Appendectomy versus antibiotic treatment for acute appendicitis
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
Is antibiotic treatment as effective as surgical appendectomy (laparoscopic or open) in patients with acute appendicitis on recovery within two weeks, without major complications (including recurrence) within one year.
Background
Acute appendicitis is the most common cause of acute abdominal pain.The life time risk of appendicitis is approximately 7‐8%, with the highest incidence in the second decade (Addis 1990).
Although the etiology of acute appendicitis is poorly understood, it is probably in the majority of cases caused by an obstruction of the lumen (Addis 1990). The luminal obstruction can be caused by fecaliths, lymphoid hyperplasia, foreign bodies, parasites and both primary (carcinoid, adenocarcinoma, Kaposi sarcoma and lymphoma) and metastatic (breast and colon) tumours. Once appendiceal obstruction occurs, the continued secretion of mucus results in elevated intraluminal pressure and luminal distention. This eventually exceeds capillary perfusion pressure, which leads to venous engorgement, arterial compression and tissue ischemia. As the epithelial mucosal barrier becomes compromised, luminal bacteria multiply and invade the appendiceal wall, which causes transluminal inflammation (Birnbaum 2000). The most common bacteria that can cause acute appendicitis are intestinal bacteria as the Escherichia Coli and bacteria belonging to the Bacteroides Fragilis group (Bennion 1990, Rautio 2000). Continued ischemia results in appendiceal infarction and perforation. However some studies report that non‐complicated and complicated (perforated) appendicitis are different entities, and that many cases of acute appendicitis will resolve spontaneously (Mason 2008).
Fitz described in 1886 the signs and symptoms of acute and perforated appendicitis, outlined the progression from acute lower quadrant inflammation through peritonitis and iliac fossa abscess formation, and recommended early appendectomy if there were signs of spreading peritonitis or of clinical deterioration. Since McBurney in 1894 defined the surgical appendectomy, the surgical removal of the diseased appendix had been the treatment of choice for acute appendicitis, because it lowered the mortality rate drastically (Birnbaum 2000, Fischer JE 2007). Surgical appendectomy is a successful treatment modality with good results, but complications are inherent to any operation. The most common complications are wound infection, intraabdominal abscess and ileus caused by intraabdominal adhesions, which vary in frequency between open and laparoscopic appendectomy (Sauerland 2004, Nakhamiyayev 2009). The overall complication rate for open and laparoscopic appendectomy are respectively 11.1% and 8.7%, with a mortality rate less than 0.5% (Guller 2004). The clinical diagnosis of acute appendicitis remains difficult. Today’s diagnostic tools like ultrasonography (US) and computed tomography (CT) reduce the number of negative findings at surgery (Randen v 2008).
Despite surgical appendectomy is the standard, several investigators have studied the conservative antibiotic treatment of acute appendicitis with good results. They described a low morbidity, low mortality rate and a recurrence rate between 7‐15% (Coldrey 1959, Hansson 2009).
Even though non‐operative management with antibiotics of other intraabdominal inflammations (uncomplicated diverticulitis, salpingitis, and neonatal enterocolitis) are now established, the non‐operative management of appendicitis remains largely unexplored (Mason 2008). Because the high prevalence of acute appendicitis, change of treatment modality will have major implications. A review of all available data is necessary.
Objectives
Is antibiotic treatment as effective as surgical appendectomy (laparoscopic or open) in patients with acute appendicitis on recovery within two weeks, without major complications (including recurrence) within one year.
Methods
Criteria for considering studies for this review
Types of studies
We will include randomised or quasi‐randomised clinical trials (RCT or qRCT) in which antibiotic treatment is compared with appendectomy in patients with suspected appendicitis. Superiority and non‐inferiority RCT's will be included. Quasi‐randomised trials are those with a non‐random non‐concealed allocation (e.g. simple alternation, date of birth, hospital admission number). We will not apply language restrictions. We will exclude studies which primarily focus on complications of acute appendicitis like abscesses and/or perforations.
Types of participants
All patients with suspected acute appendicitis will be included. There will be no restrictions for age or gender. If possible we will perform a subgroup analysis for age and gender.
Types of interventions
Antibiotic treatment (intravenous or oral) compared with surgical appendectomy (open or laparoscopic).
Types of outcome measures
Primary outcomes
-
Cured within two weeks (free of symptoms such as abdominal pain, fever, infection parameters), without major complications (including recurrence) within one year.
Secondary outcomes
-
Major complications defining the need of further invasive treatment or prolonged admission (e.g. abscesses, ileus, deep wound infection, recurrence, (re)operation, secondary perforation)
-
Minor complications (e.g. negative appendectomy, diarrhoea, superficial wound infection)
-
Duration of hospital stay (in days)
-
Period of sick leave (in days)
-
Cost effectiveness
Search methods for identification of studies
The following electronic databases will be searched:
-
Cochrane Libary (including CENTRAL)
-
MEDLINE, from 1966 to October 2009
-
EMBASE, from 1980 to October 2009
The search terms that will be used are: appendicitis [MeSH]) or appendic* or appendicitis acuta, antibiotic* therap* or antibiotic* treatment*, appendectom* [Mesh] or appendicectom*.
The exact search protocol, designed by the CCCG Trial Search Coordinator Susse Wegeberg, is described in appendices 1‐3. For the MEDLINE and EMBASE search, filters are added to find RCTs and CCTs.
The reference lists of potentially relevant articles will be screened for further potentially relevant citations.
Data collection and analysis
Selection of studies
Two authors (IW and DdH) will examine all citations and abstracts derived from the electronic search strategy and independently select trials that will be included in the review. Disagreement will be resolved by discussion. Full copies of all (potentially) relevant studies will be obtained. A standardized form will be used to assess eligibility of trials for inclusion.
Data extraction and management
Data from the studies will independently be extracted by two authors using standardised data forms, according to the criteria as described in the Cochrane Handbook (Higgins 2008). Disagreement will be resolved by discussion.
Measures of treatment effect
A non‐inferiority analysis will be made, we consider a minimum detectable difference of 20% clinical relevant. Percentage cured in the antibiotic treatment group will be calculated in each trial and after pooling the summary data with confidence interval (CI) 97,5% (one‐sided) and compared with the 20% difference. In case of non‐inferiority the minimal number to treat will be calculated with a confidence interval (CI) of 95% (two‐sided) and a power of 80% using the online binary outcome equivalence trial power (sample size) calculator as programmed on SealedEnvelope.com (SealedEnvelope).
As described in Chapter 9 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008), we will calculate the mean difference and the standardized mean difference (SMD) for continuous data.
Dealing with missing data
We will try to obtain the missing data by contracting the trial author. When this is not possible we attempt to make assumptions of the missing data and perform a sensitivity analysis to assess how sensitive the results are to reasonable changes in the assumptions that are made and address the potential impact of missing data on the findings of the review in the discussion (Higgins 2008).
Assessment of heterogeneity
Heterogeneity will be assessed using chi‐square test with P‐value of <0.05 considered as statistically significant. I2 will be assessed to quantify the degree of heterogeneity present. A fixed effects model will be used when there is no or little heterogeneity (I2 < 30%), and random effects model will be used when there is statistically significant moderate heterogeneity (I2 up to 50%), and no pooling when substantial heterogeneity is present (I2 >75%). Sources of heterogeneity will be identified by subgroup analysis. And we will describe the probable cause for heterogeneity.
Assessment of reporting biases
We will look for evidence of collection by study investigators of a small number of key outcomes that are routinely measured in the area in question, and report which studies report data on these and which do not. When there is a suspicion or direct evidence for selective outcome reporting, we will ask the trial author for additional information (Higgins 2008).
Data synthesis
Any data which had been reported in the included studies and are relevant to the primary and secondary outcomes of this review will be analysed by two authors (IW and DdH) using RevMan 5.0 and reported according to Cochrane Collaboration criteria (RevMan 2008).
Subgroup analysis and investigation of heterogeneity
If possible we will apply a subgroup analysis for age and gender. We will investigate the way in which the diagnosis had been made, because this can be a source for heterogeneity.
Sensitivity analysis
A sensitivity analysis will be performed using RevMan 5.0, to determine whether poor quality studies, studies with doubtful eligibility or which look like an outlier have to be included or excluded.
