Postoperative Adhesions: From Formation to Prevention

Zeynep Alpay, MD; Ghassan M. Saed, PhD; Michael P. Diamond, MD


Semin Reprod Med. 2008;26(4):313-321. 

In This Article

Clinical Aspects

Laparotomy Versus Laparscopy

In clinical practice, the type of surgical approach is debated as an important factor that influences the development and extent of postoperative adhesions. However, it has been reported in a comprehensive literature review that the rate of adhesion development after gynecologic surgery is between 55% and 100%, regardless of the surgical approach, laparotomy or laparoscopy.[18] In an animal study that compared adhesion development after CO2 laser treatment between laparotomy and laparoscopy, the mean postoperative adhesion scores were similar in both groups.[19] In a prospective randomized human study, adhesion development was evaluated in 105 patients with ectopic tubal pregnancy. They were treated with either laparotomy or laparoscopic intervention. Significantly less adhesion development was observed in the laparoscopy group compared with the laparotomy group during second-look laparoscopies, but the degree of adhesions were not different.[20] However, in another study, adhesion development after ovarian endometrioma resection by laparotomy has been found to result in less frequent adhesions than did laparoscopy (54.2% vs. 37.2%).[1] In a meta-analysis, Wiseman et al searched the incidence of adhesion development after abdominopelvic surgery in clinical reports from 1966 to 1996. Analysis included the incidence of adhesions (present or absent), type of adhesions (de novo or re-formed), type of surgical procedure (laparotomy vs. laparoscopy) and the intraperitoneal instillates used in the surgery (saline vs. Ringer's lactate). When comparisons were made between de novo type 1A and 1B adhesions formed at both laparotomy and laparoscopy, adhesion-free outcome (number of adhesion-free sites/total sites) was higher for de novo 1A adhesions (72% for laparotomy vs. 82.4% for laparoscopy) than for de novo 1B adhesions (20% laparotomy vs. 36% laparoscopy). The adhesion-free outcome for de novo 1B adhesions in surgical sites undergoing laparotomy was higher than that for re-formed adhesions (45.2% vs. 26.6%, respectively; p < 0.00001). In sites undergoing laparoscopy, the incidence of de novo 1B adhesions was 37.2%, whereas the incidence of re-formed adhesions was 14.3% (p = 0.092). Adhesion-free outcome for de novo 1B adhesions at sites undergoing laparotomy was greater without crystalloid use than with use (45.2% vs. 20%, respectively; p = 0.018). In laparoscopic procedures, adhesion-free outcome for de novo 1B adhesions was greater in patients who did not receive Ringer's lactate than in those who received it (71.7% vs. 25%, respectively; p = 0.00003).[21]

The Operative Laparoscopy Study Group has assessed the frequency and severity of adhesion re-formation and de novo adhesion formation after operative laparoscopy by a prospective, multicenter study. Operative laparoscopy was performed including adhesiolysis in 68 patients, followed by a second-look procedure within 90 days. Whereas a significant reduction of adhesions was noted after laparoscopic adhesiolysis (52% in total mean adhesion score), at the second-look laparoscopy, however, 97% of the patients had adhesion re-formation.[18] Adhesion re-formation was seen at 66% of the sites where adhesions were lysed during the initial surgery, whereas de novo adhesion development was observed in 12% of patients.[18] Second-look laparotomies have shown an incidence in patients of 51% for de novo adhesion development after reproductive pelvic surgeries.[22] In conclusion, adhesion re-formation is common after both laparoscopy and laparotomy; however, laparoscopy may be associated with a reduced risk of de novo adhesions in nonoperative sites in comparison with laparotomy.[23]

Despite the fact that there is no clear-cut evidence to support the preventive nature of laparoscopic surgery on the development of adhesions, various advantages of this type of surgical approach have made it more popular. Smaller skin incision size, early return of bowel function and ambulation, less bleeding, and shorter hospital stay are all advantages of a laparoscopic approach. Additionally, there is the potential for less exposure to foreign bodies such as glove powder, laparotomy pads, and contact with the tissue during laparoscopy. The pneumoperitoneum maintained during laparoscopic surgery helps control hemostasis by the tamponade effect. However, it is associated with hypoxic, hyperbaric, and acidotic environment that may potentially accelerate the ischemic injury to the mesothelial cells of the peritoneum, which may contribute to development of adhesions.[24] Additionally, such pressure may transiently suppress venous bleeding or oozing from veins and low-pressure vessels. In fact, Ott et al have demonstrated a detrimental effect of insufflation on peritoneal architecture.[25] Consistent with this observation, Molinas' group has demonstrated that insufflation per se can cause adhesion formation and does so in a time and intra-abdominal pressure related manner.[26]

The type of surgery is another factor in the development of postoperative adhesions. Myomectomies are associated with adhesion development regardless of the type of surgical approach. The risk of adhesion is more than 90% and 70% after abdominal and laparoscopic myomectomy, respectively.[27] It is also known that the location of the myomectomy scar may affect the outcome; the incidence is highest (94%) with posterior uterine wall myomectomies but lower (56%) with anterior uterine wall incisions.[28,29] One case-control study compared 16 abdominal myomectomies with 16 laparoscopic myomectomies and found reduced adhesion scores and less adhesion development with laparoscopic myomectomy.[30] Another study has described that suturing the myomectomy incision site increased the incidence of adhesions.[31] It is also know that the length of the incision affects adhesion development. In the Seprafilm Adhesion Study Group, it has been shown that the larger the incision length, the more extensive the adhesion development.[28] Thus, there should be a randomized controlled study to compare laparotomy and laparoscopy groups with the same-size uterus, fibroids, and incisions to achieve comparability.

Peritoneal Closure Versus Nonclosure

In gynecologic surgery, closure of peritoneum in postoperative adhesions prevention has been another topic of debate for years. Because there are a limited number of studies with conflicting results on this subject, the practices of many gynecologists depend on their personal experiences. In a large retrospective, nonrandomized study, patients who had laparotomy, of which 165 had nonclosure and 168 had closure of the peritoneum during reproductive surgery with Pfannenstiel incisions, were evaluated. There was no difference in the incidence of adhesion formation to the anterior abdominal wall in the two groups after second-look laparoscopies (22% in closure group vs. 16% in nonclosure group).[32] In this study, the second-look laparoscopies were offered only to the patients who did not achieve pregnancy 6 months after their first surgery. However, in another study, closure of pelvic and periaortic peritoneum significantly increased the risk of postoperative adhesion formation to the pelvic sidewall in patients who underwent radical hysterectomy and pelvic and periaortic lymphadenectomy for ovarian cancer (8.9% in closure group vs. 5.8% in nonclosure group; p < 0.01).[33] These patients had one course of chemotherapy before the second-look laparotomies. The extensive oncologic surgery involving retroperitoneal space and the difference in the nature of the oncologic patients and their healing process[34] compared with the nononcologic surgery patients and the chemotherapy that they received could be the confounding factors contributing to different outcome.

Closure of the peritoneum during cesarean section is also a controversial issue. Whereas some groups found that closure of the peritoneum helps prevent future adhesion formation,[35,36,37] others proposed that closure of the peritoneum significantly increased the incidence of adhesion formation.[38] Komoto et al compared the adhesion development in 124 patients who had peritoneum closure and nonclosure in prior cesarean sections during their repeat cesarean sections. A total of 63 patients were reoperated and evaluated for adhesion development in their second cesarean sections. They reported that the incidence of adhesion development in the closure group was significantly higher than in the nonclosure group (10 of 27 in closure group vs. 2 of 23 in nonclosure group; p < 0.05).[38] They also reported that severe adhesions that required adhesiolysis prior to uterine incision were encountered more frequently in the closure group (6 of 27) than in the nonclosure group (0 of 23) (p < 0.05). Anterior adhesions were seen more in the closure group (9 of 27) than in the nonclosure group (2 of 23) (p < 0.05). No difference in the extent of adhesions in the two groups was observed.

In a different prospective, randomized controlled study, 360 patients undergoing primary lower segment cesarean section were divided into three groups. The visceral and parietal peritoneums were left unclosed in group A. The parietal peritoneum was closed in group B, and both visceral and parietal peritoneums were closed in group C. The patients had a standard surgical technique in terms of suture material, uterine closure, anesthesia type and medication, and postoperative care. The patients were also controlled for their age, indication for the cesarean section, and maternal variables including previous medical and surgical history, infections, and complications. These patients were followed for 16 years, and 65 of 360 patients (20 in group A, 20 in group B, and 25 in group C) underwent second cesarean section during this period. The adhesion scores of these patients were compared in their second cesarean sections. There was no statistical significance in adhesion formation between the groups. However, they found more severe adhesions in patients in group C (3 of 25) than in both group A (1 of 20) and group B (1 of 20). When patients who had mild (< 3 cm band), moderate (> 3 to 5 cm band), and severe (> 5 cm band and difficult for lysis) adhesions were evaluated, it was found that the mild adhesions were mostly along the incision of the abdomen, always involving the omentum, and the severe adhesions were extended to the lower uterine segment and bladder wall area.[39]

In contrast, another study followed 45 patients who were randomized as peritoneum closure (24 patients) versus nonclosure (21 patients) during their primary cesarean section. All patients had the same surgery, in terms of the type of suture material used and the type of uterine closure, and had no difference in the incidence of postoperative morbidity. The adhesion scores of the patients having their second cesarean section (total of 31 patients: 18 in closure group and 13 in nonclosure group) were evaluated. It was found that in the nonclosure group, 7 patients (54%; p = 0.05) developed adhesions of which 2 of them had severe adhesions (not easily removable adhesions between uterine incision and bladder), whereas in the closure group, 3 patients (15%; p = 0.05) had developed adhesions that were mild in nature (easily removable between uterine incision, omentum, and rectus muscle).[37]

Furthermore, Lyell et al evaluated 173 patients who were enrolled to have their first repeat cesarean section in a prospective cohort study.[35] The adhesions were compared between patients with prior cesarean sections who had closure of parietal peritoneum versus patients with the parietal peritoneum left open. Patients with prior closure had significantly fewer dense and filmy adhesions between omentum and fascia (10% vs. 30%; p = 0.03), omentum and uterus (10% vs. 23%; p = 0.042), and fascia and uterus (12% vs. 27%; p = 0.021) (overall: 52% vs. 73%; p = 0.006). Adhesions at bowel or other pelvic structures were not found to be significantly different between the two groups. When the incidence of dense adhesions was exclusively compared, the results between omentum and fascia were significant in the closure group (30% vs. 45%; p = 0.043).[35] Another factor that could explain the differences in the results of the above-mentioned studies may also be attributed to the increase in plasminogen activator activity (PAA) levels in pregnancy. This was proposed to be a protective mechanism for adhesion development. Because there are limited studies on this subject, further investigation focusing on the question of whether the different outcomes result from the peritoneal closure during cesarean section, the physiologic changes during pregnancy or puerperium, or from other unknown factors is certainly needed.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.