Introduction

Enterotomy is the most commonly performed surgical procedure of the intestine.¹ They are commonly performed for the removal of foreign material, as well as to obtain full-thickness samples for biopsy and culture. They are also performed in the repair of trauma or full-thickness intestinal ulcerations and perforations.

Effective closure is required to avoid complications of leakage and dehiscence, which can lead to septic peritonitis. Compromised healing causing life-threatening complications, increases medical costs, prolongs hospitalization stays, and decreases quality of life.² In veterinary medicine, enterotomy closure has been described using suture and stapling devices. Many suture closure variations have been described using different types of suture material and patterns.¹ Although inverting and double-layer closures have been described, a single-layer simple interrupted or simple continuous appositional pattern is currently recommended.^1,3 Two-layer closure causes increased tissue damage with delayed healing in comparison with single-layer closures.^4–6 Single-layer appositional closure provides the best apposition of submucosa leading to improved healing.^7–13 Commonly utilized appositional patterns include simple interrupted and simple continuous suture patterns, and both are recommended for intestinal closure.^14,15

The modified Gambee suture pattern is purported to minimize mucosal eversion, thereby decreasing adhesion formation, and improving healing.¹⁶ Although not a common problem in canine patients, the presence of ischemia, foreign bodies, hemorrhage, and infection can increase the risk of adhesion formation.^17,18 Enhanced healing seen with improved apposition of the holding layer of the intestine promotes primary healing, and may decrease complications seen with enterotomy closure, including adhesions and stenosis.^9,12,13 The use of skin staples for closure has been evaluated for closure of small intestinal anastomosis; they are known to be utilized by surgeons for enterotomy closure, and have been reported in the literature.^19,20

To date, no comparison has been made in the effectiveness of different closure methods to prevent leakage following enterotomy in canines. The aim of this study was to compare the intraluminal pressures at the time of leakage between the modified Gambee, simple interrupted, simple continuous suture pattern, and skin staples for closure of enterotomy sites. A secondary aim was to compare closure time between the four closure methods. Our primary hypothesis was that there would be no difference in initial leak pressure or maximum leak pressure between different closure techniques. We also hypothesized that using skin staples would reduce closure time in comparison with suture closures.

Materials and Methods

The entire jejunum was harvested from the cadavers of 12 healthy dogs immediately following euthanasia with IV infusion of pentobarbital-phenytoin sodium^a, for reasons unrelated to this study, as approved by the Institutional Animal Care and Use Committee. Jejunal segments were excluded if gross abnormalities were present. The jejunum was divided into four 14-cm segments, which were randomly assigned to a treatment group of modified Gambee, simple interrupted, simple continuous, and skin staple closure using a random integer set generator (www.random.org). A 3-cm enterotomy incision was made along the antimesenteric border. All sutured enterotomies were closed with 3-0 polydioxanone swaged onto an included SH half-circle 26-mm tapered needle^b in the assigned suture pattern.

Suture bites were placed 2–3 mm apart and 2–3 mm from the incision edge.^14,15 The modified Gambee suture pattern used in this study was a simple interrupted appositional pattern in which the suture entered the serosa on one side of the incision and exited the mucosal surface, penetrated the mucosa on the same side of the incision and exited out of the submucosa of the cut surface to be approximated, went across the incision into the wound edge opposite and entered the opposing submucosal layer, exited down into gut lumen, entered back through the mucosa and exited through the wall to the serous surface and was tied across the incision (Figure 1). No attempt was made to prevent mucosal eversion when performing the simple interrupted and simple continuous closures. Skin staples^c were placed less than 3 mm apart as previously described by Coolman.¹⁹ A single individual performed all closures (N.R.K.). Time to perform each closure was recorded.

View Figure

• Download as Powerpoint
• Download Figure

Testing of the leak pressure was modified from a previous study.²¹ Briefly, Foley catheters (20-French) were inserted 3 cm into both ends of the intestinal segments and inflated. Doyen clamps were used to clamp the mesenteric border against the Foley catheter to ensure no leakage occurred from the ends of the intestinal segments. A digital pressure transducer^d was attached to one Foley catheter, while colored isotonic fluid was delivered at a constant rate of 378 mL/hr with an infusion pump^e into the other Foley catheter (Figure 2). The transducer was connected to a data acquisition system^f, and was calibrated before each experiment. The enterotomy site was monitored for leakage as the dye containing saline solution was infused. Initial leakage pressure was defined as the pressure at which the dye-containing saline solution was first observed to appear on the serosal surface at the enterotomy site. Once the initial leakage pressure was recorded, testing was continued until the intraluminal pressure reached a plateau, the maximum pressure for the sensor system was achieved, or there was a catastrophic failure of the intestine or closure; this was defined as the maximum leak pressure. Catastrophic failure was defined as an acute drop of intraluminal pressure. The initial leak pressure and the maximum leak pressure were recorded. The method of failure at both the initial leak pressure and the maximum pressure was subjectively assessed to be at a suture or staple hole, the suture knot, or the incision itself.

View Figure

• Download as Powerpoint
• Download Figure

Data were analyzed using a commercially available statistical program^g. The data were tested for normality and a two-way analysis of variance was applied to test the differences among the four suture groups with respect to time, initial leak pressure, and maximum leak pressure. A Tukey’s multiple comparison test was performed for comparison between the groups for each of the three variables. A Kruskal–Wallis test was performed to evaluate the general differences between the groups and Wilcoxon rank-sum test was used for post hoc testing of pairwise comparisons to further assess the effect of time. Significance was set at P < .05.

Results

All dogs were healthy intact female beagle dogs with a mean age of 26.5 ± 9.3 mo, and a mean weight of 13.6 ± 3.5 kg. No gross abnormalities were present in any of the intestines and all intestinal segments were included in the study. There was a significant difference in time of closure among the four suture groups. The skin staple closure was quickest to perform, with a mean closure time of 49.8 ± 8.8 s, followed by the simple continuous closure at 207.1 ± 43.9 s, simple interrupted pattern at 262.2 ± 46.5 s, and modified Gambee pattern at 356.1 ± 60 s. The time for skin stapled closure was faster than all other closures and the modified Gambee suture was slower than all other closures (P < .001). There was no significant difference in time between simple interrupted and simple continuous closures (P = .07; Figure 3).

View Figure

• Download as Powerpoint
• Download Figure

The modified Gambee pattern had the highest initial leak pressure of 50.7 ± 15.4 mm Hg, followed by the simple interrupted pattern at 38.6 ± 15.9 mm Hg, the simple continuous pattern at 30.4 ± 11.8 mm Hg, and skin staple closure at 26.4 ± 25.2 mm Hg. The initial leak pressure for the modified Gambee pattern was significantly higher in comparison with skin staple closures (P = .01; Figure 4). There were no other significant differences between suture patterns in regard to initial leak pressure.

View Figure

• Download as Powerpoint
• Download Figure

The mean maximum leak pressure for the modified Gambee suture pattern was 123.7 ± 28.8 mm Hg; for the simple interrupted suture pattern, it was 81.4 ± 40.4 mm Hg; for the simple continuous suture pattern, it was 75.7 ± 35.7 mm Hg; and for the skin staple closure, it was 37.9 ± 37.2 mm Hg. The maximum leak pressure for the modified Gambee suture pattern was significantly higher than the simple continuous and skin staple closures (P = .009 and P = .0249; Figure 5). There were no other significant differences between suture patterns in regard to maximum leak pressure.

View Figure

• Download as Powerpoint
• Download Figure

No samples leaked at the cuffs. Methods of failure included leakage at the suture hole, or leakage at the incision. Most failures occurred at the suture hole for the modified Gambee closure (83%). The skin staple group samples all failed at the incision line, not at the staple entry site, with several samples leaking immediately (pressure of 0 mm Hg). The simple interrupted and simple continuous closures both had leakage at the suture hole and incision.

Discussion

Enterotomy is frequently performed for the removal of foreign bodies that have caused obstruction of the small intestines.¹ Healthy dogs are estimated to have a jejunal intraluminal pressure of 6–25 mm Hg during peristalsis.²² All closure groups in this study had mean initial leakage pressures greater than 25 mm Hg, suggesting that all techniques may be adequate for closure of enterotomies. However, there was great variability in initial leak pressure with the skin staple closure group, including several that failed below 10 mm Hg (Figure 4). Additionally, several samples leaked immediately with skin staple closure, suggesting that closure in this manner may not be adequate to prevent leakage, which could lead to sepsis in a clinical case. Although this was an ex vivo study, if these findings correlate in vivo, not all enterotomy closures performed with skin staples may withstand normal jejunal intraluminal pressure. Leak testing the intestine following closure as described by Saile et al. can be used to detect leakage at pressures lower than expected to occur during normal peristalsis, although no association has been found between leak testing and risk of enteric leakage.²³

Subjectively, most failures occurred at the needle hole rather than at the incision itself for suture closure methods. The clinical relevance of this leakage is unknown. The amount of leakage may be related to needle and suture size. Future studies comparing different needle size and suture size may show a difference in where failure is most likely to occur. We utilized 3-0 suture in this study as it is commonly used in practice and has been used in previous studies similar to this one. However, it is important to recognize that in smaller patients 4-0 is likely a more appropriate suture size to select when closing an enterotomy. It is possible that leakage at a needle hole has little clinical effect on risk of failure of the enterotomy site leading to significant patient complications, such as septic peritonitis. The skin staple closure group had the most catastrophic failures, with pressure dropping suddenly to zero, and all failed at the actual incision line. Fibrin clots form rapidly in the gastrointestinal tract.^24,25 In ex vivo specimens, no fibrin seal is formed at the enterotomy site, so leak pressure testing is evaluating only the tissue and closure pattern. In a clinical case, a fibrin seal would be expected to form rapidly and contribute to closure strength.

The modified Gambee suture pattern is purported to have less mucosal eversion compared with the other suture patterns used. The benefit of decreased mucosal eversion is a better serosal seal and decreased adhesion formation, as evidenced in the human literature, and experimental animal models.^15,16 The modified Gambee closure had a higher mean initial leak pressure than all other closure types, which was significantly higher than skin staple closure (Figure 4). The modified Gambee suture pattern also and had a statistically higher maximum leak pressure than the simple continuous and skin staple closure (Figure 5). Clinically, initial leak pressure is more relevant, and all hand-sewn closure methods had initial leak pressures greater than physiologic pressures seen in the small intestine (6–25 mm Hg).²² The maximum leak pressures seen here are within the range of previously reported values, although direct comparison with other studies is difficult because most evaluated enterectomy models.^{9,11,19,23,26} The differences seen in leak pressure between studies may be attributable to variance in testing methods and method of intestinal closure.

Limitations of this study include the number of samples tested, and the use of cadaveric tissue rather than live tissue. The effects of tissues undergoing peristalsis and the passage of food boluses, which places the closure under additional stress, could not be evaluated. In addition, when performing enterotomies tissue is often abnormal, and even though no gross lesions were present in any of the specimens, histopathologic evaluation was not performed to evaluate for microscopic changes. The effects of this on bursting strength of various closure methods are unknown. Furthermore, no control group was tested to ensure that tissue breakdown had not occurred, although this is unlikely because the samples were tested within 12 hr of harvest. Finally, the effect of tissue handling could not be evaluated.

It is difficult to evaluate eversion in a cadaveric model, making conclusions on degree of eversion challenging. Evaluating individual closures for degree of eversion would have allowed for an objective evaluation of eversion between suture types. Future studies should include larger sample sizes to avoid a potential type 2 error between modified Gambee and simple interrupted initial leak pressure. A continuous modified Gambee pattern could be evaluated to determine if there is improved initial leak pressure, and faster closure time, compared with other hand-sewn closure techniques. The simple continuous pattern was the most difficult pattern to perform consistently, which may have affected the results of leak pressures found in this study. Placement of the modified Gambee suture pattern was relatively easy to perform, although more time-consuming, with the submucosa being easily visualized as a thin white line. The use of smaller needle types than the one utilized in this study could affect initial leak pressure, or method of leakage. The needle type was selected because it is commonly used in many companion animal practices, and has been used in previous studies evaluating enterotomy and enterectomy closure. Most importantly, in vivo assessment is needed to determine if there is any difference between closure method and long-term outcome in clinical patients, including animals with histopathologic changes of the gastrointestinal tract.

Conclusion

Results of the present study indicate that a modified Gambee pattern showed a significantly higher initial leak pressure than the skin staple closure, and significantly higher maximum leak pressure than the simple continuous and skin staple closure tested. Skin staple closure was found to be fastest to complete, as expected, which is consistent with the findings of previous studies comparing skin staple closure with hand-sewn closures in an enterectomy model.¹⁹ Although skin staples were faster to place than suture closures, their use resulted in leakage below the normal pressure in healthy jejunum with peristalsis in several samples tested. Additionally, skin staple samples showed catastrophic failure with leakage noted immediately with testing, which may suggest an increased risk of sepsis from leakage in vivo.