Diverticular disease is a significant problem in developed countries, with a prevalence of diverticulosis of 65% to 80% of the population by 80 years of age.1 Of those affected, 1% to 4% will develop an episode of acute diverticulitis,2 with an estimated 1 in 2000 receiving an operation.3 Peritonitis caused by perforated diverticulitis is classified, according to Hinchey criteria, into purulent or fecal, Hinchey grade III and IV, respectively.4 Current best practice for these grades of perforated disease is emergency surgery, with colonic resection and stoma (Hartmann procedure) being the preferred method by most surgeons. This treatment is, however, associated with significant morbidity and mortality. Mortality rates as high as 27% have been reported.5
The high morbidity and mortality of sigmoid resection has led to consideration of alternative approaches to the management of Hinchey grade III patients. The fact that at operation the perforation has often self-sealed suggests that suction of contamination followed by a wash-out of the peritoneal cavity laparoscopically may be sufficient treatment and a valid alternative to sigmoidectomy, thereby avoiding the risks associated with major laparotomy, stoma creation, and subsequent reversal. This nonresectional laparoscopic approach was described in 1996 by O'Sullivan et al.6 Subsequent systematic reviews of several case series indicated that this alternative approach is effective and may reduce mortality to less than 5% and lead to lower morbidity and requirement for stomas.7–9 More recently, 2 randomized controlled trials (RCTs)10,11 published data contradicting the findings of the earlier case series and systematic reviews, whereas a third RCT has provided evidence that lends support to the use of lavage.12,13
The objective of this systematic review and meta-analysis was to consider the evidence for the use of laparoscopic lavage in the management of Hinchey grade III diverticulitis, and clarify its role.
The review employed the methodology suggested by Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)14 guidelines when performing and reporting this systematic review and meta-analyses.
Criteria for Considering Studies for This Review
It included all RCTs that compared laparoscopic lavage with primary resection in Hinchey grade III diverticulitis. Studies in which laparoscopic lavage was used as part of a 2-step management scheme that resulted in primary resection were not included. Primary outcome was the rate of early reintervention, whereas secondary outcomes were Intensive Care Unit (ICU) admission postoperatively, 30-day mortality, 90-day mortality, and the presence of a stoma at 12 months.
A search was performed of MEDLINE from 1946 to present (via OVID), Cochrane Database of Systematic Reviews (via OVID) and Cochrane database of Registered clinical trials (via OVID) and EMBASE (via OVID) using a combination of key search terms. These included “laparoscopy” AND, (“primary resection” OR “Hartmann procedure”, OR “sigmoidectomy”), AND “Diverticulitis”, AND “Peritonitis” AND “therapeutic irrigation” or “lavage” AND “randomized controlled trial” any derivatives of those terms. Only publications in English or with an available translated version were included. The first search was run December 2015, repeated February 2016, and updated May 2016.
In addition, clinicaltrials.gov was searched for proposed or ongoing trials using the search terms “laparoscopic lavage”, “diverticulitis” and “sigmoidectomy”. One further trial was identified in the enrolling stage. A hand-search of reference lists of identified studies and systematic reviews for additional relevant articles was also performed.
Data Collection and Analysis
One review author (JM) examined the titles and abstracts of the studies identified through the outlined search strategy. When studies could not be excluded on the basis of title and abstract full-text papers were obtained. Studies identified were discussed with a second review author (TE). The original study authors were contacted to obtain unpublished data and for clarification when needed to include studies. The decisions made on all studies were documented and presented in the PRISMA flow chart (Fig. 1).
Data Extraction and Management
A single review author (JM) extracted data from each study. This was reviewed in discussion with a second author (TE). Using the template provided by Cochrane,15 a data extraction template was adapted for this review. Data were extracted regarding participant demographics, American Society of Anesthesiologist physical status (ASA) score, disease status, intervention (type), and postoperative outcomes (ICU admission, reintervention, 30 and 90-day mortality, and stoma rates at 12 months) using predesigned forms.
The statistical software Review Manager version 5.316 was used to conduct the meta-analysis. Relative risks (RRs) were presented for the dichotomous variables indicating the relative benefits of laparoscopic lavage compared with primary resection. Heterogeneity was assessed using a statistical test for heterogeneity (P < 0.10 was considered significant) or the I2.17 Because there was moderate heterogeneity between the studies, random-effects models were used for all meta-analysis results. Heterogeneity was further explored using subanalyses that were defined a priori. Due to the small number of studies included in the review, meta-regression was not undertaken. The quality of the included studies was assessed in relation to their risk of bias using the criteria described by the Cochrane Handbook of Systematic Reviews of Interventions.15 The risk of bias was evaluated as “high,” “low,” or “unclear” in relation to the following areas: sequence allocation, allocation concealment, participant blinding, assessor blinding, follow-up completion, intention-to-treat analyses, and selective reporting. Funnel plots were used to consider the potential for publication bias.
Figure 1 shows the PRISMA flow chart of our search strategy. We identified 24 full-text articles of potential relevance, of which 20 observational studies were excluded. Three RCTs met the inclusion criteria.10–13 This included 4 full-text articles, with 1 RCT publishing short and long-term results in separate articles. All included trials had separately published trial protocols18,19 and/or were published at www.clinicaltrials.gov. This search resulted in the inclusion of 370 patients, 191 of whom underwent laparoscopic lavage.
The characteristics and findings of the excluded studies are shown in Table 1.6,20–38 These studies included patients with Hinchey grades III and IV, and found a significant range of reintervention rates varied considerably, ranging from 0% to 34.3% after laparoscopic lavage. Overall, 906 patients were included in these studies, with 33 (3.60%) patients requiring reintervention.
The characteristics and findings of the included studies are shown in Table 2.10–13 Follow-up in all studies was 12 months. Two of the studies randomized only patients with Hinchey grade III diverticulitis after laparoscopic inspection of the abdomen, whereas the Scandinavian Diverticulitis trial (SCANDIV) randomized patients preoperatively. One study allowed only Hartmann procedure in the resection arm, whereas primary resection with anastomosis was used as an alternative either based on surgeon preference in the second trial or according to trial randomization protocols in the third trial.
Table 3 shows the baseline characteristics of the included studies. The sample size of the studies varied from 83 (43 and 40 in each arm) through to 197 (101 and 96 in each arm). The majority of patients who underwent laparoscopic lavage had an ASA score of II compared with those in the resection group who were more evenly distributed between ASA scores of II and III. For the majority of patients in both arms, this was their first presentation of diverticulitis.
The risk of bias analysis is shown in Table 4. Bias was well-managed in all studies and assessed to be of good quality. Assessor blinding was poorly achieved in all studies, with only 1 of the 3 studies reporting any method in which the original case report forms were reviewed for discrepancies. All studies showed power calculations, with only 1 study failing to reach the calculated sample size due to early termination for safety concerns.
The forest plot for rates of reintervention (reoperation and percutaneous drainage) within 30 days postoperatively for each study is shown in Fig. 2. In all, there were 45 reinterventions in the laparoscopic lavage group and 13 in the resection group, giving overall rates of 23.6% and 7.3%, respectively. The overall analysis shows that the RR of reintervention was 3.03 [95% confidence interval (CI) 1.16–7.89] for those who underwent laparoscopic lavage. In the laparoscopic lavage group, 37.8% of all reinterventions occurred without general anesthetic (GA), and the majority of these were percutaneous drainages. There was evidence of statistical heterogeneity between studies (I2 = 60%). This was explored with subgroup analysis, and with exclusion of treatment of acute diverticulitis laparoscopic lavage vs. resection (DILALA),12,13 due to differences in the control procedure and being potentially underpowered, this removed all of the heterogeneity (I2 = 0%). The funnel plot did not suggest evidence of publication bias.
The secondary outcomes for the included studies are shown in Table 5. Postoperative ICU admission was higher for the resection group (21.2% compared with 17.3%), although this did not reach statistical significance (RR 0.85, 95% CI 0.40–1.78). Mortality at 30 days and 90 days was comparable between both groups (RR 1.34, 95% CI 0.37–4.79; and RR 0.86, 95% CI 0.40–1.84, respectively). In the 2 studies which reported stoma, 10% and 20.5% of patients were living with unreversed stomas in the laparoscopic lavage and resection groups, respectively, at 12 months (RR 1.90, 95% CI 0.76–4.79).
The present meta-analysis includes all RCTs published to date comparing laparoscopic lavage and primary resection for management of Hinchey grade III diverticulitis. It has shown an increased risk of postoperative reintervention within the first 30 days for Hinchey grade III diverticulitis managed with laparoscopic lavage. This did not result in an increased mortality rate at both 30 and 90 days when compared with resection. In addition, there was a trend toward reduced stomas at 12 months in those who underwent laparoscopic lavage, although this did not reach statistical significance.
The question of whether to resect or leave the perforated colon in situ in the setting of purulent peritonitis associated with diverticulitis is, in fact, not a new one. A large, multicenter study was conducted in France between 1989 and 1996.39 This compared open surgery, wash-out, and formation of a defunctioning loop stoma (with no bowel resection) versus a Hartmann procedure. The primary end point was the development of localized or generalized postoperative peritonitis. Patients who underwent a Hartmann procedure had a better outcome.
It appears clear from this meta-analysis, as was evident in the original French study, that leaving the diseased colon in situ is associated with persistent intra-abdominal sepsis, as evidenced by the significant increase in reintervention required in the laparoscopic lavage group. However, the clinical significance of these reinterventions deserves closer examination. The reinterventions undertaken in the resection group all required a GA and the associated risks, whereas in the lavage group, 37.8% of reinterventions occurred without a GA, mostly consisting of percutaneous abscess drainage. In fact, when reopreration was considered separately (excluding percutaneous abscess drainage), there was no significant difference between the 2 treatment groups. That the persistent intra-abdominal sepsis associated with laparoscopic lavage could be dealt with simply in many cases is also supported by the fact that the increase in early reintervention did not translate into increased short-term mortality or ICU admissions. These secondary outcomes in the present study should be interpreted with caution as none of the individual trials were powered to detect a significant difference. For example, based on the figures in the present analysis, in relation to the 90-day mortality (baseline mortality 0.084%), a sample size of 4660 patients in each group would be required to have 80% power to show a 20% increase in mortality. Some 244 patients in each group would be needed to show a doubling in mortality at 90 days. Despite this, a previous population-based study of 2400 patients demonstrated32 ICU admissions of 11.6 % and 3.0% in the resection and laparoscopic lavage groups, respectively, also suggesting that persistent sepsis could be managed safely without resection.
In addition to persistent intra-abdominal sepsis, nonresection risks leaving a perforated tumor in situ. In all the included trials, some patients proved to have colonic or rectal malignancies, some of which were left behind at the lavage procedure. Leaving a perforated tumor in situ with a lavage procedure might impair the long-term prognosis for these patients and represents another factor yet to be investigated in this setting. If laparoscopic lavage has been performed, then colonoscopy should still be routinely performed postoperatively to exclude malignancy.
Nonresection also exposes the patient to the risk of recurrent diverticulitis in the affected segment in the longer term. None of the included studies have sufficient length of follow-up to assess the magnitude of this risk and how it might impact on the long-term utility of laparoscopic lavage. A previous systematic review of nonrandomized studies of laparoscopic peritoneal lavage found that 51% overall underwent subsequent elective resection.40 However, the policy of elective resection was institution-dependent. In those series where elective resection was not performed, recurrence appeared low. For example, Myers et al31 reported recurrence in only 2 of 88 patients treated with laparoscopic lavage without subsequent elective resection albeit with short (median 36 months) follow-up. Hence, whereas nonresection exposes the patient to some risk of recurrence, the magnitude of this risk may not justify elective resection, and further long-term data from the randomized trials included here may clarify this.
Despite these risks, one of the major proposed advantages of laparoscopic lavage is avoidance of a stoma. Stomas are not only associated with reduced quality of life, but require a second, major, and potentially morbid operation to reverse.41 The morbidity and mortality associated with Hartmann reversal is not insignificant, with reported rates up to 41% and 4%, respectively.42,43 This morbidity was not taken into account in the present analysis, and subsequent studies will need to consider whether the morbidity of Hartmann reversal offsets any increased rate of early reintervention. Some 35% of Hartmann will not be reversed leading to persistent morbidity and reduction in quality of life.44 Whereas there was no significant difference in stoma rates at 12 months in the 2 groups in this analysis, there was a trend to reduced rates in the laparoscopic lavage group. Data on stoma rates from the SCANDIV trial were not available, so it is likely analysis of this secondary outcome is underpowered to show a significant difference. Pooled data on quality of life could not be included in this meta-analysis, although 2 of the randomized trials did measure this and separately did not show a difference in the 2 treatment arms.
There were a number of limitations to this systematic review. We found there was significant heterogeneity between the studies. This was expected due to the differences in study design, exclusion criteria and surgical techniques employed (Table 2). Lavage technique varied between studies, with 1 trial setting a maximum volume of fluid to be used compared with a minimum amount and requiring return of clear fluid; this may have resulted in inadequate lavage for some patients resulting in poorer outcomes. Studies included in this analysis were of high quality for all bias criteria apart from assessor blinding, with only 1 study reporting any form of independent review. It was noted that in all the included studies, the clinician responsible for the patient's care and ultimately the decision for reintervention was aware of the randomization outcome and this may have introduced some degree of bias.
The generalizability of the results presented in this analysis also deserves further consideration. All included studies had the weakness that a large number of patients with perforated diverticulitis could not be included, introducing potential selection bias. All studies reported the Hinchey score for patients that were not considered or were not suitable for inclusion and the proportion of Hinchey IV patients was higher in this group, indicating that the most severely ill patients were not included. The reported ASA score for patients outside the SCANDIV trial was higher than that for included patients. There is reason to believe that this was the case in the other two trials and also the included patients in these trials had lower ASA scores than in SCANDIV. It is also important to note that at the other end of the spectrum of severity of perforated diverticulitis there may be a subset of patients with less severe Hinchey grade III that could be managed with intravenous antibiotics, watchful waiting, and potentially percutaneous drainage.45 There is the possibility that a number of patients included in these trials and previous case series may have settled without operative intervention. If laparoscopic lavage is to be considered a viable alternative to resection, then the subset of patients to which it is applicable needs further definition.
Future prospective cohort studies with short and long-term follow-up are required to improve the understanding of laparoscopic lavage in Hinchey grade III diverticulitis. A core outcome set (COS) needs to be discussed and agreed upon to ensure future data are comparable and allows for a more complete analysis. The minimum COS should include postoperative ICU admissions, reinterventions (both with and without GA), patients with stoma and percentage reversed with associated outcome data, and mortality within 30 days and 12 months. Further study also needs to be done to identify patient factors that may impact on the success of laparoscopic lavage. Possible candidates include preoperative factors such as ASA scores, age, BMI, and CT severity, and intraoperative factors such as the Manheim Peritonitis Index. It would also be of benefit to consider data of ineligible patients excluded from the trials to allow for better comparison. We are currently awaiting the results of Laparoscopic Lavage for Acute Non-Faeculant Diverticulitis (LapLAND)46 and the observational study, which has not yet started recruiting, laparoscopic-lavage observational study (LLOS).47
This review did not confirm the results found in previous case series and systematic reviews. Laparoscopic lavage was found to have a significantly higher rate of reintervention postoperatively, although mortality at both 30 and 90 days was found to be similar to those who underwent resection. The role of laparoscopic lavage in Hinchey grade III diverticulitis is yet to be clarified. Further studies are required to identify patients who will benefit from lavage as first-line management.
1. Weizman AVNG. Diverticular disease: epidemiology and management. Can J Gastroenterol
2. Shahedi K, Fuller G, Bolus R, et al. Long-term risk of acute diverticulitis among patients with incidental diverticulosis found during colonoscopy. Clin Gastroenterol Hepatol
3. Janes S, Meagher A, Frizelle F. Elective surgery after acute diverticulitis. Br J Surg
4. Hinchey E, Schaal P, Richards G. Treatment of perforated diverticular disease of the colon. Adv Surg
5. Toorenvliet BRSH, Schoones JW, Hamming JF, et al. Laparoscopic peritoneal lavage for perforated colonic diverticulitis: a systematic review. Colorectal Dis
6. O'Sullivan GCMD, O’Brien MG, Ireland A. Laparoscopic management of generalized peritonitis due to perforated colonic diverticula. Am J Surg
7. Alamili M, Gögenur I, Rosenberg J. Acute complicated diverticulitis managed by laparoscopic lavage. Dis Colon Rectum
8. Cirocchi R, Trastulli S, Desiderio J, et al. Treatment of Hinchey stage III–IV diverticulitis: a systematic review and meta-analysis. Int J Colorectal Dis
9. Toorenvliet B, Swank H, Schoones J, et al. Laparoscopic peritoneal lavage for perforated colonic diverticulitis: a systematic review. Colorectal Dis
10. Schultz JYS, Yaqub S, Wallon C, et al. Laparoscopic lavage vs primary resection for acute perforated diverticulitis: The scandiv randomized clinical trial. JAMA
11. Vennix S, Musters GD, Mulder IM, et al. Laparoscopic peritoneal lavage or sigmoidectomy for perforated diverticulitis with purulent peritonitis: a multicentre, parallel-group, randomised, open-label trial. Lancet
12. Angenete E, Thornell A, Burcharth J, et al. Laparoscopic lavage is feasible and safe for the treatment of perforated diverticulitis with purulent peritonitis: the first results from the randomized controlled trial DILALA. Ann Surg
13. Thornell A, Angenete E, Bisgaard T, et al. Laparoscopic lavage for perforated diverticulitis with purulent peritonitis: a randomized, controlled trial. Ann Internal Med
14. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med
15. Higgins JP, Altman DG, Gøtzsche PC, et al. The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ
16. Review Manager (RevMan) [computer program]. Version 5.3. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration; 2014.
17. Higgins J, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med
18. Swank HA, Vermeulen J, Lange JF, et al. The ladies trial: laparoscopic peritoneal lavage or resection for purulent peritonitis A and Hartmann's procedure or resection with primary anastomosis for purulent or faecal peritonitis B in perforated diverticulitis (NTR2037). BMC Surg
19. Thornell AAE, Gonzales E, Heath J, et al. Treatment of acute diverticulitis laparoscopic lavage vs. resection (DILALA): study protocol for a randomised controlled trial. Trials
20. Bretagnol F, Pautrat K, Mor C, et al. Emergency laparoscopic management of perforated sigmoid diverticulitis: a promising alternative to more radical procedures. J Am Coll Surg
21. Edeiken SM, Maxwell RA, Dart IVBW, et al. Preliminary experience with laparoscopic peritoneal lavage for complicated diverticulitis: a new algorithm for treatment. Am Surg
22. Faranda C, Barrat C, Catheline J-M, et al. Two-stage laparoscopic management of generalized peritonitis due to perforated sigmoid diverticula: eighteen cases. Surg Laparosc Endosc Percutan Tech
23. Favuzza J, Friel JC, Kelly JJ, et al. Benefits of laparoscopic peritoneal lavage for complicated sigmoid diverticulitis. Int J Colorectal Dis
24. Franklin ME Jr, Portillo G, Treviño JM, et al. Long-term experience with the laparoscopic approach to perforated diverticulitis plus generalized peritonitis. World J Surg
25. Galleano R, Di Giorgi S, Franceschi A, et al. Two-stage laparoscopic management of complicated acute diverticulitis. Initial experience. Annali italiani di chirurgia
26. Karoui M, Champault A, Pautrat K, et al. Laparoscopic peritoneal lavage or primary anastomosis with defunctioning stoma for Hinchey 3 complicated diverticulitis: results of a comparative study. Dis Colon Rectum
27. Lam H, Tinton N, Cambier E, et al. Laparoscopic treatment in acute complicated diverticulitis: a review of 11 cases. Acta Chir Belg
28. Liang S, Russek K, Franklin ME Jr. Damage control strategy for the management of perforated diverticulitis with generalized peritonitis: laparoscopic lavage and drainage vs. laparoscopic Hartmann's procedure. Surg Endosc
29. Lippi C, Beatini L, Cervia S, et al. Laparoscopic lavage and drainage in the surgical treatment of diverticular disease complicated by peritonitis. Chirurgia italiana
30. Mazza D, Chio F, Khoury-Helou A. Conservative laparoscopic treatment of diverticular peritonitis. Journal de chirurgie
31. Myers EHM, O'sullivan GC, Kavanagh D, et al. Laparoscopic peritoneal lavage for generalized peritonitis due to perforated diverticulitis. Br J Surg
32. Rogers AC, Collins D, O'Sullivan GC, et al. Laparoscopic lavage for perforated diverticulitis: a population analysis. Dis Colon Rectum
33. Taylor CJLL, Ghusn MA, White SI. Perforated diverticulitis managed by laparoscopic lavage. ANZ J Surg
2006; 11:962–965. 76.
34. White SI, Frenkiel B, Martin PJ. A ten-year audit of perforated sigmoid diverticulitis: highlighting the outcomes of laparoscopic lavage. Dis Colon Rectum
35. Zonča P, Ihnát P, Peteja M, et al. [Laparoscopic lavage and drainage in the management of acute diverticulitis: is it time to move on?]. Rozhledy v chirurgii
36. Lee EC, Murray JJ, Coller JA, et al. Intraoperative colonic lavage in nonelective surgery for diverticular disease. Dis Colon Rectum
37. Zonča P, et al. “[Laparoscopic lavage and drainage in the management of acute diverticulitis: is it time to move on?].”. Rozhledy v chirurgii: mesicnik Ceskoslovenske chirurgicke spolecnosti
38. Da Rold AR, Guerriero S, Fiamingo P, et al. Laparoscopic colorrhaphy, irrigation and drainage in the treatment of complicated acute diverticulitis: initial experience. Chir Ital
39. Zeitoun G, Laurent A, Rouffet F, et al. Multicentre, randomized clinical trial of primary versus secondary sigmoid resection in generalized peritonitis complicating sigmoid diverticulitis. Br J Surg
40. Afshar S, Kurer M. Laparoscopic peritoneal lavage for perforated sigmoid diverticulitis. Colorectal Dis
41. Wood C, Lawton S, Downing A, et al. Quality of life of colorectal cancer survivors in England: report on a national survey of colorectal cancer survivors using Patient Reported Outcome Measures (PROMs). 2015. Avaliable at http://eprints.nottingham.ac.uk/32442/
42. Banerjee S, Leather A, Rennie J, et al. Feasibility and morbidity of reversal of Hartmann's. Colorectal Dis
43. Pearce N, Scott S, Karran S. Timing and method of reversal of Hartmann's procedure. Br J Surg
44. Maggard MA, Zingmond D, O’Connell JB, et al. What proportion of patients with an ostomy (for diverticulitis) get reversed? Am Surg
45. Eglinton TW, Frizelle FA. Trials of surgical treatment of acute perforated diverticulitis: finding what they look for. Ann Internal Med
47. Laparoscopic-lavage Observational Study (LLOS) [database online]. Insert City of Publication Here see notes; 2016.