Introduction
1. Clinical presentation
A 74- year old woman was at the ER due to a painful stomach. She only has a light form of asthma in her medical history. For the last 2 months, she often had pain in the left lower side of the abdomen and often felt feverish. Since this week the pain had become progressively worse. This morning she could not handle the pain and came to the ER, she cannot exactly locate the pain. Physical examination showed a painful woman with a temperature of 38.7 degrees. Her pulse is 105/min regular and blood pressure is normal. Blood tests showed a CRP of 111 and leukocytes of 34.
A sonography fast showed fluid in the peritoneal cavity. A CT scan was made afterwards, there were a few diverticles with a thickened wall, perforation of a diverticle in the sigmoid colon could be seen. The CT scan also showed a bit of free air and fluid in the peritoneal cavity. Diverticulitis Hinchey 3 was suspected.
The patient was treated by going under operation the same day. She got a sigmoid resection with a colostoma, Hartmann procedure.
Also, the golden standard in the treatment of acute diverticulitis Hinchey classification 3 is either Hartmann procedure or sigmoid resection with primary anastomosis.
I was wondering if there were other treatments for acute diverticulitis Hinchey classification 3 that were less invasive. When looking at the literature, it said that laparoscopic lavage was also a treatment option for Hinchey classification 3.
So, what I wanted to know was if laparoscopic lavage was a better treatment option for patients in this group than resection of the sigmoid colon. The morbidity and mortality for the current standard treatments is very high, 20-40% and 10-20%. (1) I was wondering if these 2 factors are lower when treated with laparoscopic lavage. Also, the difference and severity of complications is something I want to know.
2. Clinical question
PICO question: Is laparoscopic lavage a better treatment option for acute diverticulitis with peritoneal peritonitis than sigmoid resection with anastomosis or Hartmann procedure regarding mortality, morbidity and complications?
P: Patients with acute diverticulitis, Hinchey 3
I: Laparoscopic lavage
C: Sigmoid resection with primary anastomosis or Hartmann procedure
O: Mortality, complications, morbidity
3. Guidelines
According to the protocol of VUmc, acute diverticulitis with purulent peritonitis is treated with either sigmoid resection with primary anastomosis or with the Hartmann procedure. Laparoscopic lavage as treatment is not mentioned in the protocol. (2)
The national guidelines say that for Hinchey 3 and 4 the golden standard is: sigmoid resection with primary anastomosis and Hartmann procedure. It is said that sigmoid resection has a high morbidity and mortality. Another option that was given for the treatment in only the Hinchey 3 group is laparoscopic lavage with intravenous antibiotics. The guideline recommends that when treated with this option, patients should be included in a study since the current grade of evidence is not very high. There is low evidence of the effect of treatment in Hinchey grade 4. (3)
4. Methods
Search strategy
The aim was to identify systematic reviews that considered the difference between laparoscopic lavage and sigmoid resection with stoma or primary anastomosis in the treatment of acute diverticulitis Hinchey classification 3. To identify relevant publications, I searched in the database of Pubmed and Conchrane library. Only studies written in English and published in the past 10 years were included.
Furthermore, the Mesh terms that was used were: ‘laparoscopy’ AND (‘primary resection’ OR ‘Hartmann procedure’, OR ‘sigmoidectomy’), AND ‘Diverticulitis’, AND ‘Peritonitis’ AND ‘therapeutic irrigation’ or ‘lavage’.
Selection criteria
Studies that were included, if they were written in English, a systematic review design, published in the past 10 years and assessed the difference between laparoscopic lavage and resection of the sigmoid.
Studies that were excluded, if anything other than systematic reviews, non-English publications and publication of the article was more than 10 years ago.
After applying the selection criteria, 9 articles are included. After reading the title and abstracts of the remaining reviews only 1 article was relevant to answer my question.
The article which I chose was: ‘Laparoscopic lavage versus resection in perforated diverticulitis with purulent peritonitis: a meta-analysis of randomized controlled trials.’ (4) This article was published in 2016. Other reviews found were much older or did not answer the review question.
‘
5. Critical appraisal
In the introduction, the relevance of the article is broadly described. The reason why they did the meta-analysis and the difference in mortality and morbidity between both treatments is explained. The question that this meta-analysis tries to answer is the difference in the safety and feasibility of laparoscopic lavage compared to sigmoid resection. This was not done according to the PICO question.
Search strategy
The Search strategy was handled well. Different databases were researched. Until March 2016 Pubmed, EMBASE, Medline, the Cochrane central register of controlled trials and Cochrane database of systematic reviews were researched. Furthermore, the references of selected articles were also reviewed. Also, the terms that were used to search for the articles is written in the method section.
Of the 1439 items found through database searching only 3 studies were included in this meta-analysis. (5,6,7)
Selection procedure
The selection procedure was handled well, the method section contained clear inclusion and exclusion criteria in the search of studies. Furthermore, two different investigators independently performed a systematic research.
Moreover, the inclusion criteria for this meta-analysis were prospective clinical trials which included patients suspected to have perforated purulent diverticulitis who were randomized to either have laparoscopic lavage or bowel resection. Excluded were case reports, letters, reviews, meta-analysis, retrospective studies and non-English studies. Sadly, not all languages were reviewed. Non-English articles were excluded from the research. So, some data could be missing if those articles were excluded. However, if the researcher does not understand the study it can be interpreted wrong and lead to distortion of the results
Heterogeneity
Only a few characteristics of the original studies are stated in the study. Also, when looking at characters there are a lot of differences between the 3 RCT’s that are chosen. The primary outcome, inclusion and exclusion criteria were all different between the studies. The clinical heterogeneity was handled using the random effects model. The statistical heterogeneity was handled well. The statistical test they used was described, they were valid tests and in the results section, you can see they were correctly used (I^2 inconsistency test and forest plots).
Quality test
The quality test was handled well. The studies were selected by two independent reviewers. The 3 studies chosen were both evaluated on the quality level of the study. This was done using the guidelines of Cochrane collaboration. They reviewers looked at 6 items: randomization, proper allocation concealment, homogeneousity of groups at baseline, description of drop outs and loss at follow up, documentation of eligible criteria and whether intention to treat analysis was conducted. If more than 4 items were positive than the RCT was of high quality and 3 or less of low quality. All 3 included studies were of high quality. Both reviewers agreed that the RCTs were of high quality.
Important characteristics of included trials
The meta-analysis contains a table with the most important basic characteristics of each RCT. The table contains information like the number of participants in each study, what treatment was done in each study and the inclusion and exclusion criteria. The limitation of each study was also mentioned. So, I think they included all the important characteristics of the included trials.
Transparent presentation of the results
The individual results of the studies are expressed in a standard way. Outcomes were expressed as odds ratio with a 95% confidence interval. They were calculated with the fixed effects and random-effects model. The results were displayed in a forest plot. The results are presented transparently. However, it should be noted that in the discussion section the author wants the reader to take into consideration the small number of patients that were included in the RCTs and that one RCT was largely underpowered (5).
6. Results
In this meta-analysis 3 RCTs were evaluated containing 315 patients who were enrolled.
The primary outcome was the difference in mortality, morbidity and complications between laparoscopic lavage and sigmoid resection. For mortality at index they included only 2 RCT’s which contained 232 patients. (5,6) Of the 120 patients that got the laparoscopic lavage 4 died and 4 of the 112 in the resection group. No significant difference between laparoscopic lavage and the resection group was found, p=0.92. In the same studies the number of re-operations was also evaluated. In the laparoscopic lavage group 17,5 % needed a re-operation because of failure of the treatment. Laparoscopic lavage had a higher rate of re-operations compared to bowel resection, n=21 vs 6, a statistical significance was reached p= 0.006.
Furthermore, two RCT’s (5, 7) in this meta-analysis also looked at the difference in mortality 12 months after the treatment (n=191), statistical difference still was not found p= 0.51. In presence with a stoma there also was no statistical difference p=0.27. Reoperations occurred more often in the laparoscopic group than in resection, p=0.004.
Also, the morbidity between the 3 studies were reported until 90 days after the treatment. In total 315 patients were included. An increased morbidity with statistical significance (P=0.05) could be seen in the laparoscopic lavage group.
Lastly in the meta-analysis they also looked at the different kind of complications published in the 3 RCT’s such as intra-abdominal abscess, incidence of wound infection, pneumonia, urinary tract infection and heart and lung complications. A higher incidence of intra-abdominal abscess is found in the laparoscopic lavage group with a statistical significance of p=0.0003. Laparoscopic lavage is also associated with a lower incidence of wound infection, p= 0.0009. There was no statistical difference between both treatments if looked at the other complications.
7. Overall
The conclusion of this meta-analysis is that more studies are needed to evaluate the safety and feasibility of laparoscopic lavage in acute diverticulitis with purulent peritonitis. In this review, sigmoid resection and laparoscopic lavage showed equal results in term of mortality at index and 12 months after treatment. However, laparoscopic lavage is associated with a higher number of intra-abdominal abscess and re-operations (at index or until 12 months after treatment). The laparoscopic lavage group also had a higher morbidity postoperatively compared to sigmoid resection.
The patient I described got treated with the Hartmann procedure. Since there is no difference in mortality it does not matter which treatment the patient got. In my opinion, she got the right treatment since the incidence of morbidity and complications this meta-analysis showed was very high.
The national guidelines say that the golden standard is Hartmann procedure or sigmoid resection with primary anastomosis. In the guideline, it is stated that laparoscopic lavage is an alternative treatment for acute diverticulitis with purulent peritonitis. It says that in a small group of patients with acute diverticulitis Hinchey 3 they found out that laparoscopic lavage was an effective and safe treatment option. (3) The evidence does not really support the guidelines. The mortality is the same as sigmoid resection in laparoscopic lavage, but it has a higher number of morbidity.
The doctor and hospital have more experience with the Hartmann procedure or sigmoid resection with primary anastomosis. So that is why he chose for the Hartmann procedure. The patient said she left the choice with the doctor. She only wanted to go home fast. Consequently, this preference does not influence the choice of the treatment.
8. Bottom line
The outcome of the meta-analysis says that there is no difference in laparascopic lavage and sigmoid resection for the treatment of acute diverticulitis with purulent peritonitis in terms of mortality. However, a higher rate of morbidity, higher incidence of intra-abdominal abscess as complication could be seen in laparoscopic lavage. The guidelines say that the golden standard is the Hartmann procedure or sigmoid resection with primary anastomosis and laparoscopic lavage is an alternative. It is also recommended that if the patient wants to have laparoscopic lavage he/she should participate in a study. The doctor treated this patient with the Hartmann procedure. That is because he has more experience in doing this kind of operation.
The patient did not really have a preference. No definitive high-grade conclusion can be made. We need more research, including a higher number of participants and the follow- up should be longer.