Frequency of Dehiscence in Hand-Sutured and Stapled Intestinal Anastomoses in Dogs

Performance time and leak pressure of hand-sewn and skin staple intestinal anastomoses and enterotomies in cadaveric cats

OBJECTIVE

To compare time to construct completion and leak testing between hand-sewn and skin staple anastomoses and enterotomies in cats.

STUDY DESIGN

Ex vivo, randomized study.

ANIMALS

Fresh feline cadavers (n = 20).

METHODS

Jejunal segments (8 cm) were harvested and tested on the same day as euthanasia. From each cadaver, one segment was randomly assigned to control (C), hand-sewn enterotomy (HSE), and skin staple enterotomy (SSE) groups, and two segments were randomly assigned to hand-sewn anastomosis (HSA) and skin staple anastomosis (SSA) groups. Construct completion time, initial leak pressure (ILP), and maximum intraluminal pressure were compared. Leakage location was reported.

RESULTS

Mean time (s) ± SD was longer (p < .001) for HSA (317.0 ± 50.9) than SSA (160.8 ± 13.1) and for HSE (172.0 ± 36.5) than SSE (20.3 ± 5.0). ILP (mean ± SD) for C (600.0 mmHg ±0.0) was higher (p < .001) than all constructs. ILP (mean ± SD) for SSA (124.2 mmHg ±83.7) was not different (p = .49) than HSA (86.1 ± 51.9), but HSE (200.3 ± 114.7) was higher (p < .001) than SSE (32.2 ± 39.7). Immediate leakage from the center of enterotomy closure was observed in 7/20 SSE.

CONCLUSIONS

HSA construct completion took twice as long as SSA with no difference in intraluminal pressures. Although HSE construct completion took 8x as long as SSE, HSE had higher intraluminal pressures.

CLINICAL SIGNIFICANCE

In cats, SSA may be an alternative to HSA for intestinal anastomosis, but SSE is not recommended as an alternative to HSE for intestinal enterotomy closure.

Gastrointestinal thickness, duration, and leak pressure of five intestinal anastomosis techniques in cats

OBJECTIVE

To compare time to construct completion and resistance to leakage for five intestinal anastomosis techniques in cats and to report normal feline gastrointestinal thickness.

STUDY DESIGN

Experimental study.

SAMPLE POPULATION

Grossly normal intestinal segments (n = 120) from 10 fresh cat cadavers.

METHODS

A total of 8 cm segments of fresh feline cadaveric intestine were collected, and mural thickness was recorded. Segments were randomly allocated between a control group (n = 20 segments) and five treatment groups (20 segments/group with 2 segments/construct = 10 constructs per group): (1) hand-sewn anastomosis - simple interrupted (HSA-SI), (2) hand-sewn anastomosis - simple continuous (HSA-SC), (3) functional end-to-end stapled anastomosis (FEESA), (4) functional end-to-end stapled anastomosis with oversew (FEESA-O), (5) skin stapled anastomosis (SS). Time to construct completion, leakage location, initial leak pressure (ILP), and maximum intraluminal pressure (MIP) were compared.

RESULTS

Mean mural thickness ± SD (mm) for the stomach, duodenum, jejunum, and ileum were 1.66 ± 0.28, 2.05 ± 0.18, 2.28 ± 0.30, and 2.11 ± 0.39, respectively. ILPs (mean ± SD) for HSA-SI (165 ± 122 mmHg), HSA-SC (149 ± 83), FEESA-O (63 ± 25, FEESA (84 ± 59), SS (77 ± 56), and control segments (>500) were compared. There was no statistically significant difference in ILP (p > .08) or MIP (p > .084) between any treatment groups. Nonoversewn FEESAs were 2.4 times faster to perform compared to oversewn FEESA and SS groups, and 4.7 times faster than HSA (p < .001).

CONCLUSION

All anastomosis techniques provide resistance to leakage that is supraphysiological to that of the normal maximum intraluminal pressure. HSA take longer to complete than stapled anastomoses.

CLINICAL SIGNIFICANCE

All anastomotic techniques may be appropriate in cats. Hand-sewn anastomoses result in a longer surgical time.

Chitosan thread in the healing of the cecal wall of rabbits (Oryctolagus cuniculus) submitted to cecorrhaphy

The development and evaluation of synthesis materials are crucial to reducing the morbidity and magnitude of post-enterorrhaphy surgical complications. Despite the possibility of production, chitosan thread has not yet been used in enterorrhaphy, and its effects on intestinal healing have not been evaluated. Therefore, this study aimed to evaluate the effects of chitosan thread on the intestinal wall repair of rabbits submitted to cecorrhaphy. For this, 42 rabbits were allocated into two groups with 21 animals. One group was submitted to cecorrhaphy with chitosan suture thread (CG) and the other with poliglecaprone suture thread (PG). The occurrence of postoperative complications, the intensity of edema, cellular response, formation of granulation tissue, as well as the deposition and maturation of collagen fibers, and the intensity of vascular endothelial growth factor (VEGF-α) expression, were evaluated during the intestinal wall repair process. The evaluations occurred on the 5th, 15th, and 25th postoperative (PO) days. The animals did not develop peritonitis, but adherence was observed in six animals from CG and seven from PG, with no difference between groups. The polymorphonuclear infiltrate showed higher intensity and higher amount of type III collagen fibers in CG on the 15th PO day. In contrast, a lower amount of type I collagen fibers was observed in CG samples on the 25th PO day. Therefore, the chitosan thread used for cecorrhaphy in rabbits results in minimal postoperative complications, presents biocompatibility, and bioactively assists the tissue repair process of the cecal wall, inducing minimal tissue reaction, stimulating the deposition of type III collagen fibers in the proliferative phase, with sustained VEGF-α expression, but with reduced deposition of type I fibers, indicating a delay in collagen maturation.

Evaluation of automated staple sizes on gastric layer incorporation and intraluminal pressure for partial gastrectomy closure in an ex vivo canine model

OBJECTIVE

To evaluate the performance of automated staple sizes on a cadaveric canine partial gastrectomy model.

METHODS

Stomachs were transected through the gastric body axis and randomly allocated to two closure groups: Group B, thoracoabdominal (TA) stapler 3.5 mm staple cartridge (blue); Group G, TA stapler 4.8 mm staple cartridge (green). After construct completion, leak testing was performed for both groups and compared. Initial leakage pressure (ILP), maximal leakage pressure (MLP) and leakage location were recorded. Staple lines were evaluated by direct observation and fluoroscopy to assess sub-mucosal layer incorporation and staple conformation. Staple shape was classified as optimal or suboptimal. Significance was set at p less than 0.5.

RESULTS

Following gastrectomy, the mean double gastric wall thickness was 7.82 ± 2.05 mm at the gastric body. Mean ILP was significantly lower in groups G (17.13 ± 1.19 mmHg) compared to group B (50.46 ± 6.03 mmHg, p = 0.0013). Similarly, mean MLP was significantly lower in group G (21.41 ± 1.39 mmHg) compared to group B (64.61 ± 10.21 mmHg, p < 0.0001). Although group G had higher percentage of B-shaped staple formation compared to group B, this was not significant (group G; 92.38%, group B; 54.56%; p = 0.054).

CONCLUSION

The 3.5 mm TA staple cartridge (blue) achieved superior bursting pressures compared with the 4.8 mm TA staple cartridge (blue) for the closure of a canine partial gastrectomy model. Both staple sizes incorporated all gastric layers. No differences were noticed in optimal staple conformation between groups. In vivo investigation is warranted to evaluate the use of different staple sizes on gastric tissue perfusion, successful healing and post-operative stasis and dehiscence.

Clinical outcomes of the use of unidirectional barbed sutures in gastrointestinal surgery for dogs and cats: A retrospective study

OBJECTIVE

To report the clinical outcomes of gastrointestinal surgery using unidirectional barbed sutures in single-layer appositional closure in dogs and cats.

STUDY DESIGN

Retrospective and descriptive study.

SAMPLE POPULATION

Twenty-six client-owned dogs; three client-owned cats.

METHODS

Medical records of dogs and cats that received gastrointestinal surgery closed with unidirectional barbed sutures were reviewed to collect information on signalment, physical examinations, diagnostics, surgical procedures, and complications. Short- and long-term follow-up information was collected from the medical records, the owners, or the referring veterinarians.

RESULTS

Six gastrotomies, 21 enterotomies, and nine enterectomies were closed with a simple continuous pattern with unidirectional barbed glycomer 631 sutures. Nine dogs had multiple surgical sites closed with unidirectional barbed sutures. None of the cases in the study developed leakage, dehiscence, or septic peritonitis during the 14-day short-term follow up. Long-term follow up information was collected for 19 patients. The median long-term follow-up time was 1076 days (range: 20-2179 days). Two dogs had intestinal obstruction due to strictures at the surgical site 20 and 27 days after surgery. Both were resolved with an enterectomy of the original surgical site.

CONCLUSION

Unidirectional barbed suture was not associated with a risk of leakage or dehiscence after gastrointestinal surgery in dogs and cats. However, strictures may develop in the long term.

CLINICAL SIGNIFICANCE

Unidirectional barbed sutures can be used during gastrointestinal surgery in client-owned dogs and cats. Further investigation of the role of unidirectional barbed sutures leading to abscess, fibrosis, or stricture is necessary.

A quantitative evaluation of the effect of foreign body obstruction and enterectomy technique on canine small intestinal microvascular health

OBJECTIVE

To investigate sidestream dark field (SDF) videomicroscopy as an objective measure of intestinal viability and determine the effects of enterectomy techniques on intestinal microvasculature in dogs with foreign body obstructions.

STUDY DESIGN

Prospective, randomized, clinical trial.

ANIMALS

A total of 24 dogs with an intestinal foreign body obstruction and 30 systemically healthy dogs.

METHODS

An SDF videomicroscope imaged the microvasculature at the site of the foreign body. Subjectively viable intestine received an enterotomy whereas nonviable intestine received an enterectomy using a handsewn (4-0 polydioxanone, simple continuous) or a functional end-to-end stapled technique (GIA 60 blue, TA 60 green) was used on an alternating basis. The microvasculature adjacent to the enterectomy was interrogated. Quantitative measures of microvascular health were calculated for each site and compared with healthy dogs.

RESULTS

Microvascular density (mean ± SD) at the site of obstruction (140.84 ± 77.40) was lower than healthy controls (251.72 ± 97.10, p < .01). There was no difference in microvascular parameters (density or perfused boundary region, PBR) between obstructed dogs with subjectively viable and nonviable intestine (p > .14). The density (p = .66) and PBR of microvessels (p = .76) adjacent to the sutured enterectomy or TA green staple line did not differ.

CONCLUSION

Sidestream dark field videomicroscopy can identify obstructed intestine and quantitate the severity of microvascular compromise. Handsewn and stapled enterectomies equally preserve perfusion.

CLINICAL SIGNIFICANCE

Stapled enterectomies do not lead to greater vascular compromise than handsewn enterectomies.

Associations of abdominal discomfort and length of clinical signs with surgical procedure in 181 cases of canine small intestinal foreign body obstruction

BACKGROUND

Gastrointestinal foreign bodies are a common indication for abdominal exploratory surgery.

OBJECTIVES

The objective of this study was to evaluate the relationship of pre-operative abdominal discomfort and duration of clinical signs with surgical resolution of canine small intestinal foreign body obstructions (SIFBO).

METHODS

We performed a retrospective study of 181 canine abdominal exploratory surgeries for confirmed SIFBO at two referral hospitals. Animals were categorized into five surgical groups (gastrotomy after manipulation into the stomach, enterotomy, resection-and-anastomosis [R&A], manipulated into colon, already in colon) and further grouped by whether entry into the gastrointestinal tract (GIT) was required.

RESULTS

Abdominal discomfort was noted in 107/181 cases (59.1%), but no significant differences in abdominal discomfort rates were present among the surgical groups or between GIT entry and no entry groups. Clinical sign duration was associated with surgical procedure; median durations were R&A = 3 days (range, 1-9), enterotomy = 2 days (range, 1-14), gastrotomy = 2 days (range, 1-6), already in colon = 1.5 days (range, 1-2), and manipulated into colon = 1 day (range, 1-7). In a pairwise comparison, differences in the duration of clinical signs were found for obstructions manipulated into the colon versus R&A, gastrotomy versus R&A, and in colon versus R&A. When patients were grouped according to GIT entry, cases with entry had a longer duration of clinical signs (median = 2 days [range, 1-14] versus 1 day [range, 1-7], respectively).

CONCLUSIONS

Abdominal discomfort was not associated with surgical complexity; however, the duration of clinical signs was associated with surgical complexity, with longer duration being associated with entry into the GIT and R&A. Despite statistical significance, the maximum difference of 2 days between surgical groups is unlikely to be clinically relevant.

Case report: Ultrasonographic assessment of early leakage in intestinal sutures in dogs

Intestinal suture dehiscence is one of the most feared complications following gastrointestinal surgery in both human and veterinary medicine, increasing the morbidity and mortality of these patients. Clinical and laboratory early signs of septic peritonitis are not always easily identifiable while prompt treatment should help decrease postoperative morbidity and mortality. The aim of this study is to describe the ultrasonographic (US) features of confirmed leakage of intestinal sutures (LIS) and to evaluate if this imaging technique can be useful as noninvasive tool for the early diagnosis of LIS. Seven dogs developed LIS in a range of three-four days after gastrointestinal surgery and four of these developed a second dehiscence. On B-mode ultrasonography, all intestinal surgical sites were identified and characterized by a bowel focal thickening with reduced or absent wall layering and the presence of hyperechoic, double-walled foci at regular intervals (suture material). Furthermore, hyperechoic linear interfaces associated with dirty acoustic shadowing and comet-tail artifacts crossing the intestinal wall to free-float in peritoneal cavity or in a saccate collection have been documented. On the basis of these preliminary results, canine abdominal ultrasound seems to be a useful diagnostic technique for post-operative monitoring of patients undergoing intestinal surgery, allowing early detection of signs of a LIS, before the patient develops clinical signs of septic peritonitis.

Intestinal dehiscence and mortality in cats undergoing gastrointestinal surgery

OBJECTIVES

The aim of this study was to determine the incidence of and risk factors for both gastrointestinal (GI) incisional dehiscence and mortality in a large cohort of cats undergoing GI surgery. We hypothesized that cats with preoperative septic peritonitis (PSP), systemic inflammatory response syndrome (SIRS) or sepsis would have higher GI dehiscence and mortality rates than unaffected cats.

METHODS

A medical records search identified cats with surgically created, full-thickness incisions into their stomach, small intestines or large intestines. Preoperative data, including signalment, clinical signs, comorbidities, surgical history, current medications, presenting physical examination findings, complete blood counts and serum biochemistry values, were collected. It was determined whether or not cats had PSP, SIRS or sepsis at admission. Intraoperative data, final diagnosis and postoperative variables such as vital parameters, bloodwork and (if applicable) the development of GI dehiscence or mortality were noted. Postoperative follow-up of at least 10 days was obtained in survivors.

RESULTS

In total, 126 cats were included. One cat developed GI dehiscence following complete resection of a jejunal adenocarcinoma. Twenty-three cats (18.2%) died within 10 days of surgery. Cats with PSP (P = 0.0462) or that developed hypothermia 25-72 h postoperatively (P = 0.0055) had higher odds of mortality in multivariate analysis. Cats with PSP had 6.7-times higher odds of mortality than cats not diagnosed with PSP.

CONCLUSIONS AND RELEVANCE

In cats receiving GI surgery, the incidence of GI incisional dehiscence was <1%. Cats with PSP had a higher likelihood of mortality. SIRS was a common finding in cats with septic peritonitis, but was not associated with mortality. Postoperative mortality during the home recovery period might be significant in cats. Future studies evaluating postoperative mortality in cats should consider extending the research period beyond the date of discharge.

Influence of barbed suture oversew of the transverse staple line during functional end-to-end stapled anastomosis in a canine jejunal enterectomy model

Objective

To evaluate the influence of barbed suture oversew of the transverse staple line during functional end‐to‐end stapled anastomosis (FEESA) in dogs.

Study design

Randomized, experimental, ex vivo.

Animals or sample population

Grossly normal jejunal segments from 14 adult canine cadavers.

Methods

Ninety‐eight jejunal segments (n = 14/FEESA group, n = 14 controls) were harvested and randomly assigned to a control group, FEESA + monofilament suture oversew, FEESA + unidirectional barbed suture oversew or FEESA + bidirectional barbed suture oversew. Oversew techniques were performed using a Cushing suture pattern. Initial (ILP) and maximum leakage pressure (MLP), repair time (s), and location of observed leakage were recorded.

Results

No differences were detected in ILP (p = .439) or MLP (p = .644) respectively between experimental groups. Repairs times using barbed suture were ~ 18% faster (~25 s faster; p < .001) compared to monofilament suture. There was no difference between barbed suture types (p = .697). Mean ILP (p < .001) and MLP (p < .0001) were 6.6x and 5.1x greater respectively in the control group. Leakage location occurred predominately at the crotch of the FEESA in all groups.

Conclusion

FEESAs closed with a transverse staple line oversew using barbed suture, regardless of barb orientation, were completed faster and resulted in similar resistance to anastomotic leakage compared to monofilament suture.

Clinical significance

Oversewing the transverse staple line following FEESA using barbed suture offers similar resistance to anastomotic leakage, and may be associated with decreased surgical times in dogs compared to monofilament suture. Further studies are necessary to determine the benefits of barbed suture use in both open and laparoscopic gastrointestinal surgical applications following FEESA in dogs.

Ex vivo comparison of different thoracoabdominal stapler sizes for typhlectomy in canine cadavers

OBJECTIVE

To determine the influence of staple size on leakage pressure of typhlectomy sites in canine cadavers.

STUDY DESIGN

Randomized, experimental cadaveric study.

ANIMALS

Twenty-four fresh canine cadavers.

METHODS

Ileocecocolic segments were exteriorized following right paracostal laparotomy after euthanasia. Cecal base length and wall thickness were measured. Each cecum was randomly assigned to 1 of 3 groups (TA 30 V3 2.5 mm, TA 60 3.5 mm, and TA 60 4.8 mm). The cecal base was stapled and the cecum was removed. A 10 cm segment including the stapled cecal excision site was tested for initial leak pressure.

RESULTS

The mean ± standard deviation body weights across the groups were 18.7 ± 6.1 kg, 16.2 ± 7.5 kg, and 14.2 ± 5.5 kg for the TA 30 V3 2.5 mm, TA 60 3.5 mm, and TA 60 4.8 mm groups, respectively (P = .48). There were no differences for mean cecal base length or wall thickness. Mean initial leak pressure (ILP) across groups was 182 ± 111 mmHg (TA 30 V3 2.5 mm), 112 ± 57 mmHg (TA 60 3.5 mm), and 77 ± 60 mmHg (TA 60 4.8 mm) (P = .78).

CONCLUSION

Each stapler size that was evaluated resulted in a mean ILP in excess of typical intraluminal pressures under normal circumstances. There were no differences among groups.

CLINICAL SIGNIFICANCE

The results of this cadaveric study support the use of any of the stapler sizes evaluated in similarly sized dogs. A prospective study is needed to be able to correlate stapler size and clinical outcome.

Influence of crotch suture augmentation on leakage pressure and leakage location during functional end-to-end stapled anastomoses in dogs

OBJECTIVE

To determine the influence of anastomotic crotch suture augmentation on leakage pressures and leakage location following intestinal functional end-to-end stapled anastomosis (FEESA) in dogs.

STUDY DESIGN

Ex vivo, randomized, experimental.

SAMPLE POPULATION

Chilled jejunal segments from 3 adult dogs.

METHODS

Jejunal specimens were tested within 24 hours of collection. A FEESA was performed and randomly assigned to 1 of 4 treatment groups (n = 12/group): (1) no crotch suture (NCS); (2) simple interrupted crotch suture (SICS); (3) two simple interrupted crotch sutures (TCS) placed laterally on opposing jejunal limbs; (4) simple continuous crotch suture (SCCS) augmentation. Crotch sutures were performed using 3-0 USP polydioxanone. Initial (ILP) and maximal (MLP) leakage pressures (Mean ± SD mm Hg) and leakage location were recorded and compared between groups.

RESULTS

Initial leakage pressure was greater after placement of TCS (37.8 ± 6.4, P < .039) and SCCS (47.6 ± 11.0, P < .002) than NCS (27.1 ± 2.5) and SICS (33.0 ± 6.0). Maximal leakage pressure was greater in specimens including SICS, TCS, and SCCS than those without crotch suture augmentation (P < .043). Leakage occurred at the anastomotic crotch in 8/12 NCS, 6/12 SICS, 11/12 TCS, and 12/12 SCCS constructs (P < .001).

CONCLUSION

Augmentation of FEESA with TCS and SCCS increased ILP and decreased the occurrence of leakage from the anastomotic crotch, while all methods of anastomotic crotch augmentation increased MLP.

CLINICAL SIGNIFICANCE

Augmenting the FEESA with crotch suture(s) improved the resistance of the jejunal anastomosis to leakage in normal cadaveric segments. Placing 2 crotch sutures or use of a simple continuous pattern for anastomotic augmentation appeared to be superior to the placement of a single suture.

Volume of saline (0.9% NaCl) solution required to reach maximum peristaltic pressure in cadaveric intact jejunal specimens from dogs of various sizes

OBJECTIVE

To compare the volume of saline (0.9% NaCl) solution required to reach a maximum intraluminal peristaltic pressure of 25 mm Hg in dogs of various sizes.

SAMPLES

25 grossly normal jejunal segments from 6 canine cadavers < 20 kg (small dogs) and 25 segments from 5 cadavers ≥ 20 kg (large dogs).

PROCEDURES

Jejunal specimens were obtained within 1.5 hours after euthanasia. Harvested tissue was transected into 12-cm-long segments, mesentery was trimmed, and each segment was measured from the antimesenteric to mesenteric serosal edges. A 10-cm segment was isolated with Doyen forceps, securing a pressure sleeve within the lumen. Intraluminal saline was infused, and the volume was recorded when a pressure of > 25 mm Hg was achieved. Data were analyzed only from specimens in which the pressure remained between 24 and 26 mm Hg for > 5 seconds.

RESULTS

Mean ± SD intestinal measurement for large dogs (17.82 ± 1.44 mm) was greater than that for small dogs (12.38 ± 1.38 mm) as was the volume of saline solution infused (17.56 ± 7.17 mL vs 3.28 ± 1.41 mL, respectively). The volume infused increased by 1.31 mL (95% CI, 1.08 to 1.18) for every 1-mm increase in intestinal measurement and by 1.06 mL (95% CI, 1.052 to 1.068) for every 1-kg increase in body weight.

CONCLUSIONS AND CLINICAL RELEVANCE

The volume of saline solution used for intestinal leak testing should be determined on the basis of patient intestinal measurement or body weight. In vivo studies are necessary to establish the optimal volume for intestinal leak testing.

Intraoperative surgeon probe inspection compared to leak testing for detecting gaps in canine jejunal continuous anastomoses: A cadaveric study

OBJECTIVE

To compare the detection of gaps in jejunal continuous anastomoses by probing (PT) versus leak testing (LT).

STUDY DESIGN

Experimental study.

ANIMALS

Normal jejunal segments (n = 24) from two fresh canine cadavers.

METHODS

Intestinal segments were randomly selected by four volunteers who created six simple continuous anastomoses: two constructs using standard technique with 2-3 mm suture spacing (ST, 8 total), and four constructs with one 4-mm gap (IG, 16 total). All 24 anastomoses were examined with PT (1 volunteer) prior to LT (3 volunteers). LT was performed within a maximal peristaltic pressure range (34-54 cmH₂ O). The presence and location of anastomotic leaks (L_A ) detected with LT were compared with gaps detected with PT.

RESULTS

Nineteen out of twenty-four (79.2%) samples had at least one probe drop, and four out of twenty-three (17.4%) samples leaked. LT results from one sample were excluded due to iatrogenic suture failure. PT was 100% sensitive (95% CI: 51.01, 100%) with a 100% negative predictive value (95% CI: 56.55, 100%) at detecting gaps compared with LT. The location of all L_A was accurately identified with PT and confirmed during LT.

CONCLUSION

PT was highly sensitive at detecting gaps compared with LT. All L_A were accurately detected by PT. None of the segments negative during PT leaked.

CLINICAL SIGNIFICANCE

The diagnostic value of PT appears adequate to recognize gaps in cadaveric small intestinal anastomoses. PT offers surgeons an alternative to detect suture gaps that may contribute to postoperative dehiscence of intestinal anastomoses.

Comparison of patient outcomes following enterotomy versus intestinal resection and anastomosis for treatment of intestinal foreign bodies in dogs

OBJECTIVE

To quantify the relative risk of intestinal dehiscence in dogs undergoing intestinal resection and anastomosis (IRA), compared with enterotomy, for surgical management of small intestinal foreign bodies, and to evaluate the association between nasogastric tube placement for early enteral nutrition (EEN) and hospitalization time.

ANIMALS

211 dogs undergoing 227 surgeries for intestinal foreign body removal.

PROCEDURES

Medical records were reviewed for dogs undergoing a single-site sutured enterotomy or IRA for foreign body intestinal obstruction between May 2008 and April 2018. Multivariable logistic regression was used to quantify the association between surgical procedure and dehiscence. Multiple linear regression was used to quantify the association of nasogastric tube placement with total hospitalization time.

RESULTS

Dehiscence rates were 3.8% (7/183) and 18.2% (8/44) for enterotomy and IRA, respectively. Overall dehiscence rate for all surgeries was 6.6% (15/227). The odds of intestinal dehiscence for IRA were 6.09 times (95% CI, 1.89 to 19.58) the odds for enterotomy. An American Society of Anesthesiologists score > 3 (OR, 4.49; 95% CI, 1.43 to 14.11) and an older age (OR, 1.02 [95% CI, 1.01 to 1.02] for each 1-month increase in age) were significantly associated with greater odds of intestinal dehiscence, regardless of surgical procedure. Placement of a nasogastric tube was not associated with intestinal dehiscence or decreased total hospitalization time when controlling for the year of surgery.

CONCLUSIONS AND CLINICAL RELEVANCE

Patients undergoing IRA were at a significantly higher risk of intestinal dehiscence, compared with patients undergoing enterotomy. Although this finding should not be used to recommend enterotomy over IRA, this information may be useful in guiding owner expectations and postoperative monitoring.

Ex vivo comparison of leak testing of canine jejunal enterotomies: Saline infusion versus air insufflation

OBJECTIVE

To compare the ability to detect leakage of enterotomy closures by intraluminal injection of saline or air.

STUDY DESIGN

Ex vivo study.

ANIMALS

Grossly normal jejunal segments (n = 60) from five fresh canine cadavers.

METHODS

Eight-centimeter jejunal segments were randomly assigned to two control (saline control [n = 5], air control [5]) and two treatment groups (injection of saline [n = 25] or air [25] after enterotomies were closed in a simple continuous pattern using 4-0 glycomer 631). Initial leak pressure (ILP, mean ± SD), maximum intraluminal pressure (MIP), and leakage location were compared. For all air insufflation constructs, the volume of air insufflated at the time of initial leakage was recorded.

RESULTS

The ILPs of control segments did not differ whether injected with saline (405.71 ± 56.97 mmHg) or air (376.84 ± 42.54, p = 1.00). Enterotomy closures leaked at lower pressures when injected with air (ILP: 68.52 ± 6.56) compared to saline (87.76 ± 5.20, p = .03). Similar results were obtained when comparing MIPs. A moderate association (r = .51) was identified between volume of air infused and ILPs. The strength of the association improved when stratified by cadaver. The location of leakage-most commonly suture tracks-was identified for all air constructs and for 14 of 25 saline constructs.

CONCLUSION

Enterotomy closures leaked at lower pressures after intraluminal injection of air compared to saline.

CLINICAL SIGNIFICANCE

Intraoperative leak testing of small intestinal enterotomy closures may be rendered more sensitive and precise by the use of air instead of saline as the infusate.

Evaluation of intraoperative leak testing of small intestinal anastomoses performed by hand-sewn and stapled techniques in dogs: 131 cases (2008-2019)

OBJECTIVE

To compare the rate of postoperative dehiscence on the basis of intraoperative anastomotic leak test results (ie, positive or negative for leakage or testing not performed) between dogs that underwent hand-sewn anastomosis (HSA) or functional end-to-end stapled anastomosis (FEESA) of the small intestine.

ANIMALS

131 client-owned dogs that underwent 144 small intestinal anastomoses (94 FEESA and 50 HSA).

PROCEDURES

Medical records were searched to identify dogs that had undergone a small intestinal anastomosis (HSA or FEESA) from January 2008 through October 2019. Data were collected regarding signalment, indication for surgery, location of the anastomosis, surgical technique, the presence of preoperative septic peritonitis, performance of intraoperative leak testing, development of postoperative dehiscence, and duration of follow-up.

RESULTS

Intraoperative leak testing was performed during 62 of 144 (43.1%) small intestinal anastomoses, which included 26 of 94 (27.7%) FEESAs and 36 of 50 (72.0%) HSAs. Thirteen of 144 (9.0%) anastomoses underwent dehiscence after surgery (median, 4 days; range, 2 to 17 days), with subsequent septic peritonitis, including 10 of 94 (10.6%) FEESAs and 3 of 50 (6.0%) HSAs. The incidence of postoperative dehiscence was not significantly different between FEESAs and HSAs; between anastomoses that underwent intraoperative leak testing and those that did not, regardless of anastomotic technique; or between anastomoses with positive and negative leak test results. Hand-sewn anastomoses were significantly more likely to undergo leak testing than FEESAs. Preoperative septic peritonitis, use of omental or serosal reinforcement, preoperative serum albumin concentration, and surgical indication were not significantly different between anastomotic techniques.

CONCLUSIONS AND CLINICAL RELEVANCE

Performance of intraoperative anastomotic leak testing, regardless of the anastomotic technique, was not associated with a reduction in the incidence of postoperative anastomotic dehiscence.

Gastrointestinal thickness, duration, and leak pressure of six intestinal anastomoses in dogs

OBJECTIVE

To compare leak pressures and construct completion time of six intestinal anastomoses and report normal canine gastrointestinal thickness.

STUDY DESIGN

Experimental study.

ANIMALS

Grossly normal jejunal segments (n = 140) from 10 fresh canine cadavers.

METHODS

Gastrointestinal thickness was recorded. Eight-centimeter cooled canine cadaveric jejunal segments were randomly assigned to a control group (20 segments) and six treatment groups (20 segments/group [10 constructs/group]): (1) handsewn anastomosis (HSA), (2) functional end-to-end stapled anastomosis (FEESA)-blue thoracoabdominal (TA; FEESA-TAB), (3) FEESA-green TA (FEESA-TAG), (4) FEESA TA-gastrointestinal anastomosis (GIA), (5) FEESA with suture oversew (FEESA-O), and (6) skin staples (SS). Construct assembly time, initial leak pressure (ILP), maximum intraluminal pressure (MIP), and leakage location were compared.

RESULTS

Initial leak pressures (mean ± SD) for control (308.38 ± 115.91 mm Hg), HSA (41.96 ± 15.97), FEESA-TAB (31.71 ± 15.71), FEESA-TAG (27.24 ± 14.11), FEESA-GIA (25.62 ± 11.22), FEESA-O (31.01 ± 17.38), and SS (44.42 ± 28.88) groups were compared. No difference in ILP (P > .24) or MIP (P > .17) was detected between treatment groups. Sutured anastomoses took up to 10 times longer to complete (P = .0025). The stomach, duodenum, jejunum, and ileum mural thicknesses (mean ± SD) were 3.99 ± 0.44 mm, 2.34 ± 0.16, 2.49 ± 0.28, and 2.30 ± 0.31, respectively.

CONCLUSION

The ILP of all anastomoses exceeded maximum intraluminal peristaltic pressures. Stapled anastomoses were faster to complete.

CLINICAL SIGNIFICANCE

All anastomoses may be considered when performing an intestinal resection and anastomosis, with stapled anastomoses resulting in a shorter surgical time. Canine intestinal thickness may warrant use of a larger staple size.

Intestinal incisional dehiscence rate following enterotomy for foreign body removal in 247 dogs

OBJECTIVE

To report the intestinal incisional dehiscence rate following enterotomy for foreign body removal in dogs.

ANIMALS

247 client-owned dogs with intestinal foreign bodies treated with enterotomy between November 2001 and September 2017.

PROCEDURES

Medical records were reviewed, and data were collected regarding signalment, history, surgery, clinicopathologic findings, hospitalization, intestinal incisional dehiscence, and survival to hospital discharge. Dogs were grouped according to whether intestinal incisional dehiscence occurred (dehiscence group) or did not occur (nondehiscence group) following enterotomy, and the rate of dehiscence for the total number of enterotomies during the study period was calculated. Univariable analysis was performed to identify variables associated with intestinal incisional dehiscence.

RESULTS

8 of the 247 (3.2%) dogs had preoperative septic peritonitis, and all 8 dogs survived to hospital discharge. Incisional dehiscence occurred following 5 of the 247 (2.0%) enterotomies, and only 2 dogs in the dehiscence group did not survive to hospital discharge. Duration of hospitalization was longer for dogs in the dehiscence group than for dogs in the nondehiscence group.

CONCLUSIONS AND CLINICAL RELEVANCE

Results indicated that enterotomy for intestinal foreign body removal had a lower rate of dehiscence in dogs during the study period, compared with rates previously reported; however, the low rate should not be used as a reason to perform an enterotomy rather than an enterectomy when needed. Surgeons should thoroughly evaluate the bowel and perform an enterotomy only when indicated.

Evaluation of short-term outcomes and potential risk factors for death and intestinal dehiscence following full-thickness large intestinal incisions in dogs

OBJECTIVE

To determine complication rates for dogs in which full-thickness large intestinal incisions were performed, assess potential risk factors for death during hospitalization and for intestinal dehiscence following these surgeries, and report short-term mortality rates for these patients.

ANIMALS

90 dogs.

PROCEDURES

Medical records of 4 veterinary referral hospitals were reviewed to identify dogs that underwent large intestinal surgery requiring full-thickness incisions. Signalment, history, clinicopathologic data, medical treatments, surgical procedures, complications, and outcomes were recorded. Descriptive statistics were calculated; data were analyzed for association with survival to discharge (with logistic regression analysis) and postoperative intestinal dehiscence (with Fisher exact or Wilcoxon rank sum tests).

RESULTS

Overall 7-day postoperative intestinal dehiscence and mortality rates were 9 of 90 (10%) and 15 of 90 (17%). Dogs with preoperative anorexia, hypoglycemia, or neutrophils with toxic changes and those that received preoperative antimicrobial treatment had greater odds of death than did dogs without these findings. Preexisting colon trauma or dehiscence, preexisting peritonitis, administration of blood products, administration of > 2 classes of antimicrobials, positive microbial culture results for a surgical sample, and open abdominal management of peritonitis after surgery were associated with development of intestinal dehiscence. Five of 9 dogs with intestinal dehiscence died or were euthanized.

CONCLUSIONS AND CLINICAL RELEVANCE

Factors associated with failure to survive to discharge were considered suggestive of sepsis. Results suggested the dehiscence rate for full-thickness large intestinal incisions may not be as high as previously reported, but several factors may influence this outcome and larger, longer-term studies are needed to confirm these findings.

A Review of Normal Intestinal Healing, Intestinal Anastomosis, and the Pathophysiology and Treatment of Intestinal Dehiscence in Foreign Body Obstructions in Dogs

Small intestinal anastomoses are commonly performed in veterinary medicine following resection of diseased or devitalized intestinal tissue. Traditionally, suture has been employed to anastomose intestinal ends. However, use of intestinal staplers has become increasingly popular due to the ability to produce a rapid anastomosis with purported superior healing properties. Under normal conditions, intestinal healing occurs in three phases: inflammatory, proliferative, and maturation. Dehiscence, a devastating consequence of intestinal anastomosis surgery, most often occurs during the inflammatory phase of healing where the biomechanical strength of the anastomosis is almost entirely dependent on the anastomotic technique (suture or staple line). The resulting septic peritonitis is associated with a staggering morbidity rate upwards of 85% secondary to the severe systemic aberrations and financial burden induced by septic peritonitis and requirement of a second surgery, respectively. Intraoperative and postoperative consideration of the multifactorial nature of dehiscence is required for successful patient management to mitigate recurrence. Moreover, intensive postoperative critical care management is necessitated and includes antibiotic and fluid therapy, vasopressor or colloidal support, and monitoring of the patient's fluid balance and cardiovascular status. An understanding of anastomotic techniques and their relation to intestinal healing will facilitate intraoperative decision-making and may minimize the occurrence of postoperative dehiscence.

Influence of oversewing the transverse staple line during functional end-to-end stapled intestinal anastomoses in dogs

OBJECTIVE

To determine the effect of oversewing the transverse staple line after functional end-to-end stapled intestinal anastomoses (FEESA) on canine jejunal leakage pressures.

STUDY DESIGN

Experimental, ex vivo, randomized study.

SAMPLE POPULATION

Jejunal segments from three adult canine cadavers.

METHODS

Jejunal segments were harvested within 2 hours of euthanasia and anastomosed (24 jejunal segments per group, consisting of two segments per construct with n = 12/ group). Constructs were then randomly assigned to receive FEESA alone, FEESA + Cushing oversew, or FEESA + simple-continuous oversew of the transverse staple line with 3-0 polydioxanone. Results for initial leakage pressure (ILP) and maximal leakage pressure (MLP) and initial leakage location (LL) were compared between groups.

RESULTS

Mean ILP was 1.8-fold higher for FEESA + Cushing oversew (62.4 ± 7.8 mm Hg) compared with FEESA alone and FEESA + simple-continuous oversew (P < .001). Mean MLP were higher for both oversewn techniques compared with FEESA alone (P < .001). Oversewing the transverse staple line with either pattern increased mean MLP by 1.4-fold compared with FEESA alone. Leakage occurred at the level of the transverse staple line in nonoversewn constructs (P < .001).

CONCLUSION

Oversewing the transverse staple line after FEESA increased MLP and decreased the occurrence of leakage at this location. Oversewing with a Cushing pattern increased ILP compared with oversew with a simple-continuous pattern.

CLINICAL SIGNIFICANCE

Our results provide evidence to support oversewing the transverse staple line after FEESA. Doing so may reduce the occurrence of postoperative dehiscence. These findings warrant additional focused investigation in vivo through a prospective randomized clinical trial.

Ex vivo comparison of leakage pressures and leakage location with a novel technique for creation of functional side-to-side canine small intestinal anastomoses

OBJECTIVE

To determine the ability of functional side-to-side small intestinal anastomoses (FSS-SIA) created with an electrothermal bipolar vessel sealing (EBVS) device to resist leakage.

STUDY DESIGN

Experimental, ex vivo.

SAMPLE POPULATION

Jejunal segments (n = 130) from 10 healthy canine cadavers.

METHODS

Four types of anastomoses were created (two segments/construct and 15 constructs/group): EBVS (group A), EBVS + transverse stapling (group B), stapled (group C), and EBVS + suture augmentation (group D). Initial leakage pressure (ILP), initial leakage location (ILL), and maximal intraluminal pressure were compared between groups, and five group A constructs were analyzed histologically.

RESULTS

Initial leakage pressure was greater in group D than in groups A, B, and C (P < .011). There was a difference in ILL among groups (P = .003). Leakage occurred at the side-to-side intestinal anastomosis fusion line in 13 of 15 (87%) constructs for groups A and B and in nine of 15 (60%) constructs for group D. Maximal intraluminal pressure was greater in group C than in groups A, B, and D (P < .004). Histological examination was consistent with collagenous fusion without cavitation defects.

CONCLUSION

Functional side-to-side small intestinal anastomosis was consistently achieved with an EBVS device. Augmentation of EBVS anastomoses with simple interrupted sutures along the anastomotic fusion line increased ILP compared with stapled anastomoses.

CLINICAL SIGNIFICANCE

Despite the success and feasibility of creating an FSS-SIA with an EBVS device, additional in vivo studies are required to determine the effectiveness of intestinal fusion prior to clinical implementation.

Influence of closure technique on leakage pressures in an ex vivo canine typhlectomy model

OBJECTIVE

To determine the influence of three closure techniques on leakage pressures of canine typhlectomies.

STUDY DESIGN

Experimental, ex vivo.

SAMPLE POPULATION

Grossly normal cecal segments from 24 adult canine cadavers.

METHODS

Typhlectomies were assigned to one of three closure techniques: simple continuous closure with a Parker-Kerr pattern with 4-0 polydioxanone (group 1), closure with a 60-mm gastrointestinal stapler loaded with a 3.8-mm staple cartridge (group 2), and placement of a Cushing suture to augment the stapled closure (group 3). The median (range) of initial leakage pressure (ILP) and maximum leakage pressure (MLP) was compared between groups along with leakage location.

RESULTS

Typhlectomies in group 3 leaked at higher ILP (310 mm Hg; 188-310) than those in groups 1 (43 mm Hg; 31-80) and 2 (109.5 mm Hg; 68-173; P < .0001). Maximum leakage pressure were greater in group 3 than in other groups (P < .0001). Leakage was detected in 20 of 24 (83%) typhlectomies, located at the incisional line in 13 of 24 (54%) specimens and from suture holes in seven of 24 (29%) specimens.

CONCLUSION

Placement of a Cushing suture pattern to augment stapled incisions improved the ability of typhlectomies to sustain pressure compared with sutured or stapled cadaveric specimens alone.

CLINICAL SIGNIFICANCE

These results provide evidence to support placement of a Cushing suture pattern to augment the staple line for typhlectomies in dogs, although in vivo studies are required to determine the clinical significance of these findings.

Initial leak pressures of four anastomosis techniques in cooled cadaveric canine jejunum

OBJECTIVE

To compare the effectiveness of four different intestinal anastomosis techniques at preventing leakage after enterectomy.

STUDY DESIGN

Experimental study.

SAMPLE POPULATION

Grossly normal jejunal segments (N = 70) from three fresh canine cadavers.

METHODS

Eight-centimeter cooled canine cadaveric jejunal segments were randomly assigned to a control group (six segments) and four treatment groups (16 segments each [two segments/anastomotic construct]): (1) handsewn anastomosis (HSA), (2) barbed suture anastomosis (B-HSA), (3) stapled functional end-to-end anastomosis (SFEEA), and (4) stapled functional end-to-end anastomosis with an oversew (SFEEA-O). Control segments and anastomotic constructs were infused intraluminally to the point of leakage. Initial leak pressures were recorded and compared.

RESULTS

Initial leak pressures (median + range) for jejunal control segments, HSA, B-HSA, SFEEA, and SFEEA-O were 331.88 mmHg (range, 315.34-346.64), 35.17 (20.29-56.24), 24.99 (6.08-38.64), 28.77 (18.80-85.09), and 35.92 (12.05-80.71), respectively. No difference was detected between leak pressures of anastomosed segments (P = .35), all of which were more variable and lower than those of intact segments.

CONCLUSION

No difference in initial leak pressures was detected between the four anastomosis techniques tested in cooled canine cadaveric jejunum.

CLINICAL SIGNIFICANCE

All four anastomosis techniques evaluated in this study may be suitable in dogs.

Intestinal surgery in small animals: historical foundations, current thinking, and future horizons

Intestinal wounds require precise closure after intestinal biopsy, enterotomy, or enterectomy in small animals. Preexisting factors such as intra-abdominal sepsis and hypoalbuminemia as well as poor surgical technique increase the risk of intestinal dehiscence, with considerable negative impact on patient morbidity and mortality. Live dog studies have demonstrated the dangers of mucosal eversion especially in the septic abdomen. Approximating patterns preserve luminal diameter, heal optimally, and have equal bursting strength compared with inverting patterns after 24 hours. Simple interrupted and simple continuous suture patterns and disposable skin staples are established alternatives for manual wound closure. Knotless quilled suture currently used in laparoscopic gastropexy techniques shows bursting strength equal to monofilament sutures in dog cadaveric intestine. Dehiscence rates with hand sewn vs titanium automated stapling anastomosis are similar in uncomplicated cases; however, auto stapling devices may be the preferred method of anastomosis when preexisting abdominal sepsis is present and when patient size allows it. Regardless of the technique, current standard of care involves leak testing and omental wrapping, followed by early postoperative feeding. The past decade has ushered in an exciting new era of laparoscopic assisted techniques that have the potential to reduce postoperative pain and patient morbidity. An understanding of these applications will establish the future of minimally invasive small animal intestinal surgery for veterinary specialists. In summary, surgeons have a variety of methods at their disposal for optimal clinical outcome in small animal intestinal surgery.

Evaluation of suture reinforcement for stapled intestinal anastomoses: 77 dogs (2008-2018)

OBJECTIVE

To determine the influence of oversewing a transverse staple line in functional end-to-end stapled intestinal anastomoses (FEESA) in dogs.

STUDY DESIGN

Retrospective observational study.

SAMPLE POPULATION

Seventy-seven client-owned dogs that underwent 78 FEESA reinforced (n = 30) or not reinforced (n = 48) with suture at the transverse staple line.

METHODS

The medical records database was searched and reviewed for dogs that had undergone a FEESA between January 2008 and September 2018. Data were collected regarding signalment, body weight, clinical presentation, indication for surgery, serum albumin, presence of septic peritonitis, previous surgeries, surgical techniques (ie, oversew, crotch suture, omental wrap, omental patch, serosal patch), histopathology results, and postoperative outcome.

RESULTS

The only differences identified between groups consisted of higher preoperative albumin (2.89 ± 0.56 vs 2.34 ± 0.62 g/dL; P = .006) and lower postoperative dehiscence rate (0/30 vs 7/48; P = .028) in dogs with an oversewn FEESA. Oversewing the FEESA was identified as the significant factor in a model with oversewing and preoperative albumin fit to the outcome of dehiscence (oversew P = .010, albumin P = .761). The location of the dehiscence was specified in four of seven dogs, all along the transverse staple line. Patterns used for oversew were unspecified (n = 11), simple continuous (8), Cushing (4), simple interrupted (2), cruciate (1), interrupted horizontal mattress (1), and Lembert (1).

CONCLUSION

Oversewing the transverse staple line in FEESA was associated with a reduced occurrence of postoperative dehiscence.

CLINICAL SIGNIFICANCE

Our results provide evidence to support additional investigation of suture reinforcement (oversewing) at the transverse staple line of FEESA to reduce postoperative dehiscence.

Intra-abdominal complications following intestinal anastomoses by suture and staple techniques in dogs

OBJECTIVE To compare the incidence of intra-abdominal complications in dogs following resection and functional end-to-end stapled anastomosis (FEESA) versus anastomosis with an end-to-end sutured technique for treatment of enteric lesions. DESIGN Multicenter, retrospective descriptive cohort study. ANIMALS 180 dogs. PROCEDURES Medical records of dogs undergoing intestinal resection and anastomosis at 3 nonaffiliated private practice specialty centers were retrospectively reviewed. Preoperative clinical variables, indication for surgery, surgical technique (sutured end-to-end anastomosis vs FEESA), and evidence of postoperative anastomosis site leakage (dehiscence) were recorded. Variables of interest were analyzed for associations with dehiscence. RESULTS Dehiscence rates of sutured and stapled anastomoses were 12 of 93 (13%) and 4 of 87 (5%), respectively; odds of postoperative dehiscence were significantly lower for dogs with FEESAs than for dogs with sutured anastomoses (OR, 0.28; 95% confidence interval, 0.09 to 0.94). Among dogs that underwent surgery for treatment of intestinal dehiscence after surgery at another facility, subsequent dehiscence developed in 3 of 5 with sutured anastomoses and 0 of 11 with stapled anastomoses. Dehiscence rates varied significantly among clinics. No other variable was associated with risk of dehiscence. Eleven of 16 dogs with dehiscence were euthanized without additional surgery. Impaction at the anastomosis site was identified months or years after surgery in 3 dogs (4 anastomosis sites) that had FEESAs. CONCLUSIONS AND CLINICAL RELEVANCE Odds for dehiscence were significantly greater for sutured end-to-end anastomoses than FEESAs, and dogs undergoing surgery for previous dehiscence were significantly more likely to experience a subsequent dehiscence with a sutured anastomosis. However, variability of procedure types and dehiscence rates among clinics suggested further research is needed to confirm these findings. Obstruction at the anastomosis site was identified as a potential long-term complication of FEESA.

The use of disposable skin staples for intestinal resection and anastomosis in 63 dogs: 2000 to 2014

OBJECTIVE

To describe the use of disposable skin staples for intestinal resection and anastomosis in dogs and report associated dehiscence and mortality rates.

METHODS

Retrospective evaluation of medical records of dogs that underwent intestinal resection and anastomosis using disposable skin staples between 2000 and 2014. Data regarding patient signalment, indication for surgery, location of the resection and anastomosis, number of procedures performed, evidence of peritonitis at the time of surgery, surgeon qualifications, dehiscence, and mortality were obtained from the medical records. Mortality was defined as failure to survive beyond 10 days following resection and anastomosis.

RESULTS

The overall mortality rate of patients undergoing intestinal resection and anastomosis was 12·7% (8/63). The most common indication for resection and anastomosis was neoplasia (20/63 [31·7%]), followed by foreign body removal (19/63 [30·2%]). The overall dehiscence rate was 4·8% (3/63). No difference in mortality associated with indication for surgery, whether multiple procedures were performed, surgeon qualifications, or evidence of peritonitis at the time of surgery was identified.

CLINICAL SIGNIFICANCE

In this retrospective study, the overall mortality and dehiscence rates using disposable skin staples were similar to previously reported outcomes following resection and anastomosis.