The use of Antibiotics in treatment of Uncomplicated/Mild Diverticulitis
Abstract:
Until recently the prescribing of antibiotics has been the cornerstone of treatment of uncomplicated/mild diverticulitis. Recommendations for their use has been advised by expert opinion, guidelines and in textbooks. However recent evidence has suggested that diverticulitis may be a more inflammatory process rather than infectious and that routine use of antibiotics the treatment of diverticulitis should be re-evaluated. This has led to contrasting guidelines due to the emergence of new evidence. This aim of this review is to assess the literature currently available and attempt to establish whether antibiotics contribute to recovery from uncomplicated diverticulitis.
Methods:
PUBMED database were searched using the keywords: Diverticulitis, Acute Diverticulitis, Diverticulitis AND antibiotics, Diverticulitis AND treatment, Diverticulitis AND management, Acute diverticulitis AND antibiotics, Acute diverticulitis AND treatment, Acute Diverticulitis AND management. Studies from 2011- 2017 were included in the search.
A total of 5 relevant recent studies were identified for review, 2 Randomised Control Trials, one without blinding, a cohort study, a prospective observational study and a retrospective case controlled study. Older studies were found however their poor quality of study design rendered them not useful for the purposes of this review.
Introduction:
Diverticular disease which encompasses diverticulosis, the presence of diverticula, an out-pouching or protrusion of mucosa through the colonic wall(1) and diverticulitis, an inflammatory process of the diverticula which is associated with fever, leukocytosis and pain is a common disease among the population of the developed world. These outpouchings occur in the mainly in the descending and sigmoid colon however they are not exclusively found in these areas.
The aetiology of diverticulosis it not well understood and is thought to be a multi-factorial process, changes in colonic pressure, motility and the structure of the colonic walls as well as a link with low-fibre dietary intake have all been suggested as potential causes.(2)
Localised inflammation of diverticula is known as acute diverticulitis. While its aetiology is unclear it has been posited that stasis, obstruction, ischemia or local changes in bacterial microflora may play a role.(3) More recent evidence suggests inflammation may play a role in early pathogenesis, this is based on the presence of inflammation in colonic diverticula without evidence of clinical diverticulitis.(4)
Diverticular disease is a common gastrointestinal disease. In those over the age of 65 the prevalence of diverticular disease is greater than 65%.(5) Increasing age shows a rise in the prevalence of the disease, with 5% of the population between 50 and 60 affected, while 50% of those between 80 and 90 affected and up to 60% of those greater than 90 years of age.(6) Diverticular disease poses a significant healthcare burden and in the last decade there have been increases in rates of clinical visits and hospital admissions for diverticular disease and its complications despite advances in surgery and intensive care.(7) Diverticular disease is one of the leading causes of health spending among gastrointestinal diseases.(7)
There are several classification or grading methods used for acute diverticulitis. The European Association for Endoscopic Surgery grade acute diverticulitis as ‘complicated’ or ‘uncomplicated,’ (8) while the Ambrosetti computed tomography (CT) criteria uses ‘mild’ or ‘severe,’ (9) or alternatively the use of the modified Hinchey classification(10). It has been reported that 0-10% of patients admitted present with complicated disease which would require surgery or percutaneous drainage and so conservative treatment is the management of choice in the majority of patients.(11) Complicated diverticulitis includes complications such as perforations, fissure, obstruction and bleeding.
Bowel rest, intravenous fluids and antibiotics have long been the accepted treatment for uncomplicated diverticulitis.(11) While there is much research into the effect of treatment in complicated disease likely due to its more serious nature there is a scarcity of research on the topic of treatment in uncomplicated disease. As previously outlined up to 90% of patients presenting present with uncomplicated disease.(11) The use of antibiotics for acute uncomplicated diverticulitis has been rationalized by the idea that diverticulitis has an infectious aetiology and therefore antibiotic use will lead to a reduction in complications.(12) The lack of understanding of the aetiology of acute uncomplicated diverticulitis and the emergence of evidence suggestive of an inflammatory aspect to the aetiology (4) has led to questioning of the routine use of antibiotics for its treatment and recent evidence has demonstrated disease resolution without the use of antibiotics.(3, 13) The current use of antibiotics for treatment of acute uncomplicated diverticulitis(AUD) is based mainly on expert opinion which is Class 1C low quality evidence and medical dogma. The obvious lack of evidence and guidance for clinicians has led to an increase in investigation in this field in an effort to produce more high quality evidence.
Results:
A retrospective case-control study was performed at two different hospitals in the Netherlands. The period evaluated was between Jan 2001 to December 2007. In 2011, de-Korte, et al. performed this study to assess outcomes in patients with imaging confirmed (CT/Ultrasound) mild acute uncomplicated diverticulitis when treated with and without antibiotics. 272 patients were identified, 191 were observed and antibiotic treatment was not prescribed while 81 were treated with antibiotics. Treatment failure was defined as the need for urgent or emergent surgery and/or the need for percutaneous drainage of abscesses because of clinical deterioration. The groups were comparable at baseline with respect to age, sex, comorbidities, use of non-steroidal anti-inflammatory drugs(NSAIDS), steroids and aspirin. C-reactive protein and white cell count did not differ between the groups on admission however there were more patients in the antibiotic group with a temperature above 38.5 on admission (8% vs 19%, p=0.014). Risk of recurrence was increased in the antibiotic group however this finding did not reach statistical significance (p=0.880). The only factor which was found statistically to increase the risk of recurrence was NSAIDS use (p=0.037). There was no statistically significant difference in treatment failure between the observed group and the antibiotic group.(14) This study while it does begin to question conventional treatment of AUD it does have some design flaws. Patients in the study were not randomized therefore there may an element of selection bias as well as the higher temperature found in patients in the treatment group indicating possible selection bias for this group.
Chabok, et al. (2012) aimed to investigate the need for antibiotic treatment for patients with CT confirmed uncomplicated diverticulitis. The trial included a 12-month follow-up with patients. This was a multicenter randomized control trial without blinding performed across 10 surgical departments in Sweden and 1 in Iceland. Uncomplicated diverticulitis was defined as left lower quadrant pain confirmed with CT. If a patient had an abscess, fistula, sepsis or free air under the diaphragm it was classed as complicated and they were excluded from the trial. 623 patients were recruited and were randomly assigned to the treatment group in which they received IV antibiotics followed by a 7-day course of oral antibiotics or control group who received IV fluids only. No significant differences were found in sex, age, comorbidities, body temp, white cell count or c-reactive protein on admission. Pain was recorded on a visual analogue scale (VAS) and abdominal tenderness graded 0-4 on palpation. No significant differences were found in VAS score (p=0.253), temperatures normalized similarly between groups after 2 days (p=0.343), there was a significant difference in abdominal tenderness on day 2 (p=0.041) with a mean difference of 0.8 from baseline in the control group and 1 in the treatment group. Median hospital stay was 3 days in both groups. There were no significant differences seen in rates of complication in the treatment group vs. the control group. 9 patients in total developed complications, 3 from the treatment group and 6 from the control group. On follow-up at 12 months 93 patients reported recurrent diverticulitis however there was no difference in the rates of recurrence between the groups. Symptoms of abdominal pain and changes in bowel habit at 12 months did not differ between groups. The group concluded that antibiotics do not affect recovery nor do they prevent recurrence of diverticulitis and so should be reserved for the treatment of complicated dicerticulitis.(15) While many patients within this study would have been eligible for outpatient treatment they were treated as inpatients and received IV fluids or antibiotics. This causes the study to have less weight and makes it difficult to generalize its findings.
Isacson et. al (2014) performed a retrospective population-based cohort study at a hospital in Sweden. The aim was to evaluate the need for antibiotic treatment in patients with AUD and the consequences of applying a no antibiotic policy. Medical records for patients admitted with AUD in the year 2011 were carefully reviewed. 195 patients with AUD were identified. 182 patients were treated on an inpatient basis while 13 on an outpatient basis. Of the 182 treated within the hospital 165 (91%) did not receive antibiotics while 17 (9%) did. Of the 165 treated without antibiotics 6 patients (4%) required readmission. 1(6%) of the 17 patients treated with antibiotics required readmission. All 13 patients treated on an outpatient basis without antibiotics had no complications.(16)
In light of previous findings Isacson et. al (2015) performed a prospective observational study at 2 Swedish hospitals from March 2012 to December 2013 and May 2012 to August 12. This study was the first of its kind designed to assess outpatient treatment for AUD without the use of antibiotics. Patients with CT confirmed AUD were selected for the study. Patients were asked to keep a personal journal recording pain score, temperature, oral intake of food and drink, bowel habits and use of analgesics. A nurse contacted the patient daily and follow-up with a surgeon was carried out at 1 week and 3 months. Treatment failure was defined as readmission within 1 month. 155 patients were selected for the study, 101 patients were female and 54 were male. The mean age of the patients was 57.4 years. At the time of diagnosis c-reactive protein and white cell count were elevated but normalized in 84% of the patients. 30% of the patients required the use of analgesics. The mean pain score at 3 days was reported to be 1.8. 151(97.4%) patients were successfully managed as outpatients. 4 (2.6%) patients failed treatment and required readmission within the 1-month time period. Of those who failed management 3 patients had complications, two with perforation and one with an abscess the 4th patient had no discernable complication on imaging. When compared to antibiotic use the readmission rate in this study is very similar. When antibiotic treatment is prescribed readmission rates are reported at 2.5%. (17)
One of the most recent studies performed in a Norwegian teaching hospital by Brochmann et. al (2016) was designed to assess outcomes for patients treated without antibiotics for AUD following a policy change within the hospital, implementing an approach of non-antibiotic treatment in cases of AUD confirmed with imaging(CT). All patients with CT-confirmed AUD were admitted for observation and supportive care. This study was a retrospective single-centre cohort study performed at the hospital from 1st January 2013- 30th June 2014. 244 patients were admitted in this time period with AUD. 177(81%) patients received no antibiotic treatment as per the new policy being implemented. 67(19%) received antibiotics on admission, 47 of these were considered a policy violation while the other 20 satisfied exclusion criteria that deemed antibiotics necessary. Of the 177 treated without antibiotics there were 7(3.9%) management failures, 5(2.8%) patients requiring antibiotics due to deterioration and 2(1.1%) readmissions within 1 month. There was 1(<1%) complication reported within the group, a fistula. 8 patients (5%) experienced a recurrence of diverticlitis within 1 year and 2(1%) had elective surgery. The 20 patients whom met the exclusion criteria and therefore received antibiotics 6(30%) patients developed complications. Of the 47 patients who received antibiotics due to a policy violation there were no reported complications among this group.(18)
Daniels et. al (2017) performed a large multi-centred randomized control trial across 22 clinical sites in the Netherlands to analyse the effectiveness of an antibiotic vs non-antibiotic treatment strategy for the treatment of first episode of AUD. The eligible patients had CT proven, primary left-sided, uncomplicated, acute diverticulitis. The patients were randomly assigned to a treatment or control group. Variables analysed included time to recovery during 6 months of follow-up, readmission rate, complicated, ongoing and recurrent diverticulitis, sigmoid resection and mortality. 528 patients were included in the trial. The median time to recovery was 12 days for the treatment group and 14 days for the control group (p=0.151) showing no significant statistical difference between the groups. No significant differences between groups were found for other varibales analysed including; complicated diverticulitis (2.6% vs 3.8% respectively; p=0.337), ongoing diverticulitis (4.1% vs 7.3%; p=0.183), recurrent diverticulitis (3.0%vs 3.4%; p=0.494), sigmoid resection (2.3% vs 3.8%; p=0.323), readmission (12.0% vs 17.6%; p=0.148), adverse events (54.5% vs 48.5%; p=0.221) and mortality (0.4% vs 1.1%; p=0.432). Duration of hospital stay was significantly shorter in the observation group (2days vs 3 days; p=0.006).(19)
Discussion:
Antibiotics have long been the cornerstone of management for acute uncomplicated diverticulitis. This treatment strategy has been largely based on expert opinion with a distinct lack of evidence for their use. The emergence of resistance to antibiotics is one of the reasons for a reassessment of this treatment choice. A review performed by de Korte et. Al (2011) (20) found that there were no randomized or prosepective studies available on the effect of outcome at that point in time. Since then better quality evidence on the topic of antibiotic use in AUD have emerged. The studies evaluated by de. Korte et. al (2011), while not considered to be high quality evidence demonstrated that use of antibiotics in the treatment of AUD may not provide any clinical benefit and these findings acted as a catalyst for more recent studies. The more recent studies have provided more concrete evidence. de korte et. al (2011) concluded that there was no difference in treatment failure rates when patients were treated with and without antibiotics, an incidental finding of this study showed that NSAID use increased the risk of recurrence of AUD. A multicenter randomized control trial carried out by Chabok et. al (2012) echoed these findings, the median hospital stay time was 3 days for both groups while no difference in rates of complication were found. A retrospective population based cohort study by Isacson et. al (2014) found that the rate of readmission for the non-antibiotic group was 4% compared to 6% for the antibiotic group. It must be noted that the number of patients in the antibiotic group was significantly lower than the non-antibiotic group 17 vs 165 respectively and so it is difficult to draw conclusions from this result. Isacson et. al (2015) then performed a prospective observational study which was the first study designed to assess outpatient treatment for AUD without the use of antibiotics. Failure of treatment was defined as readmission within one month and this study furthered the case for treatment of AUD without antibiotics. A failure of treatment rate of 2.6% in the group treated without antibiotics as outpatients which is almost exactly the same as the rate of failure seen when antibiotics are the treatment of choice 2.5%. Brochmann et. al (2016) performed a restrospective study and while the results showed a slightly higher rate of treatment failure (3.9%) when compared to the findings of Isacson et. al (2015) the use of conservative treatment is still justified. Daniels et. al (2017) performed a large multi-centre randomized control trial further investigating antibiotic vs conservative treatment in AUD. Results of this study shoed no significant difference in time to recovery or other varibales analysed. Duration of hospital stay was shown to be significantly lower in the non-antibiotic group vs antibiotic group 2 vs 3 days repectively.
Conclusion:
Antibiotics have been the cornerstone of treatment of acute uncomplicated diverticulitis despite a scarcity of evidence showing that they improve patient outcomes. A review of the treatment guidelines for the use of antibiotics in the treatment of AUD indicates that there are conflicting ideas with regards antibiotic use. The American Society of Colon and Rectal Surgeons, European Association for Endoscopic Surgery and the World Society for Emergency Surgery all recommend routine use of antibiotics for uncomplicated diverticulitis. The Association of Surgeons of the Netherland and the Danish Surgical Society do not recommend the routin use of antibiotics in treatment of uncomplicated diverticulitis while the American Gastroenterolical Society recommend selective rather than routine use of antibiotics. (21, 22). Recent evidence would suggest that routine antibiotic use in AUD does not show improved outcomes and that they should be reserved for use in the case of complicated diverticulitis or cases of deteriorating uncomplicated diverticulitis. However, it must be acknowledged that more high quality evidence is required. Evidence has shown differences in anatomic disease location of diverticulitis in different areas of the world (23) and so more studies must be performed in different geographical locations before the results of these studies can be generalized and ultimately lead to concrete guidelines with regards the routine use of antibiotics for treatment of uncomplicated diverticulitis.