Abstract
Background: Asymptomatic Diverticulosis (the presence of outpouchings of the colonic mucosa) affects a third of the US adult population over the age of 60. This condition often (10-40%) leads to more acute disease such as infection and perforation, which carry significant risk of morbidity and mortality. The causes of diverticulosis remain unknown and poorly researched, but it is primarily a disease of industrialized, western populations. Current theory links low dietary fiber intake, constipation, physical inactivity, and a diet high in red meat and fat to the condition. Surprisingly, a similar study to this one, failed to show an association between these factors and diverticulosis. Therefore we sought to expand on those findings with more variables and a larger, multicenter data set.
Methods: We performed a cross-sectional analysis of 2804 adult patients undergoing colonoscopy at 11 sites throughout the United States and Puerto Rico. Patient population and data came from subjects enrolled in the Vitamin D/Calcium Polyp Prevention Study, a randomized, double-blind, placebo-controlled trial. At original enrollment, data on demographics, medical history, lifestyle habits, and diet were collected. Hemorrhoids were also studied because they are thought to share similar risk factors with diverticula (increased pressure in the large intestine and rectum). Cases are defined as patients with any diverticula (D) identified at colonoscopy, whereas controls had no diverticula reported (C). Multivariate analyses were performed using logistic regression to assess the association between potential risk factors and case-control status.
Results: Subjects with diverticula were younger (56.6 vs. 59.9 years, p<0.0001) and had a slightly lower BMI (28.8 vs. 29.7, p<0.0001). We found a very strong positive correlation between hemorrhoids and diverticula upon colonoscopy (Chi-squared test: 57.05, p<0.0001; Odds Ratio 1.81). A weak correlation exists between genders, with men slightly more likely than women to have diverticula (Chi-squared test: 3.87, p=0.049; Odds Ratio 1.17). There was no statistically significant difference in fiber intake between subjects with and without diverticula (15.28 vs 14.78 grams, p<0.088). We also found no correlation between activity level and presence of diverticula (Chi-squared test: 0.575 p=0.75).
Conclusions: Our study found no link between diverticulosis and fiber intake, activity level, or increased BMI despite finding a greater odds of diverticula in subjects with hemorrhoids. This is in opposition of current theory on the genesis of diverticulosis and brings to question the pathophysiology of the condition.
