Experimental Metabolic Surgery: Justification and Technical Aspects
- First Online:
- Cite this article as:
- Sabench Pereferrer, F., Hernàndez Gonzàlez, M. & Del Castillo Déjardin, D. OBES SURG (2011) 21: 1617. doi:10.1007/s11695-011-0367-4
- 196 Views
Metabolic surgery is a surgical strategy which has shown great potential in the treatment of diseases which may be associated with morbid obesity. It must be developed on the basis of both animal and clinical research. The objective of this study is to set out the various options in experimentation animals and the technical characteristics in operations, and the specific animal care undertaken by our group.
We identified and reviewed the key points to be considered in animal handling during interventions such as sleeve gastrectomy, Roux-en-Y gastric bypass, ileal transposition and duodenal exclusion.
The technical variations found at experimental level are due to the pouch capacity for the Roux-en-Y gastric bypass. Intestinal anastomosis is the variable with the greatest differences found between the various working groups. Ileal transposition is a technique that is undergoing constant review, and the results differ substantially depending on the animal model chosen, and are also metabolically effective in animals with a normal weight. Duodenal exclusion by means of a physical barrier has not been studied sufficiently but could be a pre-operative support for weight loss.
There are experimental technical discrepancies and further studies are necessary to ascertain their efficiency. Metabolic surgery currently complements bariatric surgery and justifies the appearance of new experimental studies. The animal models chosen are very important as only very specific study models will be used in cases in which the technique is sufficiently validated by the research team, as the results to be assessed depend on this.
KeywordsIleal transpositionSleeve gastrectomyDuodenal exclusion
The clear increase in the prevalence of morbid obesity and metabolic syndrome in recent years and the still invisible “pool” of a population of ill children represent a major challenge for today’s health system. In this context, metabolic surgery has consolidated itself not only because of its benefits regarding weight, but also regarding the mechanisms regulating intake and resolution of the main comorbidities accompanying morbid obesity. In the field of bariatric surgery, there have been numerous experimental studies endorsing a specific technique, produced at the same time as clinical studies. This may be due to the lack of studies in humans, or to the need to complete ethically dubious studies in humans. Thanks to the various study models, experimental research has been able to specify its study objectives, which in this case would be to ascertain the metabolic effects of a specific technique. In metabolic surgery, experimental surgery began to assume a major role between 2000 and 2002 thanks to the studies of Cummings et al.  and the experimental studies by Xu et al. , which established the need to investigate hypothetical metabolic-surgical links. In view of this paradigm shift, scientific companies were forced to add the concept of “metabolic surgery” to their name. Today, within the concept of metabolic surgery in its strictest sense, there are three different situations, with three different groups of techniques: the first group includes classic bariatric surgery techniques, which are the current benchmark and which act as the gold standard for most patients. This is true of the Roux-en-Y gastric bypass (RYGB), Scopinaro’s biliopancreatic diversion and the duodenal switch. In these techniques, resolution of diabetes mellitus type 2 (T2DM) is too early to be attributed only to weight loss according to some authors, and they suggest that other factors have an impact on glucose homeostasis . The second group includes techniques such as ileal transposition, which have been constantly reviewed since the 1980s, and are still being reviewed thanks to the antidiabetic and anorexigenic action of the LPG-1 it entails . Finally, there is a third group of techniques that has appeared recently, having been developed from classic techniques and a consequence of their adjustment to their metabolic nature. These include sleeve gastrectomy, thanks to its relationship with ghrelin, jejunal duodenal bypass and duodenal exclusion. The latter came about due to the hypothesis of Rubino F. and Marescaux M, who observed that duodenal exclusion led to an improvement in T2DM, entirely regardless of weight loss . This would therefore prevent the production of mediators or enteric signals creating insulin resistance in diabetics (the foregut hypothesis) . Meanwhile, contact by practically undigested food with the distal intestine would create or produce enteric signals or substances which would improve sensitivity to insulin such as LPG-1 (the hindgut or early ileal stimulation hypothesis). This can be seen in the studies by De Paula  in which results with ileal transposition achieve a normalisation of glycosylated haemoglobin in 70% of patients.
In view of all these new changes, experimental surgery has had to adapt in order to be able to endorse or complete some of these techniques, and even to focus existing studies on the classic techniques in this new direction. Obese rats or those with an associated metabolic syndrome are used to that end, in order to produce a model accurately reflecting the original pathology. Large animals such as pigs or dogs are mainly used to perfect learning curves or to validate specific endoscopic techniques, new instruments or various means of access to the abdominal cavity (NOTES, SILS) [8, 9].
Due to the recent boost in experimental surgery thanks to the constant development of new techniques, our aim in this study is to set out the various alternatives currently available in animal experimentation, and the main technical characteristics of the operations and specific animal care in the metabolic surgery field that we have been undertaking in our centre.
Before considering any study, it is very important to ascertain and choose the right animal model based on the objectives and results desired. The most commonly used rats are various types of obese rats, although non-obese rats such as the Sprague–Dawley are also used. This rat was produced in 1925 by Robert W. Dawley from a hybrid male rat and a female Wistar rat. It is called OFA (Oncins France Strain A). They are pheotypically albino, medium-sized and with a high growth rate. They are highly resistant to surgical aggression and can be used to standardise and refine techniques. Another advantage is that they can be subjected to a fattening protocol by means of a cafeteria diet or hypercaloric food in the following proportion: 45% lipids, 45% carbohydrates and 10% proteins.
Evolution of weight in a 4-week cafeteria diet protocol
Mean ± DS
Weight (gr) non obese
Weight (gr) obese
Sprague–Dawley rat by cafeteria diet
318.9 ± 9.4
316.2 ± 5.1
357.9 ± 8.2
385.3 ± 7.8
381.6 ± 10.2
420.4 ± 6.6
402.5 ± 2.9
455.2 ± 9.5
420.5 ± 9.7
481.3 ± 7.5
A variation of the above is Zucker Diabetic Fatty (ZDF) rats (Fig. 1b). These are denoted as ZDF/Gmi-fa/fa. They represent a diabetes mellitus type 2 animal model that is not insulin-dependent. It is a Zucker rat with the same mutation, which was found to be diabetic in 1974 in a non-consanguineous colony (Dr. Walter Shaw—Indianapolis). The consanguineous ZDF line was characterised in 1985 and produced by Genetic Model Inc. There is a non-obese control model (lean fa/+). They express the characteristics of diabetes mellitus type 2 and its resistance to insulin. They present hypercholesterolemia, hypertriglyceridemia, slow healing of wounds and hydronephrosis. Their absolute weight in relation to their age is lower than the Zucker rat due to their extreme diabetic nature. The main disadvantages are their high price, an intense polydipsia (double water bottle in housing), an unpleasant smell and high post-surgical mortality . This fragility may to some extent restrict their use in the surgical laboratory. Nonetheless, they are used because they provide quite a precise approximation to obesity associated with Diabetes type 2 in humans.
From the point of view of metabolic surgery on non-obese animals, and considering the studies by Rubino F and Marescaux M, we used the Goto–Kakizaki rat (GK/ToshiCskCrljCrl) , created in Japan from endogamous cross-breeding of Wistar rats but with glycaemias at the upper threshold of normality (>137 mg/dl). The control model is the Wistar rat. These animals present an increase in peripheral resistance to insulin and alterations in lipid metabolism. They are not obese rats but metabolically reproduce the diabetes mellitus type 2 models. The defects in glucose-stimulated insulin secretion, peripheral insulin resistance, hyperinsulinemia and hyperglycemia are seen as early as 4 weeks after birth. Other characteristics are neuropathy, osteopathy, retinopathy and nephropathy. Various papers have studied the mechanism by which this cell disorder is produced in the Beta cells. In response to hyperglycemia, it has been shown that apoptosis increases in these cells as oxidative stress increases . It has been deduced that the development of diabetes type 2 in this animal model is multifactorial. It includes a genetic susceptibility to express a pancreatic cell dysfunction, a decrease in the neogenesis of B cells that is programmed after being transmitted over various generations. As well as a persistent response to hyperglycemia there is a decrease in B-cell differentiation associated with an inflammatory state that leads to the abovementioned oxidative stress, islet fibrosis and a disorder in their vascularization . Goto–Kakizaki rats are very useful in metabolic studies, but are expensive because very few laboratories produce them.
In an attempt to create colonies that are more suited and adapted to the polygenic inheritance of obesity, two lines were developed from Sprague–Dawley rats which can provide a good inference of metabolic syndrome in situations of obesity and non-obesity. They have been produced by Charles River Lab® since 2008. They are: The Obese-Prone rats (OP-CD). These rats were developed from a line of SD rats by selecting future breeders with accelerated weight gain. They become obese when fed high-fat diets and the subpopulation of non-responders was eliminated. They develop a polygenic obesity despite having a fully functioning leptin receptor. The characteristics are: obesity, metabolic syndrome, hypertension, hyperinsulinemia and insulin resistance. The subpopulation of non-responders became a new line: the Obese-Resistant rat (OR-CD). The characteristics of these rats are the same, but they are a non-obese model. They are also the control model for the obese-prone rat. They can be very useful for determining the effects of metabolic surgery in non-obese but sick populations.
Another obese model was recently created by crossing obese Sprague–Dawley rats with insulin resistance resulting from polygenic adult-onset obesity and Zucker diabetic fatty-lean rats that have defective pancreatic β-cell function but normal leptin signaling. This model is known as the UC Davis-T2DM (University of California) rat and it is more similar to clinical T2DM. UCD-T2DM rats develop T2DM when fed a standard chow diet .
Some new models also reflect metabolic syndrome. They all have hyperinsulinemia, insulin resistance and obesity, but are selectively used to study the cardiovascular and renal component of metabolic syndrome. Briefly, these models are the following: The ZSF1 rat. This model was created from the ZDF rats and develops an important nephropathy and congestive cardiac insufficiency. The Dahl/Salt rat: this model develops hypertension and renal failure when the diet is rich in salt. The Spontaneously Hypertensive Rat (SHR): created from Wistar rats, this is the most commonly used genetically hypertensive rat model. The Spontaneously Hypertensive Stroke Prone rat: created from Wistar rats, if their diet is rich in salt they develop brain haemorrhage at 16–18 weeks of age.
Main rodents used in experimental metabolic surgery
Sprague–Dawley (OFA) or Wistar
Possibility of fattening protocol
Continuous pre-operative monitoring of weight and calorie intake
Resistant to surgical aggression
Zucker Diabetic Fatty (Gmi-fa/fa)
Type 2 Diabetes established model
High post-surgical morbi-mortality
Excessive hyperglycemia and low pancreatic reserve
Not obese. T2DM ++
No obesity. Useful to explain the metabolic effects (weight loss independent)
Obese-Prone rat (OP-CD)
Few studies and insufficient data
Polygenic obesity rat model
Obese-Resistant rat (OR-CD)
Few studies and insufficient data
Hyperinsulinemia and Insulin resistance
Obese ± T2DM +
Late hyperglycemia from week 18 onwards
General Care of the Animals
Before and after surgery, the animal is housed according to government legislation and the regulations of the Federation of Laboratory Animal Science Associations. In our case, the animals are housed in the Faculty of Medicine at Rovira i Virgili University in accordance with the applicable laws for animal care (Official Bulletin of the Government of Catalonia 2,073, 10/07/1995). It is advisable to use young animals (not older than 13–14 weeks old), and male animals, so that hormonal influences do not distort the results. A highly appropriate point from which blood can be extracted before the operation is the external jugular path, because it is easy to access, clean and a high volume can be extracted (1.5 to 2 ml). This point of access is useful precisely because of the volume that it enables to be extracted. Some commercial models have a catheter already incorporated in the jugular vein to facilitate the extraction process (Harlan® or Charles River®). The extraction requires a silicon catheter (Silastic®) with an internal diameter of 0.05 cm and an external diameter of 0.1 cm. This catheter is treated with EDTA to prevent the blood from coagulating prematurely. Throughout the process, we use a surgical stereoscopic microscope (Olympus®). With the animal anesthesized and lying supine, a small left or right laterocervical incision is made and the external jugular vein is dissected and proximally ligated (Fig. 1c). Under ×2 magnification with the stereoscopic microscope, the vein was punctured and cannulated with the silicon catheter (Fig. 1d), and the blood was drawn using the Vacutainer® system. Once the extraction has been completed, the catheter is withdrawn and the vein is ligated distally with a 4/0 silk tie. The skin is closed using discontinuous stitches with polyglycolic acid suture.
It is advisable to wait 1 week between the extraction of blood and the surgical operation, so that the animal’s normal blood volume is restored, although 2 ml of glucosaline serum is administered subcutaneously by protocol after extraction (Fig. 1e). To monitor the weight, it is advisable to use animal growth curves through twice-weekly weighing before surgery and daily after surgery. Food is administered and weighed daily, and what the animal rejects must be monitored. Metabolic cages are very useful for this. The energy contribution of the intake (kcal) can subsequently be calculated, and its weight curves produced. Indirect calorimetry is advisable before and after the surgical operation, as this confirms the change in energy expenditure and its relationship with the calculated intake. The standard maintenance feed used is Panlab® AO4 3,173 kg, the qualitative composition of which is carbohydrates 60.5%, proteins 15.4%, fats 2.9%, fibre 3.9%, minerals 5.3% and humidity 12%. Chlorinated water is administered ad libitum, as is the liquid diet during the post-surgery period, with the administration of nutrition formulae such as Resource 2.0® by Nestlé Corp. (2 kcal/ml) for 72 h. Glucose can be monitored weekly before and after surgery by a blood sample obtained from the coccygeal vein in the tail, and ascertained using a conventional glucometer. With the animal immobilized by conventional means but with its tail free for purposes of manipulation, we apply a rubber tourniquet at the base of the tail, thus engorging the veins. The coccygeal vein is punctured with a 22 G needle, which produces a drop of blood that enables the glucose level to be determined at once. Glucose curves (intravenous glucose tolerance test (IVGTT)) can also be obtained by continuous monitoring in fasting conditions or after controlled intake to determine their variability. Glucose curves can be obtained by continuous infusion of insulin and glucose. The plasma levels of parameters such as leptin  or GLP-1  can also determined in the same way. Inadequate insulin secretion to compensate for insulin resistance followed by eventual β-cell decompensation is the key to the pathogenesis of clinical T2DM. Demonstrating it by using the IVGTT to take longitudinal measurements of glucose and insulin can give an idea of just how severe the alterations are. This method can also be used to test resistance to insulin and the ability to reduce insulin in people who are at risk of conditions other than diabetes, such as high blood pressure, and to determine how well a treatment improves insulin resistance or insulin secretion.
After the surgical operation, the monitoring period of the animals can vary from 2 weeks to several months, taking into account that 10 days for a rat is equivalent to approximately a 1-year period of time in a human being . An overdose of intraperitoneal anaesthetic is very effective for killing, and reduces animal stress to the minimum. A second sample of blood is extracted beforehand by direct intra-cardiac means. Samples of hepatic tissue, subcutaneous and visceral fat, among many others, can be extracted during the surgery and at sacrifice.
The surgical operations were carried out in the experimental surgery laboratory equipped to that end belonging to the Surgery Department of the Faculty of Medicine at the Rovira i Virgili University (Reus, Spain). The anaesthetic administered was intraperitoneal Zoletil®, which combines tiletamine with zolazepam at a dose of 30 mg/kg, with a wide safety margin. All the equipment is adapted to the size of the animal. A binocular microscope was used for the small intestinal anastomoses when vision was limited. During the surgical operation, the animal’s temperature was monitored every 20 min and the anaesthetic dose was repeated by 50% if the animal woke up or showed signs of doing so (positive plantar reflex). The laparotomies were closed with a 3/0 continuous absorbable polyglactin suture. To keep the animal hydrated during the immediate and late post-operative period, 4 cc of 5% glucose–saline solution was injected every 24 h for the first 48 h. Throughout the whole of the post-operative period, the animals were provided with chlorinated water ad libitum. The laparotomy wound was treated topically with povidone iodine every 24 h.
Sleeve Gastrectomy (Fig. 2)
Roux-en-Y Gastrojejunal Bypass (Fig. 3)
Ileal Transposition (Fig. 4)
Duodenal Exclusion (Fig. 5)
Characteristics of pouch and Roux limb length in experimental RYGB
Small intestinal division
Roux limb length (cm)
Xu et al. 2002 
2 double rows of titanium staple lines reinforced with interrupted polyglactine sutures.
16 cm distal to ligament of Treitz.
Meguid et al. 2004 
Sprague–Dawley diet-induced obesity
Row of titanium staple lines
30 cm below the ligament of Treitz
Sprague–Dawley diet-induced obesity
Gastric transection at ridge that separates the forestomach and glandular stomach
15–18 cm distal to ligament of Treitz
Suzuki et al. 2005 
Sprague–Dawley diet-induced obesity
Triple rows of titanium staple lines
30 cm below the ligament of Treitz
Inoue et al. 2007 
Fischer rat (non obese)
5 mm downward from the gastroesophageal junction
10 cm distal to ligament of Treitz
Furnes et al. 2008 
Esophagus anastomosis to the proximal jejunum
2–3 cm distal to ligament of Treitz
Tichanski et al. 2008 
Sprague–Dawley diet-induced obesity
10% (1–2 ml)
10 cm distal to duodenum
Wolff et al. 2009 
GIA stapler 45 mm
Biliopancreatic limb of 15 cm
Sabench et al. 2009 
Gastric transection 6–8 mm under the ridge that separates the forestomach and glandular stomach. Manual suture (5/0 polypropylene)
15 cm distal to ligament of Treitz.
Stearns et al. 2009 
Pouch includes glandular mucosa under the ridge
Biliopancreatic limb of 16 cm
Li et al. 2010 
Goto Kakizaki rats
Shin et al. 2010 
Sprague–Dawley diet-induced obesity
Triple rows of titanium staple lines
Biliopancreatic limb of 40 cm
As regards ileal transposition, the distances established between the ileal section (10 cm from the ileocecal valve) and the length of the segment (10 cm) are corroborated by studies beginning in 1982. This technique began on an experimental basis, in order to clarify the cause that determined the decrease in intake after a jejeunoileal bypass. During the weight loss process after a jejeunoileal bypass, there is a clear decline in intake which is caused by the sensation of discomfort after meals (diarrhoea, nausea...) and also by ileal overstimulation of nutrients, which triggers internal signals that inhibit the feeling of hunger. The results were a significant decline in intake and weight loss, as well as an increase in pancreatic mass and a decline in glycemia, which opened a research line into the hormonal mechanisms involved in these findings (Hypothesis of Koopmans et al.) . Atkinson R et al.  set out the same results in his work on rats in 1982, but with transposition at jejunum level. Ferri et al.  in 1983, observed a marked increase in plasma enteroglucagon and the ileal cells producing it. Research is currently being done on its metabolic component, with no direct effect on weight observed in Kakizaki rats, but there is a direct effect on hyperglycemia, with an increase in pancreatic ß cells in some animals with the pancreatic reserves almost exhausted [30, 31]. For this reason, it is clinically important to consider the Peptide C and HOMA index plasma levels very carefully, because they provide information on the secretion of insulin, its level of resistance and the pancreatic reserves (Homeostasis Model Assessment HOMA-IR: (insulin in fasted conditions × glucose in fasted conditions) / 405, with the insulin expressed in uU/ml and the glucose in mg/dl (405 must be replaced by 22.5 when the glucose is expressed in mmol/L). However, when applied to animals that are obese due to a cafeteria diet or Long–Evans rats, the effects on weight and intake are significant [26, 32, 33]. Herein lies the importance of the animal model chosen, depending on the objectives of the study. This technique has been reviewed constantly since then, and it has been performed clinically in some Latin American countries with an excellent learning curve. A BMI of less than 30 kg/m2 but with an associated metabolic syndrome is also applied due to this technique obtaining beneficial effects regardless of the weight loss obtained .
A removable, reversible and endoscopic device (Endobarrier® GI Dynamics) is being considered with regard to duodenal exclusion, and it is in the experimental phase from the clinical and animal point of view. The studies published consider this mechanism as a pre-operative medium for 3 months before surgical operation [35, 36]. The weight loss percentage (EWL%) is 20% and it is more effective than a low calorie diet before surgery. However, it is not free from secondary effects, such as vomiting or local intestinal inflammation in the implantation area, requiring withdrawal in some cases. In experimentation animals, the study by Aguirre et al.  uses rats obese from hypercaloric feeding and places the 10-cm long polyethylene device from the duodenum in a caudocranial direction. It improves glycemia and glycosylated hemoglobin, with intake reduced by 28%. Our technical experience suggests that it should be inserted from the stomach, as the figures for intestinal necrosis from the duodenum were high (50%) in the process prior to standardisation of the technique . In the preliminary results of the initial series performed in rats fattened by a cafeteria diet, the difference in weight increase between the sham group and the intervention group were statistically significant, although the metabolic potential of the technique has yet to be determined.
By way of summary and final comment, there are two well-known hypotheses that explain how type 2 diabetes can be resolved: the hindgut hypothesis (diabetes control results from the more rapid delivery of nutrients to the distal small intestine, which increases GLP-1 plasma levels and stimulates insulin secretion) and the foregut hypothesis (exclusion of the proximal small intestine reduces the secretion of anti-incretin hormones). Gastric bypass can also bring about significant improvement in hepatic insulin sensitivity, most likely through reduced hepatic gluconeogenesis and without affecting insulin sensitivity. Gastric bypass promotes intestinal gluconeogenesis and stimulates the hepatoportal glucose sensor, while the lack of gluconeogenetic response is associated with absence of the anti-diabetic effects following the operation. This finding suggests that intestinal gluconeogenesis is involved in the improvement of glucose homoeostasis after RYGB .
Metabolic surgery is a major breakthrough for resolving T2DM and metabolic syndrome and has shown great potential for its resolution in patients with poor pharmacological control. This must all be supported by both animal and clinical research, in terms of randomised clinical trials and comparative studies between different animal models. These studies must be complementary rather than exclusive, and must take advantage of the benefits provided by animal experimentation, such as the homogeneity of the sample and the reproducibility of results. The animal model chosen to study is very important, as only very specific models will be used in cases in which the technique is sufficiently validated by the research team. Goto–Kakizaki rats are models for diabetes mellitus type 2 without associated obesity and will therefore be very useful in metabolic studies on surgery in situations of non-obesity. The other models are useful for determining how weight can be by factors other than metabolism. The Zucker rat is a good model for this. Rats that are fattened on a cafeteria diet have the advantage that they are highly resistant and the genetic component is pushed into the background. They are also an excellent basis on which to ground training in experimental surgery. Obese-Prone and Obese-Resistant rats can be a good model of the polygenic inheritance of metabolic disease. Work protocols in the laboratory and the animal unit must be constantly followed by the entire team in order to obtain good results.