Impact of Deserosalization on Small Bowel Anastomosis Healing in Swine: A Pilot Study

Materials and Methods

The Institutional Animal Care and Use Committee and the Veterinary Authorities of East Attica Region (Ref. number 2990) approved experiments described herein, under Greek Law No. 160, A-64, May 1991, which adheres to the Guiding Principles in the Care and Use of Animals approved by the Council of the American Physiological Society and complies with European Union regulations (EU Directive 2010/63/EU), ARRIVE guidelines, and the Guide for the Care and Use of Laboratory Animals (26).

Animal care and preparation. The seven female Landrace swine used had a weight range of 30-32 kg and were 4-5 months old. Animals were fed with a standard diet and observed for a minimum of 1 week to ensure a good state of health and acclimatization. After 12-hour fasting and ad libitum water access, the animals were prepared for the experimental operation.

Anesthesia and monitoring. All animals were initially sedated by intramuscular injection of ketamine hydrochloride (10 mg/kg), midazolam (0.5 mg/kg), and atropine (0.05 mg/kg). Subsequently, they were transferred to the operating table, induced to anesthesia with an i.v. bolus propofol (2 mg/kg) dose via the marginal auricular vein, and intubated with a cuffed endotracheal tube (4.0-5.0 mm). A volume-controlled ventilator (AM 5000 Major International, Alpha Delta Ventilator; SIARE, Bologna, Italy) was used with initial settings of 15 ml/kg tidal volume, 14 breaths/min, and 60% oxygen. The settings were adjusted to maintain normoxia and normocapnia. Continuous infusion of 2% propofol at 150 μg/kg/min and fentanyl citrate at 50 μg/kg/h maintained adequate anesthetic depth and analgesia. A bolus dose of cisatracurium (0.15 mg/kg) was repeated as needed. Vital signs monitoring was continuous. An i.v. dose of 750 mg cefuroxime was administered.

Surgery. Surgery was performed under aseptic and antiseptic techniques. The abdominal cavity was accessed via a 15 cm midline incision. During the abdominal exploration, the stomach, duodenum, and proximal jejunum were identified. Subsequently, 20 cm distally to the duodenum, three different types of jejuno-jejunal side-to-side anastomoses were performed twice at 20-cm distance from each other. Thus, each animal received six anastomoses in total. The three types of anastomoses were the following: A: No removal of the serosal layer (serosa-serosa), B: the serosal layer was removed only from the proximal jejunal helix before the anastomosis (serosa-subserosa), and C: the jejunal serosaI layer was removed from both the proximal and distal jejunal helices before the anastomosis (subserosa-subserosa). The type order (proximally to distally) of the six anastomoses per animal was defined according to the following rules: a) The three most proximal anastomoses were allocated to include all the possible permutations for the three types at least once; the permutation to be represented twice was randomly selected; b) the three most distal anastomoses were allocated to include all possible permutations for the three types at least once; the permutation represented twice was randomly selected; c) the combination of rules “a” and “b” to create the final order was random. The data regarding the order of types of anastomoses per animal were recorded.

The jejunal area of deserosalization had an elliptical shape with a 4-cm-wide major axis along the jejunal antimesenteric border and a 2-cm-wide minor axis (Figure 1A and B). A side-to-side jejuno-jejunal two-layer anastomosis was fashioned using 5-0 absorbable monofilament sutures (polydioxanone, PDS II; Ethicon Inc., Johnson & Johnson Medical Devices Companies, Somerville, NJ, USA). Interrupted sutures were used for the outer layer, and continuous suturing was used for the inner layer of the anastomosis. The external sutures of the dorsal anastomosis aspect were placed along the edge of the deserosalized area before the jejunal lumens of both the helices were opened (Figure 1C). Then a continuous inner suture was fashioned incorporating the submucosa and the muscle layer, in a 5 mm thickness and 5 mm step. Finally, outer sutures of the ventral anastomosis aspect were placed along the edge of the deserosalized area once again. The abdominal wall was closed with a single-layer of nylon 2-0 suturing. The skin was approximated with interrupted stitches.

Postoperative period. A liquid diet was resumed 6 h postoperatively, and was advanced to a free one at POD 2. On POD 2, blood samples were obtained for white blood cell count, C-reactive protein, albumin, and total protein levels. On POD 7, animals underwent explorative laparotomy. The abdomen was examined for any macroscopic evidence of a leak. Following the in vivo BP measurement, each anastomotic segment was resected en bloc and divided into two symmetrical parts on the antimesenteric plane. One part was fixed in 10% formaldehyde solution, and one was preserved in liquid nitrogen. Terminal procedures were performed under general anesthesia.

Bursting pressure measurements. The resistance of the intestinal wall to increasing intraluminal pressure until the disruption of the anastomotic line is defined as BP and can be measured in vivo or in vitro (24). In our study, the bowel was in vivo externally obstructed 3 cm above and below the anastomosis using 2 Overholdt clamps. Two 14-Fr catheters were inserted into the jejunal lumen and secured on the proximal and distal end of the anastomosis. The proximal catheter was used for the infusion of diluted water-soluble iodinated radiopaque contrast (Gastrografin; Bayer Hellas Commercial & Industrial S.A., Marousi, Athens, Greece) at a rate of 3 ml/min, under continuous videofluoroscopy. The distal catheter was used for pressure monitoring. The pressure at which contrast liquid was detected outside the enteral lumen was recorded as the anastomosis BP.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Presentation of the deserosalization technique. A and B: Schematic presentation of the place and size of the deserosalization area for each jejunal helix. C: In type B anastomosis, the serosal layer from only one side of the jejunal helix was removed. The black asterisk indicates the serosal surface; the white asterisk indicates the subserosa area; the arrowhead indicates the external sutures of the dorsal aspect of the anastomosis, which were placed along the edge of the deserosalized area before the jejunal lumen opening.

Tissue hydroxyproline concentration. Hydroxyproline is almost unique to collagen. Up to 14% of dry collagen weight is composed of hydroxyproline (27). Thus, the tissue concentration of hydroxyproline is used as an estimate of collagen content. Anastomotic tissue levels of hydroxyproline in the 42 liquid nitrogen-preserved samples were measured using an enzyme-linked immunosorbent assay (ELISA) kit (BG HYP ELISA Kit; BlueGene Biotech Co., Ltd., Shanghai, China). Cubic tissue specimens (0.5×0.5×0.5 cm) were excised from the anastomotic site, rinsed in ice-cold phosphate-buffered saline (pH 7.2; PBS), and weighed. The tissues were minced and homogenized in 0.5 ml PBS with a glass homogenizer on ice. The suspension was subjected to two freeze-thaw cycles, and the homogenates were centrifuged at 2,000 × g for 15 min. The total protein content of the supernatants was determined following Bradford's assay, while aliquoted samples kept at −80°C until analysis. The ELISAs were performed in triplicates, as per the manufacturer's instructions (limit of quantification= 0.003 nmol/ml, limit of detection=0.001 nmol/ml).

Histological evaluation. Five micrometer-thick sections were prepared for histological examination after dehydration and paraffin embedding. Sections were stained with hematoxylin-eosin stain (HES) and Masson's trichrome stain (MTS) and evaluated under a light microscope (Eclipse E200; Nikon Corporation, Tokyo, Japan) by two pathologists. The HES colors cell nucleus blue and other eosinophilic structures in various shades of red, pink, and orange. MTS colors collagen as blue or green, cytoplasm as red, and nuclei as black.

We used a modified scoring system (28) as a semi-quantitative method for the histological assessment. It included four parameters of acute inflammation (i.e. polymorphonuclear neutrophil infiltrate, edema, hemorrhage, and necrosis), as for the calculation of the total histological acute inflammation score (HAIS), and three parameters of tissue repair (i.e. fibroblast proliferation, collagen density, and neovascularization), as for the calculation of the histological repair score (HRS). The above-mentioned parameters were assessed using the following scoring system for each one: 0, None present/minimal; 1, mild; 2, moderate; 3, marked. The interpretation of collagen density was based on the intensity and depth of distribution within tissue sections according to the following scoring values: 0, none; 1, superficial; 2, superficial to mid-layers; 3, all layers. The HAIS (range=0-12) and HRS (range=0-9) were calculated for each specimen.

Statistical analysis. Statistical analysis was conducted either with the Statistical Package for the Social Sciences (SPSS, version 23.0; IBM, Armonk, NY, USA) or with GraphPad Prism v. 7 for Macintosh (GraphPad Software, San Diego, CA, USA). For normally distributed data, t-test or one-way ANOVA was used. The non-parametric alternatives Mann–Whitney U-test and Kruskal–Wallis test were performed when assumptions underlying parametric tests were violated. For all statistical analyses, a p-value of less than 0.05 was considered significant.