Single-Anastomosis Duodeno-Ileal Bypass with Sleeve Gastrectomy Model in Mice

Summary

Single-anastomosis duodeno-ileal bypass (SADI-S) is an emerging bariatric procedure with important metabolic effects. In this article, we present a reliable and reproducible model of SADI-S in mice.

Abstract

Obesity is a major health issue worldwide. As a response, bariatric surgeries have emerged to treat obesity and its related comorbidities (e.g., diabetes mellitus, dyslipidemia, non-alcoholic steatohepatitis, cardiovascular events, and cancers) through restrictive and malabsorptive mechanisms. Understanding the mechanisms by which these procedures allow such improvements often require their transposition into animals, especially in mice, because of the ease of generating genetically modified animals. Recently, the single-anastomosis duodeno-ileal bypass with sleeve gastrectomy (SADI-S) has emerged as a procedure that uses both restrictive and malabsorptive effects, which is being used as an alternative to gastric bypass in case of major obesity. Thus far, this procedure has been associated with strong metabolic improvements, which has led to a marked increase in its use in daily clinical practice. However, the mechanisms underlying these metabolic effects have been poorly studied as a result of a lack of animal models. In this article, we present a reliable and reproducible model of SADI-S in mice, with a special focus on perioperative management. The description and use of this new rodent model will be helpful for the scientific community to better understand the molecular, metabolic, and structural changes induced by the SADI-S and to better define the surgical indications for clinical practice.

Introduction

Obesity is an emerging and endemic situation with increasing prevalence, affecting approximately 1 in 20 adults worldwide¹. Bariatric surgery has become the most effective treatment option for the affected adults in recent years, improving both weight loss and metabolic disorders^2,3, with variable results depending on the type of surgical procedure used.

There are two main mechanisms that are implicated in the effects of the bariatric procedures: restriction that aims to increase satiety (such as in the sleeve gastrectomy (SG) where 80% of the stomach is removed), and malabsorption. Among the procedures that imply both restriction and malabsorption, the single anastomosis duodeno-ileal bypass with sleeve gastrectomy (SADI-S) has been proposed as an alternative to the Roux-en-Y gastric bypass (RYGB), in which a weight regain is observed in approximately 20% patients^4,5. In this technique, a sleeve gastrectomy is associated with a small bowel rearrangment, dividing it into a biliary and a short common limb (one-third of the total small bowel length) (Figure 1A). Technically, the SADI-S has the advantage over the RYGB of requiring only a single anastomosis, reducing the operation time by approximately 30%. In addition, this method preserves the pylorus, which helps to reduce the risk of peptic ulcer disease and limits anastomotic leakage. The SADI-S is also associated with a high rate of metabolic improvement, strongly favoring its use during the last few years^6,7.

Since metabolic effects have become increasingly foundational to bariatric procedures, elucidating their mechanisms seems crucial. Therefore, the use of animal models for bariatric procedures is of utmost importance to better understand their metabolic effects and the cellular and molecular pathways involved⁸. These models contributed, for example, to a better understanding of the change in food intake after SG or RYGB in a controlled environment⁹ and to the study of glucose or cholesterol fluxes through the intestinal barrier^10,11; these informations are rarely available in clinical studies. This knowledge could help to define their optimal surgical indications. We previously described mouse models of SG and RYGB¹². However, despite its promising results in clinical practice, the SADI-S has only been developed and described in rats^13,14,15. However, given its genetic malleability, the mouse model has been useful in the past to study the various metabolic effects of such procedures^16,17,18, and a SADI-S mouse model could be useful to evaluate effects of SADI-S despite the technical difficulty.

In this article, we describe the adaptation of the SADI-S procedure in mice (Figure 1B) in a reproducible manner. Special attention is given to the description of perioperative care.

Protocol

This protocol has been approved by the local French ethical committee for animal experimentation (Comité d'éthique en expérimentation animale; reference CEEA-PdL n 06).

1. Pre-operative preparation

1. Add gel diet food to the normal diet 3 days before the surgery. Fast the mice 6 h before the surgery.
2. Induce anesthesia with 5% isoflurane (1 L/min) in a dedicated chamber with oxygen (1 L/min). Inject the mice subcutaneously with buprenorphine (0.1 mg/kg), amoxicillin (15 mg/kg), metoclopramide (1 mg/kg), meloxicam (1 mg/kg) and iron (0.5 mg/kg).
3. Shave the first 2/3 parts of the mouse's abdomen beginning from the xiphoid process using an electric razor. Disinfect the mouse's abdomen in two steps using an iodine polyvidone solution .
4. Place the mouse supine on a dedicated heat pad covered with a clean underpad. Maintain anesthesia using a nose cone with 2%-2.5% isoflurane (0.4 L/min) with oxygen (0.4 L/min). Use a toe-pinch test to confirm the depth of anesthetization.
5. Cover the mouse in a sterilized plastic wrap. In order to apply hyperextension on the mouse's abdomen, fix the lower paw and use a 1 mL syringe or equivalent placed behind the mouse's back. Cut an opening in a sterile compress with the size of the future incision, and use it as an operating field to cover the mouse. The general installation is shown in Figure 2A.
6. Before the surgery, use a face mask, a scrub cap, and sterilized gloves. Use sterilized instruments for the surgery.

2. The SADI-S protocol

1. Median laparotomy
   1. Under a binocular microscope (8x magnification), perform a median laparotomy with scissors or a scalpel by opening the abdominal skin from the xiphoid process to the middle of the abdomen. Ensure that the xiphoid process and the musculoaponeurotic layer are visible (Figure 2B).
      NOTE: Administer bupivacaine (3 mg/kg) subcutaneously at the surgical site 5 mins before making the skin incision.
   2. Open the abdominal wall along the linea alba with scissors between the abdominal muscles. Be careful not to enter the thoracic cavity (Figure 2C).
2. Duodenal exclusion
   1. Gently mobilize the duodenum from the abdominal cavity using a moistened cotton swab to see its anterior and posterior sides. Localize the main bile duct, which is immediately visible under the binocular microscope on the posterior side of the lesser omentum and the duodenum (Figure 3A, black arrows).
   2. Proximally from the main bile duct, visualize an area between the duodenal arteries under the binocular microscope (Figure 3A,B, blue dotted circles). Penetrate this area using curved micro forceps from one side of the duodenum to the other, and perform a duodenal ligation between the arteries using a 6-0 non-absorbable suture (Figure 3C-E). Be careful not to ligate the branches of the duodenal arteries.
3. Sleeve gastrectomy
   1. Mobilize the stomach from the abdominal cavity using a moistened cotton swab and a non-traumatic clamp. Separate the stomach from the surrounding organs using micro scissors: separate the greater omentum, cut the short gastric arteries (branch of the splenic artery) between the stomach and the spleen, and the lipoma linking the stomach to the lower part of the esophagus (Figure 4A,B).
   2. Using micro scissors, perform a 5 mm gastrotomy by opening the fundus and remove the residual food using a cotton swab (Figure 4C, arrow). Rinse the gastrotomy site with sterile saline solution (37 °C) to avoid contamination from the removed gastric contents. 
   3. Apply surgical clips (medium size, 5.6 mm) along the stomach's greater curvature to exclude approximately 80% of the stomach. Two clips are sufficient. Remove the excluded stomach by cutting it with micro scissors (Figure 4D-G).
   4. Anchor the surgical clips to ascertain impermeability by performing a running suture (8-0) from the beginning to the end of the stomach resection (Figure 4H).
4. Duodeno-ileal anastomosis
   1. Under the binocular microscope, visualize the last ileal loop, which is situated just before the caecum (Figure 5A). Gently mobilize the small intestine outside the abdominal cavity from the last ileal loop. Lay out the small bowel, as displayed in Figure 5B, so that the last ileal loop is located on the left side. Using a previously sized suture cord, measure 10 cm (approximately 1/3 of the total length of the small bowel) from the last ileal loop; this will be the site of the future anastomosis.
   2. In order to ensure that the future biliary limb comes to the anastomosis site from its left side, make a large loop of the small intestine around the site of the future anastomosis. Using micro scissors, perform a 4 mm enterotomy by opening the small bowel at this point (Figure 5C-E). Rinse the enterotomy site with sterile saline solution (37 °C) to avoid contamination.
   3. Perform a 4 mm enterotomy on the excluded part of the duodenum, immediately after the pylorus, between the stomach and the ligation performed in step 2.2.2 (Figure 5F). Place an absorbable 5 mm x 5 mm hemostatic collagen compress to favor homeostasis.
   4. Using a non-absorbable 8-0 suture, perform a side-to-side duodeno-ileal anastomosis. Begin with the posterior side anastomosis, followed by the anterior side anastomosis (Figure 5G-I).
5. Abdominal closure
   1. Display the small bowel in the abdominal cavity so that the biliary limb comes to the anastomosis from the superior-left side of the abdomen and the common limb falls to the lower part of the abdomen.
      NOTE: Lavage the abdomen three times with approximately 5 mL of sterile 0.9% saline solution (37 °C). Then, suction the fluid from the abdomen to remove residual gastrointestinal fluid and digested food to avoid bacterial infection and subsequent abdominal inflammation. 
   2. Rehydrate the mouse with 500 µL of 37 °C saline solution by applying it directly into the abdominal cavity using a 1 mL syringe.
   3. Close the musculoaponeurotic layer using a single 6-0 non-absorbable running suture. Close the abdominal skin using 6-0 non-absorbable separated sutures (Figure 5J,K).

3. General postoperative care

1. After stopping the isoflurane, let the mouse wake on the heat pad under 0.4 L/min O₂ insufflated with the nose mask. When fully awakened, which can be ensured by complete motor recuperation, place the mouse alone in a cage in a 30 °C incubator. Leave the mouse in the 30 °C incubator for 5 days (no specific condition for gas or humidity).
   NOTE: The cage should be warmed beforehand.
2. Allow free access to water immediately after surgery. Add vitamin supplements, including vitamins B1, B9, B12, and liposoluble vitamins (A, D, E, K), to water (800 mg/180 mL of water) until the end of the protocol.
3. Maintain analgesia by subcutaneous buprenorphine injections (0.1 mg/kg) twice a day from day 1 to day 3, once a day afterward until day 5. Continue amoxicillin (15 mg/kg), meloxicam (1 mg/kg) and metoclopramide (1 mg/kg) subcutaneous injections once a day until day 3. Provide subcutaneous injections of iron (0.5 mg/kg) once a day until the end of the protocol.

4. General measurements and euthanasia

1. Weigh the mice every day until postoperative day 5. Then weigh on day 7, and then weekly.
2. To measure daily food intake, place one mouse per cage. Place a known weight of a solid diet and measure the weight of the solid diet remaining after 24 h. Measure food intake on day 3, 4, 5, 7, and then weekly.
3. Euthanize the mice by cervical dislocation under general anesthesia (5% isoflurane (1 L/min) with oxygen (1 L/min)) with subcutaneous injection of buprenorphine (0.1 mg/kg) after cardiac left atrium incision for blood sampling (500 to 600 µL of blood).
4. Measure the blood hemoglobin concentration using an automatic hematology analyzer requiring 20 μL of blood.

Representative Results

Learning curve
The learning curve for this model is displayed in Figure 6. A progressive decrease in the operating time is observed, reaching approximately 60 min of surgery after 4 weeks of intensive training (Figure 6A). The 5-day postoperative survival also improved with time, reaching 77% during regular practice (Figure 6B). The most frequent causes of mortality were anastomotic leaks and an afferent loop syndrome resulting in biliary peritonitis. We observed no death later in the first month with the technique described in this manuscript. Of note, previous experiments performed without anchoring surgical clips with running sutures led to clip migration in two-thirds of the cases, resulting in one death by small bowel occlusion at 31 days. These results emphasize that mastering this model requires intensive training.

General parameters
Mice with a C57BL6/J background were randomly assigned to the SADI-S group (n = 9; 5 males, 4 females) and the sham control group (n = 4; 2 males, 2 females). Between the SADI-S mice and the sham mice, the mean pre-operative weight (27.9 g ± 0.98 g vs. 28.5 g ± 2.4 g) and age (14.8 weeks ± 7.2 weeks vs. 18.7 weeks ± 10.3 weeks) were not significantly different. One mouse died after SADI-S at postoperative day 4 from an anastomotic leak and was therefore excluded from the following analysis. SADI-S mice experienced significant weight loss in comparison with the sham control mice from the fourth postoperative day: 21.7 g ± 1.6 g versus 29.0 g ± 0.7 g (p = 0.0081) (Figure 7A). Daily food intake (14 days) significantly increased in SADI-S mice (4.4 g ± 0.1 vs. 2.9 g ± 0.6 g per day, p = 0.027) (Figure 7B).

Mice were sacrificed 28 days after surgery. One mouse in the SADI-S group, which did not display significant weight loss, appeared to have duodenal repermeabilization. No such event was observed in the other 7 mice. As displayed in Figure 7C, the hemoglobin concentration was not significantly different from the sham control mice in the SADI-S group after iron supplementation.

Figure 1: Representation of single anastomosis duodeno-ileal bypass with sleeve gastrectomy (SADI-S). (A) In humans, the duodenum is cut proximally from the main bile duct. A latero-terminal duodeno-ileal anastomosis is performed with the remnant duodenum, defining a biliary limb (before the anastomosis) and a common limb which measures one-third of the total length of the small bowel (after the anastomosis). (B) In mice, the duodenum is excluded by ligature proximally to the main bile duct, and a latero-lateral duodeno-ileal anastomosis is performed. The figure was created with BioRender.com and Servier Medical Art templates which are licensed under a Creative Commons Attribution 3.0 Unported License; https://smart.servier.com/. Please click here to view a larger version of this figure.

Figure 2: Mouse installation for SADI-S. (A) General installation. (B) Skin opening from the xyphoid process (sternal base) to the middle of the abdomen. (C) Musculo-aponeurotic layer and peritoneal opening. Please click here to view a larger version of this figure.

Figure 3: Duodenal exclusion. (A) Avascular window between duodenal arteries (blue dotted circle) on the posterior side of the duodenum, localized before the main bile duct (black arrows). (B) Avascular window between duodenal arteries (blue dotted circle) on the anterior side of the duodenum. (C,D) Duodenal exclusion using 6-0 non-absorbable suture. (E) Final view of excluded duodenum. Please click here to view a larger version of this figure.

Figure 4: Sleeve gastrectomy. (A) Greater omentum removal. (B) Incision of short gastric arteries. (C) Initial gastrotomy (blue arrow). (D-G) Stomach cardiac region removal using two surgical clips. (H) Surgical clips anchoring using 6-0 non-absorbable suture. Please click here to view a larger version of this figure.

Figure 5: Duodeno-ileal anastomosis. (A) Identification of the last ileal loop (asterisk). (B) Count 10 cm (one-third of the total length of the small bowel) from the last ileal loop (asterisk) to the site of the future anastomosis (blue arrow). (C,D) Small bowel rotation around the site of the future anastomosis (blue arrow). (E) Ileal enterotomy. (F) Duodenotomy (white arrow). (G-I) Side-to-side anastomosis in two layers between the duodenotomy (white arrow) and the ileal enterotomy (blue arrow). (J) Musculo-aponeurotic layer closure. (K) Skin closure. Please click here to view a larger version of this figure.

Figure 6: The SADI-S procedure learning curve. (A) The effect of training on the duration of the operation. Data are presented as the mean value ± SEM. (B) The effect of training on five-day survival. Data are presented as percentages. Please click here to view a larger version of this figure.

Figure 7: General parameters after SADI-S. (A) Postoperative body weight, (B) food intake measured for 24 h at day 14, and (C) blood hemoglobin concentrations were compared between SADI-S and sham control mice. Data are presented as the mean ± SEM. Statistical comparisons were made with two-way ANOVA (with Sidak's multiple comparisons test) or Mann-Whitney non-parametric tests. * p < 0.05; ** p < 0.01. Please click here to view a larger version of this figure.

Discussion

Bariatric surgeries, whose techniques are constantly evolving, appear to be currently the most effective treatment for obesity and associated metabolic comorbidities^3,19,20. The SADI-S procedure, firstly described in 2007⁴, is a promising procedure associated with greater metabolic effects than other malabsorptive surgeries. Animal models, particularly mice that allow the rapid generation of genetically modified models, are strongly needed to fully understand the mechanisms underlying these improvements. Here we describe a reliable and reproductible model of SADI-S in mice.

The first critical step of the SADI-S procedure is the exclusion of the duodenum, allowing only the bile and pancreatic secretions to travel into the duodenum and the first two-thirds of the small intestine. In humans, the duodenum is cut, allowing end-to-side duodeno-ileal anastomosis⁴. In the rat SADI-S model described by Montana et al.¹⁵, exclusion of the duodenum by a non-absorbable suture or surgical clamp is imperfect in a few cases, resulting in duodenal repermeabilization (i.e., reintroduction of the bolus into the original digestive tract). However, a section of the duodenum followed by end-to-side anastomosis is difficult to transpose in mice, leading us to prefer duodenum ligation. Indeed, the short length of the duodenal vessels limits the duodenal mobilization if the duodenum is completely transected, making it difficult to perform termino-lateral anastomosis. Initial experiments (data not shown) showed high mortality, even with trained and skilled experimenters. Only one case of repermeabilization has been observed in this study. Special attention must be given to the duodenal arteries during this step. Circumferential devascularization of the duodenum leads to death in all cases, but mice can be expected to recover from a small devascularized area caused by distal vessel ligation. The anatomical variability of the duodenal vascularization in mice prevents us from describing a constant localization to perform this exclusion. However, 0.5 cm of the duodenum after the pylorus must be available to permit the end-to-side anastomosis.

Another critical step when performing the anastomosis is to display the bowel in a way that the biliary limb comes to the site of the duodeno-ileal anastomosis from the left side. Otherwise, the food will oppose the bile flow, causing the biliary limb to distend, the bile to diffuse to the abdominal cavity, and the mice to die from biliary peritonitis around postoperative day 2. This condition resembling an afferent loop syndrome²¹ can be prevented by performing a loop of the small intestine centered on the anastomosed zone of the ileum. This is necessary because, contrary to humans, the caecum is positioned on the left side of the abdomen in 80% of cases in mice²².

In humans, the common limb measures approximately 250 cm to limit malnutrition, which corresponds to approximately one-third of the total length of the small bowel²³. Prior to surgeries, we measured the total length of the small intestine of the mouse model under similar feeding conditions (C57BL6/J under a chow diet) to determine the size of the common limb. As the small bowel length could vary between mice of different genetic backgrounds or following different feeding conditions, we strongly encourage future surgeons to perform a pilot study to measure the bowel size. The same size should be used for each mouse of the same background, as systematically exteriorizing the totality of the bowel for a complete measurement during surgery should be avoided (as there is increased risk of dehydration, hypothermia, and visceral injury).

The sleeve gastrectomy is part of the original SADI-S technique, allowing restriction in addition to malabsorption⁴. Several models of sleeve gastrectomy in mice are available in the literature^12,24,25,26. The use of surgical clips instead of sutures alone allows a significant gain of time²⁴ and reduces blood loss, two necessary conditions for surgical success. Anchoring the surgical clip using a 8-0 running suture prevented intragastric clip migration in all cases in our experiment. By removing the cardiac region, this technique allows the removal of about 80% of the stomach¹². In this model, however, SADI-S was associated with overfeeding compared with the sham control mice, aiming (probably) to compensate for the malabsorption caused by bowel derivation. Other models suggested that sleeve gastrectomy in mice preferentially modified the food intake behavior instead of the absolute quantity of food ingested per day in the long term^11,26. This limited restrictive effect is a limitation of this model.

This protocol has a 75% survival rate. It is worth noting that the 5-day survival was a strong predictor for long-term survival, as no late death occurred during our experiment. No anastomotic stenosis was observed. However, reaching this survival rate required at least 3 weeks of intensive microsurgical training by an experimenter specialized in animal surgery; the increased survival over time correlated with a decreased operating time. Perioperative care is one of the keys to the success of this technique. A strict analgesic protocol is needed in addition to systematic antibiotic-based therapy, and alimentation must be introduced progressively, using only a gel diet for 3 days. As previously described¹², supplementation with vitamins B1, B9, B12, and liposoluble vitamins (A, D, E, K) is necessary after malabsorptive surgery, as well as iron supplementation, which prevented anemia in our experiment, but has not been described for the SADI-S model yet¹⁵.

In conclusion, SADI-S can be transposed successfully in mice, with a few modifications from its description in humans. This technique requires training and a strict perioperative protocol. Adapting this surgery to mice could allow for a better understanding of the mechanisms underlying the strong metabolic effect of this promising procedure in comparison with former models and could help to better define its surgical indications.

Materials

Name	Company	Catalog Number	Comments
Agagani needle 26 G	Terumo	050101B	26 G needle
Betadine dermique	Pharma-gdd	3300931499787	Povidone solution
Betadine scrub	Pharma-gdd	3400931499787	Povidone solution
Binocular microscope	Optika Microscopes Italy	SZN-9	Binocular stereomicroscope
Buprecare	Animalcare	3760087151244	Buprenorphin
Castroviejo, straight 9 cm	F.S.T	12060-02	Micro scissors
Castroviejo, straight 9 cm	F.S.T	12060-02	Needle holder
Chlorure de sodium Fresenius 0.9%	Fresenius Kabi	BE182743	NaCl 0.9%
Clamoxyl	Med'vet	5414736007496	Amoxicilline
Cotton buds	Comed	2510805	Cotton swabs
Element HT5	Scilvet	Element HT5	Automated hematology analyzer
Emeprid	CEVA	3411111914365	Metoclopramid
Extra Fine Graefe Forceps, curved (tip width: 0.5 mm)	F.S.T	11152-10	Surgical forceps
Extra Fine Graefe Forceps, straight (tip width: 0.5 mm)	F.S.T	11150-10	Surgical forceps
Fercobsang	Vetoprice	QB03AE04	Iron, multivitamins and minerals
Forane	Baxter	1001936060	Isoflurane
Graefe forceps, straight (tip width: 0.8 mm)	F.S.T	11050-10	Forceps
Graphpad Prism version 8.0	GraphPad Software, Inc.	Version 8.0	Software for statistical analysis
Heat pad	Intellibio innovation	A-2101-00300	Heat pad
Incubator	Bioconcept Technologies	Manufactured on demand	Incubator
Lighting	Optika Microscopes Italy	CL-30	Lighting for microscopy
Ocrygel	Med'vet	3700454505621	Carboptol 980 NF
Pangen 2.5 cm x 3.5 cm	Urgovet	A02978	Haemostatic collagen compress
Prolene 6/0	B.Braun	3097915	Optilene 6/0 (0.7 metric) 75 cm 2XDR13 CV2 RCP, suture cord
Prolene 8/0	Ethicon	8732	2 x BV175-6 MP, 3/8 Circle, 8 mm,  suture cord
Scissors	F.S.T	146168-09	Surgical scissors
Sterile compresses	Laboartoire Sylamed	211S05-50	Non-woven sterile compressed
Terumo Syringe	Terumo	50828	1 mL syringe
Titanium hemostatic clip	Péters Surgical	B2180-1	Surgical clip
Vannas Wolff	F.S.T	15009-08	Micro scissors
Vita Rongeur	Virbac	3597133087611	Vitamin supplementation
Vitaltec stainless	Péters Surgical	PB 220-EB Medium	Surgical clip applier