Introduction

Gastric dilatation-volvulus (GDV) is a commonly encountered condition in large-breed dogs and has a recorded mortality rate of 4.3% to 26.8%.^1–5 Treatment involves rapid medical stabilization, surgical gastric repositioning, and gastropexy to prevent recurrence. Among various gastropexy techniques, incisional gastropexy (IG) is one of the most easily performed and commonly used methods.^6–8 While gastropexy is typically effective in preventing GDV,^7–9 failure of IG has been reported. Hammel et al¹⁰ reported recurrence of GDV after IG in a Rottweiler. Hoogzand¹¹ and Theisens¹² reported failure rates of 6.9% and 9% in 2016 and 2017, respectively.

Since IG can fail, a more secure method may be helpful to avoid future GDV. In 1 case, failure was proposed to have resulted from chronic stretch of the original gastropexy site, resulting in ineffective adhesion formation.¹⁰ In a report of circumcostal gastropexy, security was improved by engagement of gastric submucosa.¹³ Similarly, a modification of incisional gastropexy was developed whereby 2 simple interrupted full-thickness sutures, 1 cranial and 1 caudal to the primary gastropexy incision, are added. These 2 sutures penetrate into the gastric lumen to ensure engagement of the gastric submucosa, the strongest layer of the stomach.^13,14 Adding these 2 sutures may improve the strength of gastropexy and potentially minimize the failure of IG.

The objective of this study was to use tensile testing to compare the acute strength of standard incisional gastropexy (SIG) and modified incisional gastropexy (MIG) as described above. It was hypothesized that (1) MIG would fail at higher force than SIG and (2) failure would be manifested as stretching or tearing of gastric tissue rather than stretching or tearing of abdominal wall muscle or suture breakage.

Materials and Methods

Forty-one pigs were euthanized due to reasons unrelated to this study. The pigs were euthanized between 4:00 pm and 6:00 pm on 3 different consecutive days (day 1, n = 14; day 2, 14; and day 3, 13). Their entire stomachs and right abdominal walls were harvested and stored overnight at 5 °C in lactated Ringer’s solution (LRS). Amounts of time in the chiller were similar across all samples (17 hours ± 30 minutes). Each pair of stomach and body wall was placed in 1 ziplock bag (Ziploc; SC Johnson & Son Inc) and labeled with a number corresponding to the number identification of that pig. Numbers were randomized and evenly distributed into 2 groups. Four samples (2 MIG and 2 SIG) were excluded from surgery due to poor sample quality. The matched tissues in each group were assigned to receive either SIG or MIG by the same person, a board-certified small animal surgeon, in the morning following tissue harvest. The 2 groups were labeled 1 and 2.

The procedures of SIG and MIG were as follows: a 3-cm full-thickness incision in the transversus abdominis muscle was made with a No. 10 scalpel blade. Then, Metzenbaum scissors were used to make a 3-cm incision in the serosa and muscularis of the stomach on the ventral surface of the pyloric antrum. This incision was extended by blunt separation of the muscularis from the submucosa with Metzenbaum scissors followed by a linear cut until the gastric seromuscular incision was the same length (3 cm) as the abdominal wall incision. The measurements of these incisions were recorded (Figure 1). The first bite of suture (0 polypropylene) was placed from the caudal portion of the abdominal wall incision to the left portion of the gastric incision and tied so that the strands were located caudally. The free end of the suture was left long enough to eventually be tied again and tagged with a hemostatic forceps. Then, the dorsally located transversus abdominis and gastric seromuscular edges were apposed with a simple continuous suture pattern. After reaching the cranial extent of the incisions, the simple continuous pattern was continued in a caudal direction, apposing the ventrally located transversus abdominis and gastric seromuscular wound edges until the free end of the original knot was reached. The suture line was completed by removing the hemostatic forceps and tying (3 square knots) to the free end. For those receiving MIG, 2 additional simple interrupted sutures (0 polypropylene) were placed, 1 immediately cranial and 1 immediately caudal to the continuous suture line. These sutures were placed full thickness into the stomach (to ensure submucosa was included) and transversus abdominis muscle. A new pack of 0 polypropylene was used for each gastropexy specimen. The specimens were placed in individual bags containing LRS for transport to the testing laboratory. All specimens were tested on the same day the gastropexies were performed.

View Full Size

Incisional gastropexy using porcine tissue and 0 polypropylene suture. A—The transversus abdominis muscle was incised to create a 3-cm incision, and Metzenbaum scissors were used to make an incision of the same length in the serosa and muscularis of the stomach on the ventral surface of the pyloric antrum. B—The dorsally located transversus abdominis and gastric seromuscular edges were apposed with a simple continuous suture pattern from caudal to cranial. C—The standard incisional gastropexy was completed by apposition of the ventral transversus abdominis and gastric seromuscularis incised edges. D—The modified incisional gastropexy was completed by adding the cranial (white arrow) and caudal (black arrow) full-thickness simple interrupted sutures. Cranial is to the top of the photos.

Citation: Journal of the American Veterinary Medical Association 261, 9; 10.2460/javma.23.03.0141

• Download Figure
• Download figure as PowerPoint slide

For biomechanical testing, the stomach and attached section of abdominal wall, including the gastropexy site, were removed from the LRS. A tensile test machine (eXpert 2600; Admet Inc) was used to measure the force change and breaking force of the suture or tearing force of the sutures when abdominal and gastric tissue were stretched away from each other. The abdominal wall tissue was secured in the top grip (surface area, 2,280 mm²/face), and the gastric tissue was secured in the bottom grip with the same surface area (Figure 2). The top grip moved upward at a constant rate of 20 mm/min. The test process was stopped after the point of failure when the maximum tension was reached. Work done to reach failure, defined as area under the curve with failure load as the end point, was calculated for each sample (Figure 3). The MTESTQuattro software (MTESTQuattro version 5.07.07; Admet Inc) was used to record the generated tension (N). A curve reflecting the change in tension and a corresponding comma-separated values file containing information regarding position and load were produced by the software. The entire process was videotaped for each specimen using a smartphone (iPhone 8; Apple Inc). After each test, specimens were examined to determine whether the mode of failure was caused by suture breakage, untied knot, or tissue tearing.

View Full Size

Porcine stomach (in bottom grip) and abdominal wall (in top grip) in position on the tensile testing apparatus (A) before distraction and (B) after maximum tension was reached.

Citation: Journal of the American Veterinary Medical Association 261, 9; 10.2460/javma.23.03.0141

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Representative tensile curve of standard incisional gastropexy (SIG) in sample number 4 versus modified incisional gastropexy (MIG) in sample number 13. The black arrow indicates the point of failure of sample number 13. The gray shadow indicates the work of sample number 13 to reach point of failure.

Citation: Journal of the American Veterinary Medical Association 261, 9; 10.2460/javma.23.03.0141

• Download Figure
• Download figure as PowerPoint slide

Results

Forty-one samples were harvested. Four samples were excluded due to poor sample quality (incompletely harvested tissue making performing gastropexy impossible). Thirty-seven samples were tested mechanically, and all data (SIG, n = 19; MIG, 18) were included. There was no significant difference between body weights of sample sources between groups (SIG, 14.66 ± 2.80 kg; MIG, 12.78 ± 2.15 kg; P = .1) (Figure 4). The maximum load (N) at failure and the work (N·mm) performed are recorded elsewhere (Supplementary Table S1). The MIG group failed at a higher load (SIG, 52.33 ± 7.61 N; MIG, 67.16 ± 11.61; P = .00003) and required more work to fail (SIG, 19.867 ± 5.07 N·mm; MIG, 32.10 ± 10.63 N·mm; P = .00004) compared to the SIG group (Figure 5). MIG strongly predicts higher failure load (R2 = 0.4; P = .00016) and higher work to failure (R2 = 0.4; P = .00013). All failures were caused by tearing of gastric tissue. No suture breakage or knot untying was observed.

View Full Size

Box plot showing body weights (in kilograms) of all sample sources. The horizontal line represents the median value, the upper and lower limits of the box represent the IQR (25th to 75th percentile), and the upper and lower whiskers represent the highest and lowest values.

Citation: Journal of the American Veterinary Medical Association 261, 9; 10.2460/javma.23.03.0141

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Violin plot comparing SIG (n = 18) versus MIG (19) for (A) failure load (N; P = .00007) and (B) failure work (N·mm; P = .00003). The width of shading represents the number of samples, the horizontal line represents the median value, the upper and lower limits of the box represent the IQR (25th to 75th percentile), and the upper and lower whiskers represent the highest and lowest values. Linear regression was performed using SIG or MIG as predictor variable, load (C; R2 = 0.4; P = .00016) and work (D; R2 = 0.4; P = .00013) as dependent variables, and body weight as adjusting variable. Shading represents the 95% CI.

Citation: Journal of the American Veterinary Medical Association 261, 9; 10.2460/javma.23.03.0141

• Download Figure
• Download figure as PowerPoint slide

Discussion

Among the reports of GDV occurrence after incisional gastropexy or other gastropexy techniques,^10–12,16 none document the exact reason for failure, but failure could be attributed to breakage of suture material, knot failure, or tearing of tissue. In 1 case report,¹⁰ stretching of the gastropexy site was suggested to be the cause of failure. In the study reported here, all failures were caused by gastric tissue tearing instead of suture or knot failure; therefore, one might attribute gastropexy failure to stretching or tearing of tissue rather than suture material breakage. One might argue that the observed difference in failure load between SIG and MIG may be attributed to the 2 additional sutures instead of the engagement of submucosa; however, it has been reported that the number of suture bites does not affect the strength of gastropexy.¹⁷ Instead, the increased failure load of MIG is more likely caused by the engagement of submucosa, where more collagen is incorporated than when only serosa and mucosa are engaged.

Multiple reports have evaluated biomechanical properties and failure loads of different gastropexy techniques either on sacrificed live animals or cadavers.^18–23 However, no optimal force to failure for gastropexy has been determined. The preadhesion failure load ranges from 20 to 80 N among different gastropexy techniques, and postadhesion failure load up to 123.61 N has been observed with circumcostal gastropexy. No report of postadhesion failure load of incisional gastropexy is available.

Polypropylene was the chosen suture for this study, whereas polydioxanone is another popular suture for incisional gastropexy. It is unlikely that the loads to failure would differ between polypropylene and polydioxanone in an acute study like this one, since failures were due to tissue tearing and these 2 sutures are similarly strong.²⁴ However, differences in absorption, reduction in tensile strength, and relative knot security could affect long-term gastropexy strength. A prospective live-animal study would be necessary to determine such differences between polypropylene and polydioxanone. In this present study, we observed that all failures were caused by tearing of tissue instead of suture breakage or knot untying. As a result, no significant difference is expected per acute strength between polypropylene and polydioxanone.

Limitations of this study were related to the use of cadaveric tissue. Whether the increased strength of MIG compared to SIG would decrease the gastropexy failure rate cannot be concluded from this study. Also, complications of MIG could not be determined in this study. For example, in the live animal, penetration of the gastric lumen could facilitate infection into the peritoneum across suture placements. A large-scale retrospective or prospective study would be required to compare the outcomes and complications of dogs receiving SIG and MIG. Another limitation was the inability to perform surgery and biomechanical testing immediately after collection of samples. Storage at 5 °C overnight may attenuate the biomechanical properties of tissue and affect test results. However, all samples were processed under similar conditions (storage time and temperature) to minimize the effect of storage conditions. Furthermore, testing of similarly stored intestinal and esophageal samples have yielded interpretable results.^25–27 Therefore, the observed differences in failure load and work should be valid. Lastly, this study did not compare the strength of SIG and MIG during healing of the gastropexies. Although biomechanical testing of healing gastropexies would yield more directly applicable results, the sacrifice of animals to do so could not be justified. However, the superior acute strength of MIG suggests that MIG would be more secure than SIG during the healing process. Also, it is possible to make additional full-thickness bites as a part of simple continuous suture. Future study is required to compare such technique with SIG and MIG as performed in this study.

In conclusion, failure of acutely performed SIG and MIG occurs due to tearing of gastric tissue, and MIG fails at higher forces than SIG.

Supplementary Materials

Supplementary materials are posted online at the journal website: avmajournals.avma.org