Temporal changes in the management of diverticulitis

Purpose

This study was designed to evaluate temporal trends in the use and type of operative and non-operative interventions in the management of diverticulitis.

Methods

A retrospective cohort using a statewide administrative database was used to identify all patients hospitalized for diverticulitis in the state of Washington (1987–2001). Poisson and logistic regression were used to calculate changes in the frequency of hospitalization, operative and percutaneous interventions, and colostomy over time.

Results

Of the 25,058 patients hospitalized non-electively with diverticulitis (mean age 69 ± 16, 60% female) there were only minimal changes in the frequency of admissions over time (0.006% increase per year-IRR 1.00006 95% CI 1.00004, 1.00008). The odds of an emergency colectomy at initial hospitalization decreased by 2% each year (OR 0.98 95% CI 0.98, 0.99) whereas the odds of percutaneous abscess drainage increased 7% per year (OR 1.07 95% CI 1.05, 1.1). Among patients undergoing percutaneous drainage, the odds of operative interventions decreased by 9% compared to patients who did not have a percutaneous intervention (OR 0.91 95% CI 0.87, 0.94). The proportion of patients undergoing colostomy during emergency operations remained essentially stable over time (range 49–61%), as did the proportion of patients undergoing prophylactic colectomy after initial non-surgical management (∼10%).

Conclusions

There was a minimal increase in the frequency of diverticulitis admissions over time. A rise in percutaneous drainage procedures was associated with a decrease in emergency operative interventions. The proportion of patients undergoing colostomy remained stable, and there does not seem to be a significant increase in the use of one-stage procedures for diverticulitis.

Introduction

Most data regarding diverticulitis management comes from single-institution case series and is therefore limited by selection and publication along with scope. Although there have been obvious changes in the management of diverticulitis over the last 20 years [computed tomography (CT) scanning, percutaneous drainage, better oral antibiotics and pain management, etc.] it is less clear to what extent these have impacted important outcomes. For example, there has been a growing body detailing primary resection and anastomosis [1, 2] in patients with diverticulitis as an alternative to Hartmann’s Procedure (HP). Over the same time period, primary anastomosis has become the standard of care in patients with traumatic perforations of the colon [3]. It is unclear to what extent primary anastomosis and colostomy avoidance has been incorporated into the care of patients with diverticulitis.

The purpose of this study was to evaluate temporal trends in the use of surgical and percutaneous interventions in the management of diverticulitis. Specifically, our goal was to define operative management patterns in the population at large and given the trend toward primary anastomoses in other areas of colonic surgery, to test the hypothesis that the use of colostomy during non-elective operations has decreased over time.