Meckel's diverticulectomy: a multi‐centre 19‐year retrospective study

Meckels diverticulum (MD) causes a number of acute surgical pathologies and can contain ectopic tissue with the surgical aim to resect all ectopic mucosa. This has traditionally implied a small bowel resection (BR); though contemporary literature has demonstrated Meckel's diverticulectomy to be safe. The aim of this study was to determine optimal resection strategy, and assess MD histopathological features and their relationship to outcomes.


Introduction
Meckels diverticulum (MD) is a congenital anomaly of the gastrointestinal tract occurring in 1%-3% of the population. 1 It is a blind intestinal pouch located on the anti-mesenteric border of the ileum, in the axis of the superior mesenteric artery, and is embryologically derived from the incomplete atrophy of the omphalomesenteric (vitelline) duct during early gestation. 1 Although the majority of MD remain asymptomatic, some have potential to cause acute surgical pathologies, including gastrointestinal bleeding, intestinal obstruction and perforation. MD are typically lined by ileal mucosa, but can contain ectopic mucosa (usually near the tip) which is most commonly gastric origin, but can be of duodenal, colonic, pancreatic, Brunner's glands, hepatobiliary and endometrial origin. 2 In diverticula that are short and broad based, this ectopic tissue may theoretically extend into the base and ileum, with the potential for complications related to acid secretion and malignant transformation. Therefore, open segmental small bowel resection with primary anastomosis is commonly performed 3 but not without consequence as has been found to be associated with higher rates of postoperative morbidity when compared to stapled diverticulectomy. 4 The surgical objective in resection of MD has been to resect all ectopic mucosa without increasing patient operative morbidity. However, in the literature there have been no reported cases of complications or malignancy from remnant ectopic mucosa in the ileum following diverticulectomy. Laparoscopic stapled Meckel's diverticulectomy is technically safe, 5 in both emergency and incidental settings, 6 and therefore avoids the morbidity associated with open surgery and small bowel resection. The aim of this study is to determine and compare the characteristics, and management of MD that underwent diverticulectomy or bowel resection within the Northern Sydney Local Health District experience. Secondary outcomes include assessment of MD histopathological features in the context of perforation, malignancy and complications.

Patient selection and data collection
The Northern Sydney Local Health District surgical pathology database was interrogated between the years of 1998 to 2017 to identify patients who had undergone surgery for MD. This included data from seven hospitals. The clinical records of these patients were then retrieved on the electronic medical records from year 2012 to 2017 and by paper medical records prior to year 2012.
Clinical and pathological data fields collected are summarized in Table 1. A symptomatic MD was defined as having a pathological process that correlated with the pre-operative symptoms. An asymptomatic MD was defined as being identified incidentally at the time of operation where the patient did not manifest symptoms that contribute to the pre-operative diagnosis.
Research protocol was approved by the Northern Sydney Local Health District ethics committee as a negligible/low risk project. Data was de-identified and retrospectively collected, and therefore informed consent was not required from each patient.

Statistics
Descriptive statistics were performed, and analysis was conducted using an independent t-test for continuous variables and the chisquared test for categorical variables to compare patient population characteristics. Univariate regression was performed for risk factor analysis. A P < 0.05 was considered statistically significant. IBM SPSS Statistics version 28 7 was used for analysis.

Results
A total of 160 patients underwent resection of MD. The mean age of patients with MD was 43.6 years (SD 21.5 years). Overall, MD was more common in males (71.2%) than in females (28.8%). The mean age for patients who underwent Meckel's diverticulectomy was 39.2 years, and a M:F ratio of 2.9:1, compared to mean age of 47 years and M:F of 2.2:1 for patients who underwent small bowel resection. In this cohort, 92 (57.5%) patients presented with symptoms attributed to MD pathology whereas 68 patients (42.5%) had MD removed after it was found incidentally. A total of 70 patients (43.7%) underwent Meckel's diverticulectomy and 90 patients (53.6%) underwent small bowel resection. Full comparisons are provided in Table 1.

Presenting symptoms
Demographics of the 92 patients with symptomatic MD identified the most common presentation was abdominal pain followed by obstruction (Table 1). No significant difference was identified between symptomatic and asymptomatic patients for age, gender, ectopic tissue, diverticulectomy use, complications, or malignancy rates (Table S1). Presence of ectopic tissue was significantly associated with symptomatic patients (P = 0.025).
Sixteen asymptomatic patients underwent elective resection. There was no significant difference in gender (P = 0.388), complication rate (P = 0.119) or ectopic tissue (P = 0.589) when compared to the symptomatic cohort. An older mean age of 62.6 years was significant P < 0.001 (SD 15.8). Right lower quadrant abdominal pain was the most common presenting complaint for patients who underwent Meckel's diverticulectomy (40%) and obstruction in those undergoing bowel resection (27.8%) ( Table 1).

Histopathology findings
In the stapled diverticulectomy cohort, the mean length of MD (Table 2) was 36.03 mm (SD 18.7), mean width was 18.81 mm (SD 8.7), and length to width ratio of stapled diverticulectomy was 2.12 (SD 1.3). In comparison, the small bowel resection cohort demonstrated a MD with mean length of 37.47 mm (SD 18.3), width of 23.61 mm (SD 18.5), and length: width ratio of 1.84 (SD0.9).
Overall, acute pathology was demonstrated in 88 patients (55.3%) ( Table 1). Fifty percent of the Meckel's diverticulectomy cohort demonstrated abnormal results, compared to 60% of those who underwent small bowel resection. The most common finding in both groups was Meckel's diverticulitis (20.6%), followed by ulceration (18%) and necrosis (8.8%%), occurring with similar frequencies. Ectopic tissue was present in 46 patients (28.8%) overall and was most commonly of gastric origin (54%), with a larger proportion occurring in the bowel resection cohort compared to those who underwent Meckel's diverticulectomy. Of the 92 patients (57.5%) with pathology, 48 (27.5%) had location of pathology described. This was most commonly in the tip (27.3%) in the Meckel's diverticulectomy cohort, compared to entire diverticulum involved (50%) in the small bowel resection cohort. Patients that had ectopic tissue present had significantly greater length: width ratio comparison to those that did not ( Table 2) but individually did not have a significant difference in length or width. Comparison of incidence in long MD defined as length: width ratio greater and less than 1.6 or 2.0 showed no significant difference in ectopic tissue located in the tip (P = 0.199 and P = 0.382), but more patients with ectopic tissue in MD ratio >2.0 (P = 0.043) but not >1.6 (P = 0.8).

Perforated Meckel's diverticulum
In all, 24 (15.3%) MD were perforated, of which 16 had small bowel mucosa, 5 had gastric mucosa, 2 had mixed ectopic mucosa and 1 carcinoid tissue. In patients with perforation, the average length was 42.1 mm (SD 17), width 21.3 mm (SD 10.5) and length: width ratio 2.5 (SD 1.8) narrower and longer than those without perforation (ratio 1.9, P = 0.012). Of those with identified location, tip perforation was most common. Wall perforation commonly occurred at the junction of ectopic and normal tissue.

Risk factors
The MD length: width ratio was a risk factor for perforation OR

Outcomes
A total of 15 patients (9.4%) suffered complications, with the majority (73%) in the small bowel resection group (P = 0.044). Two mortalities were encountered post small bowel resection both via open approach (Table 3). One patient with appendicitis that underwent Meckel's diverticulectomy required re-operation for neuro-endocrine tumour within the mesoappendix. Malignancy was found in seven patients (4.4%) and in one patient with perforation that underwent small bowel resection. Of these patients with malignancy, three underwent Meckel's diverticulectomy with no subsequent complications. Follow up was 12 months postoperatively.

Discussion
Meckel's diverticulectomy has been demonstrated to be a safe method in the management of symptomatic MD in both adult and paediatric cohorts. 8,9 In this cohort, the approach to MD resection was almost equally split between diverticulectomy and small bowel resection (56% vs. 44% respectively), based on surgeon preference, with the majority of Meckel's diverticulectomy performed with stapled resection (86%), utilizing a linear gastrointestinal stapler applied to the diverticulum at its base ( Fig. 1), Diverticulectomy in asymptomatic and symptomatic groups demonstrated similar outcomes and safety, even in perforation despite significant differences in MD size. Previously, two case series have demonstrated that short and broad-based diverticula have ectopic tissue present throughout the whole diverticula, with segmental small bowel resection therefore traditionally preferred surgical technique 10,11 to ensure clearance of all ectopic tissue. However, the results of this study demonstrate greater complication rates in open bowel resection compared to stapled diverticulectomy. In addition to fewer complication rates, our results support the safety of stapled diverticulectomy in the management of both symptomatic and perforated MD, as perforation is most likely at the tip, as observed in our cohort. The reported incidence of perforation associated with symptomatic MD varies between 0.5% and 12% within the literature, 1,12 compared to 15% in our study of an adult population. Perforation is commonly secondary to diverticulitis or ulceration and more rarely, trauma from a foreign body. In addition to pathology, acid secreted by ectopic mucosa (most commonly gastric), may pool in single area and cause erosion and bleeding downstream from the point of acid secretion. However, these ectopic tissue nests, present in 29% of our cohort (10%-60% in the literature), are commonly asymptomatic and. 5,13,14 Thus, the management of asymptomatic MD would suggest resection due to potential malignant transformation of ectopic heterotopia which are estimated to account for 0.5%-3.2% of complications. 12,[15][16][17] Comparable to previous reports on incidence, 5% of our cohort were associated with malignancy. 18 Of note, there were no cases of malignancy arising from ectopic mucosa within the ileum within our study cohort. Isolated resection of the MD with stapled diverticulectomy therefore represents a safe oncological approach to resection of MD. While a wedge resection (Fig. 2) may provide the benefit of mucosal visualization to ensure complete excision of ectopic tissue, a laparoscopically stapled diverticulum can be sent for frozen section analysis intraoperatively 19 to assess the base for ectopic or malignant tissue. If the resected diverticulum has only ileal mucosa at the base, simple stapled diverticulectomy should suffice 5,13,14 thereby also favourably avoiding the morbidity of open operation and small bowel resection.
Size of MD, in particular the heterogeneously described "long MD", has been a focus of interest in the management of MD as a risk factor for ectopic tissue distribution with a predominance for tip ectopic tissue described in the literature where length: width ratio is greater than 1.6 or 2.0. 10,11 This tip predominance was not evident in our cohort at either cut-off ratio, however MD ratio >2 correlated with presence of ectopic tissue but length and width independently did not. Symptomatic patients had a greater incidence of ectopic tissue which supports literature recommendations that all symptomatic patients undergo resection. Furthermore, we propose that intra-operative assessment of a MD size ratio >2, and  not MD length assessed in isolation, as a relative indication for resection. There is ongoing debate surrounding the benefits of prophylactic resection for MD when found incidentally in adults. Numerous studies advocate for the resection of MD found intraoperatively based on the risk of future complications or malignant transformation, while others advocate for a conservative approach given the morbidity associated with resection. 20,21 MD has been associated with a 4%-9% lifetime risk of complications 13,20 and a 70-fold risk of malignant transformation compared to any other site in the ileum. 19 In light of this evidence and given the overall low complication rate, this study supports laparoscopic stapled resection of MD found intraoperatively.

Limitations
As this is a retrospective analysis of records, this study was subject to recall bias. Data that was collected was limited to the recorded information at the time of surgery. In turn this resulted in incomplete data as evident in the microscopic perforation sites. Retrospective analysis restricts ability to reliability identify factors contributing to each surgeon's resection choice, thereby limiting our conclusions somewhat. Insufficient number of cases with perforation limited appropriate risk factor analysis of variables outside length: width ratio including age, type of ectopic tissue, presence of malignancy or symptomatic presentation. In addition, by the nature of how the database was queried, cases in which MD was found incidentally at time of operation but not operated on were not included.
Confounding factors of patients undergoing bowel resection including presenting pathology, operative indication and higher perforation rates, must be acknowledged, as the evidence suggests that small bowel resection is associated with increased postoperative morbidity, particularly in regard to wound infection and mechanical obstruction. 3

Conclusion
Our study demonstrated Meckel's diverticulectomy is safe compared to small bowel resection with fewer complications and complete resection of ectopic tissue. Despite higher perforation rates in MD with greater length: width ratio, no malignancy or ectopic risk was identified, supporting diverticulectomy as a safe operative approach for complete resection not requiring subsequent re-operation. Confounders of surgeon's resection preference exist, and future prospective studies including long term postoperative outcomes are needed to fully understand the risks and benefits of the surgical approach of MD.