Introduction

The incidence of diverticulosis and colorectal polyps increases as the population ages [1–3]. Diverticular disease has an overall prevalence of <10 % in individuals <40 years of age. However, it increases to 66 % by the seventh decade of life [1, 2]. Similarly, the incidence of colorectal polyps also increases with advancing age. Several studies have reported correlation between the presence of diverticulosis and colorectal polyps. The proposed common mechanisms for diverticulosis and colon polyp formation include lack of dietary fiber, increased saturated fats, and slow colonic transit time [1, 2, 4]. In animal models, diets high in fiber protect against collagen crosslinking, which has been proposed as possible mechanism for the formation of diverticulosis [5]. It has also been hypothesized that the composition of the extracellular environment of the colonocytes also plays a role [6]. Both macroscopic and microscopic alterations involved in the development of diverticular disease may also predispose the colonic mucosa for polyp formation [7]. Current evidence strongly supports the role of chronic microscopic inflammation of diverticula in the formation of colorectal carcinogenesis [8]. Histopathologic studies have shown an increased inflammatory cell density and proliferation index in the colonic mucosa of patients with diverticulosis when compared to healthy controls [9, 10].

While there have been studies assessing the association between diverticulosis and colorectal polyps, the results have been conflicting. For example, Morini et al. performed a retrospective cohort study of 630 patients, who underwent colonoscopy with symptomatic diverticular disease. They concluded that patients with diverticular disease were more likely to have adenomas in the sigmoid colon [11]. In another retrospective study of 672 patients from Japan, an increased association of colon polyps with diverticulosis was reported [12]. Similarly, Bressler et al. showed that diverticulosis was an independent risk factor for subsequent development of colorectal cancer within 3 years after a colonoscopy [13]. In contrast, a retrospective study from the Netherlands evaluated 4,241 patients and found no relationship between the presence of diverticulosis and colorectal polyps [14]. Furthermore, Kieff et al. evaluated 502 patients undergoing screening colonoscopy. They found diverticulosis was not associated with advanced proximal neoplasia; however, women did appear to have an increased risk of distal neoplasms [4]. Other studies have reported conflicting results as well [6, 15, 16].

The relationship between diverticulosis and colorectal polyps is difficult to study in view of parallel increase in frequency of both conditions, as the patient population ages. If an association exists between these two conditions, the type of cancer screening might need possible adjustment. For example, identifying diverticulosis on CT colonography or flexible sigmoidoscopy would warrant a complete colonoscopy to evaluate for the presence of polyps. Despite sharing common epidemiologic predisposing factors, the association between diverticulosis and colorectal polyps remains unclear and needs better clarification.

Aim

The primary aim of our study is to evaluate if there is any association between diverticular disease and colorectal polyps. The secondary aim is to find out if there is an association between indications for colonoscopy and incidence of colorectal polyps in patients with diverticulosis.

Materials and methods

This retrospective data collection study was conducted at a Veterans hospital. It was approved by the Institutional Review Board and the Research and Development Department of the University of South Florida and James A. Haley Veterans Hospital in Tampa, Florida.

Inclusion criteria

All consecutive patients who underwent colonoscopy for various indications between January 2009 and December 2011 were included.

Exclusion criteria

  1. 1.

    History of inflammatory bowel disease

  2. 2.

    Polyposis syndrome

  3. 3.

    Poor bowel preparation at the time of procedure

  4. 4.

    Incomplete colonoscopy

Outcome measures

The primary outcome was association between diverticulosis and colorectal polyps. The secondary outcome was association between diverticulosis and colorectal polyps according to age and indications for colonoscopy.

Data abstraction

Data were abstracted from medical records of individual patients who met the inclusion criteria. Data was collected on the presence or absence of diverticular disease and colorectal polyps. Diverticular disease was defined as the presence of three or more diverticula on colonoscopy. We also collected data on the location of diverticulosis and polyps, which were classified into right-sided, left-sided, or pan-colonic. Right-sided disease was defined as the presence of diverticulosis or polyps proximal to the splenic flexure (cecum, ascending colon, and transverse colon). Left-sided disease was defined as the presence of diverticulosis or polyps distal to the splenic flexure (descending colon, sigmoid colon, and rectum).

Polyp pathology

The histopathology of the polyps was also recorded. Hyperplastic polyps were excluded from the statistical analysis. Only polyps with the histology of “tubular adenoma and advanced neoplasia” were included in the analysis. Advanced neoplasia was defined as an adenoma with villous component of greater than 30 %, high-grade dysplasia, or adenocarcinoma.

Statistical analysis

The association between diverticulosis and colorectal polyps was assessed using chi-square test. We also used a logistic regression analysis to examine the association between diverticulosis and colorectal polyps. Data on all outcomes are summarized as odds ratio (OR) along with associated 95 % confidence intervals (CI). All statistical analysis was performed with SPSS 15.0 for Windows.

Results

A total of 2,223 patients met the inclusion criteria. The baseline characteristics of the patients and indications for colonoscopy are shown in Table 1. The groups with and without diverticulosis were similar for variables of age, gender, and indications for colonoscopy. While the median age of subjects in the diverticulosis group was 63 years, the median age of patients without diverticulosis was 59 years. There were 1,203 (54 %) patients with diverticulosis and 1,020 (46 %) without diverticulosis. Polyps were found in 1,001 out of 2,223 (45 %) patients. The most common indication for colonoscopy was screening followed by personal history of polyps and hematochezia/melena. The location of diverticulosis and colorectal polyps along with their pathology is shown in Table 2. Tubular adenoma was found in 88 % of patients, and 12 % patients had advance adenoma including adenocarcinoma. The summary result on the association between diverticulosis and colorectal polyps is shown in Table 3. Overall, there was a statistically significant association between diverticulosis and colorectal polyps (OR 1.54; 95 % CI 1.27–.80, p = 0.001). This association was found significant for all locations of polyps (right-sided, left-sided, and pan-colonic) and for all histological subtypes. There was also a statistically significant association between age, presence of diverticulosis, and colorectal polyps (OR 1.03; 95 % CI 1.02–1.04). The incidence of colorectal polyps increases as age advances in patients with diverticulosis, with the highest association in patients >70 years of age (OR 3.55; 95 % CI 2.50–5.04). There was no significant association between indication for colonoscopy and incidence of colorectal polyps in patients with diverticulosis (OR 0.98; 95 % CI 0.95–1.01). The incidence of diverticulitis was low (<1 %), and there was no association between diverticulitis and colon polyps.

Table 1 Baseline characteristics of patients who underwent colonoscopy for various indications
Full size table
Table 2 Location of diverticulosis and colorectal polyps with their histology
Full size table
Table 3 Univariate and multivariate logistic regression analysis for predictors of polyps in patients with diverticulosis
Full size table

Discussion

The findings from our study showed a significant association between diverticulosis and colorectal polyps. Patients with diverticulosis are at a higher risk for colorectal polyps as compared to those without diverticulosis. This association was found significant for all locations of polyps and all histological subtypes. Furthermore, analysis also revealed that the age is an important predictor of colorectal polyp in patients with diverticulosis. The incidence of polyps increases in these patients as age advances. This association warrants further evaluation of the proposed common mechanisms of slow colonic transit and low dietary fiber consumption, leading to the development of both diseases. The combination of the long colonic transit time with an altered colonic extracellular matrix from the decrease in dietary fiber intake may lead to genetic alterations resulting in the predisposition to form polyps and diverticulosis [7].

We have conducted this study to further evaluate the possible association of these two frequent disease entities in the colon, because the previous reported studies have had conflicting results. Our study is confined to the veteran population which has not previously been evaluated. This population is mainly older males with multiple other comorbid conditions. We have excluded patients with inflammatory bowel disease and polyposis syndrome as they are predisposed to colon cancer as a part of the disease process. Similarly, hyperplastic polyps were also excluded from the analysis and we only included “pre-malignant lesions (advanced neoplasia).”

The rate of diverticulosis and colorectal polyps in our cohort is in agreement with previous studies, showing an increase in frequency as the patient population ages. The higher OR of 1.54 is in agreement with the recent study from Japan by Hirata et al., where they had patients with more right-sided diverticular disease with exclusion of patients with prior polypectomy [12]. Our findings are also in agreement with a recent study by Rondagh et al., where they found a significant association between diverticulosis and colorectal polyps in patients <60 years of age (OR 1.87, 95 % CI 1.26–2.78) [17]. Similarly, the association of the presence of polyps in patients with diverticular disease is also in agreement with previous studies conducted by Loffeld and Van Der Putten [15] and Lee et al. [16]. In contrast, Krones et al. found a decreased rate of advanced colonic neoplastic lesions in patients with acute diverticulitis undergoing colonic resection [6]. Other studies which failed to show an association between diverticulosis and colorectal polyps were mainly focused on hospitalized patients with acute diverticulitis [18]. However, there are studies which have reported higher incidence of colorectal cancer in patients with diverticulitis [19].

Several pathology studies have shown both macroscopic and microscopic inflammation in some patients with diverticulosis. This is also regarded as a risk factor for colorectal neoplasia. Furthermore, by using proliferating cell nuclear antigen index, an upward shift of cellular proliferation has been observed in patients with diverticulosis. It has been demonstrated in patients with advanced adenoma which makes it intriguing. Other microscopic studies have yielded aberrant crypt foci in diverticulosis which are significantly increased in patients with colorectal cancer, leading credence to the concept of inflammation-carcinoma sequence [7]. Alteration in stromal matrix may promote some of the epithelial transformation through cross talk between the extracellular matrix and epithelium [20]. Several advances have been made to explain exactly how inflammation promotes carcinogenesis. Inflammation increases oxidative stress and free radicals, most of which target DNA, RNA, proteins, and lipids [21]S. Lastly, the free radicals can damage key genes, for example, p53 gene, and initiate carcinogenesis, and these have been suggested in interventional studies in animals as well, supporting the notion that inflammation provides a “congenial soil” for carcinoma [8]. Diverticulosis is a “pre-inflammatory lesion” and polyps (adenomas) are pre-cancerous lesions of the colon. Keeping this in mind, we did a subgroup analysis of patients with diverticulosis and looked into all the case with prior history of diverticulitis. The incidence of diverticulitis in our patient population was low (<1 %). We found no association between diverticulitis and the incidence of colon polyps in our patient population. This may be related to the small subset of patients with prior history of diverticulitis in our cohort.

Severe diverticular disease may also limit adenoma detection due to inability to visualize the entire mucosa, and polyps may be missed. The low adenoma detection rate (ADR) could be a limiting factor in the studies which failed to show an association between diverticulosis and colorectal polyps [14]. This could affect the colonoscopy interval in patients with severe diverticulosis, and more frequent examination may be necessary. The overall ADR in our study was 45 % (1,001/2,223), which is currently superior to the current national guideline of 25 % in men for screening colonoscopy. The higher ADR might be a reason for the significant association between diverticulosis and colorectal polyps in our cohort.

There are limitations to our study. First, it is a retrospective cohort study with a large number of male patients since it was conducted at a veterans facility. The indication of surveillance colonoscopy for history of polyps was largely self-reported by the patient. Several attempts were made to obtain the previous histology of polyps; however, due to the large number of patients with outside pathology records, this was not possible. The strength of our study is the large sample size (2,223 patients) with post procedural data available in all patients.

Conclusion

There is a significant association between diverticulosis and synchronous pre-cancerous colorectal polyps (adenomas). Patients with diverticulosis have a higher risk of colorectal polyps. The incidence of colorectal polyps increases as age advances in patients with diverticulosis. These findings are based on a large patient series (>1,000 patients per group). However, the association between diverticulosis and colorectal polyps needs further validation and recognition by a large multicenter prospective study.