Meat and meat-mutagen intake, doneness preference and the risk of colorectal polyps: The Tennessee colorectal polyp study
Article first published online: 12 MAR 2007
Copyright © 2007 Wiley-Liss, Inc.
International Journal of Cancer
Volume 121, Issue 1, pages 136–142, 1 July 2007
How to Cite
Shin, A., Shrubsole, M. J., Ness, R. M., Wu, H., Sinha, R., Smalley, W. E., Shyr, Y. and Zheng, W. (2007), Meat and meat-mutagen intake, doneness preference and the risk of colorectal polyps: The Tennessee colorectal polyp study. Int. J. Cancer, 121: 136–142. doi: 10.1002/ijc.22664
- Issue published online: 24 APR 2007
- Article first published online: 12 MAR 2007
- Manuscript Accepted: 16 JAN 2007
- Manuscript Received: 3 AUG 2006
- National Cancer Institute. Grant Numbers: P50 CA950103, R01 CA97386
- colorectal neoplasms;
Although meat intake has been fairly consistently linked to the risk of colorectal cancer, only a few studies have evaluated meat intake by doneness level and the heterocyclic amines (HCAs) and polycyclic aromatic hydrocarbons (PAHs) produced by high temperature cooking of meat in relation to colorectal adenomatous and hyperplastic polyps. We evaluated these associations in a large colonoscopy-based case-control study. Included in this study were participants with adenomatous polyp only (n = 573), hyperplastic polyp only (n = 256), or both adenomatous and hyperplastic polyps (n = 199), and 1,544 polyp-free controls. In addition to information related to demographic and other lifestyle factors, meat intake by cooking method and doneness preference were obtained through telephone interviews. Polytomous logistic regression models were used to estimate odds ratios (OR) and 95% confidence intervals for the association between exposures and colorectal polyp risks. Presence of hyperplastic polyp was found to be positively associated with high consumption of total meat (ptrend = 0.076) or red meat (ptrend = 0.060), with an approximate 50–60% elevated risk observed in the highest vs. the lowest intake group. High intake of 2-amino-I-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) and 2-amino-3,4,8-trimethylimidazo [4,5]quinoxaline (DiMeIQx) were associated with increased risk for hyperplastic polyp (ptrend = 0.036 and 0.038, respectively). With a possible exception of the intake of total well-done meats (ptrend = 0.055) or well-done red meats (ptrend = 0.074) with the risk of large adenomas, no other positive association was found specifically for the risk of adenomas with any of the exposure variables aforementioned. This study provides additional support for a positive association of high intake of red meat with colorectal adenomas, and suggests that high intake of meats and meat carcinogens may also be associated with hyperplastic polyps. © 2007 Wiley-Liss, Inc.
Meat consumption, particularly red and processed meat consumption, has been linked to the risk of colorectal cancer in many previous epidemiological studies.1–7 Initially, meat was presumed to affect colorectal cancer development by its fat content, which could influence the bile acid content of the gut. However, recent evidence suggests that other components, including mutagens formed during meat preparation, may play a role.1, 8–11 Heterocyclic amines (HCAs) and polycyclic aromatic hydrocarbons (PAHs) are formed during high-temperature cooking of meats.12–14 Well-done meat contains 10 times the concentration of HCAs compared to rare meat prepared by the same cooking method.15 These amine compounds are mutagenic in Ames/Salmonella assays and produce colon tumors in laboratory animals.8
Colorectal adenomatous polyps are well established precursors of colorectal cancer.16, 17 The associations of red meat intake with colorectal adenoma risk18, 19 and recurrence20, 21 have been evaluated in several previous studies with inconsistent results. Very few studies have used a questionnaire specifically designed to obtain information about meat intake by preparation method and doneness level.22–28 Traditionally, hyperplastic polyps were not considered to be precursors for colorectal cancer. However, it has been recently recognized that some hyperplastic polyps could be better classified as sessile serrated adenomas, which may have malignant potential through pathways other than the adenoma-carcinoma sequence.29 Very few studies have compared risk factors for adenomatous polyps and hyperplastic polyps.30–32 In this study, we evaluated meat intake, meat preparation methods, doneness preference, and intake of HCAs and benzo[α]pyrene on the risk of colorectal polyps in a large-scale colonoscopy-based case-control study.
Material and methods
Tennessee colorectal polyp study
Subjects for the present analyses were participants of the Tennessee Colorectal Polyp Study, an on-going colonoscopy-based case-control study being conducted in Nashville, Tennessee. Eligible participants, aged between 40 and 75 years old, were identified from patients scheduled for colonoscopy at the Vanderbilt Gastroenterology Clinic and the Veteran's Affairs Tennessee Valley Health System Nashville campus between February 1, 2003 and December 31, 2005. Approximately 94% of participants from the Vanderbilt clinic were from the Middle Tennessee area, and most of them (87%) were from the metropolitan Nashville area (13 counties). For participants from the Veterans' Affairs clinic, however, 62% were from the Middle Tennessee area, an additional 17% were from Kentucky, and 19% were from other Tennessee counties. Excluded from our study were participants who had genetic colorectal cancer syndromes (e.g. hereditary nonpolyposis colorectal cancer or familial adenomatous polyposis), a prior history of inflammatory bowel disease, adenomatous polyps or any cancer other than nonmelanoma skin cancer. The study was approved by relevant committees for the use of human subjects in research. Among 4,617 eligible participants, 3,083 (66.8%) provided written informed consent. Among them 2,678 (87%) completed a telephone interview and 2,355 (76%) completed a self-administered food frequency questionnaire (FFQ) developed for a similar southern United States population.33 Details about the FFQ used in this study have been described elsewhere.33 In brief, using the NHANES-III database for the southern region of the United States, a list of 262 food categories was identified, coded and reduced to 102 food items that could account for the vast majority of food groups and nutrients that have been linked to cancer risk in previous studies.
Based on the colonoscopy and pathologic findings, participants were assigned as cases with adenomatous polyp only, hyperplastic polyp only, or concurrent adenomatous and hyperplastic polyps or polyp-free controls. To be diagnosed as a control, the participant must have had a complete colonoscopy reaching the cecum and been found to be polyp-free. Based upon the reported endoscopic size, participants with an adenomatous polyp equal to or larger than 1 cm were classified as a large adenoma case.
Meat preparation questions
A telephone interview was conducted to obtain information on medication use, demographics, medical history and lifestyle including questions on usual intake frequencies, and portion size of 11 meats (hamburgers or cheeseburgers from fast food, hamburgers or cheeseburgers not from fast food, beef steaks, pork chops or ham steaks, bacon, sausage, hotdogs or franks, chicken, fish, meat gravies made with drippings and short ribs or spareribs). For each meat item, except hamburgers or cheeseburgers from fast food and meat gravies made with dripping, we also asked how often the participants consumed the meat as well as the cooking methods used, including oven-broiled or oven-baked, grilled or barbecued, pan fried, deep fried (for chicken and fish), and all other ways. For example, subjects could have responded with an answer of grilled 50% of the time and pan fried the other 50%.
Participants reported their usual preference level of meat doneness over the preceding 5 years by using a series of 3 color photographs representing increasing levels of doneness of 7 meat items; i.e., hamburger patties, beef steaks, pork chops, bacon, grilled chicken, pan fried chicken and pan fried or grilled fish. Each photograph was labeled with a number and represented a range of doneness level from medium, well-done to very well-done (hamburger patties and beef steaks) or from just done, well-done, to very well-done (all other items). Participants were asked when they usually ate each food item whether the item usually looked less than one, about the same as one, about the same as two, about the same as three, or more than 3 (5 categories), representing rare, medium or just done, well-done, very well-done and extremely very well-done. All subjects who completed the questions on food doneness had the food booklet in front of them during the telephone interview. This booklet was typically given to subjects in-person on the day of colonoscopy. At the beginning of the telephone interview, the trained interviewers verified that the participant had the interview packet in hand, including the food pictures booklet and waited to begin the interview until the participant located the booklet. In the event that a participant had misplaced the booklet, that portion of the interview was rescheduled by the interviewer for another time when a replacement booklet was received, generally within a few days. The median interval between the day of colonoscopy and interview was 13 days. The meat questionnaire, along with the food pictures, was developed initially by Dr. Sinha23 and subsequently modified for our studies.34 The original questionnaire contained a list of questions about the intake frequency for each meat item by cooking method. We asked our subjects to report the intake frequency and usual portion size of each meat item and then estimate the proportion of time the meat item was cooked by each of the cooking methods. The original questionnaire included 4 photos to represent the doneness level of hamburger and beef steaks. We used 3 photographs, excluding the first photograph which represented rare doneness level. Subjects were asked to report whether they prefer meat prepared less than the first photograph (medium done), and thus information for intake of meat at the rare doneness level was captured in our study.
Meats were categorized by type (red or processed) and preparation method to facilitate statistical analysis. The red meat category included hamburgers or cheeseburgers from fast food or not from fast food, beef steaks, pork chops or ham steaks, bacon, sausage, hotdogs or franks, and short ribs or spareribs. Processed meat included bacon, sausage and hotdogs or franks. Total meat included all red meat items, in addition to chicken and fish. High-temperature cooking methods included grilled or barbecued and pan fried. Well-done or very well-done preference levels were grouped together to calculate the total amount of well-done meat intake of grilled or barbecued and pan fried meats. Similarly, the intake amounts for rare/medium meats were also calculated. HCAs (MeIQx, DiMeIQx and PhIP), benzo(α)pyrene (BP) and mutagenic activity were estimated using the software application known as the computerized heterocyclic amines resource for research in epidemiology of disease (CHARRED, http://charred.cancer.gov/).35
Generalized linear models and Mantel-Haenszel χ2 tests were used to compare the distribution of demographic characteristics and risk factors for colorectal polyps among case and control groups as well as characteristics between controls recruited from the Vanderbilt Clinic and the Veterans' Affairs hospital with adjustment for age and sex differences. Polytomous logistic regression models were used to estimate odds ratios (OR) and their 95% confidence intervals (CI) for the association between exposures and types of polyps or between exposures and number, size and anatomical location for adenomatous polyps to compare risks among 3 outcome groups. OR were adjusted for known risk factors for colorectal polyps and variables that showed different distribution among comparison groups, i.e., age, sex, study site, education, indication for colonoscopy, smoking, alcohol consumption, body mass index, exercised regularly in the past 10 years, regular NSAIDs use and total energy intake. Energy intake level for those who did not provide FFQ information (n = 614) were input with age- (40–49/50–59/60–64/≥65) and gender- specific mean values. Sensitivity analysis was also performed excluding subjects without FFQ information with additional adjustment for calcium, vitamin D, and folic acid intake. p-values for trend were derived by using the median intake values of each quartile as a continuous variable in the model.
For the current analysis, participants were limited to those with a confirmed adenomatous or hyperplastic polyp. Excluded were participants who failed to provide information for all meat consumption and doneness questions (12 cases and 24 controls). Finally, 573 adenomatous polyp only cases, 256 hyperplastic polyp only cases, 199 cases with both adenomatous and hyperplastic polyps and 1,544 polyp-free controls were included in this analysis.
The distribution of demographic and other characteristics for 3 case groups and polyp-free controls are presented in Table I. There were significant differences in sex between controls recruited at the Vanderbilt and veterans affairs (VA) clinics. Only 7% of participants recruited at the VA clinic were female, whereas about 61% of participants at the Vanderbilt clinic were female. Compared with controls, all polyp cases were more likely to be male, a smoker, and to have lower educational attainment and lower household income. Additionally, adenomatous polyp only and concurrent adenomatous/hyperplastic polyps cases were older than controls, whereas hyperplastic polyp only cases showed a mean age similar to controls.
|Characteristic||Polyp type||Controls (N = 1,544)||p-value1|
|Adenomatous only (N = 573)||Hyperplastic only (N = 256)||Concurrent adenomatous and hyperplastic (N = 199)|
|Age (years, mean ± SD)||59.6 ± 7.5||56.7 ± 7.0||59.7 ± 6.4||57.3 ± 7.8||<0.001|
|Sex (female, %)||25.1||30.5||16.6||42.0||<0.001|
|Indications for colonoscopy (%)2|
|High school or less||29.9||38.4||36.3||27.2|
|Graduate or professional education||19.8||14.6||11.7||25.6||<0.001|
|Annual household income (%)2|
|Race (white, %)||83.4||91.8||89.4||86.3||0.005|
|Family history (%)2|
|Regular alcohol consumption (%)2|
|Body mass index (kg/m2, mean)3|
|Height (cm, mean)3|
|Regularly exercised (%)2||51.4||50.6||46.7||56.2||0.033|
|Regular NSAIDs use2|
|Total energy intake (kcal/day, mean)3|
|Folic acid intake (mg/day, mean)3|
Table II shows associations of intake amounts of meats with risk for colorectal polyps. Marginally significant positive associations were found for hyperplastic polyp cases with intake of total meat (p for trend = 0.076) and red meat (p for trend = 0.060). High intake of total meat cooked well-done or very well-done was associated with a marginally increased risk for concurrent adenoma and hyperplastic polyp (p for trend = 0.076). High consumption of total meat cooked well-done or very well-done (p for trend = 0.055) and red meat cooked well-done or very well-done (p for trend = 0.074) was associated with an increased risk for large (≥1 cm) adenomas, although the trend tests were of borderline significance. High consumption of total meat cooked rare/medium was associated with an increased risk for hyperplastic polyp (p for trend = 0.019). No apparent association was found for processed meat or red meat cooked rare/medium with any kind of polyp.
|Intake amounts (g/day)||Controls (n)||Polyp types||Size of adenomatous polyps|
|Adenoma only||HPP only||Adenoma and HPP||Large (≥1cm)||Small (<1cm)|
|n||aOR (95% CI)1||n||aOR (95% CI)1||n||aOR (95% CI)1||n||aOR (95% CI)1||n||aOR (95% CI)1|
|0–62.4||417||138||1.0 (reference)||49||1.0 (reference)||39||1.0 (reference)||35||1.0 (reference)||138||1.0 (reference)|
|62.5–99.9||402||151||1.1 (0.9–1.5)||60||1.3 (0.9–2.0)||39||1.1 (0.7–1.8)||42||1.3 (0.8–2.2)||147||1.1 (0.8–1.5)|
|100–151.9||377||134||0.9 (0.7–1.3)||65||1.3 (0.9–2.0)||58||1.3 (0.8–2.1)||41||1.2 (0.7–2.0)||152||1.0 (0.7–1.3)|
|152.0–657.0||348||150||1.0 (0.7–1.3)||82||1.5 (1.0–2.3)||63||1.3 (0.8–2.1)||51||1.4 (0.8–2.3)||161||1.0 (0.7–1.3)|
|p for trend||0.630||0.076||0.293||0.309||0.575|
|0–22.3||441||129||1.0 (reference)||42||1.0 (reference)||27||1.0 (reference)||28||1.0 (reference)||127||1.0 (reference)|
|22.4–48.8||401||145||1.1 (0.8–1.4)||55||1.2 (0.7–1.8)||46||1.3 (0.8–2.3)||46||1.5 (0.9–2.5)||145||1.0 (0.8–1.4)|
|48.9–91.2||369||140||1.0 (0.7–1.3)||77||1.6 (1.0–2.5)||57||1.5 (0.9–2.6)||42||1.3 (0.8–2.2)||155||1.0 (0.7–1.4)|
|91.3–585.8||333||159||1.0 (0.7–1.4)||82||1.6 (1.0–2.5)||69||1.5 (0.9–2.6)||53||1.5 (0.9–2.7)||174||1.0 (0.7–1.4)|
|p for trend||0.998||0.060||0.254||0.300||0.958|
|0–1.8||421||139||1.0 (reference)||55||1.0 (reference)||39||1.0 (reference)||37||1.0 (reference)||140||1.0 (reference)|
|1.9–11.3||406||143||1.0 (0.8–1.4)||60||1.1 (0.7–1.6)||34||0.8 (0.5–1.4)||37||1.0 (0.6–1.7)||137||1.0 (0.7–1.3)|
|11.4–27.9||389||124||0.7 (0.5–0.9)||66||0.9 (0.6–1.4)||55||0.9 (0.5–1.4)||41||0.8 (0.5–1.4)||142||0.7 (0.5–1.0)|
|28.0–368.1||328||167||1.0 (0.7–1.4)||75||1.1 (0.7–1.7)||71||1.1 (0.7–1.8)||54||1.2 (0.7–2.0)||179||1.0 (0.7–1.4)|
|p for trend||0.830||0.591||0.285||0.392||0.649|
|Total meat cooked well-done or very well-done|
|0–4.2||422||127||1.0 (reference)||52||1.0 (reference)||39||1.0 (reference)||38||1.0 (reference)||127||1.0 (reference)|
|4.3–18.6||396||149||1.2 (0.9–1.6)||58||1.0 (0.7–1.6)||40||1.0 (0.6–1.6)||32||0.9 (0.5–1.4)||157||1.2 (0.9–1.7)|
|18.7–40.5||393||146||1.2 (0.9–1.6)||68||1.3 (0.8–1.9)||55||1.5 (0.9–2.3)||47||1.3 (0.8–2.1)||154||1.3 (0.9–1.7)|
|40.6–514.3||333||151||1.2 (0.9–1.7)||78||1.3 (0.9–2.0)||65||1.4 (0.9–2.2)||52||1.4 (0.9–2.3)||163||1.3 (0.9–1.7)|
|p for trend||0.226||0.114||0.076||0.055||0.197|
|Total meat cooked rare/medium/just done|
|0||831||301||1.0 (reference)||128||1.0 (reference)||111||1.0 (reference)||95||1.0 (reference)||316||1.0 (reference)|
|0.1–16.2||336||140||1.3 (1.0–1.7)||52||1.2 (0.8–1.7)||45||1.3 (0.9–1.9)||38||1.2 (0.8–1.8)||147||1.4 (1.1–1.8)|
|16.3–275.4||377||132||1.1 (0.8–1.4)||76||1.5 (1.1–2.1)||43||1.1 (0.7–1.6)||36||1.0 (0.6–1.5)||138||1.1 (0.8–1.4)|
|p for trend||0.726||0.019||0.730||0.864||0.665|
|Red meat cooked well-done or very well-done|
|0||521||159||1.0 (reference)||64||1.0 (reference)||51||1.0 (reference)||41||1.0 (reference)||168||1.0 (reference)|
|0.1–9.1||307||103||1.1 (0.8–1.4)||45||1.1 (0.7–1.7)||32||1.0 (0.6–1.7)||30||1.3 (0.8–2.1)||105||1.1 (0.8–1.4)|
|9.2–23.6||374||159||1.2 (0.9–1.6)||66||1.2 (0.8–1.8)||45||1.0 (0.6–1.5)||41||1.2 (0.7–1.9)||163||1.2 (0.9–1.5)|
|23.7–498.1||342||152||1.1 (0.8–1.4)||81||1.3 (0.9–1.9)||71||1.2 (0.8–1.9)||57||1.6 (1.0–2.5)||165||1.1 (0.8–1.4)|
|p for trend||0.687||0.200||0.289||0.074||0.830|
|Red meat cooked rare/medium|
|0||1076||388||1.0 (reference)||169||1.0 (reference)||133||1.0 (reference)||115||1.0 (reference)||405||1.0 (reference)|
|0.1–15.5||219||90||1.3 (1.0–1.7)||31||1.0 (0.6–1.5)||28||1.3 (0.8–2.0)||25||1.2 (0.8–2.0)||93||1.3 (1.0–1.7)|
|15.6–267.3||249||95||1.0 (0.8–1.3)||56||1.3 (0.9–1.9)||38||1.2 (0.8–1.8)||29||1.1 (0.7–1.7)||103||1.0 (0.8–1.4)|
|p for trend||0.745||0.121||0.336||0.685||0.655|
Table III presents the association of intake amounts of HCAs, BP and mutageneity with risk for colorectal polyps. High intake of 2-amino-I-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) was associated with increased risk for hyperplastic polyp (OR = 1.6 for highest vs. lowest quartiles of intake, ptrend = 0.036). Similarly, high 2-amino-3,4,8-trimethylimidazo [4,5] quinoxaline (DiMeIQx) intake was associated with increased risk for hyperplastic polyp (OR = 1.5 for highest vs. lowest quartiles of intake, ptrend = 0.038).
|Intake amounts||Controls (n)||Polyp types||Size of adenomatous polyps|
|Adenoma only||HPP only||Adenoma and HPP||Large (≥1 cm)||Small (<1 cm)|
|n||aOR (95% CI)2||n||aOR (95% CI)2||n||aOR (95% CI)2||n||aOR (95% CI)2||n||aOR (95% CI)2|
|0–15.3||429||137||1.0 (reference)||47||1.0 (reference)||31||1.0 (reference)||37||1.0 (reference)||130||1.0 (reference)|
|15.4–39.2||408||138||0.9 (0.7–1.3)||58||1.1 (0.7–1.7)||38||1.1 (0.7–1.9)||33||0.8 (0.5–1.4)||143||1.0 (0.8–1.4)|
|39.3–83.9||366||146||1.0 (0.7–1.3)||69||1.3 (0.9–2.0)||62||1.6 (1.0–2.6)||55||1.3 (0.8–2.1)||153||1.1 (0.8–1.4)|
|84.0–846.8||341||152||0.9 (0.7–1.2)||82||1.4 (0.9–2.1)||68||1.4 (0.8–2.2)||44||0.9 (0.5–1.5)||175||1.0 (0.7–1.4)|
|p for trend||0.569||0.179||0.331||0.792||0.904|
|0–78.6||414||144||1.0 (reference)||45||1.0 (reference)||40||1.0 (reference)||50||1.0 (reference)||133||1.0 (reference)|
|78.7–186.4||394||141||1.0 (0.8–1.3)||60||1.3 (0.8–2.0)||48||1.3 (0.8–2.1)||40||0.8 (0.5–1.3)||149||1.1 (0.9–1.5)|
|186.5–357.5||377||140||1.0 (0.8–1.3)||72||1.5 (1.0–2.3)||54||1.4 (0.9–2.1)||36||0.7 (0.5–1.2)||157||1.2 (0.9–1.6)|
|357.6–3,128.9||359||148||1.1 (0.8–1.4)||79||1.6 (1.1–2.4)||57||1.4 (0.9–2.1)||43||0.9 (0.5–1.4)||162||1.2 (0.9–1.6)|
|p for trend||0.475||0.036||0.318||0.645||0.332|
|0–1.03||417||139||1.0 (reference)||45||1.0 (reference)||42||1.0 (reference)||36||1.0 (reference)||144||1.0 (reference)|
|1.04–3.17||398||148||1.1 (0.8–1.4)||59||1.2 (0.8–1.8)||38||0.8 (0.5–1.4)||43||1.2 (0.7–1.9)||143||1.0 (0.7–1.3)|
|3.18–7.04||384||135||1.0 (0.7–1.3)||71||1.5 (1.0–2.3)||53||1.2 (0.8–1.9)||44||1.2 (0.7–2.0)||144||1.0 (0.7–1.3)|
|7.05–106.3||345||151||1.1 (0.8–1.4)||81||1.5 (1.0–2.4)||66||1.2 (0.8–1.9)||46||1.1 (0.7–1.9)||170||1.1 (0.8–1.4)|
|p for trend||0.837||0.038||0.170||0.804||0.520|
|Mutageneity (revertants colony)|
|0–3,963||425||131||1.0 (reference)||47||1.0 (reference)||40||1.0 (reference)||36||1.0 (reference)||134||1.0 (reference)|
|3,964–7,836||370||167||1.5 (1.1–1.9)||62||1.5 (1.0–2.3)||44||1.3 (0.8–2.0)||46||1.5 (0.9–3.4)||165||1.4 (1.1–1.9)|
|7,837–14,183||393||131||1.0 (0.8–1.4)||67||1.4 (0.9–2.1)||52||1.3 (0.8–2.1)||39||1.1 (0.7–1.8)||144||1.0 (0.8–1.4)|
|14,184–334,481||356||144||1.0 (0.8–1.4)||80||1.5 (1.0–2.3)||63||1.3 (0.8–2.0)||48||1.3 (0.8–2.0)||158||1.1 (0.8–1.4)|
|p for trend||0.498||0.132||0.421||0.711||0.611|
|0–9.84||424||134||1.0 (reference)||42||1.0 (reference)||43||1.0 (reference)||38||1.0 (reference)||138||1.0 (reference)|
|9.85–33.16||372||152||1.2 (0.9–1.5)||69||1.5 (1.0–2.3)||50||1.0 (0.6–1.6)||54||1.4 (0.9–2.1)||148||1.0 (0.8–1.4)|
|33.17–82.66||380||153||1.2 (0.9–1.6)||62||1.3 (0.9–2.1)||48||1.0 (0.6–1.7)||36||1.0 (0.6–1.6)||165||1.2 (0.9–1.6)|
|82.67–910.7||368||134||0.9 (0.7–1.3)||83||1.6 (1.1–2.5)||58||1.0 (0.6–1.6)||41||0.9 (0.6–1.6)||150||0.9 (0.7–1.3)|
|p for trend||0.345||0.098||0.883||0.339||0.538|
Since there were differences in the meat intake levels between men and women, separate analyses were done, stratified by sex, using the same intake categories with those presented in Tables II and III. The directions for associations were very similar between men and women (data not shown). Among women, statistically significant associations were observed between high red meat intake and concurrent adenoma and hyperplastic polyp risk (ptrend = 0.026), high PhIP intake and hyperplastic polyp risk (ptrend = 0.007), and high DiMeIQx intake and hyperplastic polyp risk (ptrend = 0.034). Sensitivity analysis, excluding subjects without FFQ information, was done with adjustments for total energy, calcium, vitamin D and folic acid intake. The directions of associations were almost the same as results for all study subjects, although statistical power was reduced due to the decreased sample size. For example, OR for each quartile of intake of PhIP on hyperplastic polyp risk were 1.4, 1.3 and 1.8 for the second, third and fourth quartiles, respectively, compared to the first quartile (ptrend = 0.028).
In our study using a questionnaire designed to assess meat intake by cooking method and doneness preference, an increased risk for colorectal hyperplastic polyps was observed among participants whose consumption levels of meat or meat mutagens were high. Well-done meat intake was also associated with marginally increased risk for large adenomatous polyps.
A systematic review of the association between meat and the risk of colorectal adenomas showed nonsignificantly increased risk for colorectal adenoma in subjects whose beef or processed meat consumption was high.19 Among case-control studies, 6 studies reported no association with meat (total meat, fresh meat, processed meat, beef, red meat or chicken and fish consumption depend on studies), 3 reported an increased risk of adenomatous polyps with high red meat or beef consumption, and 3 showed an inverse association with high consumption of meat or lean meat.19, 36, 37 The sample sizes for most these studies were, however, relatively small, ranging between 49 and 488 adenoma cases. In a cohort study including 170 cases of adenoma and 7,284 adenoma-free participants, red meat intake was positively associated with the risk of adenoma.38 In a recent report from the Prostate, Lung, Colorectal and Ovarian Cancer Project (PLCO), which included 3,696 distal adenoma patients and 34,817 controls, high intake of well-done red meat and bacon/sausage was associated with increased risk of colorectal adenoma.25 Among adenoma recurrence studies, the polyp prevention trial including 1,905 subjects did not find any evidence to suggest that reduction in total meat and red meat consumption might reduce the risk of multiple or advanced adenoma recurrence during a period of 4 years.20 Robertson et al. also reported no association between total red meat intake in relation to the risk of adenoma or advanced adenoma recurrence using data from 1,520 participants in two randomized trials for colorectal adenoma recurrence.21
The inconsistency on the association between meat intake and colorectal adenoma risk may be due to differences in meat cooking and consumption habits across study populations. Meat prepared using high temperature cooking methods contains more mutagens than meat which has been microwaved or boiled. High temperature cooking methods also increase doneness from both longer cooking time and higher cooking temperature, and have been related to higher levels of mutagens.11 In support of this hypothesis, some previous studies have observed an increased risk for colorectal adenoma with high intake of well-done red meat and grilled or barbecued red meat.22, 23 However, Tiemersma et al. reported no association between cooking habits, including temperature of heat source during browning, use of a lid, addition of water during browning, and preferred color of meat surface and colorectal adenoma risk, although total meat intake was associated with an increased risk of colorectal adenoma.27
It is possible that colorectal adenoma risk may be more closely associated with specific mutagens such as heterocyclic amines and PAHs. Colorectal adenoma risk has been evaluated in relation to dietary intake of carcinogens, including 2-amino-3,4,8-trimethylimidazo[4,5-f]quinoxaline (DiMeIQx), 2-amino3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx), PhIP, BP and mutagenic activity.12–14, 22, 24, 25 The results have been inconsistent. One study found that only MeIQx was significantly associated with adenoma risk,24 and another sigmoidoscopy-based study found that BP but not MeIQx, DiMeIQx or PhIP was associated with adenoma risk.22 The PLCO study found that MeIQx was associated with an increased risk for distal colon adenoma, and MeIQx and PhIP were associated with increased risk for nonadvanced distal adenoma.25
We found a marginally increased risk of large adenoma with high intake of total meat or red meat cooked well-done or very well-done. This result is consistent with Gunter et al., who reported that high intake of barbecued red meat and benzo[α]pyrene (BP) were associated with increased risk of large adenoma, and Kono et al., who reported increased risk for large adenoma (≥5 mm) with high meat intake.22, 39 The results from the PLCO study reported a positive association between red meat with known doneness level and the risk for advanced adenoma. However, strong associations between red meat intake and meat-related mutagens with single and nonadvanced adenomas were also observed.25
The meat intake levels in our study were comparable with those from the PLCO cancer screening trial; however, the HCAs and PAHs levels in our study were much higher compared to that study.25 The differences are mainly because of a higher percentage of subjects who preferred well-done meat in our population. Whereas approximately half of red meat intake was rare/medium meat in the PLCO trial, less than 30% of red meat intake was cooked rare/medium in our population.25 The difference could be due to differences in the meat questionnaire or could reflect real differences in meat cooking habits of the 2 study populations.
Hyperplastic polyps are commonly detected polyps in the large intestine and have generally not been regarded to have malignant potential. However, recent studies have suggested that some hyperplastic polyps may develop into cancer via a different pathway than the adenoma-carcinoma sequence.29 Although the concept of the adenoma-carcinoma sequence explains the majority of colorectal cancer, ∼10–15% of colorectal cancers show microsatellite instability (MSI) and do not share common molecular pathways with the adenoma-carcinoma sequence.29, 40 This group of colorectal cancers are characterized by defective nucleotide mismatch repair, such as aberrant DNA methylation.41 It has been shown that individuals with colorectal cancers showing MSI are more likely to harbor serrated or hyperplastic polyps than individuals with cancers not showing MSI.42 There are several epidemiological studies, which have shown that the risk factor profiles for MSI tumor and microsatellite stable (MSS) tumor are different.43–46 In a study of 276 colon cancer cases, patients with MSI colon cancers had significantly higher preference for well-done red meat and high frequencies of using certain cooking methods, such as frying, barbecuing, broiling and using meat drippings compared with MSS colon cancer patients.46
Few studies have compared risk factors between adenomatous and hyperplastic polyps.31, 32, 47 Risk factors were generally similar, although different strengths of associations for several risk factors were also observed. For example, cigarette smoking has been shown to have a stronger association with hyperplastic polyps than adenomatous polyps,31, 32, 47 and patients who had concurrent adenomatous and hyperplastic polyps showed similar risk with hyperplastic polyp only patients.31 Only one study has evaluated meat intake with hyperplastic polyp risk.32 Erhardt et al. reported increased risk for hyperplastic polyps but not for adenomatous polyps with ham and sausage intake.32 The authors selected ham and sausage as a measure of red meat. In our study, high meat intake showed stronger associations with hyperplastic polyp than adenomatous polyp risks, and participants who had both adenomatous and hyperplastic polyps showed a similar risk to hyperplastic polyp only group.
Strengths of our study include a large sample size, a detailed exposure assessment, and the use of a colonoscopy-based approach in classifying case-control status. Similar to other case-control studies, recall bias could be a concern for the study. However, all of these polyps are considered benign lesions; thus, the diagnosis of polyps is unlikely to have a major impact on lifestyle changes or recall of meat intake habits. Because participants from the VA clinic are mainly veterans, the difference in sex distribution is inevitable. However, we adjusted the analyses by study site as well as sex. Relatively low response rates for the FFQ could introduce a selection bias; however, the results excluding subjects without FFQ showed very similar associations with the results from all subjects who responded to the meat questions.
In conclusion, high intake of meat cooked well-done and HCAs formed during high-temperature cooking may increase the risk of developing colorectal polyps, particularly hyperplastic polyps, and the effect of meat intake differs by size of polyp for those with adenomatous polyps. Additional studies investigating differences in risk for hyperplastic and adenomatous polyps are needed to understand the etiology of these polyps as precursors of colorectal cancers with different molecular pathways.
We would like to thank research staff and study participants for their contribution to the study, Ms. Jane Curtin for technical assistance with carcinogen calculation, and Ms. Brandy Bentley for technical assistance in the preparation of this article.
- 21Fat, fiber, meat and the risk of colorectal adenomas. Am J Gastroenterol 2005; 100: 2789–95., , , , , ,Direct Link:
- 29Hyperplastic polyps, serrated adenomas, and the serrated polyp neoplasia pathway. Am J Gastroenterol 2004; 99: 2242–55., , ,Direct Link: