Association of Colonic Diverticula with Colorectal Adenomas and Cancer

Background and Objectives: Conflicting evidence is reported regarding any association between colonic diverticula with colorectal adenomas or cancer. The present study aimed to evaluate, in a cohort of Caucasian patients, the association between colonic diverticula and colorectal polyps and cancer. Materials and Methods: All consecutive patients undergoing colonoscopy at our institution were included in the study. The presence and location of diverticula, polyps, and cancers were recorded. Histologically, polyps were classified as adenoma (with low or high dysplasia), hyperplastic, or inflammatory. The relative risk of the association of polyps and cancer with diverticula was assessed. Multiple logistic regression analyses, including age, sex, family history for colorectal cancer (CRC), and family history for diverticula, were carried out. Results: During the study period, 1490 patients were enrolled; 37.2% (n = 555) showed colonic diverticula or polyps or CRC (308 males, mean age 66 years). Particularly, 12.3% (n = 183) patients presented only diverticula, 13.7% (n = 204) only polyps or cancer, 11.3% (n = 168) both diseases, and 62.7% (n = 935) neither diverticula nor polyps and cancer. A total of 38 patients presented colorectal cancer, 17 of which had also diverticula. A significant increase in relative risk (RR 2.81, 95% CI 2.27–3.47, p < 0.0001) of colorectal adenoma and cancer in patients with colonic diverticula was found. At multivariate analysis, only diverticula resulted to be significantly associated with colorectal adenomas and cancer (Odds Ratio, OR 3.86, 95% CI 2.90–5.14, p < 0.0001). Conclusions: A significant association of colonic diverticula with colorectal adenoma or cancer was found. This implies that patients with colonic diverticula require a vigilant follow-up procedure for the prevention of colorectal cancer from those applicable to the general population.


Introduction
Colonic diverticula and colorectal polyps and cancer are common findings during colonoscopy, especially in patients over 50 years of age [1]. Most patients with diverticula of the colon remain asymptomatic throughout their lives, a condition commonly referred to as diverticulosis or asymptomatic diverticula. Approximately 20% of those with colonic diverticula develop symptoms (abdominal pain or discomfort, bloating, constipation, or diarrhea), conditions described under the general heading of symptomatic diverticular All consecutive patients who underwent colonoscopy in the period between September 2016 and September 2019 were included. Patients underwent the procedure for a variety of reasons: uncomplicated lower abdominal pain or discomfort, hematochezia, changes in bowel habit, weight loss, iron-deficiency anemia, chronic constipation, chronic diarrhea, surveillance after colonic polypectomy, and screening for CRC.
Ethics Committee Approval: The study was performed with the institutional review board's approval (Prot. N. 43958). All clinical investigations were conducted according to the principles laid down in the Declaration of Helsinki.
Informed consent: Written informed consent was obtained from all patients included in the present study.

Colonoscopy and Histopathology
Before the colonoscopy, patients received a polyethylene glycol solution for bowel preparation and intravenous diazepam or midazolam for sedation. During the study period, 1994 colonoscopies were performed. A total of 504 patients did not meet the inclusion criteria, as 311 were affected by IBD, infectious colitis, or ischemic colitis, 137 had poor bowel cleansing (Boston score 0-1) [33], 54 had incomplete endoscopic examination, and 2 had polyps close to the scar, resulting from previous polypectomies. A total of 1490 patients were therefore enrolled and included in the statistical analysis. All enrolled patients were divided into four groups according to the endoscopic findings: group A, patients with only DD; group B, patients with only polyps or CRC; group C, patients showing both DD and polyps or CRC; and group D, patients without diverticula, polyps, or cancer ( Figure 1). Data were recorded on a standard database.

Statistical Analysis
Comparisons among the groups were assessed by the chi-square test and Fisher's exact test, as appropriate, for categorical variables. The relative risk (RR), with the 95% confidence intervals of the association of polyps and cancer with diverticula, was Colonoscopy findings were documented using a standardized colonoscopy-reporting system. Colonic diverticula are characterized by the herniation of the colonic mucosa and submucosa through defects in the muscle layer at the weakest point in the colonic wall. Diverticulosis was defined as the presence of at least two diverticula, in the absence of symptoms. DD was defined as intestinal symptoms and signs related to the presence of colonic diverticula [5,34].
Colorectal polyps were classified according to their location and histopathology. We defined the proximal colon as the cecum, ascending colon, and transverse colon, including splenic flexure, distal colon as descending colon, sigmoid colon, and rectum. All polyps were removed with a forceps biopsy (polyps < 5 mm) or a diathermic loop (polyps > 5 mm), as appropriate. Multiple biopsies were taken from CRC.
Histological samples were classified as cancer, adenoma (with low or high dysplasia), hyperplastic, or inflammatory polyps.

Statistical Analysis
Comparisons among the groups were assessed by the chi-square test and Fisher's exact test, as appropriate, for categorical variables. The relative risk (RR), with the 95% confidence intervals of the association of polyps and cancer with diverticula, was assessed. Multiple logistic regression analyses, including age, sex, family history for CRC, and family history for diverticula, were carried out. All statistical analyses were performed using the STATA version 15.1 2017 (StataCorp LLC, College Station, TX, USA).
Clinical characteristics of the four groups of patients are summarized in Table 1. Histological characteristics and localization of neoplastic lesions and diverticula of Group C (patients with both polyps or CRC and diverticula) are summarized in Table 2. In this group of patients, most polyps were found in the distal colon.  In particular, inflammatory polyps were only found in the distal colon. The histological distribution of all polyps and cancers found in Group B (patients with only polyps or cancer), in Group C (patients with both diverticula and polyps or cancers), and Group B + Group C is shown in Figure 2; the most frequent histological finding was adenomatous polyps, accounting for 60% (n = 224) of all lesions. Out of all the 555 patients (Group A + Group B + Group C), 6.7% (37) of them showed a CRC; out of these, 45.9% (17) also showed diverticula. Adenomatous polyps were found in 40.4% (224/555) patients; out of these, 104 (45.4%) also showed diverticula. Hyperplastic polyps were found in 17.5% (97/555) patients; out of these, 38.1% (37) also showed diverticula. Inflammatory polyps were found in 2.5% (14/555) patients; out of these, 71.4% (10) also showed diverticula.
The RR of the association of polyps and cancer with diverticula was assessed. A significant increase in RR of the association of polyps and cancer with diverticula of the colon was found. The RR was higher when only adenomas and cancer were evaluated compared to all types of polyps and cancer (2.81 vs. 2.67, respectively) (Tables 3 and 4).
The RR of the association of polyps and cancer with diverticula was assessed. A significant increase in RR of the association of polyps and cancer with diverticula of the colon was found. The RR was higher when only adenomas and cancer were evaluated compared to all types of polyps and cancer (2.81 vs. 2.67, respectively) (Tables 3 and 4).  At multivariate analysis (Table 5), of the demographic (age, sex) and clinical parameters (family history of colonic diverticula and neoplastic lesions) considered, only diverticula resulted to be significantly associated with colorectal adenomas and cancer (odds ratio (OR) 3.86, 95% CI 2.90-5.14, p < 0.0001). Table 5. Multivariate analysis with logistic regression and adjusted odds ratios (OR). Multiple logistic regression analyses including age, sex, family history for CRC, family history for diverticula, and presence of diverticula were carried out to investigate the independent association with colorectal polyps.

Discussion
Colon cancer could be associated with a history of diverticular disease and personal history of polyps that may be detected during colonoscopy, especially in older patients (>50 years old).
Many studies have been conducted on colonic diverticula but they are heterogeneous due to the different study designs (prospective, retrospective, case-control), the different classes of patients considered (population-based, hospital-admitted, outpatients, etc.), the various methods employed for the diagnosis of diverticula (barium enema, computed tomography (CT), surgery, endoscopy), and the different diseases for which the evaluations were conducted (diverticular disease, diverticulitis, polyps, adenoma, cancer) ( Table 6) [1,[26][27][28][29][30][31][32].   Most of the evidence available in literature derives from retrospective studies. However, large prospective cohorts have been recently described in Asia [42,49,53,60,65]. Excluding data published in the form of an abstract, to our knowledge, only three prospective studies have been carried out in Caucasian patients [1,26,64] and two in mixed cohorts from the USA [32,41] (Table 6). The incidence and distribution of diverticula in the colon is different between Western countries and Asian countries. In the former, the diverticula are more frequent and develop more in the left colon, while in the latter they predominate in the right colon [2][3][4][5].
In this prospective study, including all consecutive Caucasian patients, we found a significant association of colonic diverticula with colorectal adenomas or cancer. Additionally, the majority of all polyps (including both inflammatory and adenomatous polyps) were found in the distal colon, where diverticula were also more frequently present.
A recent meta-analysis involving 29 cross-sectional studies reported that diverticula were associated with increased risk of adenomas (OR 1.47, 95% CI 1.18-1.84) and with the more lax-defined "polyps" (OR 1.95, 95% CI 1.15-3.31), but not with CRC (OR 0.98, 95% CI 0.63-1.50) [71], as reported in an old case-control study [36]. Interestingly, a subanalysis showed that diverticulosis did not increase the risk of adenomas (OR 1.34, 95% CI 0.87-2.06) in patients who underwent screening colonoscopy [71]. The latter result might be related by the younger age of patients undergoing their first endoscopic CRC-screening. However, this work shows some limitations, such as the inclusion of cross-sectional studies (in which the causal relationship is difficult to assess), studies in which polyps were defined morphologically regardless of histologic evaluation, and the inclusions of data reported only in the form of abstracts.
In contrast with the results of the aforementioned meta-analysis [71], a cross-sectional study involving 2310 patients by Rondagh et al. reported that the association of diverticulosis with colorectal polyps was influenced by patient age: only below the age of 60 was the prevalence of colorectal polyps significantly higher in patients with diverticulosis than in those without diverticulosis [1]. An association with the female gender has also been described [41].
In the present study, all subtypes of polyps have been considered. The majority of all polyps were found in the distal colon. In particular, inflammatory polyps were only found in the distal colon. Adenomas accounted for 60% of all polyps. A significant increase in relative risk (RR 2.81, 95% CI 2.27-3.47, p < 0.0001) of colorectal adenoma and cancer in patients with colonic diverticula was found. At multivariate analysis, of the demographic and clinical parameters considered, only diverticula resulted in a significant association with colorectal adenomas and cancer (OR 3.86, 95% CI 2.90-5.14, p < 0.0001).
The location of inflammatory polyps in the distal colon confirms literature data, probably due to the presence of a mucosal inflammation of this colonic tract more commonly involved by diverticula [25,[72][73][74][75]. Inflammatory polyps develop as a result of an exuberant mucosal regeneration and repair of mucosal inflammatory lesions. The inflammatory polyps are found not only in IBD but also in infectious colitis, ischemic colitis, necrotizing enterocolitis, graft versus host disease, and colonic anastomosis. In non-IBD colitis, inflammatory polyps do not represent a risk factor for CRC.
The association between colonic diverticula and colorectal adenomas and cancer could be related to the epithelial proliferation of colonic mucosa. An upward shifting of cellular proliferation in the colonic mucosa of patients with diverticula was demonstrated and compared with age-matched controls. The cell proliferation index in patients with asymptomatic diverticulosis is three-fold higher than that of healthy controls [76]. It is not clear whether this abnormal epithelial proliferation in colonic diverticula could be responsible for the development of inflammatory polyps, as well as colorectal adenomas and CRC.
In the cohort of patients considered in this study, differences in the frequency of polyps and cancer in patients with diverticulosis, SUDD, SCAD, and diverticulitis have not been detected.
It has been reported that patients with DD, but also patients with asymptomatic diverticulosis, may display microscopic or macroscopic mucosal signs of mucosal inflammation [25,72]. The relation between chronic inflammation and cancer is well known. Patients with IBD have a significantly higher risk of developing CRC than unaffected subjects, a risk that increases with the extension and duration of the disease [82][83][84]. Furthermore, the epithelial hyper-proliferation of colon crypts, observed both in DD and asymptomatic diverticulosis [85], could be a potential cellular mechanism, leading to a possible increased risk of developing adenomas and CRC in patients with colonic diverticula. It is important to assess whether this increased risk of colorectal adenomas and cancer is linkable to colonic diverticula in general or only to chronic diverticulitis or diverticula with specific genetic and epigenetic variants.
On the other hand, some studies have hypothesized a protective effect of diverticula on CRC development [44]. It should also be noted that not all forms of chronic inflammation of the colon are associated with an increased risk of CRC. Recently, it was reported that microscopic colitis, chronic colitis with the same incidence and prevalence as IBD, appeared to show a reduced risk of adenomas and CRC [86,87].
However, our study shows, some limitations. As this was a cross-sectional study, it was not possible to establish the causal relationship between DD and neoplasm, as well as the future risk of developing CRC in patients with diverticulosis.
Moreover, the indication of a surveillance colonoscopy for a history of polyps was largely self-reported by the patient. Similarly, familial history of CRC and colonic diverticula were also self-reported by patients.
An additional possible limitation is represented by the non-systematic use of validated scales, reporting the extent of diverticulosis and inflammation, or the Paris classification, as the aim of the study focused on histologic findings.
Finally, results involving patients with cancer may be hampered by the limited number of CRCs detected in our study.
However, with the aforementioned limitations, the importance of our study derives not only from being a prospective study including a homogenous population of Caucasian patients but also from a detailed endoscopic localization of diverticula, polyps, and cancers in the various tracts of the large bowel and a histological definition of the various types of polyps and cancer detected.

Conclusions
Diverticula and adenomas are frequently found during colonoscopies in adult and elderly patients; however, the evidence of the association between these two conditions is still conflicting.
In our cohort of Caucasic patients, colonic diverticula and colorectal polyps resulted in a significant association, more so for colorectal adenomas and cancer. This finding suggests that patients with colonic diverticula, harboring a higher risk for dysplastic lesions, could require more vigilant endoscopic surveillance than the general population.
However, the present study was not designed to assess follow-up risk. Thus, further prospective large-scale population studies with a long follow-up are needed to corroborate the evidence supporting the relationship between colonic diverticula and colorectal adenomas and CRC, as well as to determine the best time interval between endoscopic examinations.