Antibiotics for acute diverticulitis

Clinical question

Do antibiotics change clinical outcomes for patients with acute uncomplicated diverticulitis?

Bottom line

For nonseptic immunocompetent patients with acute uncomplicated diverticulitis, antibiotics do not alter early complication or recurrence rates.

Evidence

• Patients were immunocompetent and had symptoms compatible with acute uncomplicated diverticulitis (confirmed on a computed tomography scan) without being septic or critically ill. They were randomized to about 7 days of antibiotics (cephalosporin-metronidazole or amoxicillin–clavulanic acid) or placebo or no antibiotics.

• Systematic review (3 RCTs, n=1329)¹:

  • - 30-day complications (eg, abscess, perforation, obstruction, fistula): 1.5% versus 1.3% (no antibiotics) and not statistically different.

  • - Long-term (2- to 11-year) risk of recurrence was about 24% for both groups.

• 3 largest, highest-quality RCTs:

  • - 623 hospitalized adults from Sweden with first or recurrent acute uncomplicated diverticulitis²: At 1 year, there were no statistical differences in complications during hospitalization (1.0% vs 1.9% [no antibiotics]), median hospital stay (3 days each), or recurrence (about 16% each). At 11 years (about 90% of patients),³ there were no differences in recurrence or surgery.

  • - 528 adults from the Netherlands with a first diverticulitis episode⁴: At 6 months, there were no statistical differences in median recovery time (12 vs 14 days [no antibiotics]) or readmission (12% vs 18% [no antibiotics]). At 24 months (about 90% of patients),⁵ there were no differences in recurrence, complications, or surgery.

  • - 480 adults from Spain in the emergency department with diverticulitis⁶: At 3 months, there were no statistical differences in hospitalization (5.8% vs 3.3% [no antibiotics]) or emergency surgery.

• Limitations: Some RCTs were unblinded.^2,6

Context

• Guidelines suggest not routinely using antibiotics in immunocompetent, non–medically frail patients.^7,8

• Diverticulitis rates are increasing (especially in those <50 years).⁹

• Genetics appear to be involved in about 50% of cases.⁸

• Nuts, seeds, or popcorn do not appear to influence diverticulitis.⁸

• The risk of uncomplicated colorectal cancer is about 0.5% (similar to asymptomatic controls) and risk of complicated colorectal cancer is about 8%.¹⁰

• The risk of complicated diverticulitis is highest during the first episode.⁹

• Recurrence rates are about 17% after the first episode and about 44% after the second episode.⁹

Implementation

The triad of abdominal pain, fever, and elevated white blood cell count is not sufficient to diagnose diverticulitis.¹¹ A computed tomography scan is the best test to confirm diagnosis (positive likelihood ratio=94, negative likelihood ratio=0.1)^7,12 and differentiate uncomplicated from complicated diverticulitis. Due to the likelihood (8%) of complicated diverticulitis being cancerous,¹⁰ colonoscopy (after 6 to 8 weeks) should be considered.¹² Colonoscopy is not required to rule out cancer in uncomplicated diverticulitis.¹²