The Histologic Effects of Retention Sutures on Wound Healing in the Rat

A s the reported incidence of wound dehiscence in patients who have undergone major abdominal operations is often in the range of 1–3%, there has been great interest in devising methods to prevent “the burst abdomen.”^1,2 A primary prophylactic measure designed to reduce tension on wound edges involves the use of retention sutures. When first described by Reid in 1933, retention sutures were silver wire sutures placed with wide tissue bites through all layers of the abdominal wall and tied just tightly enough to allow approximation of tissue.³ Today, retention sutures are made of heavy, nonabsorbable synthetic material and often include an added buttress device for skin protection.¹ Despite the widespread utilization of retention sutures, their use is controversial. Many studies have discussed closure with retention sutures,^3–6 suggested ways to enhance their efficacy,^7–13 and characterized their clinical side effects.^1,2 The authors’ study was designed to determine the histologic effect of retention sutures on the inflammatory, proliferative, and maturation stages of wound healing in the rat. Methods Four groups of 24 Sprague-Dawley rats, weighing approximately 250 g each, were used in this study. The experimental protocol was approved by the university animal research committee, and throughout the study, all university guidelines for the use of animals in research were strictly followed. After administration of intramuscular anesthesia with ketamine and xylazine, the animals underwent a procedure to create full-thickness skin defects. In Group 1, a square defect with 6-cm sides was created on the dorsum of each rat and was closed with evenly spaced simple 3-0 nylon sutures only. Group 2 animals were similar to those in Group 1; however, their wounds were closed with 2-0 nylon retention sutures as well as with the evenly spaced simple 3-0 nylon sutures. In Group 2, 1 retention suture was placed 2 cm from each wound end, thus leaving 2 cm between the 2 retention sutures. Group 3 rats were identical to Group 1 animals, except the initial defect created was 7 cm wide rather than 6 cm, thus placing more tension on the closure. Group 4 animals were identical to those in Group 2, except they also had 7-cm initial defects created. To standardize the defects, a square template with 6-cm sides was used to trace the planned defect on all animals in Groups 1 and 2, whereas a template 6 cm long by 7 cm wide was used to trace the planned defects onto rats in Groups 3 and 4. To further standardize surgical technique, the same surgeon placed and tied all sutures in all groups. Six animals from each group were sacrificed and studied at 2, 4, 7, and 25 days postoperatively. These time periods correspond to 4 critical time periods in wound healing: the inflammatory phase, the transition from the inflammatory phase to the proliferative phase, the proliferative phase, and the maturation phase, respectively. At the time of harvesting, skin specimens along the incision line of each rat were processed with hematoxylin and eosin staining for the study of neutrophils, macrophages, lymphocytes, collagen, and fibroblasts. Cell types were differentiated and counted by a well-trained veterinarian pathologist using a high-power microscope. Evidence of infection and necrosis was assessed. A total of 96 slides (1 from each rat) were reviewed. All 96 rats tolerated the initial procedure without any difficulties. There were no postoperative complications. Histologic analysis of each slide was performed by the same veterinarian pathologist. The pathologist performed histologic analyses of the tissue samples taken from the wound edges and was blinded as to the group from which each slide was taken. The hematoxylin- and eosin-stained slides were evaluated for numbers of neutrophils, macrophages, lymphocytes, and fibroblasts. Each cell category was classified on a scale of 1 to 4 where 1 equaled a minimum, 2 equaled a mild, 3 equaled a moderate, and 4 equaled a large amount of cells per high-power field. In addition, each slide was examined for collagen density, collagen arrangement, and the histologic presence or absence of infection and necrosis. Several high-power fields were examined in each specimen. To help maintain uniformity during the evaluation process, the same veterinarian pathologist reviewed and graded all slides using an estimate of cell numbers to categorize cells into the scale of 1 to 4 as previously described. Increased concentrations of neutrophils were important in identifying areas of infection, and nonviable tissue was important in identifying areas of necrosis. Collagen arrangement was not assessed until Days 7 and 25, because the authors thought that Days 2 and 4 might be too early in the wound healing process to make meaningful collagen arrangement assessments. A statistical analysis was performed on the data collected. Analysis of variance using a 4 x 2 x 2 model was used to evaluate interactions between the data collected on 4 different days (Postoperative Days 2, 4, 7, and 25) with 2 different wound widths (6 cm and 7 cm) and 2 different suture types (simple sutures with and without retention sutures). Analysis of collagen density (with and without retention sutures) and the number of neutrophils, macrophages, lymphocytes, and fibroblasts seen was completed with a 2-tailed, 2-sample t-test. The 2-tail Fisher’s exact test was used to analyze the histologic evidence of wound infection, wound necrosis, and scattered collagen formation with the presence or absence of retention sutures. Results There was a statistically significant increase in inflammatory cell counts (Figures 1–3) on Postoperative Day 4 in wounds closed with retention sutures versus wounds closed without retention sutures (neutrophils p 4,5,6,8 One study suggests the use of retention sutures in patients with 3 or more risk factors.⁶ Though patients with these risk factors may have an increased likelihood of wound dehiscence, the cause of the dehiscence is believed to be mechanical. Studies suggest that wound dehiscence is primarily due to pressure necrosis.^8,13 Pressure is exerted on wounds by retention sutures and causes tissue ischemia, leading to weaker wound tensile strength, necrosis, and possible dehiscence.⁸ It has been shown that wounds closed with retention sutures are most likely to dehisce on Postoperative Days 8–10.^2,5,6,8 Since this dehiscence occurs so early in the wound healing process, it can be assumed that the weakness is secondary to intrinsic tissue factors (once knot slippage has been ruled out) rather than any defect in the strength of the sutures themselves. Many researchers have proposed new techniques to reduce the deleterious pressure of retention sutures while retaining their purported utility in enhancing wound stability. One prospective study suggests that maintaining a suture tension of 300 g, far less than most surgeons typically use, decreases the wound dehiscence rate significantly.⁸ Another study recommends the use of adjustable ties that allow the surgeon to tighten or loosen the retention sutures over the course of wound healing.⁷ A sidewinder technique, which spreads tension over a larger surface area, has been proposed,¹⁰ while bolsters have been suggested as part of a method to equally distribute wound holding force.⁹ Despite the controversy regarding the role of retention sutures, they remain in common use. One study suggests that as many as 10,000 patients per year in Germany are treated with retention sutures even though their side effects can include increased perioperative pain, decreased mobility, intestinal damage, and local complications, such as skin maceration and infection.¹ In addition, this study did not find evidence of a beneficial effect of retention sutures to prevent or treat abdominal fascial dehiscence. On a histologic level, the present study found several potentially deleterious effects associated with the use of retention sutures. The authors noted a significant increase in inflammatory response and histologic evidence of infection in wounds closed with retention sutures on Postoperative Day 4. Also, the presence of greater numbers of neutrophils, macrophages, and lymphocytes was significant. This may have represented a foreign-body reaction to the retention sutures. The increased inflammatory response with retention sutures on Postoperative Day 4 may have also resulted in greater numbers of other cell types in the wound, including the increased number of fibroblasts observed. Therefore, the inflammatory response alone may not have inhibited wound healing, particularly as it was associated with increased numbers of fibroblasts. Since the inflammatory response was also associated with an increased cellular infiltrate consistent with early infection, it did not appear to help wound healing in this model. By Day 7, this abnormally increased inflammatory response in wounds closed with retention sutures seemed to have subsided, as there were no longer any statistically significant increases in the numbers of cell types compared to wounds closed without retention sutures. Retention sutures also appear to have a long-term effect on rat wound remodeling, as the authors noted a significant increase in scattered (disorderly) collagen fibers on Postoperative Day 25 in wounds closed with retention sutures. Thus, although the wounds containing retention sutures had increased numbers of fibroblasts on Postoperative Day 25, the collagen produced was less likely to be orderly. Of note, this long-term increase in the number of fibroblasts was likely due to a separate phenomenon than the increase of fibroblasts and other cell types noted on Day 4, as this response on Day 4 was attributed to retention suture-related inflammation that subsided by Day 7. The long-term lack of organized collagen arrangement noted on Day 25 may provide a rationale for the increased risk of incisional herniation that has been found in wounds utilizing retention sutures.^2,11 Further studies that compare the tensile strengths of wounds closed with and without retention sutures may help determine the best method of closure to prevent incisional hernias. Studies that look at the specific effects of retention sutures on fascia may be beneficial as well. Wound healing studies involving retention sutures in higher mammals with skin that more closely resembles human skin, such as the pig, will also be beneficial in extrapolating results to humans. Rats have looser skin than humans, so wound closures of any length are more likely, in general, to have greater tension in humans than in rats. Conclusion Although retention sutures are commonly used as a prophylactic measure against abdominal dehiscence at surgical sites, caution is warranted in their use. The present study suggests that retention sutures increase the inflammatory response and localized infection incidence in postoperative wounds in the early stages of rat wound healing. In addition, retention sutures promote a scattered pattern of collagen formation at rat wound sites in the long term. These effects may inhibit normal wound healing and decrease the tensile strength of scar at an incision site. The results in this rat model, therefore, suggest that the use of retention sutures may not be supported, but further conclusions will depend on additional studies in higher mammals. Acknowledgements The authors gratefully acknowledge the Auto Suture Company for donating the sutures used in this study; Dr. Stephen Brunnert, a veterinarian pathologist, for assisting in the histologic analysis of the specimens; and Dr. Daniel F. Heitjan and Guagong Ma for their aide in the statistical analysis of the data.