in these cases, sleeve gastrectomy was further applied as a definite procedure for morbidly obese patients. In 2008, sleeve gastrectomies represented 5.4% of all bariatric procedures performed worldwide.4 From 2008 to 2011 the total number of sleeve gastrectomies performed increased tremendously from 18 098 to 94 689, and the case numbers are expected to increase further over the next few years. In the Asia–Pacific region, sleeve gastrectomies already outnumber RYGBs in 2011; 55% of all bariatric procedures performed.
This popularity is mainly caused by the surgical simplicity of sleeve gastrectomy compared with the surgically more demanding RYGB, and by a wide range of procedure-related advantages: (1) the preserved integrity of the pyloric region might prevent dumping syndrome from developing and (2) the biliary tract remains accessible for endoscopy or endoscopic retrograde cholangiopancreatography (ERCP). Furthermore, fewer nutrient deficiencies are observed after sleeve gastrectomy than after RYGB.52,53 The other main factor promoting sleeve gastrectomy as the sole bariatric procedure was excellent short-term results for weight loss,54–56 initially comparable to gastric bypass.
As dumping syndrome caused by post-operative hyperinsulinaemic hypoglycaemia is a common observation after RYGB,10 this might serve as an argument to choose sleeve gastrectomy instead of RYGB as the bariatric procedure. However, dumping has also been observed after sleeve gastrectomy. In two series published recently by Papamargaritis et al.57 and Tzovaras et al.,58 dumping syndrome was found in 29% and 40% of sleeve gastrectomy patients, respectively. Hypoglycaemia, confirmed by an oral glucose tolerance test (OGTT), was also found in 33% of patients at 6 months after a sleeve gastrectomy, but the mechanisms leading to hypoglycaemia after sleeve gastrectomy might be very different from those in RYGB.
Nutritional deficiencies are a common observation in morbidly obese patients and following bariatric surgery. Owing to a more significantly altered gastrointestinal tract, gastric bypass is assumed to result in a much higher incidence of macro- and micronutrient deficiency than sleeve gastrectomy. In a recent cross-sectional long-term survey on micronutrient deficiencies, Alexandrou et al.59 found vitamin B12 deficiency to be significantly higher in only RYGB patients, while no significant difference was found for iron or folate deficiency or anaemia.
In addition to excellent short-term results for weight loss,54–56 several studies provide data on significant weight regain in the mid- or long-term follow-up after sleeve gastrectomy. Compared with an EWL at 3 years of 73%, Himpens et al.60 observed a significant weight regain to an EWL of 57% at 6 years after surgery, which corresponds to the 5-year EWL data presented recently by Alexandrou et al.,61 or cumulative data of the 4th International Consensus Summit on Sleeve Gastrectomy, published by Gagner et al.62 For a follow-up of up to 7 years, Diamantis et al.63 found a further EWL decrease to a mean of 43% for a total number of 133 patients. Thus, it is not proven, so far, that sleeve gastrectomy actually results in stable long-term weight loss success.
In the case of weight regain, conversion from sleeve gastrectomy to RYGB64 or duodenal switch65 are the procedures of choice. Furthermore, a conversion from sleeve gastrectomy to an OLB can be performed44 but for this approach a reshaping of the enlarged sleeve should be mandatory to avoid biliary reflux.
Conversion from sleeve gastrectomy to gastric bypass has been performed in only a small number of patients64 because of intractable symptomatic gastro-oesophageal reflux (GERD), but the incidence of relevant GERD following sleeve gastrectomy might be underestimated among bariatric surgeons. Himpens et al.60 found relevant de novo GERD, more than 6 years after surgery, in one out of five sleeve gastrectomy patients, which corresponds to the 5-year observations of Bohdjalian et al.66 Owing to these findings, sleeve gastrectomy should not be offered to patients diagnosed with Barrett’s mucosa, found in the pre-operative gastroscopy.
Leaks at the proximal staple line remain the complication most difficult to treat following sleeve gastrectomy. While a median leak rate of 1.1% was published within the cumulative data at the 4th International Consensus Summit on Sleeve Gastrectomy,62 Aurora et al.67 found a mean leak rate of 2.4%, pooling data from 29 studies (a total of 4888 patients). In this study, a higher leakage rate was observed in super-obese patients and when a bougie of less than 40 French (Fr) was used for sleeve calibration, while staple height and the use of buttressing material did not affect the leak rate. Treatment of proximal staple line leaks often requires multiple placement of self-expanding stents68 and can end up with total gastrectomy with an oesophagojejunal reconstruction.69
In 2008, Vidal et al.70 presented same type 2 diabetes mellitus remission rates (84.6% vs. 84.6%). for gastric bypass and sleeve gastrectomy in a non-randomized, prospective 12-month study.
A more comprehensive view on this topic is provided by the prospective randomized STAMPEDE trial, comparing intensive medical therapy, sleeve gastrectomy and gastric bypass, including 50 patients for each treatment modality. Schauer et al.2 found that 37% of the sleeve gastrectomy patients reached HbA1c levels of less than 6.0% within 1 year, while the mean HbA1c dropped from 9.5% to 6.6%. Those who used insulin at baseline encompassed 44% of all patients, while only 8% of patients were on insulin for 1 year. Overall, these results seem to be comparable to gastric bypass, but patients needed a much higher quantity of oral antidiabetic medication to reach their goals. These 1-year results were further confirmed by the recent 3-year update32 of the trial.
Biliopancreatic diversion was developed by Scopinaro et al.71 in 1976, combining a distal gastrectomy with a remnant gastric pouch of 200–500 ml with a distal bypass resulting in a very short common limb length of 50–75 cm. After division of the jejunum 250 cm proximal to the ileocaecal valve, the distal part of the jejunum is anastomosed to the stomach and the proximal part of the jejunum is anastomosed to the ileum 50–75 cm proximal to the ileocaecal valve. The limitation of the alimentary limb together with the very short common limb results in the malabsorption, or ‘limited absorption’ according to Scopinaro et al.71
In the long-term follow-up after BPD, an extremely stable weight loss, with nearly no weight regain, is observed.72 Furthermore, the metabolic effects of BPD are impressive. In 2005, Scopinaro et al.73 published a 10-year recovery rate of hyperglycaemia of 97% after BPD, and in a randomized prospective study published in 2012 by Mingrone et al.,3 BPD patients with a remarkable long duration of type 2 diabetes mellitus (at least 5 years) were able to discontinue their medication within 15 days of surgery. At 2 years, type 2 diabetes mellitus remission was observed in 95% of the patients, together with a decrease of HbA1c from 8.6% to a mean of 4.9%.
Among all bariatric procedures, BPD achieves the highest type 2 diabetes mellitus remission rates and the most stable weight loss, but because of its malabsorptive effect diarrhoea, anaemia, protein deficiency, iron deficiency and flatulence74 are common complications that limit its application to a larger number of patients. In particular, protein deficiency can trigger chronic liver failure, sometimes even requiring liver transplantation.75 Thus, adherence to follow-up examinations and compliance to the supplementation are as important as careful patient selection for BPD.
While in Europe the numbers of OLB in 2011 already represented 4.3% of all bariatric procedures performed, the BPD (together with the duodenal switch) lost from 4.9% to 2.1% in from 2008 to 2011. Thus, long-limb OLB might replace BPD in the near future.
Weight loss mechanisms in bariatric surgery consist of more than restriction or malnutrition as additional effectors were found in alterations of gut hormones,13 bile acids,16 gut microbiota,18 taste and food preference.20 To prevent microvascular and macrovascular type 2 diabetes mellitus complications in the long-term follow-up, bariatric surgery should be performed early in patients with type 2 diabetes mellitus, preferably within 1 year of type 2 diabetes mellitus diagnosis.6 Hyperinsulinaemic hypoglycaemia, assessed by continuous glucose monitoring, is more common after RYGB than expected.10 Gastric banding has completely lost its popularity because of poor long-term weight loss results. Sleeve gastrectomy has already outnumbered RYGB in some countries,12 but frequent redo bariatric surgery due to weight regain is to be expected.63 OLB might become the next procedure of choice because of excellent weight loss success and low incidence of biliary reflux.37 Prospective randomized trials have proven the superiority of bariatric surgery over intensive medical treatment in type 2 diabetes mellitus,2,3,32 while RYGB provides more control of type 2 diabetes mellitus than sleeve gastrectomy.32