Bariatric metabolic surgery

Summary

According to the WHO, the worldwide prevalence of obesity body mass index (BMI) ≥30 kg/m² nearly doubled between 1980 and 2008, with 10% of men and 14% of women and a total of more than half a billion adults (aged >20 years old) being classed as obese. At least 2.8 million people die each year worldwide as a result of being overweight or obese, usually from the inevitable related comorbidities. It has been reported that approximately 65% of the world’s population inhabits countries where overweight and obesity are responsible for higher mortality than underweight.

The recently published Interdisciplinary European Guidelines on Metabolic and Bariatric Surgery note that despite the WHO stating that excess weight is considered the fifth leading risk for deaths worldwide, it has not yet been possible to successfully curb the obesity epidemic. Moreover, severe obesity (BMI>35 kg/m²) represents a rapidly growing segment of the epidemic in which the negative effects on health and disability are especially marked.

Excess weight drastically elevates a person’s risk of developing a number of non-communicable diseases, such as diabetes, hypertension, stroke, dyslipidaemia, sleep apnoea, cancer, non-alcoholic steatohepatitis, and other serious comorbidities. The WHO emphasises that 44% of type 2 diabetes mellitus, 23% of ischaemic heart disease and around 7−41% of certain cancers are attributable to overweight and obesity. In the majority of European countries, overweight and obesity are responsible for about 80% of cases of type 2 diabetes, 35% of cases of ischaemic heart disease and 55% of cases of hypertensive disease among adults. Additionally, a range of debilitating conditions such as osteoarthritis, respiratory difficulties, gallbladder disease, infertility, and psychosocial problems, among others, which lead to reduced life expectancy, quality of life and disability, are extremely costly in terms of both absence from work and use of health resources. Noteworthy, the lifespan of severely obese individuals is decreased by an estimated 5–20 years depending on gender, age, and race.

Weight loss and maintenance are the obvious aims of bariatric surgery. The ideal bariatric operation should be easy and quick to perform, with minimal perioperative and long-term complications. It should result in excellent weight loss and indefinite weight maintenance in the vast majority of operated patients, necessitating minimal follow-up care, and be equally suitable for all patients, independent of individual characteristics such as compliance, age, and different degrees of overweight. It is immediately apparent that such a surgical procedure does not exist, and therefore we must carefully evaluate all the different features and characteristics of each operation and of each patient to try to select the procedure that provides the greatest probability of long-term success for a particular patient, with the best cost/benefit ratio. Bariatric surgery has proven to be the most effective mode of treatment for morbidly obese patients, with recent long-term studies providing evidence of a substantial reduction of mortality in bariatric surgery patients, as well as a decreased risk of developing new health-related comorbidities. Furthermore, a reduction in the use of healthcare services and therefore a reduction in direct healthcare costs was also observed. Bariatric surgery is an established and integral part of the comprehensive management of morbidly obese patients.

Keywords:
obesity – diabetes − surgical treatment of diabetes

Introduction

During the evolution of mankind, humans have generally lived in an environment where food was scarce and had to be procured by means of physical exertion. Therefore, owing to the necessity to survive in conditions of starvation, all mechanisms that favour energy saving and oppose energy consumption were selected, thus constituting a genetic tendency to weight gain. Conversely, any weight excess beyond the optimal physical shape for hunting or gathering reduced the probability of survival and reproduction. Body weight was therefore the result of a very careful balance between the genetic characteristics of humans and the environment, and the latter changed slowly enough to allow human genetics to adapt to it. As humans began to influence the environment around them, the speed of this change did not allow the possibility to adapt. For example, food became abundant and easy to obtain independent of any physical exertion. The condition of excessive weight not only ceased to constitute a difficulty for survival, but was in fact protected by the environment, which in the form of medical advances, reduced the impact of the subsequent morbidities. Under these conditions, there was evidently no defence against the development of obesity, which increased dramatically. As these conditions are an essential part of western civilisations, the more civilisation progresses, the more obesity prevalence increases, with no possibility of a cure or prevention. The worldwide prevalence of obesity (BMI≥30 kg/m²) nearly doubled between 1980 and 2008. In 2008, 10% of men and 14% of women were obese – a total of more than half a billion adults (aged >20 years old). The differences among countries (26% in the Americas and 3% in southeast Asia in 2008) have generally reduced as western civilisation progressively expands. At least 2.8 million people die each year worldwide as a result of being overweight or due to obesity repated co-morbidities [1].

Brief history of obesity surgery

Where medical treatment fails, there is often a possibility that surgery will succeed. The first attempt to treat obesity surgically was carried out by Victor Henriksson in Sweden in 1952, who resected 1 m of small bowel [2]. There was no weight loss, but this opened the way for other attempts. The jejunocolic shunt, in which the proximal jejunum was temporarily joined to the transverse colon to obtain a weight reduction [3], was followed by the jejunoileal bypass (JIB), a simple shortening of the small bowel obtained by anastomosing the end of the proximal jejunum to the side of the distal ileum, thus leaving a long intestinal blind loop. The jejunoileal bypass was pioneered by Howard Payne [4,5], and despite many severe complications, jejunoileal bypasses were executed in more than 100,000 cases in the 1970s. In 1980, Halverson demonstrated in a 5‑year follow-up study that besides complications, a substantial late weight regain occurred [6], and the operation was abandoned. Meanwhile, two new procedures had been proposed. The Roux-en-Y gastric bypass (RYGB); (Mason, 1967, Fig. 1) [7] was the first food-limiting procedure which, in its laparoscopic version (Wittgrove, 1994, LRYGB, Fig. 2) [8], became the most commonly performed operation in the 1990s. The biliopancreatic bypass subsequently re-named diversion (BPD; Scopinaro, 1979, Fig. 3) [9,10] was a malabsorptive operation free from all the issues of JIB. Both of these procedures are still in use. However, several procedures that were developed are no longer in use today such as the gastric “restrictive procedure” Gomez “horizontal gastroplasty” (1979, Fig. 9) [11], which was used only for a short amount of time, and to Mason vertical banded gastroplasty (VBG, 1980, Fig. 4A) [12], which was the most commonly performed obesity operation for more than 10 years but was eventually abandoned for late weight gain. In 1986, Kuzmak proposed the adjustable silicone gastric banding (ASGB or AGB) [13]. The non-adjustable version developed by Wilkinson in 1981 (Fig. 5) [14] had essentially been ignored for many years before, and this also occurred with the laparoscopic version, which was proposed by Fried in 1993 [15]. After a long period of very limited success (about 10 years), AGB, which was the first almost totally reversible operation, approached and even temporarily surpassed the popularity of LRYGB owing to the development of the laparoscopic technique (LAGB) in 1994 (Belachew, LAGB, 1994) [16]. This procedure even attracted surgeons who had never been interested in bariatric surgery previously. It was around the end of the last century that the term ‘obesity surgery’ was replaced by “bariatric (greek βαρύσ = heavy) surgery” in an attempt by obesity surgeons to change people’s perception of these surgical procedures and their use in the treatment of obesity.

In 1994, Hess conceived and later on published (1998) his BPD with sleeve gastrectomy and duodenal switch (BPD/DS), or simply duodenal switch (DS) [17], hich was pioneered by Marceau [18,19,20] and first performed laparoscopically by Gagner (2000) . This modification was obtained through vertical sleeve-shaped gastrectomy, with removal of at least 80% of the fundic–antral greater curve portion of the stomach, preservation of the pylorus, and lengthening of the common limb. The “American version of the BPD” obeys the same rules as the standard operation and has largely the same results and complications. It replaced the standard BPD completely in the USA, and partially in other countries, but did not increase the number of operations performed annually as a percentage of the total. The sum of the two versions remains around 5% [22].

In 2007, Gumbs reviewed the short-term weight loss results of sleeve gastrectomy (SG) when used as a two-stage procedure in superobese patients, with subsequent addition of BPD or RYGB [23].

Subsequently, laparoscopic sleeve gastrectomy (LSG) was proposed as a stand-alone procedure [24]. An advantage of LSG over LAGB is that a strict follow-up for the attainment of a good initial weight loss and maintenance is not required. Despite its irreversibility and the frequent occurrence of leaks, LSG was an immediate success and its use was widespread long before the 5‑year results were known. However, it did not completely replace LAGB largely because LAGB is reversible and (correctly) considered to be the safer operation. According to the most recently published data (Buchwald, 2013) [22], out of a total of 340,768 bariatric operations performed worldwide in 2011, 46.6% were RYGB, 27.9% SG, and 17.8% AGB. Global trends from 2003 to 2008 to 2011 showed a decline in RYGB (65.1% to 49.0% to 46.6%), an increase followed by a steep decline in AGB (24.4% to 42.3% to 17.8%), and a marked increase in SG (0.0% to 5.3% to 27.9%). Given the above trends, it is quite possible that SG has now reached or even surpassed the use of RYGB.

It is noteworthy that none of the aforementioned procedures were considered experimental and their use was not restricted to clinical studies until a minimum follow-up was reached. On the contrary, there are currently new procedures for which their use is restricted until their effectiveness and safety is fully proved, despite many reports of very successful series. Examples include the mini-bypass (MGB) and gastric plication (GP), which resembles SG but offers many advantages such as its safety profile and reversibility.

Finally, there were several obesity surgical procedures that had very limited use, such as stereotaxic hypothalamus surgery (Quaade, 1974) [25], truncal vagotomy (Kral, 1978) [26], gastric wrapping (Wilkinson, 1981) [14,27], and digestive adaptation (Santoro S, 2006) [28], among others. Other procedures are rarely performed owing to their experimental nature, such as different types of gastric electric stimulation (Cigaina, 2002 [29], (Prager, 2006) [30], the use of a intraluminal barrier, in which a short intraluminal sleeve anchored to the pylorus delays the meeting of food with biliopancreatic juices [31], and ileal interposition (De Paula, 2006) [32]. Although these procedures are biologically interesting, their importance in a clinical setting is limited at present.

Metabolic surgery

Metabolic surgery, a term used for the first time by Henry Buchwald over 50 years ago [33], can be considered as part of the more generally defined functional surgery, which consists of “a surgically induced anatomic modification which provokes either the reduction or the annulment of the altered function which causes the disease, or a functional change of opposite direction able to counteract partially or totally the originally altered function”[33]. If that function is a metabolic function then this constitutes metabolic surgery. Examples of functional surgery where the altered function is reduced or annulled include ablation of endocrine tumours, splenectomy for idiopathic thrombocytopenic purpura, and antrectomy or vagotomy for peptic disease. In all of these cases the surgically induced anatomic change simply reduces or annuls the altered function. An example of functional surgery where the altered function is counteracted is pyloroplasty with vagotomy, where the gastric emptying problems caused by vagotomy are counteracted by the facilitated emptying provoked by pyloroplasty. However, as we will see for metabolic surgery, not all surgery with functional effects can be defined as functional surgery. If a splenectomy is performed for a spleen injury, the effects are the same, but the aim of surgery was not to obtain those effects. The term functional surgery can only be used if the indication of the operation is to obtain the functional correction.

Bariatric surgery can be described as functional surgery, in which excessive food intake is reduced or annulled by gastric restriction procedures. Alternatively, the excessive food intake can be counteracted by performing procedures that reduce intestinal energy absorption. Bariatric surgery has many beneficial metabolic effects that are purely due to weight loss. Can obesity surgery be considered as metabolic surgery? Arguments in favour are as follows: 1) obesity can generally be considered a disorder of energy metabolism, which is partially or totally resolved by bariatric surgery; 2) even if they are are mediated by weight loss, the changes eventually obtained are metabolic; 3) bariatric surgery always causes a reduction in adipose tissue, which is one of the most metabolically active tissues. However, the case against is supported by at least two very good reasons. Firstly, bariatric surgery is primarily aimed at weight reduction, while the metabolic effects are only beneficial side effects secondary to weight loss – the greater the weight loss, the greater the metabolic effects, but these effects disappear if weight is regained. In this case if the metabolic effects were not obtained, the indication to surgery would remain unchanged, whilst true metabolic surgery should primarily be aimed at correction of the metabolic alteration, and work independently of weight changes. Secondly, and most importantly, the metabolic disturbances that may accompany obesity, such as hypercholesterolaemia, hypertriglyceridaemia, insulin resistance, and type 2 diabetes mellitus, can also occur in the absence of obesity, and in these cases, a weight-reducing operation would not only be ineffective, but also potentially harmful. A true metabolic operation should be able to resolve one or more of the above conditions independently of change in the body weight (i.e. even in lean patients) through specific mechanisms of action and without causing any undue weight loss. For example, if a RYGB or a BPD, which possess these specific actions, are performed with the aim of reducing body weight, a metabolic effect may also be obtained; however, in this case, a metabolic operation has not been performed, but However, RYGB or a BPD represented in such a case a bariatric operation with additional functional results that were in fact not necessary to the success of the operation. If on the contrary an operation is performed with the primary aim of obtaining a metabolic effect, this is metabolic surgery. Therefore, the type of operation is defined by its use and not the result obtained, unless it was the aim of the operation, and thus the indication itself. This is an important point around which misunderstandings can occur and care is needed to avoid them.

To better understand this, the definition of ‘diabetes surgery’ can be deliberated. Diabetes surgery is often erroneously defined, and can be considered as both functional and metabolic surgery. To help understand what can be defined as diabetes surgery, we need to first clarify a very important concept. In overweight and obese people a continuous fight exists between beta cells (whose principal function is the production of insulin, which favours the entrance of glucose into the muscular cell for energy production, thus maintaining a normal serum glucose level) and insulin resistance (which is mainly dependent on the amount of body fat, which opposes this function). In type 2 diabetes beta cells are damaged, so their capacity to increase insulin production in response to increasing insulin resistance is not unlimited. Hyperglycaemia, and thus frank diabetes, appears when the endocrine pancreas is unable to effectively counteract insulin resistance. Therefore, the damage that has occurred to the original beta cells is smaller the higher the patient’s BMI at which the diabetes manifested. Furthermore, the higher the patient’s BMI when diabetes developed, the greater the importance of insulin resistance in the diabetes pathogenesis. On the contrary, diabetes that manifests at low BMI values indicates that the beta cells are heavily damaged and cannot effectively counteract even the small insulin resistance caused by minimal fat mass. Consequently, while any operation that reduces weight, and thereby reduces insulin resistance, can result in an improvement or remission in diabetes in morbidly obese patients, the more the patient preoperative BMI decreases, the more a specific weight-independent beneficial action on the beta cell becomes the main, if not the only, way to effectively treat type 2 diabetes.

Pure gastric restriction procedures do not have a direct action on beta cells, but act indirectly enabling the insulin resistance to be reduced as a result of the weight loss that occurs following the procedure and decreasing the beta cell glycotoxicity and lipotoxicity. Conversely, RYGB and BPD, in particular, have been proven to have a weight-independent effect on beta cell function. It has been postulated that this is mainly due to GLP‑1 [34,35], a gut hormone with beneficial effects on beta cells released by the distal ileum, which is more easily reached by food after RYGB and where the stomach directly empties after BPD.

A person’s BMI will often define the type of surgery performed and the outcome required from the bariatric procedure. It is therefore important to consider the different physiological effects and efficacy of each procedure on the different BMI ranges to ensure optimal results. In the BMI range of 35 and above, all current bariatric procedures can be performed independent of their different effectiveness on diabetes remission and despite the remission rates as indicated by Buchwald in his meta-analyses [36,37]. In fact, diabetes remission in this particular population is considered a secondary effect of an operation whose primary goal in most cases is obesity treatment. Thus, the criteria for choosing the type of surgery should be dictated by the desired effect on obesity and not on diabetes.

For most patients in the BMI range of 30 to 35, the main aim of surgery continues to be obesity treatment, even if the patient opted for surgery primarily to treat their diabetes. However, it should be noted that procedures in this BMI group are only allowed in clinical trials at present. In this BMI range, weight loss alone cannot give the required result of diabetes remission; however, a reduction in body fat accompanied by good improvement in type 2 diabetes mellitus can be considered an acceptable outcome. At the 5‑year follow-up of a perspective study of BPD and RYGB [38,39,40], very good results for obesity resolution where observed for both operations. Patients who underwent BPD had better glycaemic control owing to a continued positive effect on beta cell function. Influence on glycaemic control balanced the side effects of the operation. The choice of which operation was performed was decided by the patient and based on the different lifestyles entailed by the two procedures.

As the underlying reason for surgery in the previous BMI ranges discussed was obesity treatment rather than diabetes remission, true diabetes surgery can be considered as surgery that is performed in the BMI range of 30 and below. For patients in this BMI range the principal and only aim of surgery is to treat their diabetes, which is generally severe. The only surgical option for this group of patients is BPD, as it is the only procedure that has been shown to be effective in markedly low BMI diabetic patients. Significant improvements in beta cell function have been observed for BPD, although improved function can often be slow to return. Furthermore, BPD is the only bariatric operation which, owing to its energy absorption threshold, causes weight loss only if there is excess weight to lose. Results obtained in patients in this BMI range appear to be acceptable with diabetes considered to be controlled in 60% of patients and a HbA1c of ≤7% at the 5‑year follow-up. Improvement was observed in all other patients, despite a mean diabetes duration of 14 years and 65% of patients undergoing preoperative insulin treatment. This would indicate that surgery is a viable option in this BMI group, but diabetes remission or improvement should be the only reason to perform surgery. In conclusion, the type of surgery performed is defined by the indication for surgery and the primary result desired.

All surgeries discussed previously are included in the general frame of functional surgery. Nevertheless, functional or metabolic effects can derive from other types of surgery with completely different indications, such as ablative surgery with no functional aims. Metabolic surgery is functional surgery with the primary goal of a change in metabolic function. Therefore, there is no specific history of metabolic surgery. Since the emergence of obesity surgery, whenever a metabolic result was the primary or the only indication to surgery for that specific patient this could be considered as metabolic surgery.

Bariatric operations can always be categorised as functional surgery but they are rarely classed as metabolic surgery, as their primary aim is generally weight reduction. True diabetes surgery, that is surgery performed for the sole purpose of controlling diabetes, is both functional and metabolic surgery. In patients in the high BMI range it is represented by exceptional cases where diabetes control is the primary goal of surgery. For the majority of patients in this BMI range it is simply bariatric surgery. In patients in the 30−35 BMI range, diabetes surgery is limited to the use of RYGB and BPD when the indication is primarily glycaemic control, otherwise again it is classed as bariatric surgery.

True diabetes surgery is generally only performed in patients in the BMI range of 30 and below and is still in the experimental phase. The purpose of this surgery is to beneficially act on beta cell function without causing substantial weight loss in this patient population, and it can be done with BPD or with new types of experimental operations, such as the duodenal–jejunal bypass [41]. This procedure involves the preservation of the entire stomach with the pylorus, the anastomosis of the proximal duodenal stump with the jejunum very close to the ligament of Treitz, and the Roux-Y anastomosis of the excluded duodenum–jejunum to the side of the proximal jejunum. The results for this procedure are very poor unless a sleeve gastrectomy is added [42], which questions the validity of the standard procedure. This was also the case for the De Paula ileal interposition [32]. De Paula used a more complicated version of this procedure (2010, Fig. 6) [43], in type 2 diabetic patients with low and very low BMIs (21−34); however, as the operation entails as many as 7 suture lines, the risk of operative complications was extremely high. The Tantalus™ (gastric electric stimulation) system [30], and the EndoBarrier™ [31] were also used for the treatment of type 2 diabetes in patients with low BMIs [44,45], with poor results.

Peculiar aspects of bariatric surgery

Even if only a small minority of obese patients seek surgery, the number who are eligible for surgery is practically endless. This deeply influences obesity surgery, in that our capacity to operate on patients is much smaller than the demand.

There are a variety of different surgical techniques in bariatric surgery with varying degrees of difficulty. The simplest techniques used today are AGB and SG, and the most complex are BPD–BPD/DS and RYGB, especially if the Fobi pouch technique [46] is used. As a general rule, the easier it is to perform the technique, the poorer the weight loss results [36,37].

A strict follow-up is required for AGB to obtain satisfactory weight loss results and for malabsorptive procedures (e.g. RYGB and BPD–BPD/DS) to avoid late complications, for example micronutrient deficiencies and protein malnutrition following BPD. Today, these complications are rare but if they are incorrectly treated or not treated they can lead to rehospitalisation with total parenteral nutrition (TPN) correction and even death. Needless to say, good patient compliance is mandatory for adequate follow-up. SG requires less follow-up to obtain good weight loss results and it does not cause any major malabsorption. However, follow-up needs should not influence the choice of the bariatric operation used.

The laparoscopic approach had an enormous impact on bariatric surgery for the following reasons:

1. It contributed enormously to the widespread use of bariatric surgery. The best example of this is LAGB, which attracted several surgeons from other fields, many of whom subsequently became obesity surgeons.
2. In many cases the feasibility of using a laparoscopic approach in an operation determines whether it is chosen by the surgeon or by the patient.
3. The emergence of the laparoscopic technique meant that several procedures that had largely been abandoned gained popularity once again as the results of the procedure and the adverse effects were greatly improved. Examples of this include the non-adjustable and adjustable versions of gastric banding. Improvements to techniques that then enabled them to be used laparoscopically also resulted in a resurgence in their use. An example of this is MacLean’s vertical banded gastroplasty. To avoid the occurrence of gastro-gastric fistulas, the use of stomach division rather than stapling (Fig. 4B) was conceived by the author and this technique was subsequently published in 1993 [47]. As this technique made the VBG laparoscopically feasible, it started to be used by surgeons across Europe just as it was being rapidly abandoned in the USA.
4. It led to the development of LRYGB, which is currently the bariatric operation of choice.

Choice of operation

Weight loss and maintenance are the obvious aims of bariatric surgery. Both are influenced by many variables which, in turn, influence each other. Variables include operation characteristics (e.g. surgical technique, adaption of the human anatomy to the procedure over time, and mechanism of action and its dependence on the original anatomy maintenance), surgeon characteristics (e.g. skill, experience, availability for a lifelong follow-up, and willingness to consider satisfaction of the patient’s needs as the primary goal of the operation) and, most importantly, patient characteristics (e.g. body weight, age, sociocultural factors, financial status, mental status, family support, compliance, geographic accessibility, and availability for an indefinite follow-up). All these variables should be carefully considered when choosing the type of bariatric surgery to use. Ideally, bariatric surgery should result in the weight loss required for each patient and maintain the attained body weight indefinitely. Unfortunately, the only bariatric operation able to obtain this result is BPD, and this procedure is very demanding both for the surgeon and the patient. The procedure has numerous side effects and the potentially life-threatening complications require very careful lifelong follow-up if they are to be prevented or successfully treated. BPD should be available in all centres, however, its use should be confined to patients whose needs are greatest and for whom the benefits of this procedure outweigh the associated risks and complications (i.e. superobese patients, diabetic patients, patients who have undergone one or more failed operations, and very young patients). The use of BPD in very young patients is controversial, however, there is little doubt that resolution of obesity for their entire life (excellent weight loss results are maintained for 50 years or more with this procedure) is worth the possible risks, which are generally limited to the first postoperative year. On the contrary, a 5−6‑year weight reduction with good quality of life, resolution of comorbidities, and low risk of late complications or rehospitalisation would be a satisfactory outcome in a 60‑year-old patient, even if slow weight regain is likely after the simpler, however safer procedures.

As previously mentioned, with obesity surgery it is often found that the easier the procedure is to perform technically, the poorer the initial weight loss results and maintenance of these results. This applies to LSG, which after the initial postoperative course when leaks are possible that are generally not life-threatening, provides good weight loss with improved quality of life, even if it may last for a limited time. AGB provides a similar outcome, although the quality of life is reduced by the lack of any postcibal or appetite-reduction effect. AGB acts as a simple mechanical obstacle to food intake, with frequent vomiting as a result. Consequently, as for gastroplasty, AGB should be considered as a purely gastric restrictive operation, while both SG and RYGB, which have the same mechanism of action, should be defined as food-limiting operations. As to the choice of operation, with the exception of a few cases where a BPD or a gastric restrictive operation is indicated, a good compromise is offered by an RYGB. This procedure, although technically more difficult, offers excellent weight loss at around 2 years to morbidly obese patients (BMI≥35 or 40 kg/m²) with one or more related comorbidities) followed by a slow weight regain. The weight regain is such that at around 12 years postoperation, the weight reduction is still very good [48], and with sufficient patient compliance, an acceptable weight reduction may be maintained over many years.

The ideal bariatric operation should be easy and quick to perform, with minimal perioperative and long-term complications, result in excellent weight loss and indefinite weight maintenance in the vast majority of patients, necessitating minimal follow-up care, and be equally suitable for all patients, independent of individual characteristics such as compliance, age, and different degrees of overweight. No such surgical procedure exists, and therefore the features (e.g. benefits, risks, and results) of each procedure and the characteristics of each patient must be evaluated in order to select the procedure that will have the best outcome for that particular patient, in terms of long-term resolution of obesity and any comorbidities, with the best cost/benefit ratio. Some surgeons try to avoid this difficult choice by offering to all patients an easy and safe operation that possesses all the aforementioned qualities. A more effective and complex, but less safe procedure like RYGB, BPD or BPD/DS is then undertaken if later on the first procedure fails – this is called sequential surgery. The proposed easy and safe first operation is the AGB, but this is rapidly shifting to LSG. This would be widely acceptable if the results obtained for AGB by Favretti [49] or Fielding [50] could be achieved universally. Unfortunately, this is not the case, and this probably because most surgeons are unable to devote the necessary amount of time required for follow-up, which is essential for the sustained success of AGB. An additional important consideration is the cost of these sequential procedures, which makes it impossible to provide this concept to several hundreds or thousands of obese patients.

Moreover, it has recently been proposed that surgical indications should be extended to patients with a BMI<35 in cases in which the presence of a serious comorbidity (e.g. type 2 diabetes, arterial hypertension, or severe dyslipidaemia) indicates that a clinical benefit would be obtained from surgical treatment. This proposal, may temporarily further increase the burden on health care systems.

Obesity-related comorbidities certainly represent an additional problem for indication to bariatric surgery. Whilst their improvement or disappearance as a consequence of weight loss is an additional reason for surgery, they undoubtedly increase the operative risk. Therefore, it is recommended that comorbidities are at least temporarily controlled as much as possible prior to surgery by medical therapy, if necessary obtaining a small but quick weight reduction. This especially applies to respiratory problems, such as Pickwickian syndrome, sleep aponea, and restrictive respiratory failure. Comorbidities unrelated to obesity may not constitute a contraindication to bariatric surgery providing that they can be temporarily controlled by medical treatment.

Rise and fall of bariatric operations

Good and sustained results, and the well-being of the patient are the most important criteria when choosing the type of bariatric operation to be performed. Obesity surgery began with an operation which, in Payne’s version, had the advantage of extreme simplicity, as it essentially consisted of an easy end-to-side enteroenterostomy. This simplicity certainly contributed to its widespread dissemination in the 1970s. Postoperative management was very demanding owing to the frequent and multiple side effects and complications, which were often lethal. JIB was eventually abandoned mainly owing to the poor late weight loss results. In the 1970s gastric bypasses were also developed, and in 1979, Miller created a new version of this technique, which was very similar in structure to the procedure that was currently in use (Fig. 7) but with one very important difference in that the narrow outlet was reinforced by as many as three layers of non-absorbable separate stitches. This made it a simple and good pure gastric restriction procedure; however, it still had the same limitations as all other pure gastric restriction procedures. Edward Mason himself thought that the increasingly technical complexity of the gastric bypass, especially following its modification by Griffen from loop to Roux-Y reconstruction [51], was not worth the results. In 1971 [52], he proposed the original model of horizontal gastroplasty, where the upper and lower stomach were separated by a division with oversewing, which left a small communication channel along the greater curvature (Fig. 8). This technique was adapted by Gomez (1979, Fig. 9) [11], who obtained the division using a linear stapler where the last metal stitches, on proximity of the greater curve, had been removed. Following this adaption, horizontal gastroplasty became an extremely easy procedure to perform, and had enormous but brief success. However, it took only a few months for the weight loss failure to become apparent owing to the enlargement of the non-reinforced outlet and the occurrence of gastro-gastric internal fistulas.

It was Edward Mason, often referred to as the ‘father of obesity surgery’, who solved this problem with his VBG (1980, Fig. 4A) [12]. This procedure was widely accepted and during the 1980s and the early 1990s, represented 80% of all the bariatric surgeries performed in the USA. However, poor 5‑year weight loss results were observed with VBG and it was Mason himself who first reported this in 1992. Gastroplasty was then rapidly abandoned in favour of RYGB, which, owing to the laparoscopic adaptation developed by Wittgrove and Clark in 1994 [8], became a much more effective procedure compared with its original version. In fact, the extremely small pouch volume (15−20 ml) and the wide non-reinforced gastro-enterostomy (GEA) resulted in quick gastric emptying with rapid intestinal transit, which, in turn, caused the occurrence of postcibal syndrome with food reaching the distal portion of the small bowel, where the production of the anorexigenic GLP‑1 and PYY gut hormones is stimulated. LRYGB required no mechanical obstacle, but resulted in appetite reduction and a pleasant feeling of satiation after small amounts of food, in conjunction with a good quality of life, low operative mortality and few early or late life-threatening complications (GEA leaks, stomal ulcer and GEA stenosis). The development of the laparoscopic approach, where narrow reinforced stoma would have been very difficult to create, resulted in the creation of the most successful bariatric operation to date. In the late 1990s and early 2000s this procedure represented nearly all obesity surgeries performed in the USA (around 100,000 procedures per year) and most of those performed in countries outside of the USA [22]. The widespread use of LRYGB is not because it is seen to be fashionable or because it is a simple and convenient operation to perform, but as a result of recognition of its qualities and its results. Furthermore, owing to these qualities, LRYGB has outlasted and become more successful than procedures such as LABG and LSG.

Non-adjustable GB was first conceived by Wilkinson in 1978 [14], and this was followed by a laparoscopic version developed by Fried in 1993 [15]. However, this procedure was barely acknowledged by the bariatric surgical world. A similar reaction was seen in 1986 when Kuzmak proposed his AGB [13], which allowed almost total anatomic reversibility of the procedure. The real explosion of AGB followed the creation of the laparoscopic approach by Belachew, 1993 [16]. The reasons for this are threefold: 1) LAGB was the first truly reversible bariatric procedure where laparoscopy was feasible; 2) the laparoscopic approach was initially confined to cholecystectomy, however, the development of LAGB represented the first occasion of advanced laparoscopic surgery, thus attracting many surgeons who had never been interested in bariatric surgery before, 3) the industry realised that LAGB entailed the use of a device that had to be manufactured, promoted, and sold. For these reasons, in addition to excellent promotion, no other bariatric operation has ever had the success of the adjustable silicon gastric band or, as it is often known, the Lap-Band^®. The success of the Lap-Band and its dominance in obesity surgery was seen in France. In 1993, there was practically no obesity surgery at all in France. However, in 1997, more than 1000 French surgeons had placed at least one Lap-Band [53], and the operation became the most performed one in Europe. In 2001, the US Food and Drugs Administration (FDA) approved the device, and within a few years, the Lap-Band became the most widely used procedure in the USA as well. In 2008, the LAGB was the most commonly performed bariatric operation worldwide, and surpassed the LRYGB (42% vs 40%) [54]. In Europe, however, the use of LAGB was diminishing. But then in February 2011, the FDA approved the use of the Lap-Band in mildly obese patients (BMI 30−35) with one comorbidity, and therefore an additional 34 million people who had diabetes or simple dyslipidaemia would have access to Lap-Band, and only to Lap-Band, for surgical treatment. The Lap-Band had a high rate of failure (owing to band removal for obstinate vomiting, intragastric band migration, and other side effects and complications, or, more generally, bad quality of life and poor weight loss results), and its reversibility enabled it to be converted into any other bariatric operation. This made LAGB the ideal operation for sequential surgery.

The main problem with LAGB is that it requires a very tight follow-up to obtain good and sustained weight loss while avoiding side effects and complications. This has been proved by the excellent results reported by the few surgeons who have been able to devote the required amount of time to follow-up [49,50]. However, the amount of time required for successful follow-up is usually not possible for surgeons performing a large number of operations (this was, in my opinion, the main reason for the creation of multidisciplinary teams). Follow-up has always been the Achilles’ heel of bariatric surgeons, and a solution was eventually offered by the sleeve gastrectomy.

LSG as a stand-alone procedure (Fig. 10) probably has the same mechanism of action as RYGB, with the addition of a sharp decreased ghrelin production due to the removal of most of the ghrelin-producing tissue. The weight loss results are inferior, however, the operation had immediate enormous and rapid success, long before the 5‑year results were known. The obvious reason for this is that SG, along with AGB, has no major long-term deficiencies, and does not require a strict follow-up for weight loss result attainment and maintenance. Once the early risk of postoperative leak is overcome, the limited follow-up required means LSG is a very popular procedure. The only problem remains late weight regain, which is unavoidable as the operation is by definition a transitory one. In 2014, Diamantis [55] published a review of 16 studies, including nearly 500 long-term LSG patients, which showed that the mean loss of the initial excess weight decreased from 62% at 5 years to 43% at 7 years. However, failed LSG can be converted to many different operations, from simple re-sleeve in cases of sleeve dilation, to AGB, RYGB, and BPD to form what LSG had been originally conceived for, that is a DS. In contrast with AGB patients, SG patients are rather reluctant to be converted to other procedures, since weight loss is acceptable, comorbidities are generally resolved, and quality of life is excellent. As LSG provides satisfaction to both the patient and surgeon, it can be foreseen that it will continue to be widely performed for a long while. Gastric plication was first proposed by Tretbar in 1976 with no success [56]. In 2007, it was revived by Talebpour (Fig. 11) [57], followed by Bradnova (in 2014) [58], who demonstrated that gastric plication produced metabolic effects. Gastric plication seems to reproduce the weight loss results of SG, with the advantages of no gastric resection, fewer operative complications, and full anatomical reversibility. A large series was reported by Fried in 2012 [59], and a systematic review was published by Ji in 2013 [60], encompassing 14 studies and 1450 LGP patients in a 6‑ to 24‑month follow-up. Gastric obstruction seems to be the most common cause for readmission and reoperation (4.6%). The still limited number of cases, together with the absence of 5‑year results, means that LGP is still an investigational procedure.

The final procedure to be discussed is MGB, which should be more correctly named single anastomosis GB or loop GB. Pioneered by Rutledge in 2001 [61], MGB consists of a regular GB, where the pouch created is much longer than usual (to prevent oesophageal reflux), and the Roux‑Y anastomosis is replaced by a loop one, with the aim of sparing one enteroenterostomy (Fig. 12). The operation was severely criticised in the USA, but is being increasingly performed in many other countries, where the authors generally report the same weight loss results as those obtained from a standard RYGB, but with fewer complications and a shorter hospital stay. Interestingly, with the exception of Sanchez-Pernaute (2008) [62], none of the surgeons who have performed the MGB (including the originator) seem to have noted that the operation is a BPD, already published as such by Scopinaro in 1980 [63] as one of the possible versions of the standard BPD. In fact, all procedures that delay the meeting of food with biliopancreatic juices should be considered to be BPDs, including RYGB. The presence of energy malabsorption depends on the length of the common limb, which, in classic RYGB, owing to the very short biliopancreatic limb, can be assumed to be long enough not to cause any reduction of energy absorption. Almost no surgeons measure the small bowel when it is fully stretched, and in cases where the small bowel is short (e.g. 400 cm, which occurs relatively frequently), the 200 cm that Rutledge and most of the other authors measure distal to the ligament of Treitz (ignoring what is left distal to the GEA) could very well leave them with a common limb, which in the MGB is also the alimentary limb and whose length is not known to the surgeon, which is short enough to cause severe energy and protein malnutrition. This explains the high number of cases of protein malnutrition reported following RYGB, which would probably be even higher with MGB if the number of operations performed was greater.

Conclusions

Bariatric and metabolic surgery are relatively new disciplines and a huge number of different procedures have been proposed, each of them with a large number of modifications. Widespread application of some of these procedures has occurred, often before their efficiency has proved to be acceptable and the complication rates determined. However, some procedures have passed the test of time, are sustainable and their primary treatment role has remained intact.

Abbreviations

BMI – Body Mass Index

WHO – World Health Organisation

RYGB – Roux-en-Y Gastric Bypass

LRYGB – Laparoscopic Roux-en-Y Gastric Bypass

BPD – Biliopancreatic Diversion

VBG - Vertical Banded Gastroplasty

AGB – Adjustable Gastric Band

LAGB – Laparoscopic Adjustable Gastric Band

BPD/DS – Biliopancreatic Diversion – Duodenal Switch

SG – Sleeve Gastrectomy

LSG – Laparoscopic Sleeve Gastrectomy

MGB – “Mini Gastric Bypass”

GP – Gastric Plication

LGP – Laparoscopic Gastric Plication

GLP 1 – Glucagon-like Peptide 1

TPN – Total Parenteral Nutrition

VBG – Vertical Banded Gastroplasty

GEA – Gastro-enterostomy

FDA – Food and Drugs Administration

Prof. Nicola Scopinaro, FACS (Hon)

University of Genoa,

Italy