Experimental models of high-risk bowel anastomosis in rats: A systematic review

Abstract

BACKGROUND

Anastomotic leaks remain one of the most dreaded complications in gastrointestinal surgery causing significant morbidity, that negatively affect the patients’ quality of life. Experimental studies play an important role in understanding the pathophysiological background of anastomotic healing and there are still many fields that require further investigation. Knowledge drawn from these studies can lead to interventions or techniques that can reduce the risk of anastomotic leak in patients with high-risk features. Despite the advances in experimental protocols and techniques, designing a high-quality study is still challenging for the investigators as there is a plethora of different models used.

AIM

To review current state of the art for experimental protocols in high-risk anastomosis in rats.

METHODS

This systematic review was performed according to The Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. To identify eligible studies, a comprehensive literature search was performed in the electronic databases PubMed (MEDLINE) and Scopus, covering the period from conception until 18 October 2023.

RESULTS

From our search strategy 102 studies were included and were categorized based on the mechanism used to create a high-risk anastomosis. Methods of assessing anastomotic healing were extracted and were individually appraised.

CONCLUSION

Anastomotic healing studies have evolved over the last decades, but the findings are yet to be translated into human studies. There is a need for high-quality, well-designed studies that will help to the better understanding of the pathophysiology of anastomotic healing and the effects of various interventions.

Key Words: High-risk anastomosis, Rats, Experimental models, Bowel, Colon, Anastomotic leak, Colon resection, Inflammatory bowel disease, Intra-abdominal sepsis, Bursting pressure

Core Tip: Anastomotic leakage (AL) is a fatal complication after colorectal surgery, with high morbidity and mortality rates. AL rate is increased under emergency conditions. This review can be used as a tool to standardize and refine future research leading to studies that can be translated to human research regarding bowel anastomoses under complicated conditions.

INTRODUCTION

Colon diseases are among the most common disorders encountered by general surgeons, since more than 600000 surgical procedures are performed in the United States annually for management of colon-related disorders[1]. The most common indication for colorectal surgery is colon cancer, while other indications for resection include diverticular disease, ischemic colitis, stoma reversal or inflammatory bowel disease[2].

Despite the increased safety of colorectal surgery due to minimally invasive surgery and perioperative management advances, anastomotic leakage (AL) remains a fatal complication of colonic anastomosis, leading to increased morbidity and mortality rates, permanent stoma impaired oncological outcomes and poorer quality of life postoperatively[3,4]. AL rate varies by the level of anastomosis and is approximately 1%–3% for ileocolic anastomosis, 6%–12% for left colon anastomosis, and 3%–19% for colorectal anastomosis[5]. In general, risk factors associated with increased AL rate can be classified to preoperative, intraoperative and postoperative factors[6]. Among them, colonic resection and anastomosis in the emergency setting has been proven to be an independent risk factor for anastomotic dehiscence, as well as death after AL[7,8]. A recent prospective multi-centre study by the American Association for the Surgery of Trauma showed that anastomotic failure rate after emergent bowel resection and colo-colonic anastomosis had a failure rate of 23%, while in patients managed with an open abdomen the same rate was approximately 22%[9].

The traditional surgical dictum suggested that in emergency colectomy due to obstruction or peritonitis of large bowel origin, construction of a colostomy was imperative independently of the severity of peritonitis or the patient's condition[10]. Stomas are also associated with numerous early and late complications, as well as impaired Quality of Life and reduced rate of closure, when performed in the emergency setting[11]. However, recently emerged evidence propose the safety of anastomosis with diverting stoma under circumstances in cases of feculent or purulent peritonitis[12]. Decision regarding anastomosis in the setting of large bowel obstruction is mostly determined by the cause and the site of the obstruction[13]. In addition to peritonitis and obstruction, a series of other local and systemic conditions impair wound healing and render the construction of anastomosis perilous[14].

Animal experimental models constitute the basis of experimental study of colorectal anastomosis healing and permit monitoring of anastomotic healing with use of functional tests, clinical scores, molecular examination and histopathological examination[15]. Pommergaard et al[16], in their systematic review, evaluated the different experimental animal models that have been used for study of colorectal AL. Animal models reported in the literature include mice, pigs, rats, dogs and rabbits, with mice and pigs being the proposed by the authors experimental models for mimicking AL[16]. In addition, numerous studies have investigated the role of different potential therapeutic agents in healing of anastomosis in experimental models both under normal and pathological conditions, such as inflammation, peritonitis, obstruction, ischemia or jaundice[17]. However, pathophysiological mechanisms behind the formation of high-risk anastomoses for research aims have been scarcely evaluated and reported in the literature.

The aim of the present systematic review was to identify and classify types of experimental anastomosis that mimic high-risk colonic anastomosis in humans, in order to provide a guide for formation of standardized and easily reproducible experimental colonic anastomosis models and a translational basis for future clinical trials. The authors aspire that this review will be used as a guidance to facilitate future experimental studies, as it will give a comprehensive overview of current state of the art concerning the experimental protocols in high-risk anastomosis in rats. Experimental models to induce high-risk anastomosis are well described in the literature and will not be in the focus of the current review. On the other hand, an appraisal on the different methods of anastomotic quality assessment is lacking in the literature and will be attempted. This will hopefully help clarify questions that emerge during experimental protocol designing.

MATERIALS AND METHODS

Protocol and registration

This systematic review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, after approval of the study protocol by all authors[18]. A completed PRISMA checklist has been submitted.

Search strategy

To identify eligible studies, a comprehensive literature search (last search date as of 18/10/2023) was performed in the electronic databases PubMed (MEDLINE) and Scopus, covering the period from conception until 18 October 2023 (Supplementary material). Exported results were imported into Rayyan (Qatar Computing Research Institute, Doha, Qatar), and deduplication was performed[19]. Titles and abstracts were initially screened by two independent researchers (Georgios Ntampakis and Elissavet Anestiadou), and irrelevant articles were excluded. Potentially eligible full-text articles were screened for inclusion, according to the inclusion criteria and disagreements were resolved through discussion with participation of a third researcher (Orestis Ioannidis). In addition, a manual search was performed using the snowball methodology to identify and include any relevant studies in the list of references of the included articles.

Inclusion and exclusion criteria

Specific inclusion criteria were established prior to literature search: (1) Rats as experimental models; (2) English language; and (3) anastomotic healing as primary outcome. Studies were excluded if: (1) Were in vitro, human studies or studies in experimental models other than rats; (2) the definition of each outcome was not clearly stated; (3) incomplete information was reported; (4) they were kin studies; and (5) they were studies with no free access or no available full text.

Data extraction and synthesis

Two researchers (Georgios Ntampakis and Elissavet Anestiadou) reviewed all eligible studies. The following data were extracted and recorded: (1) First author; (2) year of publication and country of origin; (3) total number of animals used; (4) animal model (type, age, sex); (5) type of anastomosis; (6) type of intervention; (7) sampling day; (8) tests used for anastomosis assessment; and (9) characteristics of high-risk anastomosis.

Data analysis

Datasets were stratified based on the mechanism and type of high-risk anastomosis. High heterogeneity regarding anastomosis types, intervention types, method of assessment and reported outcomes of interest between the included studies rendered the conduction of a meta-analysis non-applicable. As a result, qualitative analysis of the outcomes was conducted. The provided results were assembled to identify strengths, weaknesses, and trends of each intervention.

RESULTS

As a result of our search strategy 102 studies were included in our systematic review and were categorized based on the mechanism used to create a high-risk anastomosis. The different categories of high-risk anastomosis were ischaemia, colitis, malnutrition, peritonitis, obstruction, radiotherapy, ischaemia/reperfusion injury, chemotherapy and immunosuppression. A flow diagram is shown in Figure 1.

Figure 1  The preferred reporting items for systematic reviews and meta-analyses flow diagram.

Ischaemia

Ischaemia models were some of the commonest high risk anastomosis models found. In total 15 different studies were included. Ischaemia was induced with ligation or electrocauterization of a wide segment of rat’s left colon just above the pubic symphysis. The different kinds of intervention attempted are shown in Table 1[20-40].

Table 1 Ischaemia models of high-risk anastomosis.

Ref.	Year	Sample size	Intervention	Test	Sampling day
Uzun et al[20]	2008	56	Sildenafil citrate	NOx, thiobarbituric acid reactive substances, glutathione	3/7
Karliczek et al[21]	2009	34	Visible light spectrometry evaluation	BPR in situ, tensile strength	3/7
Coneely et al[22]	2010	50	Ketotifen	BPR ex situ, HP, IL-6, VEGF	4
Karataş et al[23]	2010	24	Amelogenin	BPR ex situ, HP	4
Kaya et al[24]	2010	48	Tadalafil	BPR ex situ, HP, histology (as per Ehrlich et al[25])	4
Adas et al[26]	2011	40	Bm-MSCs 1 × 106	BPR in situ, HP	4/7
Karatepe et al[27]	2011	40	Adrenomedulin	BPR ex situ, HP, spectrophotometry, MDA, NOx, TNF-a, IL-6, VEGF, histology	3/7
Kennelly et al[28]	2011	30	Electrical field stimulation	BPR ex situ, HP, IL-6, VEGF	4
Sümer et al[29]	2011	30	Pentoxifylline and vinpocetine	BPR in situ, HP, histology (as per García et al[30])	5
Yoo et al[31]	2012	60	AdMSC 1 × 106	BPR ex situ, weight loss, macroscopic (adhesions as per van der Ham and Kort[32], strictures, ulcers), histology (as per Phillips et al[33]), wound infection, ileus, mortality	7
Wu et al[34]	2013	24	Triptolide	Histology, calprotectin, MPO, INF-g, IL-4, IL-17, TGF-b	56
Portilla-de Buen et al[35]	2014	180	Fibrin glue	BPR ex situ, macroscopic, microscopic, HP	5
Boersema et al[36]	2016	40	Hyperbaric oxygen	Macroscopic (as per Zühlke et al[37]), histology (as per Phillips et al[33]), serum creatinine	3/7
Ruiz-Luque et al[38]	2019	93	Alprostadil	BPR in situ, macroscopic (as per Knightly et al[39]), histology (as per García et al[30])	8
Kayapinar et al[40]	2021	60	CORM-2	BPR ex situ, HP, glutathione, MDA, histology (as per Ehrlich et al[25])	3/7

BmMSC: Bone marrow derived mesenchymal stem cells; AdMSC: Adipose tissue derived mesenchymal stem cells; CORM-2: Carbon monoxide releasing molecule-2; NOx: Nitric oxide; BPR: Bursting pressure; HP: Hydroxyproline; IL: Interleukin; VEGF: Vascular endothelial growth factor; MDA: Malondialdehyde; TNF-a: Tumor necrosis factor–alpha; MPO: Myeloperoxidase; INF-g: Interferon-gamma; TGF-b: Tumor growth factor-beta.

Various tests were used to assess the quality of the anastomosis, such as bursting pressure (BPR), Hydroxyproline levels (HP), different cytokines, oxidative stress markers, macroscopic and microscopic assessment using different protocols.

Colitis

In total, 6 different studies of anastomosis in colitis environment were included. In these models, Dextran sodium sulphate or 2,4,6-Trinitrobenzene sulfonic acid were used to induce acute colitis. In one study, intra-jejunal injection of iodoacetamide had been used. The different kinds of interventions can be found in Table 2[41-50]. The tests used to assess the quality of the anastomosis were: Anastomotic BPR, HP, different cytokines, oxidative stress markers, macroscopic and microscopic assessment using different protocols.

Table 2 Colitis models of high-risk anastomosis.

Ref.	Year	Colitis method	Sample size	Intervention	Test	Sampling day
Kirkil et al[41]	2008	Intra-jejunal injection of iodoacetamide	28	Endothelin receptor blockade by bosentan	BPR in situ, macroscopic (Mannheim index[42]), HP, histology (as per Mei et al[43])	4
Rijcken et al[44]	2010	7 d DSS 5%	Not mentioned	rhIGF-1	BPR in situ, microscopic (as per Phillips et al[33], Ki-67), HP, MPO	1/3/7
Myrelid et al[45]	2015	5 d DSS 3%	140	IP prednisolone, AZA, infliximab	BPR ex situ, bowel WT/length, histology, zymography	3
Alvarenga et al[46]	2019	2,4,6-TNBS	66	AdMSC 2 × 106	Histology, IL-10, IL-17, IFN-g, TGF-b, TNF-a, MMP-2, MMP-9	7
Reischl et al[47]	2021	ANXA-1 k/o mice and 7 d DSS 2%-3%-5%	Not mentioned	Ac2-26-nanoparticles	Fluorescence imaging of MMPs, histology (as per Phillips et al[33]), whole transcriptome RNA sequencing and analysis	3/7
Weber et al[48]	2023	7 d DSS 2%	84	Prednisolone	BPR in situ, macroscopic (adhesions scoring), histology (as per Philips et al[33])	3/7
Ntampakis et al[49]	2023	7 d DSS 5%	24	AdMSC 5 × 106	BPR ex situ, macroscopic (as per Bosmans et al[50]), HP, IL-6, TNF-a, VEGF	7

DSS: Dextran sodium sulphate; TNBS: 2,4,6-trinitrobenzene sulfonic acid; rhIGF-1: Recombinant human insulin-like growth factor-1; IP: Intra-peritoneal; AZA: Azathioprine; AdMSC: Adipose tissue derived mesenchymal stem cells; BPR: Bursting pressure; HP: Hydroxyproline; IL: Interleukin; VEGF: Vascular endothelial growth factor; TNF-a: Tumor necrosis factor-alpha; MPO: Myeloperoxidase; INF-g: Interferon-gamma; TGF-b: Tumor Growth Factor–beta; MMP: Metalloproteases.

Malnutrition

In total, 6 different studies of anastomosis in malnourished rat were included. Starvation for 7-15 d or 50% food restriction for 28 d were the different methods used to induce malnutrition. The different kinds of interventions attempted, can be found in Table 3[51-58]. The tests used to assess the quality of the anastomosis and animal status were BPR, tensile strength, body weight changes, nutrition markers, macroscopic and microscopic assessment using different protocols.

Table 3 Malnutrition models of high-risk anastomosis.

Ref.	Year	Colitis method	Sample size	Intervention	Test	Sampling day
McCauley et al[51]	1991	3,5% agar diet for 7 d	30	BCAA	Body WT, BPR ex situ, tensile strength, protein content, HP	-
Karahasanoglu et al[52]	1998	Low-protein diet for 10 d	40	Growth hormone	Body WT, BPR in situ, hydroxyproline	4
Salman et al[53]	2008	15 d	72	Cholerella sp. microalgae	Body WT, macroscopic (adhesions as per Ham et al[32]), BPR in situ, HP, histology (as per de Roy van Zuidewijn et al[54]), albumin, prealbumin, transferrin	3/5/7/9/11/13/15
Gonçalves et al[55]	2009	50% food restriction for 21 d	80	Pre-op nutrition	Tensile strength, histology	5
Gündoğdu et al[56]	2015	50% food restriction for 26 d	18	Pre-op nutrition	Body WT, BPR in situ, HP	4
Vizzotto Junior et al[57]	2015	Paired feeding	160	Omega-3	Body WT, tensile strength, histology	5
Danielski et al[58]	2016	50% food restriction for 26 d	45	Vitamin C	Body WT, macroscopic (as per Knightly et al[39]), histology, HP, MPO, TNF-a, nitrite/nitrate, oxidative damage	7

WT: Weight; BPR: Bursting pressure; HP: Hydroxyproline; TNF-a: Tumor necrosis factor–alpha; MPO: Myeloperoxidase; BCAA: Branch chained amino acids.

Obstruction

In total, 6 different studies of anastomosis in rats with obstruction were included. Obstruction methods were silk ligation of the distal colon or use of a silicone ring to mimic decrease in bowel diameter. There was also a study with obstructive jaundice, where the distal common bile duct was tied. Krarup et al[59] introduced a new method of inducing bowel obstruction by laparoscopic clip application to the colon. The different kinds of interventions attempted, can be found in Table 4[60-65]. The tests used to assess the quality of the anastomosis were BPR, tensile strength, hydroxyproline, body weight changes, macroscopic and microscopic assessment using different protocols.

Table 4 Obstruction models of high-risk anastomosis.

Ref.	Year	Obstruction method	Sample size	Intervention	Test	Sampling day
Törnqvist et al[60]	1990	Silicone ring 6.5 mm	54	Diverting colostomy	Tensile strength, weight, HP	2/7
Aguilar-Nascimento et al[61]	1997	Silk ligature	108	Nutritional solutions lavage of the colon	Macroscopic, histology	3/6
Erbil et al[62]	2000	Silk ligature	144	Nutritional solutions lavage of the colon	BPR ex situ, HP, macroscopic	3/6
Cağlikülekçi et al[63]	2002	Common bile duct ligation	40	rGH	BPR ex situ, macroscopic, HP, macroscopic, histology (as per Greenhalgh et al[64])	7
Lelyanov et al[65]	2004	Silk ligature	60	Sodium hypochlorite and ozone therapy	BPR ex situ, survival, macroscopic, histology	1/3/6/9/12
Krarup et al[59]	2017	Laparoscopic clip application	32	MMP inhibition	Body WT, HP, BPR in situ	3

HP: Hydroxyproline; WT: Weight; BPR: Bursting pressure; MMP: Metalloproteases; rGH: Recombinant growth hormone.

Peritonitis

In total, 29 different studies of anastomosis in rats with peritonitis were included. The different techniques to induce acute peritonitis were caecal ligation and puncture which was the commonest technique used, incomplete anastomosis, as well as faecal inoculation in the abdominal cavity. In one of the studies, Vaneerdeweg et al[66] used barium sulphate and gelatine sponges with faeces to mimic bacterial and chemical peritonitis. The different kinds of interventions attempted, can be found in Table 5[66-100]. The tests used to assess the quality of the anastomosis were BPR, tensile strength, hydroxyproline, cytokines, tPA activity, oxidative stress markers, body weight changes, macroscopic and microscopic assessment using different protocols.

Table 5 Peritonitis models of high-risk anastomosis.

Ref.	Year	Peritonitis method	Sample size	Intervention	Test	Sampling day
Vaneerdeweg et al[66]	2000	Gelatin capsule with faeces and barium sulphate	40	Gentamicin Sponges	BPR in situ, mortality, weight loss	4
Reijnen et al[67]	2002	Caecal ligation/perforation	198	Hyaluronan-based agents	tPA activity	1/3/7
Aydin et al[68]	2006	Caecal ligation/perforation	24	Laparostomy with Bogota bag	BPR in situ, HP, adhesions (as per Zühlke et al[37])	5
Li et al[69]	2006	Enterotomy	360	Fibrin glue and growth hormone	BPR in situ, HP, tensile strength, histology	1/3/5
Buyne et al[70]	2009	Faecal inoculation	148	Recombinant tPA	BPR in situ, tensile strength, macroscopic, HP	3/7
Kayaoglu et al[71]	2009	Caecal ligation/perforation	80	N-butyl-2-cyanoacrylate	BPR ex situ, macroscopic (as per Knightly et al[39]), histology	3/7
Pantelis et al[72]	2010	Caecal ligation/perforation or incomplete anastomosis	206	Collagen matrix coagulation factors I and IIa (Tachosil)	BPR ex situ, histology (as per Biert et al[73], Verhofstad et al[74], Attard et al[75]), collagen type I-II, HP	2/5/14
Rocha et al[76]	2010	Caecal ligation/perforation or incomplete anastomosis	45	Hyperbaric oxygen therapy	Total energy rupture test (tensile strength)	4
Silva et al[77]	2012	Caecal ligation/perforation or incomplete anastomosis	40	Bromopride	Macroscopic (Nair et al[78]), tensile strength (Versa test), histology, quantitative collagen analysis, HP	3/7
Holmer et al[79]	2014	Faecal inoculation	72	Collagen fleece coating	BPR in situ, histology, collagen I, III, VEGF, MMP-13	1/3/7
Camargo et al[80]	2013	Faecal inoculation	40	Peritoneal lavage with bupivacaine	Tensile strength, survival	5
Arikanoglu et al[81]	2013	Colon injury	21	Antibacterial suture	BPR in situ, HP, histology	10
Donmez et al[82]	2013	Colon injury	40	Glutamine and GH	BPR ex situ, HP	5
Senol et al[83]	2013	E. Coli inoculation	40	Fibrin glue	BPR ex situ, histology (as per Ehrlich-Hunt[25]), HP	10
Silva et al[84]	2014	Caecal ligation/perforation	80	Bromopride	MMP-1a, MMP-8, MMP-13, IL-1b, IL-6, IL-10, TNF-a, IFN-g	3/7
Erginel et al[85]	2014	Caecal ligation/perforation	40	IP O3	BPR ex situ, histology (as per Verhofstadt et al[74] and Philips et al[33]), HP	7
Pommergaard et al[86]	2014	Incomplete anastomosis	80	Tachosil coating	Tensile strength, clinical assessment	7
Ercan et al[87]	2015	Caecal ligation/perforation	40	IP L-Carnitine	BPR ex situ, histology (as per Ehrlich- Hunt[25]), HP	5
Cakir et al[88]	2015	Incomplete anastomosis	64	Sildenafil	BPR ex situ, histology (as per Phillips et al[33]), HP, MDA, GSH	3/7
Suárez-Grau et al[89]	2016	Incomplete anastomosis	56	Fibrinogen - thrombin collagen patch	Histology (as per Biert scheme[73]), macroscopic (adhesions), survival	15/30
Sozutek et al[90]	2016	Caecal ligation/perforation	50	PRP	BPR in situ, Body WT, HP, histology (as per Verhofstadt et al[74])	7
Ersoy et al[91]	2016	Caecal ligation/perforation	60	Melatonin	BPR ex situ, HP, histology, IL-6, IL-10, INF-γ, CRP	7
Çakır et al[92]	2016	Caecal ligation/perforation	18	IP O3	BPR ex situ, TNF-a, IL-1β, MDA, MPO, histology	22
Sukho et al[93]	2018	Incomplete anastomosis	60	AdMSC	BPR in situ, macroscopic (as per Verco et al[94] and Zühlke et al[37]), histology	3/7
Lorenzi et al[95]	2017	Caecal ligation/perforation	40	Omiganan	BPR in situ, histology (as per García et al[30]), HP	7
Yıldırım et al[96]	2021	Caecal puncture	21	Growth factor collagen (FGF-C), abx collagen (AB-C)	BPR ex situ, HP, macroscopic (as per Bosmans et al[50]), histology (as per Ehrlich et al[25])	7
Nakamura et al[97]	2021	Incomplete anastomosis	60	HSMM	BPR ex situ, macroscopic (as per Ham et al[32]), histology	3/5/7/14/28
Aksu et al[98]	2021	Colon injury	21	Chlorhexidine gluconate and metronidazole-soaked sponges	BPR in situ, hydroxyproline, histology	10
Yilmaz et al[99]	2021	Caecal ligation/perforation	32	Polyurethane membrane	BPR ex situ, HP, NOx, IL-6, TNF-a, tPA, macroscopic (as per Mazuji et al[100]), histology	5

BPR: Bursting pressure; WT: Weight; tPA: Tissue plasminogen activator; HP: Hydroxyproline; VEGF: Vascular endothelial growth factor; MMP: Metalloproteases; IL: Interleukin; MDA: MDA: Malondialdehyde; GH: Growth hormone; TNF-a: Tumor necrosis factor-alpha; IFN-g: Interferon-gamma; IP: Intra-peritoneal; GSH: Glutathione; MPO: Myeloperoxidase; AdMSC: Adipose tissue derived mesenchymal stem cells; HSMM: Human skeletal muscle myoblast; NOx: Nitric oxide.

Radiotherapy

In total, 8 different studies of anastomosis after radiotherapy in rats were included. In each protocol different doses of radiation were used depending on the nature of the experiment. The different kinds of interventions attempted, can be found in Table 6[101-109]. The tests used to assess the quality of the anastomosis were BPR, hydroxyproline, oxidative stress markers, body weight changes, macroscopic and microscopic assessment using different protocols. Van de Putte et al[107] used positron emission tomography/computedtomography to investigate the tropism of the AdMSCs to the anastomosis along with colonoscopy for direct assessment of anastomotic site and histological examination

Table 6 Radiation models of high-risk anastomosis.

Study	Year	Radiation dose	Sample size	Intervention	Test	Sampling day
Liu et al[101]	2001	10 Gy	74	Lactobacillus plantarum 299v	Body weight, WBC, mucosal MPO, HP, nucleotide, DNA and RNA content, colonic bacterial microflora, bacterial translocation, histology	4/7/11
Kerem et al[102]	2006	500 cGy	84	Soluble Fiber	Macroscopic (adhesions as per van der Ham and Kort[32]), BPR in situ, HP, histology (as per de Roy van Zuidewijn et al[54]), MMP-2 activity	3/7
Ozdemir et al[103]	2013	800 rad	30	Amifostine	BPR ex situ, HP, histology	5
Seker et al[104]	2014	485 cGy	60	Pycnogenol	BPR ex situ, HP, MDA, histology (as per Houdart et al[105])	3/7
Simões Neto et al[106]	2013	660 cGy	30	Fraction electron beam	BPR (not specified), histology	7
Van de Putte et al[107]	2017	27Gy	48	AdMSC	18F-FDG-PET/CT, colonoscopy, histology	32
Taşdöven et al[108]	2019	6Gy	48	Ozon PR	BPR in situ, histology (as per Houdart et al[105]), HP, MDA, MPO	3/7
Yilmaz et al[109]	2022	20Gy	32	Ozon PR	BPR in situ, macroscopic (as per Knightly et al[39]), HP, MPO, histology (as per de Roy van Zuidewijn et al[54])	5

AdMSC: Adipose tissue derived mesenchymal stem cells; PR: Per rectum; MPO: Myeloperoxidase; HP: Hydroxyproline; MMP: Metalloproteases; BPR: Bursting pressure; MDA: Malondialdehyde.

Ischaemia/reperfusion injury

In total, 10 different studies of anastomosis after ischaemia injury in rats were included. In all protocols superior mesenteric artery was clamped for 30-45 mins. In each protocol different doses of radiation were used depending on the nature of the experiment. The different kinds of interventions attempted, can be found in Table 7[110-124]. The tests used to assess the quality of the anastomosis were BPR, hydroxyproline, oxidative stress markers, cytokines, macroscopic and microscopic assessment using different protocols.

Table 7 Ischaemia/reperfusion models of high-risk anastomosis.

Ref.	Year	I/R method	Sample size	Intervention	Test	Sampling day
Terzi et al[110]	2001	SMA clampling 30 min	65	Allopurinol	BPR in situ, macroscopic (adhesions as per Knightly et al[39]), histology (as per de Roy van Zuidewijn et al[54])	3/7
Tireli et al[111]	2003	SMA clampling 30 min	20	Pentoxifiline	BPR ex situ, HP	7
Miranda et al[112]	2010	SMA clamping 45 min	45	Methylene blue	BPR ex situ, macroscopic, histology	7
Celik et al[113]	2013	SMA clamping 45 min	24	Modelukast	BPR ex situ, HP, MPO, MDA, caspase-3 activity, catalase, NOx, glutathione, SOD, TNF-a, IL-6, ALT, AST	5
Akarsu et al[114]	2017	SMA clamping 10 min	40	Simvastatin	BPR ex situ, HP	8
Özkan et al[115]	2018	SMA clamping 30 min	30	Melatonin	BPR ex situ, HP, histology (as per Nursal et al[116]), SOD, glutathione	7
Özçay et al[117]	2018	SMA clamping 45 min	40	GH	Macroscopic (as per Galili et al[118]), BPR ex situ, histology (as per Greenhalgh et al[64])	7
Sayin et al[119]	2020	SMA clamping 45 min	40	IP montelukast	BPR ex situ, macroscopic score (as per Knightly et al[39]), HP, histology (fibrosis)	7
Eryilmaz et al[120]	2020	SMA clamping 45 min	30	Hydrogen rich saline	BPR in situ, histology (as per Park et al[121] and Chiu et al[122]), TNF-a, IL-6, MDA, MPO	5
Akıncı et al[123]	2022	SMA clamping 30 min	36	Genistein	BPR ex situ, HP, SOD, glutathione, histology (as per Piroglu et al[124])	5

SMA: Superior mesenteric artery; GH: Growth hormone; IP: Intra-peritoneal; SOD: Superoxide dismutase; BPR: Bursting pressure; HP: Hydroxyproline; MPO: Myeloperoxidase; MDA: Malondialdehyde; IL: Interleukin; TNF-a: Tumor necrosis factor-alpha; I/R method: Ischaemia/reperfusion method.

Chemotherapy

In total, 9 different studies of anastomosis after different schemes of chemotherapy in rats were included. Chemotherapy includes cytotoxic agents against cancer cells and was administered IV or with intra-peritoneal infusion. The different kinds of interventions attempted, can be found in Table 8[125-133]. The tests used to assess the quality of the anastomosis were BPR, tensile strength hydroxyproline, oxidative stress markers, cytokines, macroscopic and microscopic assessment using different protocols.

Table 8 Chemotherapy models of high-risk anastomosis.

Ref.	Year	Chemo agent	Sample size	Intervention	Test	Sampling day
Nayci et al[125]	2003	IP 5-FU	40	Electromagnetic field	Tensile strength, HP	7
Cetinkaya et al[126]	2005	IP mitomycin-C	81	GM-CSF	BPR ex situ, HP, histology (as per Ehrlich et al[25])	3
Kanellos et al[127]	2006	IP 5-FU and LEV	60	Fibrin glue	BPR ex situ, HP, macroscopic (adhesions as per van der Ham and Kort[32]), histology (as per Phillips et al[33])	8
Yildiz et al[128]	2013	5-FU and 20 Gy	60	HBOT	BPR ex situ, Weight, HP, histology (Fibrosis)	5
Arapoglou et al[129]	2017	Irinotecan	40	Iloprost	BPR ex situ, macroscopic (as per van der Ham and Kort[32]), histology (as per Phillips et al[33]), HP	8
Akyuz et al[130]	2018	5-FU IV	32	Melatonin	BPR ex situ, HP, histology, TNF-a, IL-1β	7
Ocak et al[131]	2019	HIPEC with CIS	30	PRP	BPR ex situ, HP, histology (as per Verhofstad et al[74])	7
Gorur et al[132]	2020	IP 5-FU	40	PRP	Body weight, BPR in situ, HP, histology (as per Verhofstad et al[74])	7
Buk et al[133]	2020	IP OX	30	PRP	BPR ex situ, histology (as per Verhofstad et al[74]), HP	7

IP: Intra-peritoneal; 5-FU: 5-Fluorouracil; LEV: Leucovorin; HIPEC: Hyperthermic intra-peritoneal chemotherapy; CIS: Cis-platina; OX: Oxaliplatin; GM-CSF: Granulocyte–macrophage colony stimulation factors; HBOT: Hyperbaric oxygen treatment; PRP: Platelet rich plasma; BPR: Bursting pressure; HP: Hydroxyproline; IL: Interleukin; TNF-a: Tumor necrosis factor-alpha.

Immunosuppression

In total, 6 different studies of anastomosis after different schemes of chemotherapy in rats were included. Either steroids or other immunosuppression drugs were used, depending on the protocol. The different kinds of interventions attempted, can be found in Table 9[134-140]. The tests used to assess the quality of the anastomosis were BPR, tensile strength hydroxyproline, oxidative stress markers, cytokines, macroscopic and microscopic assessment using different protocols.

Table 9 Immunosuppression models of high-risk anastomosis.

Ref.	Year	Immunosuppression method	Sample size	Intervention	Test	Sampling day
Dinc et al[134]	2002	Methylprednizolone	80	GM-CSF	BPR ex situ, HP, histology (as per Ehrlich et al[25])	3
Colak et al[135]	2003	Dexamethasone	24	Trapidil	BPR ex situ, HP, histology (as per Ehrlich et al[25]), NOx, MDA	7
Sakallioglu et al[136]	2004	Dexamethasone	60	eGF	BPR ex situ, bursting site, HP, histology	7
Inglin et al[137]	2008	MMF	63	IGF-I	BPR ex situ, histology, Ki-67	2/4/6
Netta et al[138]	2014	Methylprednizolone	50	SCFA	BPR ex situ, CRP, IL-6, TNF-a	4
Karakaya et al[139]	2021	Everolimus	60	AdMSC	Macroscopic (as per Houston and Rotstein[140]), BPR ex situ, HP, histology	4/7

MMF: Mycophenolate mofetil; GM-CSF: Granulocyte-macrophage colony stimulation factors; eGF: Epidermal growth factor; IGF-I: Insulin-like growth factor; SCFA: Short chain fatty acids; AdMSC: Adipose tissue derived mesenchymal stem cells; BPR: Bursting pressure; HP: Hydroxyproline; NOx: Nitric oxide; MDA: Malondialdehyde; IL: Interleukin; TNF-a: Tumor necrosis factor-alpha.

DISCUSSION

Bursting pressure

One of the most common tests used to assess the quality of anastomosis is anastomotic BPR. In the literature, there are 2 ways of performing BPR test, either in situ (in vivo) or ex situ (in vitro). In vivo, a catheter is inserted into the rat’s rectum, dyed water is infused, and the manometer is attached more proximal to the anastomosis, recording the maximum pressure in which the anastomosis bursts. In vitro, the anastomosis is dissected away from the rat, is tied distally, and is attached to a three-way system with the manometer on one side, and the syringe with the dyed water on the other. Water is infused in the bowel segment, and maximum pressure in which the anastomosis bursts is recorded. Curran et al[141] compared the two techniques in canine small bowel and they concluded that the in vitro technique had similar results compared to the in vivo one, but it was easier to perform as the researchers do not have to carry out intensive bowel dissection. In one of our team’s previous studies we described the in vitro technique in detail[49]. Some technical pearls of the in vitro technique are that it is slightly more time consuming, and it requires careful and meticulous dissection as the adhesions formed around the anastomosis might be the factor that keeps the anastomosis patent, and extensive dissection might result in anastomotic dehiscence, rendering the specimen invalid.

Along with BPR, Sakallioglu et al[136] also documented the bursting site of the anastomosed bowel, and they found out that it is usually around the anastomosis and not on the anastomosis itself.

Tensile strength

Tensile strength is traditionally used along with BPR to assess the strength of the anastomosis. The technique includes dissecting the anastomosis out and attaching it to a device which allows application of traction force to one end of the anastomosis and recording of the force applied to the apparatus on the other end. The force at which the anastomosis is disrupted is then recorded. Either a simple commercial dynamometer[142] or more precise and expensive solutions can be used[77].

Ikeuchi et al[143] reported that there is no correlation between anastomotic bursting strength and tensile strength of anastomosis, and both tests should be used in assessing the anastomotic quality. They also suggest that minimum strength in which the anastomosis starts to rupture and maximum strength in which the anastomosis is completely ruptured should be documented.

Macroscopic assessment

Macroscopic assessment consists of both clinical observations of the rats, as well as macroscopic assessment of the anastomosis using different scales to grade it, depending on what parameters need to be observed.

Clinical parameters used, especially in malnutrition models, are weight changes of the rats, and survival curves. The general welfare of the animals, while easily appreciated by direct observation, can be considered but is not easily countable. Examples of clinical parameters are reduced mobility, fur erection, neglection of hygiene and reduced food intake. Specifically in colitis protocols, bloody diarrhea, and rectal mucosa erythema can be easily observed and colonoscopy can be used to verify intestinal inflammation before starting the experimental process[46,49].

Van de Putte et al[107] used colonoscopy to directly assess anastomotic healing internally.

Macroscopic assessment scores and what they assess can be found in Table 10. Of note, although the score by van der Ham and Kort[32] is one of the widest used for adhesion formation, van der Ham cite Houston and Rotstein[140] as the original creators of the same scoring system.

Table 10 Macroscopic assessment of the anastomosis.

Macroscopic assessment	Wound healing
van der Ham and Kort[32] score	Adhesions
Zühlke et al[37] score for adhesions	Adhesions
Mannheim index[42]	Peritonitis presence and severity
Knightly et al[39] score for adhesions	Adhesions
Bosmans et al[50] score	Anastomotic complication score
Nair et al[78] score	Adhesions
Verco et al[94] score	Abscess formation
Mazuji et al[100] score	Adhesions
Galili et al[118] score	Adhesions
Houston and Rotstein[140] score	Adhesions

All the adhesions scoring systems are similar, and all of them assess the existence or severity of the adhesions using different criteria. The score that we considered to be more complete for the assessment of a bowel anastomosis in rats is the one created by Bosmans et al[50] in an international consensus statement. This score takes into account the presence of adhesions, abscesses, anastomotic dehiscence underneath adhesions, as well as faecal peritonitis/death of the animas.

Histologic assessment

One of the most interesting findings of the current review are the different methods used in research to assess the quality of the anastomosis. As described in Table 11[144], all models have common characteristics, such as the presence of inflammatory cells in bowel tissues, fibroblastic activity, neovascularization, and collagen deposition which play a pivotal role in anastomotic healing. The oldest and most widely used model with the above characteristics is this of Ehrlich et al[25] in 1973 and later on the same model as modified by Phillips et al[33]. in 1992, both of which set the basis for histological assessment of anastomotic healing.

Table 11 Histologic assessment of the anastomosis.

Ref.	Histologic assessment
Ehrlich et al[25]	Erythrocytes, polymorphonucleated cells, mononuclear cells, fibroblasts, collagen fibers, fibrin
Houdart et al[105] and Hutschenreiter et al[144] as modified by García et al[30]	Mucosal anastomotic re-epithelialization, neovascularization, fibroblasts, fibrosis, muscle layer destruction, neutrophil infiltration, lymphocyte infiltration, histiocyte infiltration, giant cell infiltration
Ehrlich et al[25] as modified by Phillips et al[33]	Inflammatory cell infiltration, blood vessel in growth, fibroblast ingrowth, collagen deposition
Houdart et al[105]	Granulocyte infiltration, mononuclear cell infiltration, fibroblastic proliferation, focal necrosis, exudate formation
Mucosal damage index by Mei et al[43]	0: Normal mucosa, no damage on mucosal surface; 1: Mild hyperemia and edema, no erosion or ulcer on mucosal surface; 2: Moderate hyperemia and edema with erosion on mucosal surface; 3: Severe hyperemia and edema with necrosis and ulcer on mucosal surface, the major ulcerative area < 1 cm; 4: Severe hyperemia and edema with necrosis and ulcer on mucosal surface, the major ulcerative area > 1 cm
de Roy van Zuidewijn et al[54]	Re-epithelialization, regeneration of muscularis propria, mucosal muscularis propria damage, necrosis, inflammatory exudate, granulation tissue, granulocytes, macrophages, fibroblasts, granulation tissue
Greenhalgh et al[64]	Epithelization, cellular infiltration, fibroblastic proliferation, collagen deposition, neovascularization
Biert et al[73]	Necrosis, polymorphonuclear cells, lymphocytes, macrophages, edema, epithelium, submucosal - muscular continuity, neovascularization, fibrosis
Attard et al[75]	Mucosal continuity, muscular continuity, re-epithelialization, granulation tissue, polymorphonuclear cells, lymphocytes, macrophages, fibroblasts
Verhofstad et al[74]	Necrosis, polymorphonuclear cells, lymphocytes, macrophages, edema, mucosal continuity, submucosal – muscular continuity
Nursal et al[116]	Fibroblast infiltration, capillary formation, re-epithelialization, granulocyte infiltration, mononuclear cell infiltration
Park et al[121] and Chiu et al[122]	Grade 1: Normal mucosa; grade 2: The subepithelial space at the tip of the villus; grade 3: Increase in subepithelial space; grade 4: Overlapping and spills of the floor of the villus; grade 5: Disintegration of the lamina propria; grade 6: Crypt layer injury; grade 7: Transmucosal infarction and grade 8: Transmural infarction
Piroglu et al[124]	Inflammatory cell infiltration/concentration, neovascularization, fibroblastic activity, collagen fibers
Miltschitzky et al[15]	Blood vessel ingrowth, fibroblasts, collagen formation, inflammatory cell infiltration, first layer in which continuity has been restored, number of healed layers, epithelium closed, crypt architecture restored, overall healing quality

The appreciation of healing layer by layer is added by de Roy van Zuidewijn et al[54] in 1992 who added the elements of re-epithelialization, muscularis propria damage and regeneration and the presence of necrosis. Their model, despite providing information on the layer-by-layer healing and inflammatory changes of the tissues, does not address the deposition of collagen or fibroblastic activity. Other models, proposed by different teams provide a more complete approach to anastomotic healing by adding the element of connective tissue regeneration as well[73-75].

As opposed to the models described above that have a semi-quantitative approach, Park et al[121] and Chiu et al[122] suggested a more qualitative approach, with grading of mucosal layer healing. This model is used to evaluate the anastomosis after ischemia reperfusion type injury[121,122].

Miltschitzky et al[15] have a very different approach in their histology grading system which includes neovascularization, fibroblastic activity and collagen formation, inflammatory cell infiltration and a more extensive layer by layer evaluation of the anastomosis in semi-quantitative way[15]. In the authors’ view this represents the most holistic approach to anastomotic healing, is easy to use and is applicable in different types of anastomosis research.

In a few cases, as shown in Tables 1-9, researchers did mention histological evaluation in their research but they did not use any of the histological grading models described above.

Collagen formation/degradation assessment

Hydroxyproline has proven to be one of the commonest markers used in experimental protocols of anastomotic healing. With hydroxyproline we indirectly assess collagen content of anastomosis and appreciate anastomotic healing, with higher values of hydroxyproline suggesting enhanced anastomotic healing.

Matrix Metalloproteases (MMPs) and their inhibitors (TIMPs) play an important role in wound healing as they play an important role in collagen degradation, neovascularization and are regulated by chemokines and cytokines[145].

The selection and investigation of different kinds of MMPs and TIMPs depends on the nature of the research and their objectives. In the current study we identified a few authors investigating the role of different MMPs (MMP-1, MMP-8, MMP-13) that belong to the groups of collagenases which degrade triple helical collagen and MMP-2, MMP-9 gelatinases which are involved in the processes of angiogenesis and collagenesis[46,79,84,145]. Abnormal expression of MMPs can impair anastomotic healing and can be used as biomarkers in anastomosis research to evaluate the efficacy of different interventions or the effect of a condition/factor on an anastomosis.

Cytokine studies

Different kinds of cytokines can be used in research of high-risk anastomotic healing as they provide us with valuable information on the biological processes during anastomotic healing or in response to an intervention to a high-risk anastomosis. Cytokines can be measured either with ELISA or polymerase chain reaction according to local laboratory protocols.

Some of the cytokines used in research are interleukin (IL)-1b, IL-4, IL-6, IL-17, interferon gamma, tumor necrosis factor α, as pro-inflammatory cytokines to assess the severity of inflammatory response to the anastomosis after applying the stimulus and appreciating their fluctuation after the intervention[46,49]. On the other hand, increased expression of IL-10 and tumor growth factor–beta (TGF-b) which are known as anti-inflammatory cytokines can be used as a marker of effectiveness of experimental intervention. TGF-b as reported by Alvarenga et al[46] also seems to regulate the expression of certain MMPs leading to fibrosis[46].

Vascular endothelial factor is one of the cytokines that can be used to assess the anastomosis for neo-vascularization. Increased values of vascular endothelial growth factor (VEGF) will suggest increased vascularization in the anastomosis which is important for anastomotic healing. In early stages of anastomotic healing VEGF might not be significantly increased but our group showed tendency to increase in post-operative day 7 in anastomoses treated with Adipose tissue derived stem cells[49].

A meticulous study design combining the appropriate MMPs and cytokines can extract valuable information about anastomotic healing and the various signaling pathways by which the inflammatory response is regulated.

Oxidative stress

Another state of matter that can be used in research of an anastomotic healing is oxidative stress. Authors, as shown in Tables 1-8 used markers that indicate either oxidative stress damage, such as free radicals (NOx), myeloperoxidase (MPO) and Malondialdehyde, or antioxidant markers such as superoxide dismutase and glutathione. Neutrophils contain MPO and increased levels of this marker also suggests increased neutrophilic infiltration to the tissues[34].

CONCLUSION

Our review demonstrated the evolution of different high-risk anastomosis protocols in rats as well as the different techniques used to assess anastomotic healing. We emphasize the importance of systematization of research, by standardizing experimental protocols and designing high quality studies that will give us more information on the complex pathophysiological pathways of anastomotic healing. Understanding these pathways will allow us to create interventions that will attenuate the inflammation, decrease anastomotic related complications, and negate the need for diverting stomas in surgical patients.

ACKNOWLEDGEMENTS

Mrs. Mariana Tsioutsiou for grammar checks pre-admission.