Colonoscopy Screening for Colon Polyps: May be Useful at an Earlier Age in Prevention of Malignant Transformation?


 Objective

The objective of this study was to investigate whether colonoscopy screening should be performed at an earlier age than specified in the current guidelines.
Method:

Data of 1882 patients who presented to the gastroenterology clinic of our hospital and underwent colonoscopy were retrospectively evaluated. Polyps detected during colonoscopy were excised and sent to the pathology laboratory for histopathological examination. Patients were divided into two groups as < 45 yo and ≥ 45 yo and the data were compared between the groups.
Results

The most common indications for colonoscopy were found as abdominal pain in 40.6%, rectal bleeding in 18.6%, constipation in 14.2%, anemia in 10.5%, diarrhea in 3.1% and other reasons in 6.8% of the patients. The rate of polyps was statistically significantly higher in ≥ 45 yo patients (43.3%) compared to < 45 yo patients (32.1%). The rate of neoplastic polyps was statistically significantly higher in ≥ 45 yo group compared to < 45 yo group. The most common localizations of polyps were rectum and sigmoid colon in all age groups. Three endoscopists performed the procedures and one of them with more experrience detected more polyps.
Conclusion

Based on our results, we believe that onset age of screening for colon polyps should be 45 years as recommended by the updated guidelines. In addition, colonoscopy screening should be performed in young adults without alarm symptoms who present with abdominal pain. Experience of the endoscopist, bowel cleaning and duration of withdrawing fom the cecum affects the number of detected polyps.


Results
The most common indications for colonoscopy were found as abdominal pain in 40.6%, rectal bleeding in 18.6%, constipation in 14.2%, anemia in 10.5%, diarrhea in 3.1% and other reasons in 6.8% of the patients. The rate of polyps was statistically signi cantly higher in ≥ 45 yo patients (43.3%) compared to < 45 yo patients (32.1%). The rate of neoplastic polyps was statistically signi cantly higher in ≥ 45 yo group compared to < 45 yo group. The most common localizations of polyps were rectum and sigmoid colon in all age groups. Three endoscopists performed the procedures and one of them with more experrience detected more polyps.

Conclusion
Based on our results, we believe that onset age of screening for colon polyps should be 45 years as recommended by the updated guidelines. In addition, colonoscopy screening should be performed in young adults without alarm symptoms who present with abdominal pain. Experience of the endoscopist, bowel cleaning and duration of withdrawing fom the cecum affects the number of detected polyps.

Introduction:
Colon polyps are considered a bridge to colorectal cancer (CRC) because of their potential for malignant transformation. Early detection and endoscopic removal are essential in the prevention of CRC. CRC is one of the most common cancer types in Western countries, and annually 446.000 new cases are seen in Europa. Annual mortality from CRC is 214.000 for both sexes [1]. CRC cases are estimated to reach 1.36 million in men and 1.08 million in women worldwide by 2035 [2]. Recently, a change has been found in the incidence of polyps with increasing localization in the proximal colon in Western countries, which has been accompanied by the increased number of sessile serrated polyps (SSP).
Most polyps are asymptomatic and remain unrecognized, but major clinical presentations in symptomatic cases include gastrointestinal bleeding, abdominal pain, intestinal obstruction, rectal prolapsus and changed bowel habits [3]. Polyps are histologically classi ed as neoplastic and nonneoplastic [4]. Non-neoplastic colon polyps may be in ammatory hyperplastic and cystic polyps.
Whereas neoplastic polyps can be benign such as tubular, and tubulovillous adenomas or malignants, and these polyps are precursors of colon cancers. Recent studies have shown that hyperplastic polyps at the right side of the intestines may have malignant potential and should be closely followed-up [5]. These polyps have adenomatous components, they show a serrated, saw tooth like surface epithelium and have BRAF mutations [6].
Colon polyps are curable if excised. If left untreated, the patient may develop complications such as bleeding, and the condition may be fatal because of malignant transformation. Styrker et al. suggested that the risk of developing cancer because of sporadic colon polyps of 1 cm is 8% in 10 years and 24% in 20 years [7]. Therefore, removal of polyps are important to avoid the development of malignant polyps [8]. Interestingly, recent studies have reported an increase in the incidence of CRC in people younger than 40 years [9][10][11][12].
Colonoscopy is considered the best screening modality for colon cancer and adenomatous polyps by most gastroenterologists [13,14]. Current guidelines recommend to begin colonoscopy screening at 50 years old and to continue until 75 years old [15]. Given the bene ts of colonoscopy screening in adults and increasing incidence of CRC in younger ages [16], there is a debate in the literature about whether colonoscopy screening for CRC should be started at an earlier age [17]. However, there is no empirical data on the effects of colonoscopy performed in < 45 years age group.
The objective of this study was to investigate whether colonoscopy screening should be performed at an earlier age than speci ed in guidelines.

Materials And Methods
Data of patients who presented to the gastroenterology clinic of our hospital due to various reasons, had a positive stool screening test and underwent colonoscopy from 2017 through 2019 were obtained from the hospital records and retrospectively evaluated. Patients with a severe cardiac or hepatic condition, respiratory failure, hemodynamic instability and serious systemic diseases, those at advanced stage of disease, patients with active gastrointestinal bleeding and those with known colorectal cancer were excluded from the study.
Patients' demographic data such as age and gender, smoking and alcohol consumption status, presence of alarm symptoms (weight loss, anemia, etc.) and family history were recorded. In addition, indications for colonoscopy, bowel cleanliness, pathologic ndings, presence of polyps, localizations of polyps, types, number, and size of polyps and presence of dysplasia were also recorded and analyzed.
Data obtained were analyzed in details. Colonoscopy procedures were performed by three separate gastroenterologists. All polyps detected during colonoscopy were excised and sent to the pathology laboratory for histopathological examination. Polyp specimens were xed in buffered formalin. Following staining hematoxylin eosin, one or two sections of 4 µm were examined depending on the polyp size.
The patients were divided into two groups as < 45 yo and ≥ 45 yo and data obtained were compared between these two groups.
Before the beginning of the study, necessary approval was received from the local ethics committee of our hospital.

Statistical Analysis
Statistical analysis of the data obtained in this study was performed using MedCalc Statistical Software version 12.7.7 (MedCalc Software bvba, Ostend, Belgium; http://www.melcalc.org; 2013) statistical software. Continuous variables are expressed with descriptive statistics mean and standard deviation values. Categorical variables were evaluated with Chi-Square test and Fisher's Exact test, when necessary and expressed as frequency (n) and percentage (%). P < 0.01 values were considered statistically signi cant.

Results:
A total of 1882 patients who underwent colonoscopy due to various reasons in the gastroenterology clinic of our hospital from 2017 through 2019 were included in the study. Of all patients, 990 (52.6%) were male and 892 (47.4%) were female. Demographic data of groups are given in Table 1. No signi cant difference was found between the groups in terms of age and genders (p > 0.05).
The rate of smoking was found as 13.9% in ≥ 45 yo group and 15% in < 45 yo group. The rate of alcohol consumption was found as 2.5% in the ≥ 45 yo and 3.9% in < 45 yo.
When data about colonoscopy procedures were examined; family history was found in 124 (6.6%) patients and 1870 (99.4%) patients underwent their rst colonoscopy. Table 2 shows bowel cleanness status, family history, number of colonoscopies and whether colonoscopy was completed ≥ 45 yo and < 45 yo patients.
Colonoscopy examination was normal in 834 (44.3%), while pathologic ndings were found in 1048 (55.7%) patients. When all patients were evaluated; the most common pathologic ndings were found as polyps (36.9%) and hemorrhoid (9.9%). Distribution of pathologic ndings is given in Table 3 for ≥ 45 yo < 45 yo patients.
The localization of pathologic ndings was found as rectum in 20% and sigmoid colon 7.8% of < 45 yo patients.
Colonic polyps were found in 36.9% of all patients included in this study. The rate of polyps was statistically signi cantly higher in ≥ 45 yo patients (43.3%) compared to < 45 yo patients (32.1%) (p < 0.001).
When the types of polyps detected during colonoscopy were examined; neoplastic polyps were found in 292 patients, non-neoplastic polyps in 488 patients, neoplastic polyps in 88 patients, hyperplastic polyps in 490 patients, tubular polyps in 6 patients highgrade polyps with dysplasia in 50 patients and polyps with dysplasia in 96 patients. No statistically signi cant difference was found between the two groups in terms of the rate of neoplastic polyps (p = 0.904). The rate of neoplastic polyps was statistically signi cantly higher in ≥ 45 yo group compared to < 45 yo group (p < 0.001). No statistically signi cant difference was found between the two groups in terms of the rate of non-neoplastic polyps (p = 0.309).
When the localizations of hyperplastic polyps were examined; the most common polyp localizations were rectum and sigmoid colon in all age groups. Tubular polyps were found in 6 < 45 yo patients, and the localizations were found as recturm, transverse colon and sigmoid colon. Serrated adenoma was found in two of these patients. Highgrade polyps with dysplasia were found in 16 < 45 yo patients. The rate of more than 3 polyps was found as 24.9% in ≥ 45 yo patients and 18.1% in < 45 yo patients. The difference between the groups was statistically signi cant (p = 0.002).
There was a statistically signi cant difference between the number of polyps found by three examiners in ≥ 45 yo and < 45 yo patients. Accordingly, the number of polyps found by the rst examiner was signi cantly higher compared to the second and third examiners (p < 0.001). There was a statistical signi cant in the numbers of neoplastic and hyperplastic polyps found by the three examiners. The rate of neoplastic and hyperplastic polyps detection was statistically signi cantly higher in the rst examiner compared to the other examiners (for all p < 0.001). Colonoscopy durations were evaluated by an independent nurse, and the mean colonoscopy duration was found to be longer in the rst examiner compared to the other examiners.
The numbers of polyps were evaluated according to the demographic data of the patients. No statistically signi cant difference was found between sexes in terms of the number of polyps (p = 0.088).
No signi cant difference was found in polyp numbers in terms of smoking and alcohol usage (p = 0.874 and p = 0.488, respectively). Similarly no signi cant difference was found in polyp numbers in terms of the presence of family history (p = 0.222). No signi cant difference was found between the patients with and without alarm symptoms in terms of polyp numbers (p = 0.306).
According to the results of ROC analysis performed in order to determine critical age that can pose a risk for the number of polyps, 29.5 years cut-off value created a risk for high polyp number with a sensitivity of 98.6% and a speci city of 91.4% (AUC=0.590, p=0.018) (Figure 3).

Discussion:
Colorectal cancer is becoming a life threat for young adults, because it is often diagnosed in an advanced stage. In this study, we investigated whether colonoscopy screening for colonic polyps should be made at an earlier age than reported in the current guidelines. Especially since neoplastic colon polyps has malignant potential, it is important to detect these polyps in a su ciently early stage, when removal of these polyps with a simple procedure can prevent the development of colorectal cancer.
Majority of the current international guidelines recommend onset age of colonoscopy screening as 50 years in persons with an average risk. This is based on the increase of CRC incidence about 50 years old. In 2009, the diagnosis of CRC was established in persons aged 50 years and over by 90% [18]. In a study by Chen et al., one colonoscopy screening in unscreened 50-65 years old population is expected to prevent mortality from CRC by 49% [19]. In the Canadian Task Force on Preventive Health Care (CTFPEC) guidelines (Canada), colonoscopy screening onset age is recommended as 50 years, allowing postponing of screening until 60 years old [20].
In 2018, US Preventive Services Task Force (USPSTF) updated onset age of colonoscopy screening as 45 years old in average-risk patients [15]. In Saudi Arabia, onset age of colonoscopy screening is recommended as 45 years old, because the diagnosis of CRC is made in 55 years old in Saudi women and 60 years old in Saudi men [21]. Studies in the literature have reported that the incidence of CRC is increasing in parallel with the increasing incidences of sedentary lifestyle, obesity and diabetes mellitus in young adult age group [22]. In 2013, 50310 people died from CRC in the USA, and it has been reported that thousands of these people could be potentially saved with colonoscopy screening an an earlier age [23].
In studies from various countries, the incidence of CRC in persons aged under 40 years has been reported.
The incidence of CRC under 40 years old was reported as 52% in a study from Pakistan [24], 39% in a study from India [25], 36% in a study from Iran [26], and 22% in a study from Korea [27]. In a study from Turkey, the incidence of early-onset CRC was reported as 20% in persons aged under 40 years [28]. On the other hand, the incidence of CRC under 50 years old was reported as 10% in a study from the USA [29], while this rate was reported as 5% in a study from the UK [30]. It is seen that the incidence of CRC in young adults is higher in developing countries compared to the developed countries. We attribute this to more common colonoscopy screening programs in developed countries.
The European Society of Gastrointestinal Endoscopy (ESGE) recommend to evaluate bowel cleanness during colonoscopy and appropriate or good cleanliness should be reported in at least 90% of screenings [31]. In our study bowel cleanliness, which is among the factors affecting the quality of colonoscopy, was near 100% in all patients.
In a recent study, it was reported that at least 1 polyp was detected in 34.3% of patients undergoing screening colonoscopy [32]. Similarly, in our study we found at least 1 polyp in 36.9% of the patients undergoing colonoscopy.
In our study, the most common indications of colonoscopy in < 45 yo patients were found as abdominal pain by 43%, changed bowel habits by 21%, rectal bleeding by 20%, and anemia by 10%. In a study by Mikaeel et al. investigating colorectal cancer in young adults, indications of colonoscopy in patients aged under 50 years were found as rectal bleeding by 51%, abdominal pain by 16%, anemia by 16%, and changed bowel habits by 13% [33]. We believe that although the indications in our study were similar to those of the mentioned study, the differences between the rates might be resulted from different sociocultural characteristics and lifestyle between the countries.
Based on the updated recommendations in recent guidelines for performing colonoscopy at an earlier age, we divided our patients into two groups as ≥ 45 yo and < 45 yo, and evaluated the data again between these two groups. The presence of polyps and neoplastic polyps was statistically signi cantly higher in ≥ 45 yo group compared to < 45 yo groups. However, colonoscopy was performed in 1078 patients < 45 yo (52.3%) and pathologic ndings were detected in 55.3% of these patients. In a study by Peñaloza et al. in 2017, a total of 411 colonoscopies were analyzed and pathologic ndings were observed by 44%. The highest rate of pathologic ndings were found in 41-50 years age group [34].
In our study, the most common localization of pathologic ndings was rectum (19.6%) followed by sigmoid colon (11.5%). In the study by Peñaloza the most common localization was found as sigmoid colon [34]. Given that serrated adenomas are more common on the right side, the rate of pathologic ndings detected in sigmoid colon was remarkable. In addition, recent studies have shown that hyperplastic polyps at the right side of the intestines may have malignant potential and should be closely followed-up [5]. In our study, more than 5% of the hyperplastic polyps were localized in the cecum, ascending colon and transverse colon at the right side.
In overall patients, no statistically signi cant difference was found between the patients with and without alarm symptoms (weight loss, anemia, etc.). This indicated that colonoscopy screening only in patients with alarm symptoms brought the possibility of missing polyps in the other patient groups. Therefore, we think that colonoscopy screening should be performed not only in patients with alarm symptoms, but also in all persons at a su ciently early age and should be repeated with certain periods.
In our study, there was a signi cant difference between the numbers of polyps detected by the three examiners. Accordingly, the rst examiner found a higher number of polyps than the other examiners. We believe this was resulted from experience of the examiner and longer colonoscopy durations.
A ROC analysis was performed in order to determine age cut-off value that creates a risk for a high number of polyps. As a result of this analysis, 29.5 years old age cut-off value was found to create a risk for a high number of polyps by a sensitivity of 98.6% and a speci city of 91.4%. We think that this result suggests the importance of colonoscopy screening at an earlier age.
In conclusion; given the increasing incidence of early-onset CRC worldwide colonoscopy screening should be performed in young adults without alarm symptoms who present with abdominal pain. Better outcomes would be obtained with colonoscopy screening in young adults at an earlier age in terms of the prevention of cancer with detection and excision of colon polyps at a precancerous stage. In addition, colonoscopy should be performed by experienced gastroenterologists. In our study, the more experienced endoscopist detected more polyps than the others. Experience of the endoscopist, bowel cleaning and duration of withdrawing fom the cecum affects the number of detected polyps. Furthermore this screening should be performed in a longer duration in order to avoid missing of polyps. Based on our results, we believe that onset age of screening for colon polyps should be 45 years as recommended by updated guidelines.

Declarations
Ethics approval and consent to participate: Before the beginning of the study, necessary approval was received from the local ethics committee of our hospital.   ROC analysis performed to determine age cut-off that creates a risk for high number of polyps.