Approach to the patient with colonic polyps

All topics are updated as new evidence becomes available and our peer review process is complete.

Literature review current through: Oct 2015. | This topic last updated: Feb 06, 2015.

INTRODUCTION — A polyp of the colon refers to a protuberance into the lumen from the normally flat colonic mucosa. Polyps are usually asymptomatic but may ulcerate and bleed, cause tenesmus if in the rectum, and, when very large, produce intestinal obstruction. Colonic polyps are usually classified as non-neoplastic, hamartomatous, neoplastic (adenomas and carcinomas), serrated (which can be neoplastic or non-neoplastic), and submucosal (which can be neoplastic or non-neoplastic).

The different types of benign polyps that develop sporadically and their management will be reviewed here. Colon cancer screening in the average risk individual, surveillance after colorectal cancer resection, and screening in individuals with colorectal polyposis syndromes are discussed separately. (See "Screening for colorectal cancer: Strategies in patients at average risk" and "Screening for colorectal cancer in patients with a family history of colorectal cancer" and "Surveillance after colorectal cancer resection" and "Clinical manifestations and diagnosis of familial adenomatous polyposis" and "Peutz-Jeghers syndrome: Epidemiology, clinical manifestations, and diagnosis".)

NON-NEOPLASTIC POLYPS — The non-neoplastic colonic polyps may be grouped into several distinct categories:

●Hyperplastic

●Mucosal

●Inflammatory pseudopolyps

●Submucosal, some of which may be neoplastic (eg, lipomatous, leiomyoma)

●Hamartomatous

 

Hyperplastic polyps — Hyperplastic polyps are the most common polyps traditionally included in the non-neoplastic category. Because of evidence suggesting that these polyps are part of the larger group of polyps known as serrated polyps and because some hyperplastic polyps appear to progress to more advanced lesions and cancer, we have included them in the section on serrated polyps below. (See 'Hyperplastic polyps' below.)

Mucosal polyps — Mucosal polyps are small (usually <5 mm) excrescences of tissue that endoscopically resemble the adjacent flat mucosa and histologically are normal mucosa. They have no clinical significance.

Inflammatory pseudopolyps — Inflammatory pseudopolyps are irregularly shaped islands of residual intact colonic mucosa that are the result of the mucosal ulceration and regeneration that occurs in inflammatory bowel disease (IBD). These polyps are typically multiple, often filiform and scattered throughout the colitic region of the colon. They may also be more isolated and semipedunculated in areas of more active recent inflammation, and have mucus adherent to their apices (picture 1).

When pseudopolyps are in clusters, they may be associated with surrounding dysplasia, and, in this circumstance, careful attention to biopsy of the region is important. However, their presence complicates the use of surveillance colonoscopy strategies in patients with IBD, not least because the larger, inflamed polyps are quite vascular. When typical, they are best left in situ. (See "Clinical manifestations, diagnosis, and prognosis of ulcerative colitis in adults".)

Submucosal polyps — A variety of submucosal lesions, including lymphoid aggregates, lipomas, leiomyomas, pneumatosis cystoid intestinalis, hemangiomas, fibromas, carcinoids, perineuromas/fibroblastic polyps, and metastatic lesions, may impart a polypoid appearance to the overlying mucosa. The most common of these, the lipoma, can be diagnosed endoscopically because of its yellow color and softness, assessed by probing the polyp with forceps and eliciting the "pillow sign" of indentation on gentle pressure. Perineuriomas are benign tumors comprised of perineurial cells and they occasionally present as colon polyps [1]. Fibroblastic polyps have identical clinical and histologic features and some have suggested, based upon immunohistochemical staining, that they represent the same entities [2,3]. These polyps are believed to be benign and surveillance is not recommended. Endoscopic ultrasound can be useful in defining the site of origin and for biopsy of submucosal lesions if the diagnosis is in doubt. (See "Peripheral nerve tumors", section on 'Perineurioma' and "Endoscopic ultrasound for the characterization of subepithelial lesions of the upper gastrointestinal tract".)

Hamartomatous polyps — Hamartomatous polyps are polyps that are made up of tissue elements normally found at that site, but which are growing in a disorganized mass. Hamartomatous polyps have traditionally been classified as non-neoplastic but some hamartomatous polyps develop dysplasia and lead to colorectal cancer.

Juvenile polyps — Juvenile polyps are hamartomatous lesions that consist of a lamina propria and dilated cystic glands rather than increased numbers of epithelial cells (picture 2). They are usually removed because of a high likelihood of bleeding. Juvenile polyps can be diagnosed at any age, although they are relatively more common in childhood.

Juvenile polyposis coli (JPC) is often defined as the occurrence of 10 or more juvenile polyps. Approximately one-third of cases of JPC have a history of similar lesions in at least one first-degree relative, and are called familial juvenile polyposis (FJP). A subset of juvenile polyposis families has germline mutations in the SMAD4 gene on chromosome 18q21.1, or in the gene BMPR1A [4-6]. SMAD4 (also known as DPC4) encodes a mediator in the transforming growth factor-beta signaling pathway. FJP is associated with an increased risk for the development of colorectal cancer and, in some families (most notably SMAD4 gene carriers), upper gastrointestinal hamartomatous polyps and gastric cancer. Some families with juvenile polyposis due to a SMAD4 mutation also have hereditary hemorrhagic telangiectasia, which needs to be taken into consideration during polypectomy [7]. (See "Peutz-Jeghers syndrome and juvenile polyposis: Screening and management of patients and families", section on 'Juvenile polyposis syndrome and familial juvenile polyposis'.)

Peutz-Jeghers polyps — The Peutz-Jeghers polyp is a hamartomatous lesion of glandular epithelium supported by smooth muscle cells that is contiguous with the muscularis mucosa (picture 3). It is almost always associated with the Peutz-Jeghers syndrome (PJS). The polyps are usually benign, but may grow progressively and produce symptoms or undergo malignant transformation. Patients with PJS are at increased risk of both gastrointestinal (gastric, small bowel, colon, pancreas) and nongastrointestinal cancers, especially including breast cancer and testicular cancer, with a cumulative overall cancer risk of about 50 percent by age 60. (See "Peutz-Jeghers syndrome: Epidemiology, clinical manifestations, and diagnosis".)

Cronkhite-Canada syndrome — Cronkhite-Canada syndrome is a rare, nonfamilial disorder of unknown etiology associated with alopecia, cutaneous hyperpigmentation, gastrointestinal polyposis, onychodystrophy (picture 4), diarrhea, weight loss, and abdominal pain [8-10]. The polyps are hamartomas and do not appear neoplastic pathologically. Characteristic features include myxoid expansion of the lamina propria and increased eosinophils in the polyps [11,12]. There is some evidence that the disorder may be immune-mediated, since it may respond to immunosuppressive therapy and, in some patients, immunostaining of the polyps for IgG4 is positive [13]. Five-year mortality rates as high as 55 percent have been reported with most deaths due to gastrointestinal bleeding, sepsis, and congestive heart failure [14]. Treatment has included nutritional support, glucocorticoids, azathioprine, acid suppression, and antibiotics, but no specific treatment has proven to be consistently effective [13,15,16].

Other syndromes — The PTEN hamartoma tumor syndrome, which is primarily comprised of the Cowden and Bannayan-Riley-Ruvalcaba syndromes, has also been associated with hamartomatous and other histologic types of polyps. These are discussed in detail, separately. (See "PTEN hamartoma tumor syndrome, including Cowden syndrome", section on 'Cowden syndrome' and "PTEN hamartoma tumor syndrome, including Cowden syndrome", section on 'Bannayan-Riley-Ruvalcaba syndrome'.)

NEOPLASTIC POLYPS — The common neoplastic colonic polyps include:

●Serrated polyps

●Adenomatous polyps

 

Serrated polyps — Serrated polyps are a heterogenous group of polyps with variable malignant potential. They include hyperplastic polyps, traditional serrated adenomas, and sessile serrated polyps (also known as sessile serrated adenomas) [17].

Hyperplastic polyps — Hyperplastic polyps are the most common non-neoplastic polyp in the colon; because there is some evidence that some hyperplastic polyps may be the precursor of sessile serrated polyps, which can progress to cancer. They are small nodules or polypoid lesions that may be indistinguishable grossly from adenomatous polyps (picture 5).

●Histology – Hyperplastic polyps are composed of normal cellular components, do not exhibit dysplasia, and have a characteristic serrated ("saw tooth") pattern when seen sectioned along the crypt axis (picture 6). Proliferation is mainly in the basal portion of the crypt of hyperplastic polyps, a feature that can be assessed on H and E stains, which are used by pathologists to distinguish hyperplastic polyps from adenomas [18]. Three histologic subtypes of hyperplastic polyps have been described (microvesicular, goblet cell, and mucin depleted), but the separation does not have substantial clinical meaning yet and the terms are not commonly used by pathologists.

 

●Location – They are typically located in the rectosigmoid and are less than 5 mm in size [19,20].

 

●Risk of cancer – Distal small hyperplastic polyps rarely, if ever, develop into colorectal cancers (CRCs), but large hyperplastic polyps are thought to be a precursor to sessile serrated polyps, which can progress to colorectal cancer.

 

●Surveillance – Patients with small (≤10 mm) hyperplastic polyps confined to the rectum and sigmoid colon should be advised to undergo their next screening/surveillance examination in 10 years. There are no data to directly support these screening intervals but these recommendations are based on evidence that small rectosigmoid hyperplastic polyps do not appear to increase the risk of colorectal cancer [21]. This surveillance approach is consistent with the 2012 update of the United States Multi-Society Task Force guidelines (table 1) [22].

 

Multiple studies have evaluated the risk of proximal neoplasms in patients found to have distal hyperplastic polyps [23-27]. A systematic review that included 18 studies estimated that 21 to 25 percent of patients found to have a distal hyperplastic polyp had a proximal neoplasm (including 4 to 5 percent with an advanced neoplasm) [26]. In the four studies in which a colonoscopy was performed irrespective of distal findings, the relative risk of any proximal neoplasia (advanced or not) was 1.3 (95% CI 0.9-1.8). Thus, it is not clear that the risk of proximal neoplasia is elevated in this setting and, even if it is, the magnitude of risk appears to be small. The general consensus is that small left-sided hyperplastic polyps are not a significant marker of colon cancer risk and that finding them on a sigmoidoscopy is not a routine indication for colonoscopy.

However, studies suggest that serrated lesions that are larger or proximal may be associated with an increased risk of conventional adenomas:

●In a study of 1371 patients who underwent screening colonoscopy with subsequent surveillance, patients with proximal serrated polyps without dysplasia (this includes sessile serrated polyps and hyperplastic polyps) were more likely to have synchronous advanced neoplasia than patients without proximal serrated polyps without dysplasia (17 versus 10 percent) and were more likely to have three or more small tubular adenomas detected (11 versus 5 percent) [28]. The presence of a proximal or a large serrated polyp without dysplasia was associated with an increased risk of having metachronous adenomas during surveillance.

 

●In a study of 10,199 patients undergoing first-time colonoscopy, the presence of large (≥10 mm) serrated polyps was associated with an increased risk of synchronous advanced neoplasia (odds ratio 4.01), and colorectal cancer (odds ratio 2.16) [29].

 

A limitation in generalizing these studies, however, is that they included not only hyperplastic polyps, but also sessile serrated polyps, and in the case of the second study, sessile serrated adenomas. (See 'Sessile serrated adenoma/polyps and traditional serrated adenomas' below.)

Serrated polyposis syndrome — Serrated polyposis syndrome (also known as hyperplastic polyposis syndrome [HPS]) refers to a condition characterized by multiple, large and/or proximal hyperplastic polyps [30,31]. (See 'Sessile serrated adenoma/polyps and traditional serrated adenomas' below.) The World Health Organization criteria for serrated polyposis syndrome require the presence of one or more of the following:

●At least five serrated polyps proximal to the sigmoid colon, of which two or more are ≥10 mm

●Any number of serrated polyps proximal to the sigmoid colon in an individual who has a first degree relative with serrated polyposis syndrome

●Greater than 20 serrated polyps of any size, distributed throughout the colon

 

Polyps in serrated polyposis syndrome may be large and flat and are often found along the crests of haustral folds. Mucus and residue frequently adhere to the polyps, a characteristic that may make it difficult to discriminate the polyp from the surrounding normal colonic mucosa.

Unlike patients with sporadic small, distal hyperplastic polyps, patients with serrated polyposis syndrome appear to be at increased risk of colorectal cancer [32].

Molecular studies have identified high levels of gene methylation of the polyps and the colonic mucosa with patterns of gene methylation that are uncommon, if not unique, to the syndrome [33]. Cancers that occur in serrated polyposis syndrome typically follow a pathway characterized by hypermethylation of numerous genes (so-called CpG Island Methylator Phenotype [CIMP high]), and have activating BRAF mutations (table 2) [34]. (See "Molecular genetics of colorectal cancer".) Reciprocally, Kras mutations are uncommon [35].

No germline mutation that causes serrated polyposis syndrome has yet been identified. It does not appear to have a strong association with mutations in the MUTYH or MBD4 genes [30]. Patients sometimes have sufficient numbers of both hyperplastic and adenomatous polyps to qualify for the criteria for both attenuated familial adenomatous polyposis and serrated polyposis syndrome, and in these cases a search for mutations in the APC gene and especially for MUTYH mutations is warranted [30].

A family history of CRC has been reported in 33 to 59 percent of patients with serrated polyposis syndrome [30,36,37]. In a retrospective cohort study of 1639 first and second degree relatives of patients with serrated polyposis recruited from genetics clinics, the incidence of CRC and pancreatic cancer was increased in first degree relatives compared with the general population (CRC SIR first degree relative 5.2, 95% CI 3.7-7.3, pancreatic cancer SIR first degree relative 3.6, 95% CI 1.7-9.2) [38].

Management strategies for patients with serrated polyposis syndrome and their families have not been well defined. Proximal polyps should be resected completely. Where there are many such flat polyps in the thin proximal colon, the risks of endoscopic resection (perforation, bleeding) should be carefully weighed against the risks and benefits of colectomy, giving consideration to the age and comorbidities of the patient, available colonoscopic expertise, the number and configuration of the polyps, and the definitive prophylaxis provided by surgery, including minimally invasive laparoscopic surgery in centers with such experience [32].

We recommend annual colonoscopy in patients who meet the clinical criteria for the serrated polyposis syndrome [17,22]. These recommendations are consistent with the United States Multi-Society Task Force on Colorectal Cancer [22]. However, these recommendations are supported by limited evidence and other experts have recommended that colonoscopy be performed every two years in patients with serrated polyposis syndrome [39].

We suggest screening colonoscopies for first degree relatives of individuals with serrated polyposis around age 40 (or 10 years earlier than the earliest age at diagnosis of the condition in the family) [17,30]. We suggest a surveillance interval of five years if no polyps are found.

Sessile serrated adenoma/polyps and traditional serrated adenomas — Sessile serrated adenomas/polyps (SSA/Ps) are more prevalent in the proximal colon and typically lack classic dysplasia. Foci of classic cytologic dysplasia do occur in SSA/Ps and this should be noted by the pathologist as a sessile serrated adenoma/polyp with cytologic dysplasia, as this lesion is thought to be an important step toward cancer progression. In contrast, traditional serrated adenomas (TSAs) are more prevalent in the rectosigmoid and have diffuse but often mild cytologic dysplasia. Classification systems for these polyps are still evolving. Histological interpretation of these serrated polyp types differs among pathologists [40-42]. The majority of studies suggest that TSAs and SSPs have significant malignant potential and are associated with subsequent development of metachronous polyps [28,29,40,43-50]. Some studies suggest that SSA/Ps have a higher potential to develop CRC compared with adenomatous polyps (tubular, tubulovillous, or villous) [51] and others have indicated a similar or lower potential [52-54].

SSA/Ps, particularly those with foci of classic histologic dysplasia, are considered the likely precursor lesions to sporadic MSI-H colon cancer because of the similarities in their molecular profiles including methylation of genes such as DNA repair genes MLH-1 and/or 0-6-methylguanine DNA methyltransferase (MGMT) [18,55,56]. Activation of the BRAF oncogene is also a feature of SSA/Ps,as well as many hyperplastic polyps and the better studied sporadic MSI-H cancers [57,58]. Patterns of cellular proliferation and differentiation including mucus expression can also help differentiate SSA/Ps from hyperplastic polyps [18]. (See "Molecular genetics of colorectal cancer".)

Large serrated polyps have been associated with the presence of synchronous colorectal cancer [29,46]. They may also contain areas of classic dysplasia and focal cancer. In one report, 37 percent of these serrated adenomas contained areas of "significant" dysplasia and 11 percent had foci of intramucosal carcinoma (high-grade dysplasia), highlighting their malignant potential and the need for pathologic scrutiny [59].

TSAs and SSA/Ps are managed clinically like adenomatous polyps (see 'Surveillance' below). However, due to their sessile nature and indistinct borders, special care is needed to ensure their complete removal endoscopically. There is molecular and clinical evidence that these lesions, either through being missed, incompletely removed, or through a more rapid progression from adenoma to cancer, disproportionally contribute to interval colorectal cancers [60].

Adenomatous polyps — Adenomatous polyps are the most common of the classically neoplastic polyps. About two-thirds of all colonic polyps are adenomas. Adenomas are by definition dysplastic and thus have malignant potential. Most colorectal cancers arise from adenomas, but only a small minority of adenomas progress to cancer (5 percent or less). Studies reporting the average age at presentation of patients with adenomatous polyps versus colorectal cancer suggest the time for development of adenomas to cancer is about 7 to 10 years. The risk of progression is higher for advanced adenomas [61].

Definitions

●An advanced adenoma is any adenoma with high-grade dysplasia, an adenoma that is >10 mm in size, or an adenoma with a villous component. Adenomas without villous components are also referred to as tubular adenomas.

 

●A synchronous adenoma is an adenoma that is diagnosed at the same time as an index colorectal neoplasm (a pathologically more advanced lesion). Thirty to fifty percent of colons with one adenoma will contain at least one other synchronous adenoma [62].

 

●A metachronous adenoma is an adenoma that is diagnosed at least six months after the diagnosis of a previous adenoma.

 

Epidemiology — The prevalence of adenomatous polyps is variable and a number of risk factors have been identified. Older age is a major risk factor for the development of colonic adenomas. Colonoscopic screening studies in asymptomatic people suggest that the prevalence of adenomas is about 25 to 30 percent at age 50 [61,63-66]. Autopsy studies have found rates as high as 50 percent by age 70 (figure 1) [67], compared with only 1 to 4 percent in those in their twenties or thirties [66]. Advancing age is also a risk factor for right-sided polyps [68]. An increased body mass index (BMI) is associated with an increased risk of colorectal adenomas. In a meta-analysis of 36 studies, the risk for colorectal adenomas increased by 19 percent for every 5-unit increase in BMI (summary relative risk 1.19; 95% CI 1.13-1.26) [69]. Abdominal obesity, measured by increasing abdominal visceral adipose tissue volume, may be a better predictor than BMI or waist circumference in both sexes [70]. Lack of physical activity is also a risk factor [71].

Adenomatous polyps are more common in men [65,72], and large adenomas (≥9 mm) may be more common in African-Americans than Caucasians [73]. In addition, African-Americans may have a higher risk of right-sided colonic adenomas [74] and present with cancer at a younger age (<50 years of age) than Caucasians [75]. (See "Screening for colorectal cancer: Strategies in patients at average risk".)

The prevalence of advanced colonic adenomas, which are associated with a higher risk of colon cancer, is much lower than that of non-advanced adenomas, particularly in older individuals. In a meta-analysis of average risk individuals, the prevalence of advanced adenomas was 3.8 percent in patients under age 65 and 8.2 percent in older patients, whereas it was similar in the two age groups for non-advanced adenomas (17.1 and 17.3 percent, respectively) [61].

Morphologic/endoscopic classification — Polyps are further categorized as sessile, pedunculated, flat, or depressed.

●Sessile – Sessile polyps are those in which the base is attached to the colon wall.

 

●Pedunculated – Pedunculated polyps are polyps in which a mucosal stalk is interposed between the polyp and the wall (picture 5). Small polyps (<5 mm, also known as "diminutive polyps") are rarely pedunculated.

 

●Flat – Flat polyps are defined as those having a height less than one-half the diameter of the lesion. It is important to appreciate that while adenomas are most commonly found within raised lesions, up to 27 to 36 percent are relatively "flat" [52,76-78].

Some studies have found that "flat" adenomas (by the definition given above) tend to be relatively advanced histologically for their size compared with polypoid lesions [76,77,79]. However, their natural history is poorly understood [78]. The National Polyp Study suggested that patients with one or more flat adenomas at an initial colonoscopy were not at greater risk for advanced adenomas at surveillance compared with those with polypoid adenomas only [77]. On the other hand, flat lesions can be difficult to detect on colonoscopy, raising concern that such lesions may contribute to false negative colonoscopic evaluations.

 

●Depressed – Depressed lesions appear to be particularly likely to harbor high-grade dysplasia or be malignant, even if small [6,79-83]. Up to 1 percent of adenomas are depressed [52,76-78]. Although flat and depressed adenomas were initially thought to be largely a phenomenon in Asia only, they have now been shown to occur more frequently than previously recognized in western populations [79,84-87].

The Paris classification of superficial neoplastic lesions categorized polypoid lesions of the GI tract into polypoid lesions that can be either pedunculated (Ip) or sessile (Is) and nonpolypoid lesions that can be slightly elevated (IIa), flat (IIb), or slightly depressed (IIc) or excavated (III) [88]. Such a classification scheme is useful for standardizing communication in that different terminology for such lesions has been used in different countries around the world. IIc and III lesions have a high incidence of invasive cancer regardless of size. With the advent of high magnification endoscopy and visual enhancement techniques such as chromoendoscopy and optical enhancement such as narrow band imaging, it has become possible for experts to reliably distinguish the superficial pit pattern and distinguish small adenomas from hyperplastic polyps (picture 7) [89,90]. Efforts are ongoing to determine if this can be translated into community practice and endoscopy training programs [91].

Pathologic classification

Based on the glandular architecture adenomas are characterized as tubular, villous, or a mixture of the two [24]. In addition, a class of polyps known as serrated adenomas has a distinctive pathology (see 'Sessile serrated adenoma/polyps and traditional serrated adenomas' above).

●Tubular – Tubular adenomas account for more than 80 percent of colonic adenomas. They are characterized by a network of branching adenomatous epithelium (picture 8A-B). To be classified as tubular, the adenoma should have a tubular component of at least 75 percent.

 

●Villous – Villous adenomas account for 5 to 15 percent of adenomas (image 1 and picture 9). They are characterized by glands that are long and extend straight down from the surface to the center of the polyp (picture 10). To be classified as villous, the adenoma should have a villous component of at least 75 percent.

 

●Tubulovillous – Tubulovillous adenomas having 26 to 75 percent villous component account for 5 to 15 percent of adenomas.

 

All adenomas are dysplastic. Based on the degree of dysplasia, polyps are classified as having high- or low-grade dysplasia. Those with high-grade dysplasia have a higher risk of progression to cancer. Polyps with high-grade dysplasia do not contain invasive malignancy, which is defined by a breach of the muscularis mucosa by neoplastic cells. Because there are no lymphatic vessels in the lamina propria, they are not associated with metastasis. High-grade dysplasia represents an intermediate step in the evolution from a low-grade adenomatous polyp to cancer and, in some studies, is a significant risk factor for subsequent (metachronous) colorectal malignancy in the individual after its removal [92]. It should be recognized that there is substantial interobserver variability in the pathologic distinction of adenoma types particularly for tubulovillous and villous adenomas and the degree of dysplasia [93,94]. The terms "carcinoma in situ" or "intramucosal adenocarcinoma" should both be described as high-grade dysplasia [24,95].

The proliferative zone in adenomas predominates in the upper part of the crypt, although sometimes there is a diffuse pattern [18]. This is also assessable on H and E stained sections, assisting the pathologist in diagnosis.

Clinical presentation and natural history — Adenomas are generally asymptomatic and are most often detected by colon cancer screening tests. Small adenomas do not typically bleed, rendering fecal occult blood testing (FOBT) an insensitive screening method. Advanced adenomas (see below) are more likely to bleed and cause a positive fecal occult blood test, especially with immunochemical testing [96].

The growth rates of adenomatous polyps are variable and do not follow a consistent linear trend. The majority of small polyps exhibit minimal growth (averaging 0.5 mm/year). Prospective observations suggest that complete regression is uncommon [97].

When considering polyp characteristics, the risk that a particular polyp morphology will be associated with malignancy in that polyp at that time should be distinguished from the same characteristics being used as a risk factor for metachronous adenomas or cancers.

Risk factors for high-grade dysplasia and cancer — Villous histology, increasing polyp size, and high-grade dysplasia are risk factors for focal cancer within an individual adenoma [24]. Number and size are the most consistent risk factors for metachronous adenomas including histologically advanced adenomas and cancer.

●Polyp size – Adenomatous polyps >1 cm in diameter are a risk factor for containing colorectal cancer and are also a risk factor for metachronous cancer development [98]. The majority of adenomas (60 to 75 percent) are less than 1 cm at endoscopy [24,99]. These results are supported by reports which have found no increased risk of cancer among patients with small polyps treated by fulguration without subsequent surveillance [100], but up to a fourfold increase in cancer risk during follow-up of patients with polyps >1 cm in size [101]. One report found that patients with only one or two tubular adenomas less than 1 cm had a low risk of colonic neoplasia after five years of follow-up [102]. A study of pooled data from eight prospective studies reported that the absolute risk of metachronous advanced adenomas approached 20 percent in patients whose largest baseline adenoma was 20 mm in size or greater [103].

 

●Size versus villous histology – The proportion of adenomas showing advanced histologic features (high-grade dysplasia or >25 percent villous histology) increases from approximately 1 to 2 percent in small adenomas (<5 mm) to 7 to 12 percent for medium-sized adenomas (5 to 10 mm) and 20 to 30 percent for large adenomas (>1 cm) [98,104,105].

Adenomatous polyps with >25 percent villous histology are a risk factor for developing metachronous colorectal cancer [98]. In some studies, villosity also predicts metachronous advanced adenomas, although villosity is often closely linked with size and is not independently predictive in all studies.

 

●High-grade dysplasia – Adenomas with high-grade dysplasia often coexist with areas of invasive cancer in the polyp. Older age is also associated with high-grade dysplasia within an adenoma, independent of size and histology [24].

Adenomatous polyps with high-grade dysplasia and adenomatous polyps with invasive cancer (picture 10) are risk factors for developing metachronous colorectal cancer [98]. However, as size and levels of dysplasia are closely linked, most multivariate analyses are unable to separate high-grade dysplasia from size in predicting metachronous risk.

 

●Number of polyps – Numbers of adenomas at colonoscopy and cumulatively over a lifetime is the most consistent risk factor for metachronous colorectal cancer. Studies suggest that patients with a limited number (one or two) of small, tubular, adenomatous polyps removed do not have a significantly increased metachronous colorectal cancer risk. In contrast, the presence of one or more advanced adenomas predicts a higher rate of both any and advanced metachronous adenomas [92]. The risk of metachronous colon cancer (eg, diagnosed six months after removal of an index cancer or polyp) increases with the number of advanced adenomas that are present [98]. In one report, for example, patients with multiple large or villous rectosigmoid adenomas had a 6.6-fold higher risk of subsequent malignancy than the control population [92]. Another report found that large polyp size (>1 cm) and proximally located adenomas were independent predictors of recurrent high-risk adenomas [106]. A study of pooled data from eight prospective studies, proposed that absolute risks of metachronous advanced adenomas, colorectal cancer, and their combination (advanced colorectal neoplasia) within three to five years is higher (24 percent) in patients with >4 adenomas at baseline [103]. The risk for metachronous advanced adenomas increased with the number of adenomas at baseline and was 9, 13, 15, 20, and 24 percent for 1, 2, 3, 4, or ≥5 adenomas at baseline, respectively.

 

Management — The eradication of colonic adenomas is important for minimizing cancer risk and mortality. This is supported by the observation of a decrease in the incidence of colorectal cancer in the United States with the introduction of widespread screening [107].

Although the benefit of removing small polyps was not specifically addressed, the National Polyp Study reported that endoscopic removal of all adenomas was associated with a 76 to 90 percent decrease in cancer incidence when compared with three separate historic reference groups (figure 2) [108]. In addition, the National Polyp Study found that during long-term follow-up (up to 23 years), patients who had adenomas removed had an estimated 53 percent reduction in mortality due to colon cancer compared with the expected number of deaths in the general population (standardized incidence-based mortality ratio 0.47, 95% CI 0.26-0.80) [109].  (See "Colorectal cancer: Epidemiology, risk factors, and protective factors".)

However, some studies of patients with adenomas in follow-up have not demonstrated reductions in cancer risk when compared with average-risk populations, although their initial higher-risk status was likely reduced to average-risk following polypectomy [110]. In a report from three adenoma trials, colorectal cancer was diagnosed in 19 of the 2915 patients who had removal of at least one adenoma and were deemed free of remaining lesions over a mean follow-up of 3.7 years (incidence of 1.74 cancers/1000 person-years) [111].

A randomized controlled trial of screening for colon cancer through fecal occult blood testing not only demonstrated a reduced mortality from colon cancer (principally through detection of early stage and curable cancers) but also, at a later time, a reduction in the incidence of colon cancer [112]. In addition, in a population-based, case-controlled study, colonoscopy was associated with fewer deaths from colorectal cancer [113]. This association was primarily limited to deaths from cancer developing in the left side of the colon.

In a multicenter randomized controlled trial, once-only flexible sigmoidoscopy screening showed not only a reduction in mortality from distal bowel cancer (hazard ratio 0.50, 95% CI 0.42-0.59) and total colorectal cancer mortality (hazard ratio 0.69, 95% CI 0.59-0.82) after median follow-up of 11.2 years, but also a reduction in colorectal cancer incidence (through adenoma removal) (hazard ratio 0.77, 95% CI 0.70-0.84), strongly supporting the adenoma-cancer sequence as a major pathway to colorectal cancer [114]. These findings have been confirmed in the United States Prostate, Lung, Cancer, and Ovarian (PLCO) trial [115]. These findings are attributed to the detection and removal of adenomatous polyps through a screening program.

Detection — Colonoscopy is considered to be the optimal examination for the detection of adenomatous polyps, particularly in view of the ability to provide therapeutic polypectomy in conjunction with diagnosis. The superior sensitivity of colonoscopy over that of double-contrast barium enema and computed tomographic colonography is particularly evident when diminutive or flat polyps are considered, since they are unable to visualize the majority of these small adenomas [116,117]. In addition, an optimally performed double-contrast barium enema and fecal DNA test detect only approximately one-half of adenomas 5 mm or more in size that are evident on colonoscopy [118]. (See "Tests for screening for colorectal cancer: Stool tests, radiologic imaging and endoscopy" and "Overview of computed tomographic colonography".)

Colonoscopy is not a perfect screening test for detection of adenomas. One report found that the colonoscopic miss rate determined by two same-day endoscopic examinations in 183 patients was 27 percent for adenomas <5 mm, 13 percent for those 6 to 9 mm, and 6 percent for adenomas >1 cm [119]. Studies performing both CTC and colonoscopy estimate that the miss rate for colonoscopy is higher (12 percent for polyps ≥10 mm) than those found in the tandem colonoscopy trials [120]. Multiple factors likely contribute to missed polyps at colonoscopy including the quality of the prep, the training and experience of the endoscopist, and perhaps even the degree of fatigue of the endoscopist. The amount of time taken by colonoscopists during withdrawal of the colonoscope has emerged as a powerful predictor of adenoma detection rate in some studies [121]. Interval or missed cancer rates have been linked inversely to adenoma detection rates at screening colonoscopy [122,123].

In general, evidence suggests that protection afforded by colonoscopy for right-sided cancer development is significantly less than left-sided cancer in a range of settings, raising questions of the quality of colonoscopy as it relates particularly to the right colon (poor right-sided preps, incomplete colonoscopy, anatomical configurations compromising visibility, or differential biology for right-sided lesions) [113,124]. This behoves endoscopists to pay particular attention to the right colon in their practices both at initial and at follow-up examinations.

Enhanced optical technologies (chromoendoscopy, magnification endoscopy, narrow band imaging optical colonoscopy, confocal endomicroscopy) have been used to increase colorectal cancer colonoscopy detection but their ultimate value in screening for colorectal cancer has yet to be determined. Other alternative approaches to polyp detection include CT colonography and colon capsule endoscopy. (See "Tests for screening for colorectal cancer: Stool tests, radiologic imaging and endoscopy" and "Chromoendoscopy" and "Magnification endoscopy" and "Confocal laser endomicroscopy and endocytoscopy".)

Polypectomy — A variety of techniques are used during colonoscopy to remove polyps:

●Cold biopsy forceps

●Hot (monopolar) biopsy forceps

●Cold snare excision

●Standard snare excision (with monopolar cautery)

●Simple fulguration, including with argon plasma coagulation

●Piecemeal excision

●Saline assisted endoscopic mucosal resection

●Endoscopic submucosal dissection

 

Snare excision is commonly used for polyps >5 to 10 mm and offers the most complete removal of adenomatous tissue. Diminutive lesions (<5 mm) can be readily removed with either biopsy forceps or by snare polypectomy. Some advocate the use of cold forceps biopsy or cold snare excision for small polyps in the cecum and ascending colon because of the higher complication rate noted with hot biopsy in these areas [19].

Incomplete polyp resection is thought to be a substantial contributor to interval colorectal cancer. High rates of incomplete resection of small polyps have been reported in studies of polypectomy by hot or cold biopsy (17 and 29 percent, respectively) [125,126]. The rates of incomplete resection appear to be higher for sessile lesions and increases with polyp size. In one study, the rate of incomplete resection was significantly higher for polyps 10 to 20 mm as compared with polyps 5 to 9 mm (17 versus 7 percent) and for sessile serrated polyps as compared with conventional adenomas (31 versus 7 percent) [127].

●Large polyps – The optimal method of colonoscopic removal of large polyps (>10 mm) varies with the type of polyp. Large pedunculated polyps readily lend themselves to snare removal.

 

●Sessile polyps – In comparison, large sessile adenomas (>2 to 3 cm) often require piecemeal resection. The presence of a large sessile adenoma is predictive of synchronous advanced neoplasia, making it particularly important that the colon is carefully inspected [43].

 

Complete and safe removal of larger sessile lesions may be aided by the saline (or other agents that are retained submucosally for longer periods [eg, succinylated gelatin]) lift technique, which prepares the polyp for resection by first raising it from the underlying muscularis propria by injecting saline (sometimes tinged with methylene blue to highlight the planes) into the submucosa under the polyp. Vigilant follow-up, with initial colonoscopy intervals of three to six months, is warranted for these larger sessile lesions since 28 percent recur or persist, and 17 percent later degenerate into carcinoma despite apparently complete removal [128]. Endoscopic submucosal dissection allows en bloc dissection of polyps but requires specialized techniques, accessories, and expertise. (See "Endoscopic removal of large colon polyps".)

When large polyps are present in the rectum, consideration should be given to referring the patient for a transanal endoscopic microsurgery (TEMS) procedure, as there is a greater likelihood of removing the specimen in one piece, considerably enhancing subsequent decision making if cancer is found in the specimen. If resection is piecemeal, adequacy of removal may be impossible to determine pathologically, and the patient potentially could be committed to an unnecessary abdominoperineal resection. (See "Endoscopic removal of large colon polyps".)

Approach to management — If a polyp is detected by barium enema or computed tomographic colonography, colonoscopy is recommended to establish the histology, remove the polyp, and search for synchronous lesions [129,130].

●Small polyps found at colonoscopy should be completely removed by biopsy and/or by electrocautery techniques. If numerous (>20) small polyps are encountered, representative biopsies of a substantial sample of the polyp number should be obtained, as syndromic diagnosis (and therefore the gene to be analyzed) is dependent on this. Sampling should be generous, if not complete.

 

●The number of adenomas found at colonoscopy is a dominant factor in metachronous adenoma and cancer risk. Surveillance guidelines therefore take into account the number of polyps found at the most recent colonoscopy as well as their size and histologic features. (See 'Surveillance' below.)

 

●After colonoscopic removal of a large (>2 cm) sessile polyp, follow-up colonoscopy should usually be performed in three to four months (when edema has resolved) if there is concern that any portion of the adenoma has not been completely removed. If residual polyp tissue remains, it should be excised and the colonoscopy repeated in another three to six months. Surgical removal should be considered if complete resection cannot be accomplished after two to three colonoscopic attempts.

 

●High-grade dysplasia is considered noninvasive if these lesions are found within a resected polyp; they require no further therapy if the resection margins are free of neoplastic tissue [131].

 

Some adenomas with early invasive cancer also have a low (<2 percent) risk of lymphatic metastasis [132], and polypectomy alone is usually adequate for such lesions. No further treatment is necessary if specific pathologic criteria are met [129]:

●The polyp is considered to be completely excised by the endoscopist and is submitted in toto for pathological examination.

 

●The polyp is fixed and sectioned so that it is possible to accurately determine the depth of invasion, grade of differentiation, and completeness of excision of the carcinoma.

 

●The cancer is not poorly differentiated.

 

●There is no vascular or lymphatic involvement.

 

●The margin of the excision is not involved. Invasion of the stalk of pedunculated polyp, by itself, is not an unfavorable prognostic finding, as long as the cancer does not extend to the margin of stalk resection.

 

When all of these low-risk criteria are not met, the decision to proceed to surgical resection needs to be individualized, taking into account the age and comorbidity of the patient. Surgical resection is generally recommended for any invasive cancer in a sessile adenoma where the patient is otherwise well. (See "Overview of the management of primary colon cancer".)

Surveillance — Multiple studies have examined the risk of advanced colorectal neoplasia in patients with previously endoscopically resected colorectal adenomas to quantify their risk of developing a subsequent advanced adenoma or cancer [103,133]. A pooled analysis of eight prospective studies (with a total of 9167 subjects) estimated that the risk of advanced colorectal neoplasia was 12 percent during a median follow-up of four years; 58 patients (0.6 percent) developed invasive cancer [103]. The strongest risk factors were advanced neoplasia in the initial polypectomy, older age, and the number and size of prior adenomas. In a subsequent study that included 40,826 patients from a Norwegian population-based registry who had undergone resection of a colorectal adenoma during a median follow-up of 7.7 years, 1273 patients were diagnosed with colorectal cancer [134]. As compared with the general population, individuals who had undergone resection of one tubular adenoma of any size had a lower colorectal cancer mortality and those with a two or more adenomas, high-grade dysplasia, or villous histology had a higher colorectal cancer mortality (standardized mortality ratio 0.75, 95% CI 0.63-0.88 and 1.16, 95% CI 1.02-1.31, respectively).

The risk of advanced colorectal neoplasia in patients with previously endoscopically resected colorectal adenomas is the basis for the 2012 United States Multi-society Task Force (MSTF) on colorectal cancer guidelines for colonoscopy surveillance after screening polypectomy. These guidelines were endorsed by the American Gastroenterological Association, the American Society for Gastrointestinal Endoscopy, the American College of Gastroenterology, the American College of Radiology, and the American Cancer Society (table 1) [22].

●Low-risk adenomas – If only one or two small (≤10 mm) tubular adenomas are found on baseline colonoscopy, the first surveillance colonoscopy should be performed in 5 to 10 years [22]. If no adenomas are found on the first surveillance colonoscopy, a second surveillance colonoscopy should be performed in 10 years. However, if a low-risk adenoma is detected, the second surveillance colonoscopy should be performed at five years (table 3). 

In contrast to the United States MSTF guidelines, the European Society of Gastrointestinal Endoscopy (ESGE), the European guidelines for quality assurance in colorectal cancer screening and diagnosis, and Australian guidelines recommend that patients with low-risk adenomas at baseline colonoscopy participate in existing national screening programs 10 years after the index colonoscopy [135-137]. If no screening program is available, the ESGE recommends colonoscopy 10 years after the index colonoscopy.

 

●Advanced adenomas – Individuals with advanced adenomas (≥10 mm, villous histology or high-grade dysplasia) at baseline colonoscopy should undergo a first surveillance colonoscopy in three years [22]. Individuals with an advanced adenoma or between 3 and 10 adenomas on their first surveillance colonoscopy should undergo their next surveillance colonoscopy in three years (table 3). The Australian guidelines advise these patients be re-scoped at 12 months because of the high risk of missing adenomas where there is such multiplicity.

 

Results from the National Polyp Study suggested that a three-year interval for surveillance colonoscopy is safe and cost-effective for the majority of patients with advanced adenomatous polyps [108]. Longer intervals can be instituted in patients with low-risk adenomas [92].

 

It should be recognized, however, that the ability to discriminate the risk of metachronous adenomas based solely upon histologic features is limited. In one report, for example, the probability of advanced adenoma recurrence within four years was 9 percent in those with a high-risk adenoma at baseline compared with 5 percent in those with low-risk adenomas at baseline [138].

 

Among patients who had an advanced adenoma removed on a previous colonoscopy, a negative first repeat colonoscopy does not necessarily imply a good prognosis. This was illustrated in a study of 564 patients who had a colonic adenoma removed on colonoscopy followed by two subsequent colonoscopies at approximately three-year intervals [139]. Among patients in whom the first follow-up colonoscopy was negative, the findings on the initial positive colonoscopy had prognostic value for the second follow-up colonoscopy. The probability of high-risk findings was significantly increased in patients with a high-risk compared to low-risk initial adenoma (12 versus 5 percent).

 

●Multiple adenomas – The United States MSTF guidelines recommend that patients with between 3 and 10 adenomas on their first surveillance colonoscopy should undergo their next surveillance colonoscopy in three years (table 3) and that patients with more than 10 adenomas should be evaluated for a hereditary colorectal cancer syndrome and have surveillance colonoscopy in less than three years [22]. The European guidelines for quality assurance in colorectal cancer screening and diagnosis identifies patients with five or more adenomas (like the Australian guidelines) and recommends that a surveillance colonoscopy should be undertaken within 12 months to check for missed synchronous lesions [135-137]. (See "Familial adenomatous polyposis: Screening and management of patients and families", section on 'Attenuated FAP' and "Clinical manifestations and diagnosis of familial adenomatous polyposis", section on 'Diagnosis'.)

 

●Serrated adenomas/polyp (SSA/P) – Patients with SSA/P that are <10 mm and no dysplasia should undergo a repeat colonoscopy in five years. Studies have demonstrated that proximal and large SSA/P (≥10 mm) are markers of a synchronous advanced adenoma [28,29]. Thus, individuals with SSA/P ≥10 mm and a SSA/P with dysplasia should undergo a repeat colonoscopy at three years. These recommendations for surveillance are based solely on expert opinion but are consistent with United States multi-society guidelines [22].

Other expert consensus recommendations have suggested earlier colonoscopic follow-up (one- to three-year interval) in individuals with two or more SSA/P larger than 10 mm and in those with any SSA/P with cytologic dysplasia [17]. However, prospective data to support the screening/surveillance intervals are lacking. (See 'Sessile serrated adenoma/polyps and traditional serrated adenomas' above and 'Serrated polyposis syndrome' above.)

 

The risk of postprocedure hospitalization and the impact of comorbidity should be taken into account when assessing the relative risks and benefits of periodic surveillance colonoscopy in patients with advanced age or comorbidities. In a retrospective cohort study of 27,763 patients ≥50 years of age (4834 ≥75 years of age) undergoing surveillance colonoscopy for a history of CRC or adenomatous polyps at a single integrated California health system, age >75 years and a Charlson comorbidity score of 2 or higher were independently associated with a higher risk of postprocedure hospitalization (adjusted OR 1.28, 95% CI 1.07-1.53 and 2.54, 95% CI 2.06-3.14 (table 4)) [140].

Prevention — Attempts to prevent colorectal adenomatous polyps through diet and chemoprevention have been evaluated.

●Diet – A diet that is low in fat and high in fruits, vegetables, and fiber has been recommended, although based upon limited data [129]. There may be advantages with cruciferous vegetables and unprocessed forms of cereal fiber. (See "Cancer prevention", section on 'Fiber'.)

 

●Also suggested are maintenance of normal body weight through regular exercise and caloric restriction, and avoidance of smoking and excessive alcohol use, especially beer.

 

●COX-2 agents – Large trials of the COX-2 agents, celecoxib and rofecoxib, demonstrated significant reductions in advanced and metachronous adenomas, particularly at high dosages of the agents (eg, celecoxib 800 mg daily) [141-143]. However, vascular events of strokes and myocardial infarctions were significantly higher in the groups randomized to the high doses (including rofecoxib 25 mg), effectively preventing their use in chemoprevention [144].

 

●Aspirin – Aspirin also reduces the incidence of metachronous adenomas and probably cancer [145]. The role of aspirin in cancer prevention is discussed in detail separately. (See "NSAIDs (including aspirin): Role in prevention of colorectal cancer"and "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Screening and management".)

 

●Ursodeoxycholic acid – A controlled trial of ursodeoxycholic acid (a synthetic bile salt) did not find a statistically significant reduction in metachronous colorectal adenomas overall, although there was a significant reduction in the risk of adenomas with high-grade dysplasia [146].

 

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient information: Colon polyps (The Basics)" and "Patient information: Colonoscopy (The Basics)" and "Patient information: Familial adenomatous polyposis (The Basics)")

 

●Beyond the Basics topics (see "Patient information: Colon polyps (Beyond the Basics)" and "Patient information: Colonoscopy (Beyond the Basics)" and "Patient information: Flexible sigmoidoscopy (Beyond the Basics)")

 

SUMMARY AND RECOMMENDATIONS

●A polyp of the colon refers to a protuberance into the lumen from the normally flat colonic mucosa. Polyps are usually asymptomatic but may ulcerate and bleed, cause tenesmus if in the rectum, and, when very large, produce intestinal obstruction. (See 'Introduction' above.)

 

●Colonic polyps are usually classified as non-neoplastic, hamartomatous, neoplastic (adenomas and carcinomas), serrated (which can be neoplastic or non-neoplastic), and submucosal (which can be neoplastic or non-neoplastic). (See 'Introduction'above.)

 

●Hyperplastic polyps are the most common non-neoplastic polyps in the colon. Distal small hyperplastic polyps rarely, if ever, develop into colorectal cancers. (See 'Hyperplastic polyps' above.)

 

●Serrated polyposis syndrome (formerly known as hyperplastic polyposis syndrome [HPS]) refers to a condition characterized by multiple, large and/or proximal serrated polyps. Patients with serrated polyposis syndrome appear to be at increased risk of colorectal cancer. (See 'Serrated polyposis syndrome' above.)

 

●Serrated polyps are a heterogenous group of polyps with variable malignant potential. They include hyperplastic polyps, traditional serrated adenomas, and sessile serrated polyps (also known as sessile serrated adenomas). (See 'Serrated polyps'above.)

 

●Hamartomatous polyps are polyps that are made up of tissue elements normally found at that site, but which are growing in a disorganized mass. Juvenile polyps, Peutz-Jeghers polyps, and Cronkhite-Canada syndrome, are hamartomatous polyps. Patients with juvenile polyposis coli and Peutz-Jeghers syndrome are at increased risk for cancer. (See 'Hamartomatous polyps' above.)

 

●Adenomatous polyps are neoplastic polyps. About two-thirds of all colonic polyps are adenomas. Adenomas are by definition dysplastic and thus have malignant potential. (See 'Adenomatous polyps' above.)

 

●Adenomatous polyps are categorized morphologically (sessile, pedunculated, flat, or depressed) and pathologically (high- or low-grade dysplasia as well as tubular, villous, and tubulovillous). (See 'Morphologic/endoscopic classification' above and 'Pathologic classification' above.) The Paris classification has gained widespread acceptance for describing the morphology of polyps as it has a relationship with invasive cancer in the polyp.

 

●Clinically, adenomas are generally asymptomatic and are most often detected by colon cancer screening tests. (See 'Clinical presentation and natural history' above.)

 

●Villous histology, increasing polyp size, and high-grade dysplasia are risk factors for focal cancer within an individual adenoma. (See 'Risk factors for high-grade dysplasia and cancer' above.)

 

●If a polyp is detected, we recommend colonoscopy to establish the histology, remove the polyp, and search for synchronous lesions (Grade 1A). (See 'Management' above.)

 

●Recommendations for subsequent treatment and surveillance depend on the number and pathologic features of the polyp(s) (table 1). (See 'Approach to management' above and 'Surveillance' above.)

 

Use of UpToDate is subject to the Subscription and License Agreement.