A clinical decision rule to establish the diagnosis of acute diverticulitis at the emergency department

Abstract

Purpose: The aim of this study was to identify patients in whom the clinical diagnosis of diverticulitis can be made with a high certainty, distinguishing them from patients requiring imaging.

Methods: We prospectively recorded clinical features in patients with acute abdominal pain presenting at the emergency department, before they underwent imaging. We identified features significantly associated with a final diagnosis of acute diverticulitis using multivariate logistic regression analysis and developed a decision rule based on these features. We evaluated the performance of the rule in identifying patients with a high probability of having diverticulitis.

Results: In total, 112 of the 1021 patients (11%) had a final diagnosis of diverticulitis. Of the 126 patients with clinically suspected diverticulitis, 80 had a final diagnosis of diverticulitis. In 32 patients with diverticulitis as their final diagnosis, another clinical diagnosis was made. A decision rule was based on the 3 strongest clinical features: direct tenderness only in the left lower quadrant, the absence of vomiting, and a C-reactive protein >50 mg/L. Of the 126 clinically suspected patients, 30 patients had all 3 features (24%), of whom 29 had a final diagnosis of acute diverticulitis (97%; 95% CI: 83%-99%). Of the 96 patients without all 3 features, 45 (47%) did not have diverticulitis.

Conclusion: In a quarter of patients with suspected diverticulitis, the diagnosis can be made clinically based on a combination of direct tenderness only in the left lower quadrant, the absence of vomiting, and an elevated C-reactive protein. In patients without these features, imaging is required to reach adequate diagnostic accuracy.