Current status of mini-gastric bypass.

Mini-gastric bypass (MGP) is a promising bariatric procedure. Tens of thousands of this procedure have been performed throughout the world since Rutledge performed the first procedure in the United States of America in 1997. Several thousands of these have even been documented in the published scientific literature. Despite a proven track record over nearly two decades, this operation continues to polarise the bariatric community. A large number of surgeons across the world have strong objections to this procedure and do not perform it. The risk of symptomatic (bile) reflux, marginal ulceration, severe malnutrition, and long-term risk of gastric and oesophageal cancers are some of the commonly voiced concerns. Despite these expressed fears, several advantages such as technical simplicity, shorter learning curve, ease of revision and reversal, non-inferior weight loss and comorbidity resolution outcomes have prompted some surgeons to advocate a wider adoption of this procedure. This review examines the current status of these controversial aspects in the light of the published academic literature in English.

BACKGROUND

Rutledge published his initial experience with 1,274 mini-gastric bypasses (MGBs) in 2001.[1] Over the ensuing years, a number of surgeons from around the world have published extensive experience with this operation in both primary and revisional settings.[2345678910] However, most of the experience with this procedure and publications in the scientific literature come from a few select groups and many national societies do not regard it as a mainstream bariatric procedure. This poses challenges for individual surgeons and patients.

We have previously attempted to understand the academic rationale for all the controversial aspects of this procedure[11] and also performed a systematic review of the published scientific literature on this procedure.[12] Since then, several authors[579] including our own group[13] have published their experiences with this procedure. This review examines the current status of the controversial aspects of this procedure.

CURRENT STATUS OF MINI-GASTRIC BYPASS

Early safety and efficacy

It is now widely acknowledged that MGB is at least as effective as Roux-en-Y gastric bypass (RYGB), if not more effective,[314] when it comes to weight loss and comorbidity resolution. It takes less time to perform, has a shorter learning curve, and is associated with fewer major complications.[3]

In their comparative analysis of RYGB and MGB over a 10-year period, Lee et al.[3] found that at 5 years, MGB had a significantly lower body mass index (BMI) (27.7 vs 29.2) and higher excess weight loss (EWL) (72.9% vs 60.1%); there was no significant difference in the improvement of comorbidities. A randomized study from the same group[14] showed a lower complication rate with MGB (7.5% vs 20%, P < 0.05) and a higher proportion of patients achieving an EWL >50% (95% vs 75%, P < 0.05). We have also observed similar results in our practice.

Despite these obvious advantages, the uptake of MGB has been slow in many parts of the world because of several concerns that have been expressed in the past regarding this procedure. In the following paragraphs, we examine the current status of these concerns.

PATIENT ACCEPTANCE

We know from our experience[13] and our review of the literature[12] that MGB has a high patient acceptance and most patients report a significant improvement in the quality of life. Though many patients report an increased bowel frequency, it does not negatively impact the quality of life. Lee[3] found similar gastrointestinal quality of life scores between RYGB and MGB patients in their comparative analysis of the two procedures. MGB patients had a better score in abdominal pain but lower score in eating with pleasure and trouble with diarrhoea than RYGB patients.

Oesophageal/gastric cancer

It has been suggested on the basis of in vitro and animal studies that reconstruction with a loop configuration in patients undergoing gastric bypass will increase the risk of gastric and oesophageal cancers because of increased biliary reflux.[15] As we have discussed in detail previously,[11] there is no clear scientific basis to suggest this association. Duodenogastric biliary reflux is a physiological phenomenon and the presence of bile in the stomach is a very common endoscopic finding in the normal population. Most of the evidence for the role of bile as a carcinogen comes from rodent and in vitro studies and studies that can conclusively label bile as a carcinogen for gastric and/or oesophageal cancer are lacking.

There is another potential angle to this debate. It would only be necessary to find a potential etiological agent, like bile reflux, if it was evident that patients who underwent Mason's loop gastric bypass and those who are at present undergoing MGP (both different procedures but with a similar loop reconstructions) have an increased risk of gastric and/or oesophageal cancers. However this is not the case. As we have discussed before,[16] many thousands of Mason's loop gastric bypasses were performed in the 1960s and 1970s and tens of thousands of MGBs have been performed since 1997. Yet, there has been only one cancer reported in the literature in the gastric pouch[17] of a patient with Mason's loop gastric bypass (26 years after surgery) and none so far in patients with MGB. There has been one gastric cancer reported in an MGB patient but this was in the bypassed stomach and not the pouch.[18] This is significant as both these cancers have an annual incidence of 1 in a population of 7,000-10,000. We believe that with all the controversy surrounding these procedures, we would certainly have seen more reports in the literature if there was any genuine association between these procedures and gastric or oesophageal cancers.

It is perhaps time to shift the onus of evidence on to those surgeons who claim that MGB is associated with an increased risk of gastric and/or oesophageal cancer. Their refusal to put their prejudices aside may be coming in the way of giving a potentially very useful bariatric and/or metabolic procedure a fair scientific hearing.

Symptomatic (biliary) reflux

Gastro-oesophageal reflux disease (GORD) is strongly associated with obesity and a large number of patients seeking bariatric surgery suffer from it. RYGB is regarded by some as the best anti-reflux operation[19] for obese patients suffering with GORD. Sleeve gastrectomy (SG) has a complex relationship with GORD[20] where a majority of the patients report an improvement but some experience deterioration and others notice de novo reflux. Similarly, though GORD usually improves with MGB, several authors have reported troublesome reflux symptoms requiring revision to either RYGB or Braun's anastomosis.[479] It is also worth noting that reflux symptoms have also been noted after RYGB[21] and may require surgical correction. In our experience, the incidence of significant reflux requiring surgical correction appears to be lower in MGB patients than sleeve patients but higher than that seen in RYGB patients. It has an approximate incidence of 0.5% or less in our experience and that of many other surgeons who have published their data.

Lee observed no significant difference in gastrointestinal quality of life between RYGB and MGB.[3] There was no difference in the symptoms of heartburn or regurgitation between two groups. This is significant, as RYGB would be expected to cause less reflux because of a considerably smaller pouch and the Roux-en-Y configuration. At the same time, authors reported revisions of MGB to RYGB for intractable bile reflux. These data are not clearly reported in the paper but it seems that four revisions in the MGB group for intolerance were due to reflux. There were no revisions for intolerance in the RYGB group.

Tolone et al.[22] studied GORD extensively in their elegant study of patients without any preoperative GORD symptoms undergoing MGB and concluded that none of the 15 patients reported de novo GORD symptoms or oesophagitis on endoscopic examination. Interestingly, two of these patients had a small hiatus hernia (HH). Though the numbers are small in this study and there is potential for Type 2 error, these are significant findings. The authors compared the results with a control group of patients who underwent SG. As expected, intra-gastric pressure as well as the gastro-oesophageal pressure gradient increased after SG but decreased after MGB.

The authors observed that MGB led to a significant reduction in both oesophageal acid exposure and in reflux episodes, whereas SG resulted in an increase in both. More interestingly, none of the MGB patients had any “weakly alkaline reflux” before or after surgery but SG patients experienced a significantly higher total, upright, and recumbent “weakly alkaline reflux” after surgery. The authors also found normal oesophageal mucosa and gastric mucosa on endoscopy with only one patient having gastritis without evidence of any biliary gastric content. A similar gastritis was also seen in two patients after SG. Interestingly, the authors also found that neither MGB nor SG resulted in any change in LES pressure. It was a small study and the authors excluded those with GORD and/or large HH. Their findings hence, cannot be extrapolated to the entire bariatric cohort.

However, they are not alone in reporting such results. Previously, Carbajo et al.[23] had studied reflux with pH studies and endoscopic examination at 12 months and 18 months after surgery in the first 20 patients undergoing MGB at their centre and found no abnormality. It is worth noting that these authors attempted to create an anti-reflux valve between the gastric pouch and the small bowel and believed that the complete absence of reflux symptoms in their cohort was down to this technical modification. At the same time, many authors have reported persistent postoperative reflux requiring surgical correction after MGB. We, hence, need focussed studies on these patients to better understand the etiopathogenesis of GORD after MGB. Whether it is more likely in patients with pre-existing severe GORD or large HH needs to be examined, as does any correlation with pouch length.

Marginal ulcer

Reported ulcer rates after MGB vary between 0.5% and 5.0%.[71424] An ulceration rate of 2.8% seen in our systematic review[12] was not dissimilar to what is typically observed after RYGB. We have compared (data published as an abstract) marginal ulcer rates in patients undergoing MGB with that in those undergoing RYGB[25] and similar to Lee's randomised study of RYGB and MGB,[14] found no significant difference.

At the same time, studies on RYGB have found larger pouches to be associated with higher ulceration rates.[26] If these findings were extrapolated to MGB, simply on the basis of a much larger pouch, one would expect a higher ulceration rate but that has proved not to be the case. It, hence, remains to be examined if it is because of the possible buffering of excess acid with bile or due to potentially less tension on the anastomosis as a result of a longer pouch. Nonetheless, concerns that these patients will have a very high ulceration rate because of constant exposure of anastomosis to bile have not confirmed by any author so far. Similar to RYGB, ulcers in MGB also seem to be strongly associated with smoking[9] but unlike RYGB, one expects persistent ulcers to be more easily amenable to treatment because of the ease of reversal or revision of the operation. Though ulcer perforations have been reported[91324] after MGB, they seem to be restricted to smokers and can be treated by a simple closure similar to RYGB ulcer perforations.

Malnutrition

There is no doubt that approximately 0.5-1.0% of the patients develop malnutrition requiring surgical correction — Reversal or shortening of bilio-pancreatic (BP) limb, conversion to sleeve — After MGB.[234571227] We know from studies on RYGB that bypassing small bowel as BP limb[28] results in more weight loss (presumably through more malabsorption) than bypassing it as alimentary limb. This may account for the observed superior weight loss with MGB as well as the higher incidence of malnutrition with it. Small bowel bypass length has not been standardised in various patient populations undergoing RYGB. Though this causes less of a problem when small bowel is bypassed largely as an alimentary limb, when the entire bowel is bypassed as BP limb (as with MGB), even small differences in the length of BP limb may have unpredictable consequences. It is our belief that a majority of the weight loss after proximal gastric bypass, both MGB and RYGB, results from its effect on hunger and satiety through yet incompletely understood hormonal pathways and we would like to differentiate these procedures from the comparatively less popular options such as distal gastric bypass and biliopancreatic diversion/duodenal switch, which rely on malabsorption for a significant component of weight loss. We further believe that though a BP limb of 2.0 m is sufficient for this optimum hormonal effect for the majority of the population, some patients may achieve this with a shorter BP limb and with consequent reduction in malnutrition rates with MGB.

Many authors have shown satisfactory results with up to 150 cm BP limb[345102429] in patients with class II obesity. Some of these authors such as Lee[3] and Musella[5] have used a tailoring formula where patients with a BMI of 35 had a limb length of 150 cm with 10 cm per BMI point increase after that. Others such as Noun[10] used a slightly different formula where they used 150 cm BP limb for patients with BMI of 40 and increased by 10 cm for each BMI point. It is worth noting in this context that Noun reported 0.4% incidence of revisional surgery for excess weight loss compared to an approximate 1.0% incidence reported by Lee.[3] Using the same formula as Lee, Musella et al.[5] reported a 0.2% incidence of excessive weight loss. At the same time, authors such as Chevallier[9] and Kular[7] who used a fixed 200 cm noted a much lower significant malnutrition rate of 0.05% and 0.1%, respectively.

Lee et al. used a length of 120 cm in their randomised controlled trial comparing MGB and sleeve in patients with Type II diabetes mellitus and a BMI between 25 kg/m2 and 35 kg/m2.[30] Authors randomised 60 patients to either MGB or SG. The mean BMI was 30.2 in the MGB group and 31.0 in the sleeve group. At 5 years after surgery, MGB patients had a significantly lower BMI (23.3 vs 25.1). Authors found that MGB patients achieved better glycaemic control. The same group also used a BP limb of 100 cm for diabetic patients with BMI <35 kg/m2 with acceptable complication rates.[31] García-Caballero et al.[32] used a larger pouch and a BP limb of 100 cm for diabetic patients with a BMI of <30. In another study[33] involving eleven C peptide negative diabetic patients with BMI of 24-34, this group recommended using gastric pouch double the size used in obese patients and a BP limb of 100 cm for patients with BMI of 24-29 (n = 7), 120 cm for BMI of 30-32 (n = 2) and 150 cm for BMI of 33-34 (n = 2).[33] BMI stabilised at a mean of 21.6 kg/m2 in this series and no patient lost too much weight or suffered malnutrition.

Small bowel length in humans is associated with height, weight and race and it is hence, possible that different population subgroups have different lengths of BP limb that must be bypassed for optimum results with MGB. Future research should concentrate on identifying these subgroups of MGB patients who could have shorter BP limbs of 100-150 cm without any compromise in efficacy. The authors hypothesise that lighter and shorter patients might be better off with shorter limb lengths. These thoughts may be especially relevant for shorter Asian populations and for Class 1 obese patients with diabetes undergoing metabolic surgery.

Internal hernia

Internal hernia is a major long-term problem after RYGB even in the hands of surgeons who close internal defects[3435] and a recognised cause of late mortality after surgery. Lee et al.[3] found that RYGB patients had a higher chance of internal hernia and intestinal obstruction requiring revisional surgery. In fact, the authors did not report any internal hernia in MGB patients. In the entire literature, there is only one reported case of Petersen's hernia after MGB[36] in contrast to 1-6% internal hernia rates observed with RYGB.[373839] It is not inconceivable how the loop configuration will reduce sites for internal herniation. We believe that internal hernia will be a rare finding after MGB and perhaps even more importantly, this procedure will also be associated with a lower incidence of nonspecific abdominal pain compared to RYGB. Future studies will need to specifically address this point.

Acute gastric dilatation

Though most large series have not reported this, one can narrow the outlet of the bypassed stomach with MGB in an attempt to make a long enough gastric pouch and this will then result in acute gastric dilatation.[40] Surgeons, hence, must be cautious while forming the pouch in MGB and not go too close to greater curvature.

Dumping syndrome and reactive hypoglycaemia

It is believed that MGB results in less dumping and reactive hypoglycaemia compared to RYGB. Carbajo et al.[23] did not see any dumping syndrome in their patients and attributed this finding to “the constant contact of nutrients with biliopancreatic secretion”. Though we share these beliefs, this aspect needs examination in future studies. We are not aware of any report in the literature of significant reactive hypoglycaemia after MGB.

Diarrhoea

Because of the higher potential for malabsorption, MGB is more likely to be associated with steatorrhoea and diarrhoea compared to RYGB.[3] Carbajo et al.[23] observed that patients reported a mean of two to three stools per day with some fat malabsorption during the early months after surgery. However, at 1 year this normalised to one bowel motion per day with stool of normal consistency. Increased stool frequency and/or steatorrhoea in itself does not always cause any problem and may in fact account for improved gastrointestinal quality of life for those suffering from constipation and for the superior weight loss seen with MGB. This has in fact led some authors to suggest that MGB may be a better choice for those with constipation.[3] At the same time, excessive bowel frequency could compromise the quality of life even in the absence of protein calorie malnutrition. This takes us to the discussion regarding appropriate bowel bypass length with this operation. From the preceding discussions on the appropriate length of BP limb with MGB, we hypothesise that a standard 200-cm bypass length in all patients undergoing MGB is an oversimplification and some patients could achieve most of the benefits with a shorter BP limb and consequent gains in terms of troublesome malnutrition and/or diarrhoea.

Revision surgery

A certain number of patients undergoing MGB do need revisional surgery for malnutrition or troublesome reflux. We believe with further research and standardisation, it may be possible to reduce the incidence of these complications. However, it is worth bearing in mind that every single bariatric surgical procedure has a definite revision/reoperation rate attached to it. Lee found a similar revision rate (3.6% vs 2.8%, P = 0.385) between RYGB and MGB.[3] However, the reasons for revisional surgery were different between the two groups: RYGB patients experienced a higher incidence of revisions for bowel obstruction and internal hernia, and MGB patients had a higher incidence of revisions for malnutrition and “intolerance due to bile reflux”.

DISCUSSION

Despite many thousands of published cases, MGB continues to be controversial and many national societies do not regard it as a mainstream bariatric procedure. Though most institutes completely ignore this operation, others such as the Italian Society for Metabolic and Bariatric Surgery (SICOB) and British Obesity and Metabolic Surgery Society (BOMSS) suggest a careful approach to it with a close monitoring of patients. This is interesting because many other bariatric procedures have been adopted without documented results over such prolonged periods. One can only presume this is because of the concerns in the minds of many surgeons relating to troublesome symptomatic biliary reflux and/or potential risk of gastric/oesophageal cancer.

MGB has proven results when it comes to early safety and efficacy. It is non-inferior to other established bariatric procedures and is particularly suited for metabolic and diabetic surgeries.[30] Since effectively, there is only one variable, BP limb, to alter according to patient specific factors, this procedure lends itself to easy adjustments. Longer pouch and one anastomosis make it easier for newer surgeons to learn and also a viable option for technically difficult superobese patients. The ease of revision and reversal provide further safeguards against long-term problems with liver problems, persistent ulceration, or malnutrition.

This review makes the case that there is no significant basis for continued concerns regarding widespread adoption of MGB. Indeed, these concerns are non-scientific. National institutions are depriving their patients and surgeons of an effective alternative by not endorsing it. One cannot deny the fact that with our historical experience with Mason's gastric bypass, some concerns seemed justified when MGB was initially proposed. However, to continue to hold those views despite a rapidly increasing body of literature on the topic contradicting and alleviating those concerns smacks of unscientific beliefs and prejudices. It is time to move on and include MGB as a mainstream bariatric procedure. Then only can we really engage in studies and trials that will clarify an appropriate role for this procedure in the bariatric armamentarium and also help to improve it further. For example, with all the controversy of reflux surrounding it, our current approach is not to recommend MGB for patients suffering with GORD/HH. Further studies are needed to understand th esafety of MGB and a comparative analysis with other procedures in these patients.

CONCLUSION

It is no longer justified to restrict a wider adoption of MGB. National institutes should consider endorsing it as a mainstream bariatric operation. This will be a very useful addition to the existing repertoire of bariatric procedures.

Financial Support and Sponsorship

Nil.

Conflicts of Interest

There are no conflicts of interest.