Abstract
Objective
To examine whether BMI is independently related to colorectal adenomas and hyperplastic polyps.
Methods
We conducted a cross-sectional study among 1,420 asymptomatic women aged 40–79 years who had undergone complete colonoscopy. Logistic regression was used to estimate the odds ratios (OR) and the corresponding 95% confidence intervals (CI) of adenomas and hyperplastic polyps.
Results
We identified 953 women (67.1%) with no polyps, 292 (20.6%) with adenomas, and 175 (12.3%) with hyperplastic polyps. Among those with polyps, 75 women (5.3% of total women) were classified as having both adenomas and hyperplastic polyps. After adjusting for potential risk factors for colorectal cancer, BMI was related to increased risk of adenomas (OR comparing obese to normal weight women=1.57; 95% CI=1.07–2.29). Further, BMI was associated with enhanced risk of hyperplastic polyps (OR=3.76; 95% CI=2.35–6.01) and the combination of adenomas and hyperplastic polyps (OR=2.84; 95% CI=1.41–5.72).
Conclusions
Excess body mass is positively related to colorectal adenomas and hyperplastic polyps, particularly when both kinds of polyps are present in combination. Future studies should continue to delineate possible differences in potential risk factors between colorectal adenomas and hyperplastic polyps. Such work should help further elucidate the possible causes of colorectal cancer.
Keywords: Adiposity, obesity, colorectal adenoma, hyperplastic polyps, women
Introduction
Convincing epidemiologic evidence shows that greater body mass index (BMI) is related to increased risk of colon cancer [1]. Given an apparent adverse effect of adiposity on colon cancer, this factor may plausibly enhance the risk of developing adenomatous polyps, a known precursor lesion of colorectal cancer [2] and an established surrogate endpoint for colorectal cancer in mechanistic studies [3] and clinical trials [4].
Because adenomas represent the initial manifestation of the adenoma-dysplasia-carcinoma sequence [2], understanding the association between potential risk factors for this first stage of carcinogenesis is crucial for the primary prevention of colorectal cancer. To help clarify this issue, we conducted a cross-sectional of study of BMI in relation to adenomas overall and adenomas defined by stage, multiplicity, and anatomic location. In addition, we considered a separate evaluation of BMI in relation to hyperplastic polyps as a distinct endpoint because certain forms of hyperplastic polyps, such as serrated or mixed type are thought to represent precursor lesions for colorectal malignancy [5]. Hyperplastic polyps may play a role in colorectal carcinogenesis through biologic mechanisms that involve increased levels of insulin-like growth factors and insulin [5].
Our study differs from most previous investigations in presenting relations stratified by four groups of endpoints: cases with any adenomatous polyps; cases with adenomatous polyps after excluding those with hyperplastic polyps; cases with hyperplastic polyps after excluding those with adenomatous polyps; and cases with the combination of both types of polyps. We gave particular attention to the avoidance of bias and confounding by studying a population of asymptomatic women enrolled for complete screening colonoscopy, using information on body mass collected prior to the diagnosis of colorectal polyps, and controlling for a broad range of potential confounding variables.
Subjects and Methods
Study participants
The current study was conducted between July 1, 1999 and December 31, 2002 among women referred for colorectal cancer screening at four regional military centers (Bethesda, MD; Washington, DC; San Diego, CA; and Portsmouth, VA). The study enrolled asymptomatic women aged 50–79 years who were at average risk of developing colorectal cancer and asymptomatic women aged 40–79 years who had a family history of colorectal cancer. To make certain that women were asymptomatic and at average risk for colorectal neoplasia, women were excluded if they had any of the following conditions: personal history of adenomas, colorectal cancer, inflammatory bowel disease, or hereditary polyposis syndromes; rectal bleeding or hematochezia within 12 months before referral; iron-deficient anemia in the preceding 6 months; unintentional weight loss of greater than 10 pounds in the previous 6 months; or normal findings on colonoscopy or barium enema in the past 10 years or normal findings on flexible sigmoidoscopy in the past 5 years. The latter cutoffs were chosen because they correspond to the recommended time intervals for colorectal cancer screening among individuals aged 50 years or older who are at average risk for developing colorectal cancer. In addition, all participants had a blood-cell count, ferritin measurement, or fecal occult blood test completed within 6 months before referral. Among the 1483 women who met the eligibility criteria and participated in the study, we analyzed the 1420 subjects (95.8%) who had undergone complete colonoscopy to the cecum and had complete information on weight and height. Written informed consent was obtained from all study subjects. This study was approved by the institutional review board at each participating institution. Further details of the study have been reported previously [6, 7].
Identification of cases of colorectal adenoma and hyperplastic polyps
Over 99% of colonoscopic examinations were performed by gastroenterologists or colorectal surgeons. The anatomic location of all polyps was determined based on the depth of the colonoscope and anatomical landmarks. We defined proximal colorectal polyps as those located in the cecum, ascending colon, hepatic flexure, transverse colon, and splenic flexure; we defined distal colorectal polyps as those located in the descending colon and sigmoid colon. We classified rectal polyps separately.
The diameter of the polyps was estimated using a guidewire before polypectomy was performed. We classified small adenomas as those with a diameter <10mm and large adenomas as those with a diameter ≥10 mm. To classify hyperplastic polyps by size we used a cutoff of 5 mm. If multiple polyps were present, data were based on the largest. Histologic specimens from each polyp were reviewed by an expert gastrointestinal pathologist who was blinded to the initial pathological diagnosis made at colonoscopy. Adenomas were classified as tubular, villous, or mixed tubular-villous following standard criteria [8]. Grade of colorectal polyp was defined according to the most advanced histopathologic lesion. We defined advanced adenomas as those with a diameter of at least 10 mm, adenomas with villous component, or adenomas with high-grade dysplasia. Non-advanced adenomas were adenomas that did not meet those criteria. Women were considered as cases with multiple (≥ 2) polyps even if their polyps were detected in separate colorectal sites.
We assessed the relation of BMI to colorectal polyps according to four groups: cases with any adenomas; cases with adenomas after excluding those with hyperplastic polyps; cases with hyperplastic polyps only (which involved excluding those with adenomas); and cases with the combination of both types of polyps. Controls were defined as women with no adenomas or hyperplastic polyps.
Assessment of BMI
Before colonoscopy, women completed a detailed questionnaire regarding demographic, dietary, and lifestyle information, including current weight and height. Weight and height were used to calculate the BMI, which we divided into three categories that incorporated the definitions of normal weight (18.5–24.9 kg/m2), overweight (25.0–29.9 kg/m2) and obesity (30 kg/m2 or greater) proposed by the World Health Organization (WHO) [9].
Statistical analysis
We calculated means and standard errors for continuous data and proportions for categorical data. Logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (CI) for polyps, by tumor size, grade, multiplicity, and site. Analyses were adjusted for factors previously identified to be associated with colorectal adenomas in our dataset or associated with colorectal adenomas or colorectal cancer in the literature. Covariates were classified according to the following groups: age (5-year categories), study center (4 categories), race or ethnic group (White; Black; and Asian, Hispanic, and other combined), family history of colorectal cancer or colorectal polyps (yes or no), history of colonoscopy or barium enema in the previous 10 years (yes or no), history of flexible sigmoidoscopy in the previous 5 years (yes or no), smoking (never; former; current smoker of <20 cigarettes per day; and current smoker of ≥20 cigarettes per day), aspirin use (yes or no), menopausal status (yes or no), current menopausal hormone therapy (yes or no), physical activity (low; intermediate; high), and intakes of alcohol (grams/day; continuous), red meat grams/day; (continuous), and vitamin D (I.U./day; continuous). We conducted an alternative analysis using polytomous logistic regression. To evaluate whether any of the variables modified the BMI and adenoma association, we entered into the appropriate multivariate model the main effects terms for BMI and the covariate along with a term for their product, the coefficient for which was evaluated using a Wald test. All reported p-values are two-tailed. The analyses were performed using SAS software release 8.2 (SAS Institute, Cary, NC).
Results
Of the 1,420 women in the current study, 953 (67.1%) had no polyps, and 292 (20.6%) were classified as having adenomas (regardless of hyperplastic polyp status), of which 217 (15.3%) had adenomas but no hyperplastic polyps. Further, 175 (12.3%) had hyperplastic polyps but no colorectal adenomas, and 75 (5.3%) were characterized as having the combination of colorectal adenomas and hyperplastic polyps.
We first evaluated selected characteristics of the women according to polyp status to assess the potential for confounding (Table 1). As compared with women with no polyps, women with polyps were older, they had a higher BMI, and they were more likely to previously or currently smoke. In contrast, women with polyps reported less frequent use of menopausal hormones than women with no polyps. Women with adenomas only reported less frequent aspirin use than women with hyperplastic polyps only or those with no polyps. In contrast, women with hyperplastic polyps only were more likely to previously or currently smoke and to consume alcohol than women with adenomas only or women with no polyps. Women with the combination of adenomas and hyperplastic polyps were more likely to have a family history of colorectal cancer and to have formerly smoked than women with no polyps.
Table 1.
Selected characteristics of women overall and women according to the presence or absence of colorectal adenomas and hyperplastic polyps
| Variable | All women (n=1,420) | Women without polyps (n=953) | Women with any colorectal adenomas (n=292) | Women with colorectal adenomas only (n=217) | Women with hyperplastic polyps only (n=175) | Women with the combination of colorectal adenomas and hyperplastic polyps (n=75) |
|---|---|---|---|---|---|---|
| Age (years) | 58.3 | 57.5 | 60.7 | 61.1 | 58.5 | 59.6 |
| Race or ethnic group (%) | ||||||
| White | 77.2 | 76.4 | 73.9 | 73.7 | 86.9 | 74.7 |
| Black | 11.7 | 11.2 | 13.7 | 13.4 | 10.9 | 14.7 |
| Asian | 8.3 | 9.0 | 10.3 | 11.1 | 1.1 | 8.0 |
| Hispanic | 1.9 | 2.1 | 1.7 | 1.4 | 1.1 | 2.6 |
| Other | 0.9 | 1.3 | 0.3 | 0.5 | 0 | 0 |
| Family history of colorectal cancer (%) | 16.6 | 15.7 | 20.2 | 16.1 | 14.9 | 32.0 |
| Body mass index (kg/m2) | 26.5 | 26.1 | 26.9 | 26.6 | 28.2 | 28.0 |
| Physical activity (MET-hours/week) | 45.4 | 45.4 | 45.5 | 46.1 | 44.8 | 43.6 |
| Regular aspirin use (%) | 31.6 | 34.5 | 30.5 | 30.4 | 38.3 | 30.7 |
| Postmenopausal (%) | 87.3 | 86.6 | 90.4 | 90.3 | 86.3 | 90.7 |
| Menopausal hormone therapy (%) | 63.2 | 66.2 | 58.2 | 58.9 | 55.4 | 56.0 |
| Current smoker (%) | 34.8 | 32.4 | 37.3 | 37.3 | 43.4 | 37.3 |
| Past smoker (%) | 4.4 | 3.3 | 6.5 | 4.6 | 7.4 | 12.0 |
| Dietary intakes | ||||||
| Total energy (kcal/day) | 1546 | 1549 | 1482 | 1482 | 1632 | 1483 |
| Alcohol (grams/day) | 6.6 | 6.2 | 5.7 | 6.3 | 10.3 | 4.1 |
| Red meat (grams/day) | 26.5 | 26.5 | 28.2 | 27.3 | 23.5 | 30.6 |
| Vitamin D (I.U./day) | 223.5 | 222.1 | 220.2 | 240.5 | 237.5 | 163.7 |
We next examined the association between BMI and colorectal adenomas (Table 2). In age-adjusted and multivariate analyses, greater body mass was related to a statistically significant increase in adenoma risk. As compared with normal weight women, the multivariate ORs of adenomas for overweight and obese women were 1.36 (95% CI=0.99–1.88) and 1.57 (95% CI=1.07–2.29), respectively (P value, test for trend=0.02). Risk differed little according to whether adenomas were advanced or non-advanced. The positive association with body mass was strengthened when we used multiple adenomas as our endpoint (multivariate OR=2.27; 95% CI=1.17–4.41). Proximal adenomas also exhibited a strong positive relation with BMI (multivariate OR=1.85; 95% CI=1.15–2.97), whereas the association with distal adenomas was null (multivariate OR=1.09; 95% CI=0.54–2.23). There were an insufficient number of rectal adenomas to evaluate their relation with BMI.
Table 2.
Relation of body mass index to any colorectal adenomas (regardless of hyperplastic polyp status)
| Variable | Body mass index (kg/m2) | P for trend | ||
|---|---|---|---|---|
| ≤24.9 | 25.0–29.9 | ≥30.0 | ||
| Total adenomas | ||||
| No. of cases/controls | 120/477 | 103/295 | 69/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.37 (1.01–1.87) | 1.49 (1.05–2.12) | 0.03 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.36 (0.99–1.88) | 1.57 (1.07–2.29) | 0.02 |
| Nonadvanced adenomas | ||||
| No. of cases/controls | 94/477 | 74/295 | 52/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.27 (0.89–1.79) | 1.45 (0.99–2.14) | 0.12 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.29 (0.90–1.84) | 1.57 (1.03–2.34) | 0.09 |
| Advanced adenomas | ||||
| No. of cases/controls | 26/477 | 29/295 | 17/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.74 (0.99–3.06) | 1.68 (0.87–3.24) | 0.03 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.49 (0.82–2.76) | 1.52 (0.73–3.16) | 0.06 |
| Multiple adenomas | ||||
| No. of cases/controls | 25/477 | 37/295 | 23/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 2.33 (1.36–3.99) | 2.37 (1.29–4.35) | 0.002 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 2.18 (1.23–3.87) | 2.27 (1.17–4.41) | 0.007 |
| Proximal adenomas only | ||||
| No. of cases/controls | 60/477 | 53/295 | 45/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.41 (0.94–2.11) | 1.93 (1.25–2.99) | 0.007 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.34 (0.87–2.04) | 1.85 (1.15–2.97) | 0.02 |
| Distal adenomas only | ||||
| No. of cases/controls | 39/477 | 36/295 | 13/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.47 (0.91–2.37) | 0.86 (0.45–1.66) | 0.74 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.70 (1.03–2.83) | 1.09 (0.54–2.23) | 0.71 |
| Rectal adenomas only | ||||
| No. of cases/controls | 11/477 | 2/295 | 1/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 0.28 (0.06–1.26) | 0.23 (0.03–1.84) | 0.24 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 0.19 (0.03–1.06) | 0.14 (0.01–1.41) | 0.19 |
Multivariate-adjusted models included age (5-year categories), study center (4 categories), race or ethnic group (White; Black; and Asian, Hispanic, and other combined), family history of colorectal cancer or colorectal polyp (yes or no), history of colonoscopy or barium enema in the previous 10 years (yes or no), history of flexible sigmoidoscopy in the previous 5 years (yes or no), smoking (never; former; current smoker of <20 cigarettes per day; and current smoker of ≥20 cigarettes per day), aspirin use (yes or no), menopausal status (yes or no), current menopausal hormone therapy (yes or no), physical activity (low; intermediate; high), and intakes of alcohol (grams/day; continuous), red meat (grams/day; continuous), and vitamin D (I.U./day; continuous).
We repeated our analysis of BMI in relation to colorectal adenomas after excluding cases of hyperplastic polyps (i.e. excluding cases with the combination of adenomas and hyperplastic polyps) (Table 3). As compared with normal weight women, the multivariate ORs for total adenomas in overweight and obese women were 1.17 (95% CI=0.82–1.68) and 1.25 (95% CI=0.81–1.93), respectively (P value, test for trend=0.24). The weak positive and statistically non-significant association with BMI was consistent for advanced and non-advanced adenomas, and the relation was slightly stronger for multiple adenomas and proximal adenomas. Overweight versus normal weight was related to an increased risk for distal adenomas and no further increase in risk for distal adenomas was observed for obesity. Numbers of cases with rectal adenomas were insufficient for examining their association with BMI.
Table 3.
Relation of body mass index to colorectal adenomas only (excluding cases of hyperplastic polyps)
| Variable | Body mass index (kg/m2) | P for Trend | ||
|---|---|---|---|---|
| ≤24.9 | 25.0–29.9 | ≥30.0 | ||
| Total adenomas | ||||
| No. of cases/controls | 99/477 | 72/295 | 46/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.17 (0.83–1.65) | 1.19 (0.79–1.78) | 0.32 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.17 (0.82–1.68) | 1.25 (0.81–1.93) | 0.24 |
| Nonadvanced adenomas | ||||
| No. of cases/controls | 77/477 | 57/295 | 34/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.18 (0.81–1.73) | 1.14 (0.73–1.78) | 0.61 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.25 (0.84–1.85) | 1.24 (0.77–2.02) | 0.44 |
| Advanced adenomas | ||||
| No. of cases/controls | 22/477 | 15/295 | 12/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.08 (0.54–2.15) | 1.45 (0.69–3.05) | 0.16 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 0.92 (0.44–1.90) | 1.25 (0.54–2.89) | 0.25 |
| Multiple adenomas | ||||
| No. of cases/controls | 23/477 | 23/295 | 14/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.56 (0.85–2.85) | 1.55 (0.77–3.12) | 0.08 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.43 (0.75–2.71) | 1.39 (0.65–2.99) | 0.16 |
| Proximal adenomas only | ||||
| No. of cases/controls | 52/477 | 38/295 | 28/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.17 (0.74–1.83) | 1.38 (0.83–2.27) | 0.37 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.13 (0.69–1.82) | 1.36 (0.78–2.36) | 0.43 |
| Distal adenomas only | ||||
| No. of cases/controls | 28/477 | 29/295 | 11/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.62 (0.94–2.79) | 0.99 (0.48–2.05) | 0.92 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 2.03 (1.14–3.62) | 1.30 (0.59–2.87) | 0.45 |
| Rectal adenomas only | ||||
| No. of cases/controls | 10/477 | 1/295 | 1/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 0.15 (0.02–1.20) | 0.26 (0.03–2.08) | 0.33 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 0.08 (0.01–0.80) | 0.10 (0.01–1.26) | 0.19 |
Multivariate-adjusted models included age (5-year categories), study center (4 categories), race or ethnic group (White; Black; and Asian, Hispanic, and other combined), family history of colorectal cancer or colorectal polyp (yes or no), history of colonoscopy or barium enema in the previous 10 years (yes or no), history of flexible sigmoidoscopy in the previous 5 years (yes or no), smoking (never; former; current smoker of <20 cigarettes per day; and current smoker of ≥20 cigarettes per day), aspirin use (yes or no), menopausal status (yes or no), current menopausal hormone therapy (yes or no), physical activity (low; intermediate; high), and intakes of alcohol (grams/day; continuous), red meat (grams/day; continuous), and vitamin D (I.U./day; continuous).
We investigated the association between BMI and hyperplastic polyps only (Table 4). Risk of hyperplastic polyps rose sharply with increasing levels of BMI. As compared with normal weight women, the multivariate ORs of total hyperplastic polyps for overweight and obese women were 2.36 (95% CI=1.55–3.58) and 3.76 (95% CI=2.35–6.01), respectively (p value, test for trend<0.0001). A slightly stronger positive association with body mass was observed for small hyperplastic polyps (multivariate OR comparing obese with normal weight women=4.07; 95% CI=2.37–6.99) than large hyperplastic polyps (multivariate OR=3.11; 95% CI=1.34–7.21), although confidence limits overlapped considerably. Proximal hyperplastic polyps (multivariate OR=3.59; 95% CI=1.42–9.12) appeared to show a similar association with BMI as distal hyperplastic polyps (multivariate OR=3.87; 95% CI=1.87–7.97).
Table 4.
Relation of body mass index to hyperplastic polyps only (excluding cases of colorectal adenomas)
| Variable | Body mass index (kg/m2) | P for trend | ||
|---|---|---|---|---|
| ≤24.9 | 25.0–29.9 | ≥30.0 | ||
| Total hyperplastic polyps | ||||
| No. of cases/controls | 52/477 | 68/295 | 55/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 2.13 (1.44–3.16) | 2.82 (1.85–4.30) | <0.0001 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 2.36 (1.55–3.58) | 3.76 (2.35–6.01) | <0.0001 |
| Small hyperplastic polyps | ||||
| No. of cases/controls | 36/477 | 50/295 | 40/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 2.26 (1.43–3.56) | 2.94 (1.81–4.79) | <0.0001 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 2.56 (1.58–4.14) | 4.07 (2.37–6.99) | <0.0001 |
| Large hyperplastic polyps | ||||
| No. of cases/controls | 16/477 | 18/295 | 15/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.79 (0.89–3.60) | 2.48 (1.19–5.18) | 0.04 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.73 (0.80–3.73) | 3.11 (1.34–7.21) | 0.03 |
| Multiple hyperplastic polyps | ||||
| No. of cases/controls | 11/477 | 31/295 | 24/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 4.62 (2.27–9.38) | 5.74 (2.73–12.1) | <0.0001 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 5.06 (2.34–10.94) | 8.88 (3.87–20.39) | <0.0001 |
| Proximal hyperplastic polyps only | ||||
| No. of cases/controls | 15/477 | 15/295 | 11/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.57 (0.75–3.28) | 1.84 (0.82–4.13) | 0.46 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 2.19 (0.98–4.90) | 3.59 (1.42–9.12) | 0.05 |
| Distal hyperplastic polyps only | ||||
| No. of cases/controls | 18/477 | 31/295 | 20/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 2.83 (1.55–5.16) | 2.95 (1.52–5.73) | 0.0005 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 2.98 (1.57–5.64) | 3.87 (1.87–7.97) | 0.0001 |
| Rectal hyperplastic polyps only | ||||
| No. of cases/controls | 15/477 | 11/295 | 16/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.19 (0.54–2.65) | 2.95 (1.42–6.16) | 0.01 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.04 (0.44–2.44) | 2.83 (1.22–6.56) | 0.03 |
Multivariate-adjusted models included age (5-year categories), study center (4 categories), race or ethnic group (White; Black; and Asian, Hispanic, and other combined), family history of colorectal cancer or colorectal polyp (yes or no), history of colonoscopy or barium enema in the previous 10 years (yes or no), history of flexible sigmoidoscopy in the previous 5 years (yes or no), smoking (never; former; current smoker of <20 cigarettes per day; and current smoker of ≥20 cigarettes per day), aspirin use (yes or no), menopausal status (yes or no), current menopausal hormone therapy (yes or no), physical activity (low; intermediate; high), and intakes of alcohol (grams/day; continuous), red meat (grams/day; continuous), and vitamin D (I.U./day; continuous).
We examined BMI in relation to the combination of adenomas and hyperplastic polyps (Table 5). Using normal weight women as the reference group, the multivariate ORs of the combination of colorectal adenomas and hyperplastic polyps for overweight and obese women were 2.19 (95% CI=1.18–4.07) and 2.84 (95% CI=1.41–5.72), respectively (P value, test for trend=0.02). Elevated risks were observed for numerous combinations of adenomas and hyperplastic polyps, but risk estimates lacked stability, as suggested by the wide confidence intervals.
Table 5.
Relation of body mass index to the presence of both colorectal adenomas and hyperplastic polyps
| Variable | Body mass index (kg/m2) | P for trend | ||
|---|---|---|---|---|
| ≤24.9 | 25.0–29.9 | ≥30.0 | ||
| Total adenomas and hyperplastic polyps | ||||
| No. of cases/controls | 21/477 | 31/295 | 23/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 2.35 (1.31–4.19) | 2.88 (1.54–5.39) | 0.004 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 2.19 (1.18–4.07) | 2.84 (1.41–5.72) | 0.02 |
| Nonadvanced adenomas and any hyperplastic polyp | ||||
| No. of cases/controls | 17/477 | 17/295 | 18/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.64 (0.82–3.27) | 2.85 (1.42–5.69) | 0.02 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.52 (0.73–3.16) | 2.73 (1.25–5.94) | 0.08 |
| Advanced adenomas and any hyperplastic polyp | ||||
| No. of cases/controls | 4/477 | 14/295 | 5/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 5.38 (1.73–16.71) | 3.09 (0.81–11.90) | 0.08 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 5.58 (1.56–19.95) | 3.34 (0.74–15.04) | 0.14 |
| Proximal adenomas only and any hyperplastic polyp | ||||
| No. of cases/controls | 8/477 | 15/295 | 17/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 3.00 (1.25–7.21) | 5.45 (2.29–12.98) | <0.0001 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 2.80 (1.13–6.96) | 4.63 (1.81–11.83) | 0.003 |
| Distal/rectal adenomas only and any hyperplastic polyp | ||||
| No. of cases/controls | 12/477 | 8/295 | 2/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.09 (0.44–2.71) | 0.45 (0.10–2.06) | 0.28 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.14 (0.39–3.33) | 0.37 (0.06–2.45) | 0.27 |
| Small hyperplastic polyps and any adenoma | ||||
| No. of cases/controls | 16/477 | 17/295 | 15/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.73 (0.86–3.49) | 2.47 (1.19–5.16) | 0.01 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.72 (0.81–3.67) | 2.92 (1.27–6.69) | 0.01 |
| Large hyperplastic polyps and any adenoma | ||||
| No. of cases/controls | 5/477 | 14/295 | 8/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 4.38 (1.55–12.41) | 4.16 (1.32–13.09) | 0.12 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 4.68 (1.49–14.74) | 4.52 (1.27–16.10) | 0.32 |
| Proximal hyperplastic polyps only and any adenoma | ||||
| No. of cases/controls | 7/477 | 7/295 | 6/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 1.53 (0.53–4.45) | 2.08 (0.68–6.37) | 0.51 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 1.17 (0.37–3.72) | 1.59 (0.45–5.64) | 0.89 |
| Distal/rectal hyperplastic polyps only and any adenoma | ||||
| No. of cases/controls | 10/477 | 20/295 | 12/181 | |
| Age-adjusted OR (95% CI) | 1.0 | 3.31 (1.52–7.23) | 3.31 (1.39–7.89) | 0.005 |
| Multivariate-adjusted OR (95% CI) | 1.0 | 3.36 (1.46–7.71) | 3.66 (1.40–9.58) | 0.005 |
Multivariate-adjusted models included age (5-year categories), study center (4 categories), race or ethnic group (White; Black; and Asian, Hispanic, and other combined), family history of colorectal cancer or colorectal polyp (yes or no), history of colonoscopy or barium enema in the previous 10 years (yes or no), history of flexible sigmoidoscopy in the previous 5 years (yes or no), smoking (never; former; current smoker of <20 cigarettes per day; and current smoker of ≥20 cigarettes per day), aspirin use (yes or no), menopausal status (yes or no), current menopausal hormone therapy (yes or no), physical activity (low; intermediate; high), and intakes of alcohol (grams/day; continuous), red meat (grams/day; continuous), and vitamin D (I.U./day; continuous).
We performed case-case comparisons using multinomial logistic regression to assess differences in the relation with adiposity between the case groups. As compared with adenoma-only cases, the multivariate ORs comparing overweight and obesity to normal weight women for cases with hyperplastic polyps were 2.11 (95% CI=1.26–3.55) and 2.94 (95% CI=1.61–5.36), respectively. The multivariate ORs comparing overweight and obesity to normal weight women for cases with both types of polyps as compared with adenoma-only cases were 2.31 (95% CI=1.14–4.68) and 2.09 (95% CI=0.92–4.77), respectively. When compared with hyperplastic polyp-only cases, cases with both types of polyps showed no association with adiposity, although statistical power was low and risk estimates were imprecise; the multivariate ORs comparing overweight and obesity to normal weight women were 0.76 (95% CI=0.35–1.64) and 0.71 (95% CI=0.29–1.73), respectively.
The relations of BMI to colorectal adenomas and hyperplastic polyps were not modified by age, race, family history of colorectal cancer, history of sigmoidoscopy, smoking, aspirin use, physical activity, or menopausal hormone therapy (data not shown).
Discussion
In this cross-sectional study of asymptomatic women enrolled for complete screening colonoscopy, we found a statistically significant positive relation between BMI and the prevalence of colorectal adenomas. The positive association between BMI and adenomas was consistent for advanced and non-advanced lesions and it was more pronounced for multiple than individual adenomas and was more apparent for proximal than distal adenomas. The positive relation between BMI and colorectal adenomas was attenuated and became statistically non-significant after we repeated our analysis after excluding the subset of adenoma cases that also had hyperplastic polyps (i.e. those with a combination of adenomas and hyperplastic polyps). We also noted a strong positive association between BMI and the prevalence of hyperplastic polyps. In addition, BMI was positively linked to the combination of colorectal adenomas and hyperplastic polyps. Taken together, our results suggest that adiposity plays an important role in the development of both colorectal adenomas and hyperplastic polyps.
Our findings regarding BMI and colorectal adenomas are in agreement with those from abundant studies [10–36] that found a positive relation of BMI to colorectal adenomas. However, evidence has not been entirely consistent, with many previous studies [37–50] reporting no statistically significant association between the two. Our observation of a markedly stronger relation of BMI to the combination of colorectal adenomas and hyperplastic polyps than to colorectal adenomas alone suggests that some of the heterogeneity in previous studies of BMI and colorectal adenomas may be due to a potentially unaccounted for positive association between BMI and hyperplastic polyps.
Variation in results from previous studies of BMI and colorectal adenomas may also be explained by differences with respect to the endpoints considered. Evidence regarding adiposity in relation to adenoma size or degree of dyplasia shows that about half [10, 12, 15, 18, 22, 23, 27, 31, 34, 35, 39] of available investigations found a stronger positive association between BMI and large or advanced adenomas (typically defined as those with ≥10 mm diameter, villous elements, or high-grade dysplasia) than non-advanced or total adenomas. However, the findings from our and numerous other studies [14, 16, 17, 25, 28, 32, 33, 37, 42, 48] suggest no major difference between the relation of BMI to advanced adenomas to that with non-advanced or total adenomas.
We found a positive relation of BMI to proximal adenomas but no association with distal adenomas, confirming results from one study [31] that reported a stronger relation of BMI to adenomas of the proximal than distal colon. In addition, several sigmoidoscopy studies [10, 23, 34] found a positive association between BMI and adenomas of the distal colon and a weaker [23] or no association [10, 34] between BMI and adenomas of the rectum. In contrast, three previous investigations [25, 28, 42] reported similar relations of BMI to adenomas of the proximal and those of the distal colon. Additional studies with complete colonoscopy and large sample sizes are required to further clarify whether adiposity differentially affects the development of proximal versus distal colorectal adenomas.
We observed a strong positive relation of BMI to hyperplastic polyps. The positive association between BMI and hyperplastic polyps did not vary appreciably according to polyp size or anatomic location, but the relation with BMI was stronger when we considered multiple hyperplastic polyps as an endpoint. Very few studies have evaluated the relation of BMI to hyperplastic polyps, and results tend to be mixed. One study [51] reported an OR of 4.50 (95% CI=1.84–10.97) for hyperplastic polyps comparing extreme BMI quartiles. Another study [30] found no association of BMI to hyperplastic polyps among women but noted a borderline positive relation among men (OR=1.7; 95% CI=0.9–3.4; p for trend=0.04) comparing extreme BMI categories. That study [30], similar to our study of women, reported a marked positive relation of BMI to the combination of adenomas and hyperplastic polyps among men (OR=2.9; 95% CI=1.2–5.6; p for trend=0.02). Three studies [20, 30, 42] compared the relation of BMI to hyperplastic polyps to that with adenomas, one [42] of which noted a similarly null relation of BMI to hyperplastic and adenomatous polyps, the second [30] of which reported a similarly positive association of BMI to hyperplastic and adenomatous polyps, and the third [20] of which found a weaker relation of BMI to hyperplastic polyps than to adenomas.
If BMI is indeed positively related to hyperplastic polyps, the question arises as to the potential metabolic pathways through which increased body mass may favor the development of hyperplastic polyps. Data regarding this issue are very limited. One study found that circulating levels of insulin-like growth factor (IGF)-1 and insulin were positively correlated with the number of hyperplastic polyps in patients with acromegaly [52]. Hyperplastic polyps have generally been considered to be innocuous lesions. However, evidence [53] suggests that hyperplastic polyps, particularly large hyperplastic polyps, bear the potential for dysplasia and malignant potential. Perceivably, BMI may enhance colorectal cancer risk via the development of multiple hyperplastic polyps through biologic mechanisms that involve increased levels of insulin-like growth factors and insulin. In addition, hyperplastic polyps may give rise to serrated polyps and colorectal cancer by causing BRAF mutations associated with microsatellite instability [54].
Our study has numerous important strengths. The study population comprised asymptomatic women enrolled for complete screening colonoscopy, which helped avoid misclassification of non-cases who may have had a greater prevalence of asymptomatic polyps. A particular strength was the specific focus on both adenomatous and hyperplastic polyps, enabling us to compare distinct polyp types drawn from the same underlying population and examined using the same methodology. A standardized pathology review ensured comprehensive ascertainment of polyp type, which allowed us to isolate those presenting with both kinds of polyps as a distinct group. The validity of self-reported height and weight in our study should not represent a major concern because self-reported height and weight are known to be highly accurate, with correlation coefficients between self-reported and measured height and weight typically greater than 0.9 [55]. In addition, we used information on height and weight collected prior to colonoscopy to circumvent recall bias. The availability of information on a broad range of potential confounding variables minimized potential confounding.
Our study also had some limitations. The sample size was modest, which did not allow us to examine relations with great statistical power, especially in our analyses stratified by adenoma stage, multiplicity, and location. Thus, findings from our subgroup analyses should be interpreted with caution. We lacked data on serrated adenomas and we were not able to identify a sufficient number of subjects who fit the criteria for hyperplastic polyposis. Our study lacked data on central or visceral adiposity, such as waist circumference or waist-hip ratio. Observational studies that have examined BMI in relation to colorectal cancer have tended to find stronger positive associations among men than women [56]. One reason for such differences is that using BMI as a measure of adiposity may not reflect central or visceral adiposity in women.
Because participants in our study represented women who consented to participate in a colonoscopy study, our findings may not apply to the general population. Our results also may not extend to premenopausal women because the majority of participants in our study were postmenopausal. We cannot exclude the possibility that our results were affected by residual confounding by smoking or diet since those variables are measured with imprecision. A potential limitation of cross-sectional studies is determining the temporal sequence of events, but this is a minor issue in our study because it is unlikely that adenomas or hyperplastic polyps preceded or caused adiposity among women in our sample. We assessed BMI only once at the time of diagnosis, which may not have reflected BMI during the etiologically relevant time period of exposure with respect to the development of adenomas.
In summary, in this study of complete colonoscopy among asymptomatic U.S. women, BMI was positively related both to colorectal adenomas and hyperplastic polyps. A notably strong apparent adverse effect of adiposity was evident among women with both kinds of polyps. Future epidemiologic investigations should continue to identify risk factors that distinguish colorectal adenomas from hyperplastic polyps. Such work may help further elucidate the possible causes of colorectal cancer.
Acknowledgments
Financial Support: This research was supported by the Intramural Research Program of the NIH, National Cancer Institute.
Footnotes
Conflict of interest statement
All authors declare that they have no conflict of interest.
References
- 1.IARC Working Group. IARC Handbook of Cancer Prevention: Weight control and physical activity. Lyon: IARC Press; 2002. [Google Scholar]
- 2.Hill MJ, Morson BC, Bussey HJ. Aetiology of adenoma--carcinoma sequence in large bowel. Lancet. 1978;1(8058):245–7. doi: 10.1016/s0140-6736(78)90487-7. [DOI] [PubMed] [Google Scholar]
- 3.Colbert LH, Davis JM, Essig DA, Ghaffar A, Mayer EP. Exercise and tumor development in a mouse predisposed to multiple intestinal adenomas. Med Sci Sports Exerc. 2000;32(10):1704–8. doi: 10.1097/00005768-200010000-00007. [DOI] [PubMed] [Google Scholar]
- 4.Baron JA. Intermediate effect markers for colorectal cancer. IARC Sci Publ. 2001;154:113–29. [PubMed] [Google Scholar]
- 5.East JE, Saunders BP, Jass JR. Sporadic and syndromic hyperplastic polyps and serrated adenomas of the colon: classification, molecular genetics, natural history, and clinical management. Gastroenterol Clin North Am. 2008;37(1):25–46. doi: 10.1016/j.gtc.2007.12.014. [DOI] [PubMed] [Google Scholar]
- 6.Schoenfeld P, Cash B, Flood A, et al. Colonoscopic screening of average-risk women for colorectal neoplasia. N Engl J Med. 2005;352(20):2061–8. doi: 10.1056/NEJMoa042990. [DOI] [PubMed] [Google Scholar]
- 7.Woodson K, Flood A, Green L, et al. Loss of insulin-like growth factor-II imprinting and the presence of screen-detected colorectal adenomas in women. J Natl Cancer Inst. 2004;96(5):407–10. doi: 10.1093/jnci/djh042. [DOI] [PubMed] [Google Scholar]
- 8.Jass JR, Sobin LH, Watanabe H. The World Health Organization’s histologic classification of gastrointestinal tumors. A commentary on the second edition. Cancer. 1990;66(10):2162–7. doi: 10.1002/1097-0142(19901115)66:10<2162::aid-cncr2820661020>3.0.co;2-n. [DOI] [PubMed] [Google Scholar]
- 9.WHO. Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Committee. World Health Organ Tech Rep Ser. 1995;854:1–452. [PubMed] [Google Scholar]
- 10.Honjo S, Kono S, Shinchi K, et al. The relation of smoking, alcohol use and obesity to risk of sigmoid colon and rectal adenomas. Jpn J Cancer Res. 1995;86(11):1019–26. doi: 10.1111/j.1349-7006.1995.tb03015.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Anderson JC, Messina CR, Dakhllalah F, et al. Body mass index: a marker for significant colorectal neoplasia in a screening population. J Clin Gastroenterol. 2007;41(3):285–90. doi: 10.1097/01.mcg.0000247988.96838.60. [DOI] [PubMed] [Google Scholar]
- 12.Bayerdorffer E, Mannes GA, Ochsenkuhn T, Kopcke W, Wiebecke B, Paumgartner G. Increased risk of ‘high-risk’ colorectal adenomas in overweight men. Gastroenterology. 1993;104(1):137–44. doi: 10.1016/0016-5085(93)90845-4. [DOI] [PubMed] [Google Scholar]
- 13.Davidow AL, Neugut AI, Jacobson JS, et al. Recurrent adenomatous polyps and body mass index. Cancer Epidemiol Biomarkers Prev. 1996;5(4):313–5. [PubMed] [Google Scholar]
- 14.Elwing JE, Gao F, Davidson NO, Early DS. Type 2 diabetes mellitus: the impact on colorectal adenoma risk in women. Am J Gastroenterol. 2006;101(8):1866–71. doi: 10.1111/j.1572-0241.2006.00651.x. [DOI] [PubMed] [Google Scholar]
- 15.Kim SE, Shim KN, Jung SA, Yoo K, Moon IH. An association between obesity and the prevalence of colonic adenoma according to age and gender. J Gastroenterol. 2007;42(8):616–23. doi: 10.1007/s00535-007-2074-4. [DOI] [PubMed] [Google Scholar]
- 16.Larsen IK, Grotmol T, Almendingen K, Hoff G. Lifestyle as a predictor for colonic neoplasia in asymptomatic individuals. BMC Gastroenterol. 2006;6:5. doi: 10.1186/1471-230X-6-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sedjo RL, Byers T, Levin TR, et al. Change in body size and the risk of colorectal adenomas. Cancer Epidemiol Biomarkers Prev. 2007;16(3):526–31. doi: 10.1158/1055-9965.EPI-06-0229. [DOI] [PubMed] [Google Scholar]
- 18.Shinchi K, Kono S, Honjo S, et al. Obesity and adenomatous polyps of the sigmoid colon. Jpn J Cancer Res. 1994;85(5):479–84. doi: 10.1111/j.1349-7006.1994.tb02383.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tashiro M, Akiyama T, Yoshikawa I, Kume K, Otsuki M. Obesity as a risk factor for colorectal polyps in Japanese patients. Gut. 2004;53(1):156. doi: 10.1136/gut.53.1.156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wang YY, Lin SY, Lai WA, Liu PH, Sheu WH. Association between adenomas of rectosigmoid colon and metabolic syndrome features in a Chinese population. J Gastroenterol Hepatol. 2005;20(9):1410–5. doi: 10.1111/j.1440-1746.2005.03971.x. [DOI] [PubMed] [Google Scholar]
- 21.Almendingen K, Hofstad B, Vatn MH. Does high body fatness increase the risk of presence and growth of colorectal adenomas followed up in situ for 3 years? Am J Gastroenterol. 2001;96(7):2238–46. doi: 10.1111/j.1572-0241.2001.03942.x. [DOI] [PubMed] [Google Scholar]
- 22.Bird CL, Frankl HD, Lee ER, Haile RW. Obesity, weight gain, large weight changes, and adenomatous polyps of the left colon and rectum. Am J Epidemiol. 1998;147(7):670–80. doi: 10.1093/oxfordjournals.aje.a009508. [DOI] [PubMed] [Google Scholar]
- 23.Boutron-Ruault MC, Senesse P, Meance S, Belghiti C, Faivre J. Energy intake, body mass index, physical activity, and the colorectal adenoma-carcinoma sequence. Nutr Cancer. 2001;39(1):50–7. doi: 10.1207/S15327914nc391_7. [DOI] [PubMed] [Google Scholar]
- 24.Chung YW, Han DS, Park YK, et al. Association of obesity, serum glucose and lipids with the risk of advanced colorectal adenoma and cancer: a case-control study in Korea. Dig Liver Dis. 2006;38(9):668–72. doi: 10.1016/j.dld.2006.05.014. [DOI] [PubMed] [Google Scholar]
- 25.Erhardt JG, Kreichgauer HP, Meisner C, Bode JC, Bode C. Alcohol, cigarette smoking, dietary factors and the risk of colorectal adenomas and hyperplastic polyps--a case control study. Eur J Nutr. 2002;41(1):35–43. doi: 10.1007/s003940200004. [DOI] [PubMed] [Google Scholar]
- 26.Guilera M, Connelly-Frost A, Keku TO, Martin CF, Galanko J, Sandler RS. Does physical activity modify the association between body mass index and colorectal adenomas? Nutr Cancer. 2005;51(2):140–5. doi: 10.1207/s15327914nc5102_3. [DOI] [PubMed] [Google Scholar]
- 27.Jacobs ET, Martinez ME, Alberts DS, et al. Association between body size and colorectal adenoma recurrence. Clin Gastroenterol Hepatol. 2007;5(8):982–90. doi: 10.1016/j.cgh.2007.03.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kono S, Handa K, Hayabuchi H, et al. Obesity, weight gain and risk of colon adenomas in Japanese men. Jpn J Cancer Res. 1999;90(8):805–11. doi: 10.1111/j.1349-7006.1999.tb00819.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Lubin F, Rozen P, Arieli B, et al. Nutritional and lifestyle habits and water-fiber interaction in colorectal adenoma etiology. Cancer Epidemiol Biomarkers Prev. 1997;6(2):79–85. [PubMed] [Google Scholar]
- 30.Morimoto LM, Newcomb PA, Ulrich CM, Bostick RM, Lais CJ, Potter JD. Risk factors for hyperplastic and adenomatous polyps: evidence for malignant potential? Cancer Epidemiol Biomarkers Prev. 2002;11(10 Pt 1):1012–8. [PubMed] [Google Scholar]
- 31.Neugut AI, Lee WC, Garbowski GC, et al. Obesity and colorectal adenomatous polyps. J Natl Cancer Inst. 1991;83(5):359–61. doi: 10.1093/jnci/83.5.359. [DOI] [PubMed] [Google Scholar]
- 32.Terry MB, Neugut AI, Bostick RM, et al. Risk factors for advanced colorectal adenomas: a pooled analysis. Cancer Epidemiol Biomarkers Prev. 2002;11(7):622–9. [PubMed] [Google Scholar]
- 33.Wolf LA, Terry PD, Potter JD, Bostick RM. Do factors related to endogenous and exogenous estrogens modify the relationship between obesity and risk of colorectal adenomas in women? Cancer Epidemiol Biomarkers Prev. 2007;16(4):676–83. doi: 10.1158/1055-9965.EPI-06-0883. [DOI] [PubMed] [Google Scholar]
- 34.Giovannucci E, Colditz GA, Stampfer MJ, Willett WC. Physical activity, obesity, and risk of colorectal adenoma in women (United States) Cancer Causes & Control. 1996;7(2):253–63. doi: 10.1007/BF00051301. [DOI] [PubMed] [Google Scholar]
- 35.Kim S, Baron JA, Mott LA, et al. Aspirin may be more effective in preventing colorectal adenomas in patients with higher BMI (United States) Cancer Causes Control. 2006;17(10):1299–304. doi: 10.1007/s10552-006-0075-x. [DOI] [PubMed] [Google Scholar]
- 36.Wise LA, Rosenberg L, Palmer JR, Adams-Campbell LL. Anthropometric Risk Factors for Colorectal Polyps in African-American Women. Obesity (Silver Spring) 2008 doi: 10.1038/oby.2007.139. [DOI] [PubMed] [Google Scholar]
- 37.Honjo S, Kono S, Shinchi K, Imanishi K, Hirohata T. Cigarette smoking, alcohol use and adenomatous polyps of the sigmoid colon. Jpn J Cancer Res. 1992;83(8):806–11. doi: 10.1111/j.1349-7006.1992.tb01984.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Jass JR, Young PJ, Robinson EM. Predictors of presence, multiplicity, size and dysplasia of colorectal adenomas. A necropsy study in New Zealand. Gut. 1992;33(11):1508–14. doi: 10.1136/gut.33.11.1508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Kono S, Shinchi K, Imanishi K. Body mass index and adenomas of the sigmoid colon in Japanese men. Eur J Epidemiol. 1996;12(4):425–6. doi: 10.1007/BF00145309. [DOI] [PubMed] [Google Scholar]
- 40.Lieberman DA, Prindiville S, Weiss DG, Willett W. Risk factors for advanced colonic neoplasia and hyperplastic polyps in asymptomatic individuals. Jama. 2003;290(22):2959–67. doi: 10.1001/jama.290.22.2959. [DOI] [PubMed] [Google Scholar]
- 41.Manus B, Adang RP, Ambergen AW, Bragelmann R, Armbrecht U, Stockbrugger RW. The risk factor profile of recto-sigmoid adenomas: a prospective screening study of 665 patients in a clinical rehabilitation centre. Eur J Cancer Prev. 1997;6(1):38–43. doi: 10.1097/00008469-199702000-00007. [DOI] [PubMed] [Google Scholar]
- 42.Wallace K, Baron JA, Karagas MR, et al. The association of physical activity and body mass index with the risk of large bowel polyps. Cancer Epidemiol Biomarkers Prev. 2005;14(9):2082–6. doi: 10.1158/1055-9965.EPI-04-0757. [DOI] [PubMed] [Google Scholar]
- 43.Hauret KG, Bostick RM, Matthews CE, et al. Physical activity and reduced risk of incident sporadic colorectal adenomas: observational support for mechanisms involving energy balance and inflammation modulation. Am J Epidemiol. 2004;159(10):983–92. doi: 10.1093/aje/kwh130. [DOI] [PubMed] [Google Scholar]
- 44.Little J, Logan RF, Hawtin PG, Hardcastle JD, Turner ID. Colorectal adenomas and energy intake, body size and physical activity: a case-control study of subjects participating in the Nottingham faecal occult blood screening programme. Br J Cancer. 1993;67(1):172–6. doi: 10.1038/bjc.1993.30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Olsen J, Kronborg O, Lynggaard J, Ewertz M. Dietary risk factors for cancer and adenomas of the large intestine. A case-control study within a screening trial in Denmark. Eur J Cancer. 1994;30A(1):53–60. doi: 10.1016/s0959-8049(05)80019-x. [DOI] [PubMed] [Google Scholar]
- 46.Otake S, Takeda H, Suzuki Y, et al. Association of visceral fat accumulation and plasma adiponectin with colorectal adenoma: evidence for participation of insulin resistance. Clin Cancer Res. 2005;11(10):3642–6. doi: 10.1158/1078-0432.CCR-04-1868. [DOI] [PubMed] [Google Scholar]
- 47.Takemura Y, Kikuchi S, Oba K, Inaba Y, Nakagawa K. A high level of physical fitness during thirties is a negative risk factor for colonic polyps during fifties. Keio J Med. 2000;49(3):111–6. doi: 10.2302/kjm.49.111. [DOI] [PubMed] [Google Scholar]
- 48.Giovannucci E, Ascherio A, Rimm EB, Colditz GA, Stampfer MJ, Willett WC. Physical activity, obesity, and risk for colon cancer and adenoma in men. Annals of Internal Medicine. 1995;122(5):327–34. doi: 10.7326/0003-4819-122-5-199503010-00002. [DOI] [PubMed] [Google Scholar]
- 49.Sass DA, Schoen RE, Weissfeld JL, et al. Relationship of visceral adipose tissue to recurrence of adenomatous polyps. Am J Gastroenterol. 2004;99(4):687–93. doi: 10.1111/j.1572-0241.2004.04136.x. [DOI] [PubMed] [Google Scholar]
- 50.Stemmermann GN, Heilbrun LK, Nomura AM. Association of diet and other factors with adenomatous polyps of the large bowel: a prospective autopsy study. Am J Clin Nutr. 1988;47(2):312–7. doi: 10.1093/ajcn/47.2.312. [DOI] [PubMed] [Google Scholar]
- 51.Martinez ME, McPherson RS, Levin B, Glober GA. A case-control study of dietary intake and other lifestyle risk factors for hyperplastic polyps. Gastroenterology. 1997;113(2):423–9. doi: 10.1053/gast.1997.v113.pm9247459. [DOI] [PubMed] [Google Scholar]
- 52.Foltyn W, Kos-Kudla B, Strzelczyk J, et al. Is there any relation between hyperinsulinemia, insulin resistance and colorectal lesions in patients with acromegaly? Neuro Endocrinol Lett. 2008;29(1):107–12. [PubMed] [Google Scholar]
- 53.Warner AS, Glick ME, Fogt F. Multiple large hyperplastic polyps of the colon coincident with adenocarcinoma. Am J Gastroenterol. 1994;89(1):123–5. [PubMed] [Google Scholar]
- 54.O’Brien MJ, Yang S, Mack C, et al. Comparison of microsatellite instability, CpG island methylation phenotype, BRAF and KRAS status in serrated polyps and traditional adenomas indicates separate pathways to distinct colorectal carcinoma end points. Am J Surg Pathol. 2006;30(12):1491–501. doi: 10.1097/01.pas.0000213313.36306.85. [DOI] [PubMed] [Google Scholar]
- 55.Willett WC. Nutritional Epidemiology. New York, NY: Oxford University Press; 1998. [Google Scholar]
- 56.Pischon T, Lahmann PH, Boeing H, et al. Body size and risk of colon and rectal cancer in the European Prospective Investigation Into Cancer and Nutrition (EPIC) J Natl Cancer Inst. 2006;98(13):920–31. doi: 10.1093/jnci/djj246. [DOI] [PubMed] [Google Scholar]