Introduction
Diverticulosis of the colon describes a herniation of the colon wall in an area of weakness in the mesentery. Diverticulosis is extremely common, occurring in over 40% of the population over 60 years old, and its prevalence increases with age [1,2]. It is typically asymptomatic, though complications such as diverticulitis or gastrointestinal bleeding may occur.
Colorectal polyps are abnormal growths arising from the colon lumen. They vary in pathologic phenotype from benign hyperplastic polyps to adenomatous or villous polyps, considered pre-cancerous. Like diverticulosis, the incidence of colorectal polyps also increases with age. Both conditions have a higher prevalence among western and industrialized countries. These similar epidemiologic characteristics have led to the investigation of a possible shared pathophysiologic mechanism [3]. Diets low in fiber and high in saturated fats, as well as a slow colonic transit time, have been identified as common risk factors [4]. Histopathologic studies have suggested that chronic inflammation underlying diverticular disease development may also predispose colonic mucosa to polyp formation [5].
There is currently no consensus regarding the relationship between the 2 conditions. Muhammad et al determined that patients with diverticulosis have a higher risk of developing colorectal polyps compared to those without [4]. Baker et al found a lower rate of colorectal cancer (CRC), but an increased rate of polyp detection in patients with diverticulosis compared to a control population [3]. Levine et al found a lower frequency of adenomatous polyps in colonic segments containing diverticulosis and proposed a possible protective effect of diverticulosis against the development of concurrent neoplasia [6]. Our study aimed to further elucidate the significance of coexisting colonic polyps and diverticulosis, and to evaluate the proximity of diverticulosis to colon polyps and the histopathology of these polyps. Primary outcomes included the presence of colonic diverticulosis and its proximity to colon polyps, and polyp type and size (sub-centimeter vs. polyps over 10 mm in size.) Secondary outcomes included other pathologic findings of the colon.
Patients and methods
We performed a retrospective chart review of all patients who underwent colonoscopy between January 2011 and December 2020 at Saint Louis University Hospital in St. Louis, Missouri. Data were collected for Current Procedural Terminology (CPT) codes in the Provation MD Physician Documentation Software Program. All screening and diagnostic colonoscopy CPT codes in this time range were collected. Incomplete colonoscopies and flexible sigmoidoscopies were excluded. All CPT codes for diverticulosis were extracted into a list that was cross-matched with CPT codes for colon polyps and mass lesions. Duplicate colonoscopies from the same patient were excluded from this study.
The patient population included both male and female patients from 18-90 years of age who underwent colonoscopy during this time. The upper age limit of 90 years was selected because of the low number of patients receiving colonoscopies over this age. Patient characteristics such as age, body mass index (BMI), history of inflammatory bowel disease, family history of CRC, and procedure indication were documented. Each colonoscopy that matched the criteria of having both diverticulosis and polyps was evaluated to determine the spatial relationship of colon polyps with the presence of diverticula. The location of the diverticula and polyps, number of polyps, and pathology were recorded. The colon was divided into 9 segments: cecum, ascending colon, hepatic flexure, transverse colon, splenic flexure, descending colon, sigmoid colon, recto-sigmoid colon, and rectum. The left colon was defined as the rectum, sigmoid, descending, and splenic flexure, and the right colon was defined as the transverse, hepatic flexure, ascending, and cecum. If there was no segment specified, these were classified as either right, left, or pan-colonic diverticulosis. The data were then divided according to the presence or absence of polyps and diverticulosis, in relation to several patient factors including age, BMI, sex, and high-risk factors for CRC: history of inflammatory bowel disease, personal or family history of colon cancer, and history of advanced polyp. Possible confounding factors included age, sex and BMI, which were controlled for in the analysis of the association of diverticulosis and the presence of colon polyps in the adjusted odds ratio (aOR) analysis. Adjacent polyps were defined as polyps within the same colon segment as the diverticulosis, as most reports do not quantify exact distances between these 2 entities.
Statistical analysis
The incidence of diverticulosis in each of the 9 colon segments was determined, in relation to whether polyps were present in any segment during colonoscopy examination. Risk ratios and differences with 95% confidence intervals (CI) of diverticulosis incidence over polyp presence were calculated at each colon segment accordingly. Furthermore, the incidence rates of diverticulosis per segment were determined under the conditions that hyperplastic polyps or adenomas were found, respectively, and by size of polyps at each segment (size <10 vs. ≥10 mm). Patient characteristics were compared by whether diverticulosis and a polyp were found during endoscopic examination. Distribution of patient characteristics was assessed using the 2-tailed chi-square test. Risk ratios over polyp and diverticulosis presences were also assessed using the 2-tailed chi-square test. Risk difference was assessed using the Wald equality test. Multiple logistic regression models were constructed to quantify the adjusted associations of diverticulosis incidence and polyp presence, as mentioned above, after controlling for patients’ demographic and clinical factors (see Methods, third paragraph). All statistical tests were 2-tailed with an α of 0.05. All analyses were performed using SAS v9.4 (SAS Institute, Cary, NC).
Results
Our study identified 2595 screening and diagnostic colonoscopies that reported findings of colonic diverticulosis. Of these, 59 patients did not have a specified location of colonic diverticulosis and 43 patients had reported left- or right-sided diverticulosis. The baseline characteristics for this study are outlined in Table 1. Of the 2595 patients, 51% were men. The most common indication for colonoscopy was colon cancer screening (46%) and the most common location for diverticulosis was in the sigmoid colon (83%). Patient demographics were then analyzed in relation to the presence of colon polyps and, separately, colonic diverticulosis (Table 2). The presence of colonic diverticulosis itself was not statistically significant for a specific age group, personal or family history of colon cancer, or history of inflammatory bowel disease. However, a BMI <30 kg/m2 was associated with a greater incidence of diverticulosis. Demographic analysis in relation to the presence of colon polyps found that polyps were most often associated with the age group 50-69 years.
Table 1 Demographics
Table 2 Patient demographics by colon polyps and diverticulosis
With regard to the spatial relationship of colon polyps with diverticulosis, there was a statistically significant association of polyps adjacent to diverticulosis involving the rectum and transverse colon (Table 3). Of these colon segments with polyps adjacent to diverticulosis, the rectum and transverse colon had largely hyperplastic polyps (P<0.001) on pathology evaluation, while adenomatous colon polyps were more often located in the transverse colon (Table 3). Regarding polyp size, there was a statistically significant trend for transverse colon polyps adjacent to colon diverticulosis to be over 10 mm in size (P=0.0178), compared to other segments with polyps and adjacent diverticulosis (data not shown).
Table 3 Colon diverticulosis and colon polyps by pathology and location
Table 4 represents the association between advanced or adenomatous colon polyps and the incidence of CRC with the presence of diverticulosis, after adjusting for patient demographics and confounding clinical factors. The factors analyzed included patients’ age, sex, and BMI. Patients considered high risk for CRC included patients with history of inflammatory bowel disease, family history of CRC, history of adenomas or prior CRC (Table 4). For both left- and right-sided diverticulosis, patients less than 50 years old and patients over 70 had a higher incidence of adenomatous colon polyps. Patients at high risk for CRC had a greater risk of any colon polyp (P=0.048). It was noted that male patients were more likely to have adenomatous colon polyps, but overall there was no difference between men and women in the polyps found or incidence of colon cancer (aOR 1.232, 95%CI 1.051-1.443; P=0.0099). Patients with colonic diverticulosis, regardless of location, had a higher prevalence of adenomatous colon polyps. This was compared to all colon segments without diverticulosis. However, there was a lower incidence of CRC, despite the higher prevalence of adenomatous polyps (P<0.05).
Table 4 Demographics of advanced polyps with diverticulosis and incidence of colorectal cancer
Diverticulosis was then divided into right and left colon for evaluation of advanced polyps, demographics as above, and incidence of CRC. Left colon diverticulosis was associated with a higher risk of any colon polyp (both adenomatous and non-adenomatous colon polyps) (Table 4). Interestingly, there was no greater association of left colon diverticulosis with adjacent polyps and incidence of CRC (P=0.0085, aOR 1.237, 95%CI 0.900-1.273). Q. There is inconsistency between the P-value and the confidence intervals, please re-calculate. Right colon diverticulosis was not associated with a higher risk of CRC or polyps (Table 4). Univariate analysis demonstrated a significant association between left colon diverticulosis and higher rates of both adenomatous and non-adenomatous colon polyps (Table 5). CRC was not associated with colon diverticulosis, regardless of location.
Table 5 Univariate analysis: diverticulosis, colon polyps, and incidence of colorectal cancer
Discussion
Our data suggest that diverticulosis in any location in the large intestine may be associated with a greater likelihood of an adenomatous colon polyp. This remains statistically significant after controlling for patient variables, such as patients’ age, sex, BMI, and risk factors for CRC. Interestingly, the presence of diverticulosis with adjacent advanced adenomatous polyps did not yield an increased incidence of CRC. A recently published meta-analysis also demonstrated the dose-dependent relationship between BMI increase and risk of diverticular disease, while our study demonstrated that a lower BMI was associated with a greater risk of polyps with diverticulosis [7]. These findings are probably due to the smaller sample size in our study. Further limitations of our study include its retrospective methodology, where not all variables could be controlled.
In analyzing specific patient factors for both right- and left-sided diverticulosis with adjacent colon polyps, left diverticular disease was associated with both adenomatous and non-adenomatous colon polyps, but not a higher risk of CRC. Again, this is probably related to the sample size. In addition, our data illustrate that most of the diverticulosis associated with colon polyps was localized to the rectum, transverse colon and cecum. This is intriguing, as the left colon is typically the most common location for diverticular disease, which would in theory house the most polyps. However, our data show more right colon predominance for localization of both colon polyps and adjacent diverticulosis. This finding may reflect the angulation of the sigmoid colon and perhaps a greater likelihood of overlooking polyps. However, it may also reflect incorrect localization of polyps, as the prior literature indicates that gastroenterologists can be incorrect in predicting the location of the colon cancer and our data is based on gastroenterologists’ suspected location of colonic neoplasms [8].
Our research seeks to add to the body of literature evaluating the relationship of colon diverticulosis and colon polyps. Colonic diverticular disease can signify a harbinger of CRC if a thorough examination is not performed. Therefore, it is imperative to modify our clinical practice and investigate the mucosa surrounding colonic diverticulosis. To help prevent polyp formation, lifestyle modifications, such as high fiber diet, regular exercise, and healthy diet, should be recommended. Overall, more studies are needed to investigate this interesting relationship.
What is already known:
-
Diverticulosis and colon polyps are common findings during a colonoscopy
-
The presence of colon polyps can lead to an increased risk of colorectal cancer (CRC)
-
There is no consensus on the relationship between colonic diverticulosis and colon polyps
What the new findings are:
-
The presence of colonic diverticulosis in any location was associated with a greater incidence of adjacent colon polyps
-
There was a greater presence of both polyps with diverticulosis in the right colon
-
Despite the higher prevalence of colon diverticulosis with adjacent adenomatous colon polyps, there was no greater incidence of CRC