Not All Patients with Acute, Uncomplicated…

ABSTRACT & COMMENTARY

Not All Patients with Acute, Uncomplicated Diverticulitis Require Hospitalization

By Jennifer A. Best, MD

Assistant Professor, University of Washington School of Medicine, Seattle, WA

Dr. Best reports no financial relationships in this field of study.

SOURCE: Biondo A, Golda T, Kreisler E, Espin E, Vallribera F et al. Outpatient versus hospitalization management for uncomplicated diverticulitis. Ann Surg 2014;259:38-44.

Diverticulitis of the colon is a common condition frequently implicated as an admitting diagnosis among hospitalized adults in the United States and elsewhere. Most often, hospitalists initially manage this condition with intravenous antibiotics and transition to oral therapy upon improvement in clinical status, as evidenced by normalization of vital signs, diminished gastrointestinal symptoms and ability to tolerate an oral diet. Current guidelines lack clear recommendations on the ambulatory management of uncomplicated diverticulitis, which can be defined as lacking a complication such as perforation, obstruction, gastrointestinal bleeding or fistula. The majority of diagnosed episodes of diverticulitis can be classified in this fashion. Given a lack of prospective and randomized studies, it is uncertain whether acute, uncomplicated diverticulitis can safely be managed in the outpatient setting with oral agents alone. It is worth noting that only approximately 15% of patients admitted with acute diverticulitis require urgent surgical intervention during that admission.

In this trial published in the Annals of Surgery in January 2014, Biondo and colleagues evaluated the result of two different management strategies for acute, uncomplicated left-sided diverticulitis. Their randomized study was performed at five university hospitals in Spain. Potential subjects were recruited from a population of patients older than 18 years who presented to the emergency department with suspicion of diverticulitis with fever and acute abdominal pain and tenderness. These patients were evaluated initially with radiographs of the chest and abdomen to exclude other etiologies for symptoms and an abdominal CT scan with contrast. Patients were excluded from participation if they exhibited signs of complicated diverticular disease (including even small abscesses), failed to respond to initial therapy or tolerate oral intake in the ED, or carried additional comorbidities which rendered them high risk for decompensation (pregnancy, recent antibiotic use, suspicion of malignancy, immunosuppression). On-call surgeons at each site, not investigators, assumed responsibility for recruitment and randomization.

Patients were randomized to two groups. Group 1 was admitted to the hospital. Group 2 was discharged from the emergency department and contacted daily by investigators for 5 days subsequent to that discharge. All patients received an initial dose of intravenous antibiotics in the emergency department (amoxicillin and clavulanic acid; ciprofloxacin and metronidazole substituted for patients with penicillin allergy). Group 1 was then admitted to the hospital and managed with intravenous antibiotics and fluids for 36-48 hours to tolerance of oral intake and adequate pain control, followed by discharge. Group 2 was discharged directly from the emergency department with instructions to continue oral amoxicillin-clavulanic acid (or ciprofloxacin-metronidazole, if allergic). Both groups completed 10 days of total antibiotic therapy and received dietary recommendations. For the outpatient cohort (Group 2), this consisted of liquid diet with electrolyte drinks for 2 days, then escalation to low fiber. Pain was managed with paracetamol (acetaminophen). Phone calls to the outpatient subjects included assessment of temperature, diet, bowel function and pain. All patients were seen in clinic at 14 days and referred for colonoscopy to exclude neoplasm between days 45 and 60. The study's primary outcome was treatment failure of the outpatient strategy, as compared with initial hospitalization. Treatment failure was documented to have occurred in the setting of persistent pain and fever, progression to bowel obstruction or drainable abscess, surgical indication or mortality in the 60 days post-randomization. Other endpoints included quality of life (as assessed by the SF-12 tool) and a cost analysis completed at the coordinating institution, based on services for diagnosis, treatment, follow up and mean length of stay.

132 patients were ultimately randomized, and no significant differences were noted between the groups at baseline. Of this group, only 7 (5.3%) were readmitted, with no difference in readmission between the groups. No patients died and none required emergency surgery. There were no differences observed between initial inpatient and outpatient therapy related to quality of life at days 14 and 6, though within each group, quality of life improved between these two clinic visits. The cost was three times lower with outpatient therapy than with initial hospitalization (1124.70 euros less per patient — equivalent to $1532.74 US at the time of this article). The authors conclude that outpatient treatment with oral antibiotics is as safe and effective as initial hospitalization for intravenous therapy, at a lower cost and without decline in quality of life.

Though this is the first RCT to address the risks and benefits of outpatient management of acute uncomplicated diverticulitis directly, there are a number of weaknesses with this study. Subject numbers were small. Many patients were not randomized due to the presence of complicated disease — only those with confined phlegmon were included. Patients with confined small abscess were excluded for simplicity here, but this modestly more complicated population likely warrants additional study. Furthermore, a substantial number of patients otherwise eligible refused to be randomized, given the possibility of outpatient management. This reader notes the even patients randomized to group 2 received a single dose of intravenous antibiotics prior to discharge, so it remains unclear whether these patients would have recovered with a strictly oral antibiotic regimen — this too should be investigated. In summary, however, these data suggest that many patients with acute, complicated diverticulitis, as manifest by inflammation/phlegmon without abscess, may be safely treated with early discharge and oral antibiotics. On a related note, interesting data are arising about the utility and necessity of follow-up colonoscopy in these patients, but we'll save that for another review.