Comparison of single layer staple closure versus double layer hand-sewn closure for equine pelvic flexure enterotomy

Abstract

Our objective was to compare thoracoabdominal (TA Premium™ 90) stapled enterotomy closure to traditional hand-sewn closure, using time to perform the technique, luminal diameter, and bursting pressure in ex-vivo specimens. The pelvic flexures of 13 client-owned horses were harvested. Each pelvic flexure had 1 enterotomy performed; 6 were closed via staples, 7 closures were hand-sewn. Luminal diameter at the enterotomy site was assessed via contrast radiography performed pre-and post-enterotomy. Bursting pressure of the closure was assessed by continuous manometry during rapid infusion. Time to perform stapled closure was significantly shorter than hand-sewn closure (P < 0.0001). Percent reduction of luminal diameters was significantly decreased in stapled specimens (P = 0.034). There was no significant difference in bursting strength between closure techniques (P = 0.196). In conclusion, stapled enterotomy closure offers statistically significant reduction in closure time and better maintains pre-enterotomy luminal diameter without reducing biomechanical strength, compared to a double layer hand-sewn closure.

Introduction

Pelvic flexure enterotomies of the ascending colon are commonly performed in the horse during surgery of gastrointestinal conditions. The use of stapling devices in equine gastrointestinal surgery has been reported for various procedures (1,2). Stapled closure of pelvic flexure enterotomies of the ascending colon using the Thoracoabdominal (TA-90) Premium™ stapling device (Tyco; Princeton, New Jersey, USA) in a clinical setting was recently described (3). Clinical experience with this technique suggests that TA-90 stapled enterotomy closure exhibits survival and post-operative complication rates statistically equivalent to those of hand-sewn closure (4). However, no biomechanical data are available comparing stapled TA-90 enterotomy closure of the pelvic flexure with conventional double layer hand-sewn closure.

Our objective in this ex-vivo study was to compare the time to perform each technique, the percent change in colonic luminal diameter pre-and post-enterotomy for each closure technique, and the bursting pressure between the 2 techniques. We hypothesized that the TA-90 stapling device would decrease time to perform an enterotomy closure, but have similar luminal diameters and bursting strengths compared with the double layer hand-sewn closure.

Materials and methods

Thirteen client-owned horses were euthanized for reasons unrelated to gastrointestinal disease. Owners of all horses donated for this research signed a donation and euthanasia consent form. Horses were euthanized due to orthopedic conditions (n = 9), neurologic abnormalities (n = 1), chronic laminitis (n = 2), and invasive melanoma (n = 1). Horses ranged in age from 10 mo to 26 y (median 20 y). Two geldings and 11 mares were included in the study. Four Quarter horses, 2 Paint horses, 2 Thoroughbreds, 2 Belgian drafts, 2 Arabians, and 1 Welsh pony were used for this study. None of the animals were held off feed or water prior to euthanasia; nor did any of the horses exhibit clinical signs of dehydration.

Large colon segments 60 cm long, centered on the pelvic flexure were harvested from each horse post-euthanasia; these were emptied of ingesta and rinsed with water. This protocol was approved by the University of Wisconsin-Madison Research Animal Resources Center (RARC).

Luminal diameter measurements and surgical technique

Contrast radiography was performed on each colon segment pre-and post-enterotomy to calculate reduction in luminal diameter (2) with the colons distended to 13.6 cm H₂O. After instillation of 60 mL of barium sulfate suspension (E-Z-EM Canada, Liquid E-Z-Paque™; Milan, Italy) (60% weight by volume), colonic segments were occluded at each end with cable ties. Using a 14-gauge 1.5-inch needle attached via tubing to a laparoscopic insufflator (Storz Electronic Endoflator; Storz, Tuttlingen, Germany) and a compressed CO₂ tank, each colon segment was insufflated to a standardized intraluminal pressure of 13.6 cm H₂O. Intraluminal pressure at the time of insufflation was measured by the laparoscopic insufflator.

Radiographs were taken of each colon segment (Figure 1). A 6-cm longitudinal incision was made on the antimesenteric side of each distended pelvic flexure using a #10 scalpel blade. Six of the enterotomy sites were closed as previously described in a single, full-thickness layer with the 4.8 mm TA-90 stapler (Tyco) (3,4). After copious lavage, the enterotomy site was occluded using Allis tissue forceps that everted both mucosa and serosa, grasped each edge simultaneously, and was applied perpendicular to the incision. As little tissue as possible was everted while still guaranteeing continuous closure of both layers. The TA-90 stapler was placed to the luminal side of the Allis tissue forceps leaving only enough room to separate the instruments by Mayo scissor excision. The stapling device was readied to fire and centering of the staple cartridge across the entire enterotomy was assured. The stapling device was fired and the everted tissue and Allis tissue forceps excised. Further lavage was performed and the enterotomy closure was pressure tested with ingesta.

Figure 1.

Radiographs of colon segments for measurement of colonic luminal diameter and calculation of percent change in colonic luminal diameter. a) pre-enterotomy radiograph; b) radiograph of the same colon segment after enterotomy closure with the TA90; c) pre-enterotomy radiograph of colon segment for hand-sewn closure; d) radiograph of the colon segment from Figure 1c after hand-sewn enterotomy closure; e) example of radiographic measurements taken for each colon; f) example of radiographic measurement of the same colon segment from Figure 1e taken at a 90º angle.

Seven enterotomy sites were closed with 2-0 polydioxanone in a full-thickness simple continuous pattern oversewn with a Cushing pattern as previously described (5). Time to perform each enterotomy closure was recorded. All closures were performed by 1 surgical resident (JMR). After enterotomy closure, each colon segment was again insufflated to the same standardized intraluminal pressure and re-radiographed. Percent reduction in luminal diameter was calculated from pre- and post-enterotomy diameters measured radiographically. The colon segments were then deflated via needle decompression prior to analysis for bursting pressure.

Radiographic measurements were made using digital software (Phillips iSite^® Enterprise; Phillips Healthcare). Three luminal diameter measurements were made equidistant from one another on each radiograph and averaged for statistical analysis of each colon segment.

Bursting strength analysis

The bursting strength of each pelvic flexure was determined as described in a jejunojejunostomy model (6) using a fluid pump (Cole-Parmer Masterflex; Vernon Hills, Illinois, USA) to infuse the colon segment while measuring intraluminal pressure via manometry. Rapid infusion tubing connected to a 20-L carboy filled with colored water was placed into the fluid pump. The tubing was tied to a 14-gauge 1.5-inch needle placed through the wall of the oral colonic segment. Each colon was placed in a glass submersion tank filled with warm water and infused at a constant rate of 1 L/min with water dyed with food coloring or methylene blue. The pressure transducer (Posey Cufflator™; Arcadia, California, USA) was applied to the aboral colonic segment (7,8) using tubing attached via Luer lock to a 14-gauge 1.5-inch needle. Intraluminal pressure was continuously monitored via manometry; the maximal quantified measurement available on the manometer was 120 cm H₂O. Bursting pressure (cm H₂O) was defined as the pressure that caused leakage of colored solution at the enterotomy closure or bursting of the colonic wall at any site. The infusion procedure for each colon segment was digitally recorded by video.

Statistical analysis

Data were assessed for normality using the Shapiro-Wilk test. We elected non-parametric analysis for all comparisons using the Mann-Whitney U-test, with P < 0.05 deemed statistically significant. Data were reported as mean +/− standard deviation (SD).

Results

Stapled closures were performed in a mean time of 3.31 min (+/− 0.85 SD). Hand-sewn closures took a mean time of 15.8 min (+/− 1.41 SD). The time to perform stapled closure was significantly shorter than time to perform hand-sewn closure (P < 0.0001).

Mean percent reduction in luminal diameter for stapled specimens was 4.7% (+/− 2.8 SD), while mean percent change in luminal diameter for hand-sewn specimens was 11.98% (+/− 2.2 SD). The decreased percent change in luminal diameter identified in stapled enterotomy closures was statistically significant when compared to that of hand-sewn closures (P = 0.034).

Bursting pressure for 4 of the 6 stapled colons was > 120 cm H₂O (maximum measurement limitation of the pressure transducer used was 120 cm H₂O); the remaining 2 stapled colons failed at 110 cm H₂O (mean stapled bursting pressure of 116.7 cm H₂O, +/− 5.2 SD). Bursting pressure for 4 of the 7 hand-sewn colons was > 120 cm H₂O; the remaining 3 hand-sewn colons burst at 75, 105, 118 cm H₂O (mean hand-sewn bursting pressure of 111.1 cm H₂O, +/− 16.9 SD). There was no significant difference in bursting strength between the 2 methods of closure (P= 0.196). Rupture of the colon occurred adjacent to the enterotomy closure; failure of suture material or lines of staples was not observed (Figure 2).

Figure 2.

a) Photograph of typical failure of stapled enterotomy closure; b) photograph of typical failure of hand-sewn closure.

Discussion

The results of this study support our hypothesis that closure of pelvic flexure enterotomies with the TA-90 stapler was significantly faster than traditional double layer hand-sewn closure and that bursting strength was not statistically different between the 2 closure techniques. The results also illustrate that stapled closure provided a statistically significant decrease in percent change in luminal diameter compared with conventional hand-sewn closure. This finding is to be expected, since one closure is oversewn in an inverting fashion. However, we felt it was important to compare our technique to the standard of care in order to establish the effectiveness of the stapled closure.

Many of the complications associated with equine gastrointestinal surgery are closely related to surgical time. The stapled closure technique decreased enterotomy closure time but also has the potential to decrease tissue handling, subsequent iatrogenic tissue trauma, and contamination, all of which could lead to reduction in patient morbidity (9).

Anecdotal clinical data suggest that the incidence of postoperative complications in TA-90 stapled pelvic flexure enterotomy closures is statistically similar to rates in double layer hand-sewn pelvic flexure enterotomy in both survival to discharge and long-term follow-up (4). However, greater case numbers with consistent postmortem examinations at time of death are needed to ascertain differences between the two techniques in the rates of adhesion formation, impaction at the enterotomy site, and enterotomy dehiscence.

In a previous report by Young et al (1) comparing large colon enterotomy closure techniques in the equine large colon, adhesions were noted on postmortem examination in 2 horses that were euthanized 6 d after surgery. Both horses had a single layer enterotomy closure using a Utrecht pattern encompassing all layers of the colonic wall except the mucosa. Histology at the time of necropsy revealed poor apposition with this closure technique, which the authors concluded resulted in leakage and dehiscence of the enterotomy site. This inverting closure technique, though single layer, is inherently different from the single layer closure technique performed in the current study, which was everting.

An additional concern regarding adhesion formation at the stapled enterotomy site is mucosal eversion and subsequent exposure with potential contamination due to the stapling technique. Previously, stapled intestinal closures have been associated with increased complications including stenosis, adhesion formation, leakage, and hemorrhage (10). Specifically, double layer handsewn pelvic flexure enterotomy closure inverts not only the mucosal closure but also all suture material used to perform the closure, thereby reducing the risk of adhesion formation (5). However, retrospective data reported anecdotally (3) and recently corroborated by Rosser et al (4) found no increase in post-operative adhesion formation. Second look laparoscopy at varying post-operative times may be useful in determining the long-term risk of adhesion formation in clinical cases with stapled enterotomy closure.

It is important to note that the cost of TA-90 staple cartridges is substantially greater than that of suture material required to close an enterotomy. In cases where the surgeon wishes to reduce enterotomy closure time, closure with the TA-90 stapler can be done without causing an immediate difference in bursting pressure.

In conclusion, pelvic flexure enterotomy closure with a TA-90 stapling device in a single, full-thickness layer provides sufficient strength while decreasing enterotomy closure time and reducing percent change in luminal diameter compared with the traditional, hand-sewn double layer closure.