Abstract
Background
Colorectal adenomas (CRA) are critical target precursors in the carcinogenesis of colorectal cancer (CRC), and the association of dietary factors, including fruits and vegetables, with CRA is yet to be clearly understood, especially in Asian populations. It is necessary to study the association of fruit and vegetable consumption with CRA to guide interventions for the primary prevention of CRA in the general population. Therefore, this study examined the relationship of fruit and vegetable consumption with CRA among asymptomatic adults in Korea.
Methods
We identified 1,658 (61.2%, men) adults without a history of CRC who underwent colonoscopy as part of routine health screening at the Seoul National University Hospital Healthcare System Gangnam Centre from May to December 2011. Trained gastroenterologists diagnosed low- and high-risk CRA, and dietary information was assessed by trained nutritionists using a validated food frequency questionnaire (FFQ). Fruit and vegetable consumption was estimated in g/day per 1,000 kcal and stratified into quartiles. Multivariable-adjusted polytomous regression models were used to estimate the odds ratio (OR) and 95% confidence interval (CI) of CRA by quartiles of fruits and vegetable consumption at a two-sided P<0.05.
Results
Overall, 536 (32.3%) presented with CRA. The multivariable-adjusted OR (95% CI) for odds of CRA by quartiles of total fruits and vegetable consumption were 1.00, 1.19 (0.85, 1.67), 1.23 (0.88, 1.72), and 1.02 (0.71, 1.46) (P for trend =0.87) for low-risk CRA and 1.00, 1.32 (0.81, 2.13), 0.76 (0.44, 1.31), and 0.58 (0.31, 1.06) (P for trend =0.02) for high-risk CRA. Similar but statistically insignificant trends were observed for vegetables or fruits individually.
Conclusions
Higher fruit and vegetable consumption is modestly associated with a lower odds of high-risk CRA, but longitudinal studies are necessary to validate these findings.
Keywords: Colorectal cancer (CRC), optimal diet, lifestyle modification, cancer prevention
Highlight box.
Key findings
• Higher fruit and vegetable consumption was modestly associated with lower odds of high-risk colorectal adenoma (CRA), with contradictory associations regarding fermented and salted vegetable consumption in the subgroup analyses by sex, smoking, and alcohol drinking.
What is known and what is new?
• CRA accounts for a large proportion of all colorectal cancer (CRC) events worldwide, and the role of diet in the pathophysiology of CRA manifestation keeps evolving.
• Our study evaluated the relationship of fruit and vegetable consumption (taking into account primary sources of vegetables and subgroups classifications based on culinary traditions and gastronomic presentations) with CRA among asymptomatic adults in Korea, with subgroup analyses by sex, smoking, and alcohol drinking.
What is the implication, and what should change now?
• Overall, our study advanced the understanding of the significance of fruit and vegetable consumption in the CRA manifestation, in addition to providing viable information to support the efforts of researchers, clinicians, and dieticians in providing nutrition counselling, advisories, and interventions for the primordial prevention of CRC.
Introduction
Colorectal cancer (CRC) is the third most conspicuously diagnosed malignancy and a prominent cause of cancer mortality (1). CRC can potentially evolve from dysplastic premalignant epithelial lesions known as colorectal adenoma (CRA) (2) through pathophysiological sequences in the adenoma-carcinoma cascade (3). Also, CRA accounts for a large proportion of all CRC events worldwide (4), and numerous adaptable lifestyle factors, including alcohol drinking, excessive weight gain, and poor physical activity, among others, are viable targets for the primary prevention of CRA, as a nexus for the timely prevention and management of CRC (5). For example, there is substantial evidence on the significance of healthy weight management, higher physical activity, and smoking cessation, among others, is protective against CRA and CRC manifestations (6,7). However, the role of diet in the pathophysiology of CRA manifestation remains unclear and continues to evolve. While several reports have documented the significance of diet in CRC events (1,6,7), efforts to identify the role of diet in the CRA are relatively minimal. For example, diets low in milk and calcium have been linked to CRC-related disability-adjusted life years worldwide (6). Similarly, a systematic review has summarised several dietary factors, including low whole-grain and high red meat consumption, in relation to CRC incidence (1).
Nevertheless, there is limited information on the significance of fruit and vegetable consumption in the manifestation of CRA, especially among Koreans. Fruits and vegetables are rich sources of multiple nutrients with diverse subgroups and categories based on culinary presentations that are worth exploring in determining the associations with CRA. Earlier studies have documented a discreetly inverse association of fruit and vegetable consumption with CRC (8-10). However, it remains unclear whether fruit and vegetable consumption could potentially affect CRA manifestation and, by extension, modulate interactions in the adenoma-carcinoma sequence to reduce CRC risk. Given that CRA is a precursor of CRC and approximately thirty-five per cent of CRC events are attributable to dietary factors among Koreans (11), research efforts targeting the role of diet in CRA manifestation would be promising in identifying dietary drivers of CRA. This information could help design public health advisories and recommendations to guide nutritional interventions, serving as a nexus that would significantly reduce the burden of CRC and, by implication, improve the health and quality of life of populations.
Some studies in this population have reported an association between diets (including fruit and vegetable consumption) and CRA among Koreans (12,13) without exploring the significance of culinary traditions and gastronomic presentations of fruits and vegetables in these associations. Similarly, these earlier efforts did not account for the implications of sex and lifestyle factors (including smoking and alcohol drinking) on the associations. It is essential to explore the association of fruit and vegetable consumption with CRA, taking into account subgroup analyses by sex, smoking, and alcohol drinking to clarify the impact of sex-related differences and common lifestyle factors such as smoking and alcohol drinking in these associations. Researching this phenomenon is necessary to contribute important information that ensures robust and multi-stakeholder integrated interventions for early health screening of CRA and its timely prevention and management, thereby offering an opportunity to halt the pathological progression leading to the onset of CRC along the adenoma-carcinoma continuum. Therefore, this study examined the relationship between fruit and vegetable consumption and CRA among asymptomatic adults in Korea. We present this article in accordance with the STROBE reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-456/rc).
Methods
Data source and study population
The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Seoul National University Hospital (No. H-2302-016-1401). Individual consent was obtained during the administration of the food frequency questionnaires (FFQs), and the requirement for consent for retrospective extraction of data from medical records was waived. De-identified participants’ data from the retrospective cohort reported elsewhere (12) were retrieved from a secured database. In all, 2,086 participants who had undergone colonoscopic screening at the Seoul National University Hospital Gangnam Healthcare Centre, Seoul, Korea, were recruited between May and December 2011. Overall, 1,658 participants were included in the final analysis (Figure S1) after excluding participants with missing data on CRA status (n=412), implausible energy consumption (n=5), and a history of CRC (n=11).
Diagnosis of CRA
Trained gastroenterologists, in accordance to standard protocol carried out colonoscopic examinations to evaluate participants for CRA. In addition, CRA subtypes were differentiated based on the magnitude of CRA progression and the anatomic location (proximal colon, distal colon, and rectum) of the polyps. Low-risk CRA was well-defined as at least one tubular adenoma with a diameter of <10 mm and low-grade dysplasia, but high-risk CRA was well-defined as villous histology or high-grade dysplasia or a diameter that was at least 10 mm or more than three adenomas in any anatomic location (13).
Dietary assessment
The food and dietary intake of participants were assessed by qualified nutritionists using a validated 106-item FFQ during the medical examination preceding the colonoscopy. The FFQ methodology and dietary estimation have been detailed elsewhere (12,14). Information on portion sizes and frequency of consumption of food and drink items in the last 12 months was tracked under nine options: ‘never’ to ‘three times/day’, with three portion size options: ‘one-half of a standard serving’, ‘one standard serving’, and ‘one and a half standard serving’. Fruit and vegetable consumption in g/day was estimated as a function of the frequency of intake and the corresponding reported amount. The primary sources of fruits are strawberries, melons, watermelons, peaches, plums, bananas, persimmons, dried persimmons, tangerines, pears, apples, apple juice, oranges, orange juice, and grapes, as well as grape juice. In addition, primary sources of vegetables and subgroups classifications based on culinary traditions and gastronomic presentations: fermented and salted vegetables (baechu-kimchi, radish-kimchi, nabak kimchi, dongchimi, other kimchi, pickled radish, and pickled garlic), and non-fermented and non-salted vegetables including; green vegetables (baechu, spinach, lettuce, perilla leaves, ssam greens/salads, other greens, pepper leaves, cham-namul, chwi-namul, mugwort, chives, water parsley, green pepper, and zucchini) and non-green vegetables (Deodeok, balloon flower root, bean sprouts, mung bean sprouts, Gosari, sweet potato stem, cucumber, carrots, onions, squash, pumpkin, and tomatoes). Furthermore, the Korean food composition table was applied to estimate energy consumption in kcal/day from food consumed (15). Fruit and vegetable consumption in g/day per 1,000 kcal was determined by dividing the consumption in grams by the total energy consumption in kcal/day, multiplied by 1,000, and stratified into quartiles for appropriate statistical comparison.
Determination of covariates
Participants provided sociodemographic and lifestyle information using questionnaires, and medical records were retrieved. Information on age (years), sex, education (middle school or less, high school, and university education/postgraduate), alcohol drinking (never, past, and current), smoking (never, past use, and current use), and family history of CRC (no and yes) were provided by participants. Alcohol intake in g/day was estimated from the FFQ as the total sum of ethanol weight. Physical activity in metabolic equivalent of tasks (METs min/week) was calculated as the average minutes and days spent on moderate, intense, or walking activities (16). Weight (kg), height, and waist circumference (WC; cm) were assessed by trained personnel using standard methods, and body mass index (BMI) was calculated as the ratio of weight (kg) to the square of height (m) (17,18). In-body blood pressure monitors (BPBIO320; In Body Co., Ltd., Seoul, Korea) were used to assess blood pressure [systolic blood pressure (SBP) and diastolic blood pressure (DBP)]. The mean of the measurements was used to determine the blood pressure in mmHg. Laboratory evaluations (using fasting serum samples collected after a 12-hour overnight fast) including glucose (mg/dL), glycated haemoglobin (%), total cholesterol (mg/dL), triglycerides (mg/dL), and high-density lipoprotein (HDL) cholesterol (mg/dL) were carried out using a clinical chemical analyzer (ARCHITECT c16000; Abbott Laboratories, Abbott Park, IL, USA), with coefficient of variation less than 1%. Low-density lipoprotein (LDL) cholesterol (mg/dL) was calculated using the Friedewald equation (19). Metabolic syndrome was defined according to the National Cholesterol Education Program Adult Treatment Panel III criteria (20) as at least three of the following conditions: (I) abdominal obesity, defined as a WC ≥90 cm (men) or ≥85 cm (women) (18); (II) hypertriglyceridemia, defined as triglycerides ≥150 mg/dL, or current use of lipid-lowering medications; (III) high blood pressure, defined as SBP ≥130 mmHg or DBP ≥85 mmHg, or the use of antihypertensive medications (21); and (IV) diabetes was defined as fasting blood glucose ≥100 mg/dL, or the use of blood glucose-lowering drugs (22).
Statistical analysis
Participants’ baseline characteristics by CRA status (non-cases vs. CRA cases) were presented using frequencies (percentages) for categorical data and mean ± standard deviation (SD) or median [interquartile range (IQR)] for continuous data based on the distribution of the data, and has published earlier in part elsewhere (23). The odds ratio (OR) and 95% confidence interval (CI) for CRA (including low- and high-risk) were estimated by quartile distribution of fruits and vegetable consumption (with the first quartile as reference) using polytomous logistic regression models, adjusting for relevant covariates, which include age (years, continuous), education (middle school or less, high school, and university education and postgraduate), smoking status (never, past, and current), alcohol intake (g/day, continuous), physical activity (METs min/week, continuous), BMI (kg/m2, continuous), metabolic syndrome (no and yes), family history of CRC (no and yes), total energy intake (kcal/day, continuous), and total meat consumption (red meat, processed meat, and poultry combined, g/day per 1,000 kcal, continuous). In the test for linear trend, the median of the quartile distribution of fruit and vegetable consumption in g/day per 1,000 kcal was assigned in a continuous model. Subgroup analyses by sex (men or women), smoking status (never or ever), and alcohol drinking (non-drinkers or current drinkers) were carried out, and the evidence of interaction was tested using the likelihood ratio test by equating nested models with cross-product terms alongside original models without the term. SAS 9.4 (SAS Institute Inc., Cary, NC, USA) software was used for the statistical analyses at a two-sided P<0.05. However, using Bonferroni’s correction, P values in the regression model were additionally adjusted for the likelihood of inflated type I error to control for multiple testing in the subgroup analyses.
Results
Characteristics of participants
Table 1 presents the characteristics of participants by CRA status (23). A higher proportion of men had CRA compared to women. In comparison with those without CRA, participants with CRA were older and had a higher prevalence of current smoking, current alcohol drinking, and family history of CRC. Participants with CRA presented with higher BMI, WC, SBP, DBP, fasting glucose, glycated haemoglobin, triglyceride, and LDL-cholesterol levels but lower HDL-cholesterol levels than those without CRA. These trends remained independent of sex (Table S1), but women with CRA presented with a generally higher family history of CRC compared to men with CRA.
Table 1. Characteristics of participants by CRA status (23).
| Characteristics of participants | Participants without CRA (n=1,122) | Participants with CRA (n=536) | P value |
|---|---|---|---|
| Age (years) | 49.6±8.7 | 54.1±8.6 | <0.001 |
| Female | 495 (44.1) | 148 (27.6) | <0.001 |
| Education | |||
| Middle school or less | 37 (3.5) | 24 (4.7) | 0.25 |
| High school | 149 (13.9) | 81 (15.9) | |
| University education & postgraduate | 888 (82.7) | 406 (79.5) | |
| Regular cigarette smoking | |||
| Never | 610 (54.4) | 232 (43.3) | <0.001 |
| Past | 302 (26.9) | 181 (33.8) | |
| Current | 210 (18.7) | 123 (23.0) | |
| Regular alcohol drinking | |||
| Never | 305 (27.7) | 117 (22.2) | 0.052 |
| Past | 79 (7.2) | 37 (7.0) | |
| Current | 716 (65.1) | 373 (70.8) | |
| Alcohol intake (g/day) | 30.4±39.9 | 39.2±45.4 | 0.001 |
| Physical activity (METs min/week) | 1,220.9±2,475.1 | 1,348.8±2,531.0 | 0.33 |
| BMI (kg/m2) | 23.2±3.0 | 24.2±2.8 | <0.001 |
| WC (cm) | 83.9±8.2 | 86.9±7.8 | <0.001 |
| SBP (mmHg) | 114.9±13.2 | 118.6±13.0 | <0.001 |
| DBP (mmHg) | 74.4±10.4 | 77.1±10.3 | <0.001 |
| High blood pressure | 363 (32.5) | 238 (44.9) | <0.001 |
| Fasting plasma glucose (mg/dL) | 93.1±14.5 | 98.9±19.8 | <0.001 |
| Glycated haemoglobin (%) | 5.6±0.4 | 5.8±0.7 | <0.001 |
| Diabetes | 255 (22.9) | 183 (34.9) | <0.001 |
| Total cholesterol (mg/dL) | 199.7±34.4 | 202.5±35.1 | 0.13 |
| Triglyceride (mg/dL) | 97.5±63.0 | 112.8±78.4 | <0.001 |
| HDL-cholesterol (mg/dL) | 52.9±11.2 | 50.8±10.8 | <0.001 |
| LDL-cholesterol (mg/dL) | 125.3±30.8 | 128.4±29.7 | 0.049 |
| Metabolic syndrome | 183 (16.3) | 140 (26.4) | <0.001 |
| CRC family history | 61 (17.8) | 37 (25.5) | 0.053 |
Data are presented as mean ± SD or n (%). Reproduced from reference (23). High blood pressure was defined as SBP ≥130 mmHg or DBP ≥85 mmHg or the use of antihypertensive drug treatment or a history of hypertension. Diabetes was defined as fasting blood glucose ≥100 mg/dL or using glucose-lowering medications. Metabolic syndrome was defined as at least three of any of the following conditions: elevated WC (men, ≥90 cm; women, ≥85 cm), elevated triglycerides (≥150 mg/dL or drug treatment for elevated triglycerides), reduced HDL-cholesterol (men, <40 mg/dL; women, <50 mg/dL), high blood pressure or diabetes. BMI, body mass index; CRA, colorectal adenoma; CRC, colorectal cancer; DBP, diastolic blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MET, metabolic equivalent of task; SBP, systolic blood pressure; SD, standard deviation; WC, waist circumference.
Total fruit and vegetable consumption and odds of CRA
The median (IQR) distribution of total fruit and vegetable consumption by quartiles of fruit and vegetable consumption in g/day per 1,000 kcal was 73.8 (47.6–93.2) in the first quartile, 142.2 (124.9–159.8) in the second quartile, 226.3 (201.9–257.9) in the third quartile, and 407.2 (340.3–510.6) in the fourth quartile (Table 2). The distribution of fruit and vegetable consumption (by subtypes of vegetables) in g/day per 1,000 kcal in the subgroups of sex, smoking status, and alcohol drinking is detailed in Table S2. For instance, the distribution of fruit and vegetable consumption in g/day per 1,000 calories by quartiles was generally higher among women and smokers. Similar trends were observed for fruits or vegetables individually.
Table 2. ORs and 95% CIs according to fruit and vegetable consumption associated with CRA among all participants.
| Food groups | Quartiles of intakes (g per 1,000 kcal/d) | ||||
|---|---|---|---|---|---|
| Q1 | Q2 | Q3 | Q4 | P for trend† | |
| Total fruit and vegetable | |||||
| Intake (g/day per 1,000 kcal), median (IQR) | 73.8 (47.6–93.2) | 142.2 (124.9–159.8) | 226.3 (201.9–257.9) | 407.2 (340.3–510.6) | |
| Intake (g/day), median (IQR) | 129.0 (83.3–174.6) | 257.3 (209.2–330.0) | 407.2 (330.6–520.9) | 734.2 (553.0–973.1) | |
| Low-risk adenoma | |||||
| No. of cases | 99 | 104 | 109 | 89 | |
| OR (95% CI) | 1.00 | 1.19 (0.85, 1.67) | 1.23 (0.88, 1.72) | 1.02 (0.71, 1.46) | 0.87 |
| High-risk adenoma | |||||
| No. of cases | 43 | 45 | 28 | 19 | |
| OR (95% CI) | 1.00 | 1.32 (0.81, 2.13) | 0.76 (0.44, 1.31) | 0.58 (0.31, 1.06) | 0.02 |
| All vegetables | |||||
| Intake (g/day per 1,000 kcal), median (IQR) | 51.3 (33.9–65.3) | 100.2 (89.9–108.8) | 144.8 (132.7–160.5) | 223.1 (193.4–278.4) | |
| Intake (g/day), median (IQR) | 91.3 (58.6–125.4) | 184.6 (148.4–229.3) | 258.1 (216.2–315.9) | 390.0 (324.2–532.5) | |
| Low-risk adenoma | |||||
| No. of cases | 94 | 107 | 98 | 102 | |
| OR (95% CI) | 1.00 | 1.19 (0.85, 1.67) | 1.06 (0.75, 1.49) | 1.17 (0.82, 1.65) | 0.54 |
| High-risk adenoma | |||||
| No. of cases | 44 | 31 | 29 | 31 | |
| OR (95% CI) | 1.00 | 0.83 (0.49, 1.41) | 0.72 (0.42, 1.24) | 0.77 (0.45, 1.31) | 0.30 |
| Fermented and salted vegetables | |||||
| Intake (g/day per 1,000 kcal), median (IQR) | 18.5 (9.8–23.5) | 44.9 (37.7–52.3) | 74.9 (67.1–84.9) | 126.5 (109.6–157.0) | |
| Intake (g/day), median (IQR) | 30.1 (17.3–43.6) | 78.7 (63.8–106.5) | 138.9 (111.3–171.2) | 227.9 (182.1–290.7) | |
| Low-risk adenoma | |||||
| No. of cases | 97 | 91 | 107 | 106 | |
| OR (95% CI) | 1.00 | 0.90 (0.64, 1.27) | 1.05 (0.75, 1.47) | 1.09 (0.78, 1.52) | 0.42 |
| High-risk adenoma | |||||
| No. of cases | 45 | 33 | 22 | 35 | |
| OR (95% CI) | 1.00 | 0.80 (0.48, 1.34) | 0.49 (0.28, 0.86) | 0.76 (0.46, 1.27) | 0.24 |
| Green leafy vegetables | |||||
| Intake (g/day per 1,000 kcal), median (IQR) | 5.8 (3.2–7.7) | 13.1 (11.1–15.3) | 22.6 (19.9–26.9) | 47.9 (37.0–65.7) | |
| Intake (g/day), median (IQR) | 10.4 (5.8–14.0) | 23.6 (18.5–30.5) | 41.2 (33.6–52.4) | 85.4 (65.1–116.1) | |
| Low-risk adenoma | |||||
| No. of cases | 96 | 102 | 112 | 91 | |
| OR (95% CI) | 1.00 | 1.08 (0.77, 1.51) | 1.21 (0.87, 1.69) | 0.95 (0.67, 1.35) | 0.65 |
| High-risk adenoma | |||||
| No. of cases | 37 | 30 | 30 | 38 | |
| OR (95% CI) | 1.00 | 0.91 (0.87, 1.56) | 0.89 (0.52, 1.54) | 1.12 (0.66, 1.89) | 0.54 |
| Non-green leafy vegetables | |||||
| Intake (g/day per 1,000 kcal), median (IQR) | 8.4 (5.0–11.2) | 20.8 (17.6–24.2) | 37.6 (32.2–44.4) | 83.3 (65.4–116.7) | |
| Intake (g/day), median (IQR) | 14.5 (8.8–21.7) | 37.8 (29.9–49.1) | 65.2 (52.0–82.8) | 149.8 (113.0–214.8) | |
| Low-risk adenoma | |||||
| No. of cases | 104 | 96 | 104 | 97 | |
| OR (95% CI) | 1.00 | 0.90 (0.64, 1.26) | 1.11 (0.79, 1.55) | 1.04 (0.73, 1.49) | 0.62 |
| High-risk adenoma | |||||
| No. of cases | 44 | 39 | 26 | 26 | |
| OR (95% CI) | 1.00 | 0.87 (0.53, 1.43) | 0.73 (0.42, 1.27) | 0.74 (0.41, 1.31) | 0.30 |
| All fruits | |||||
| Intake (g/day per 1,000 kcal), median (IQR) | 0.0 (0.0–0.0) | 24.8 (14.9–33.7) | 81.0 (60.1–103.1) | 213.5 (162.2–295.1) | |
| Intake (g/day), median (IQR) | 0.0 (0.0–0.0) | 43.3 (27.5–63.7) | 147.9 (113.0–194.0) | 369.4 (280.8–557.5) | |
| Low-risk adenoma | |||||
| No. of cases | 148 | 61 | 101 | 91 | |
| OR (95% CI) | 1.00 | 1.11 (0.77, 1.60) | 1.12 (0.82, 1.52) | 1.02 (0.74, 1.41) | 0.99 |
| High-risk adenoma | |||||
| No. of cases | 56 | 27 | 35 | 17 | |
| OR (95% CI) | 1.38 (0.81, 2.36) | 1.18 (0.73, 1.90) | 0.65 (0.35, 1.18) | 0.12 | |
Model was adjusted for sex (male and female), age (years, continuous), education (middle school or less, high school, and university graduate and postgraduate), smoking status (never, past, and current), alcohol intake (g/day, continuous), physical activity (METs min/week, continuous), BMI (kg/m2, continuous), metabolic syndrome (no and yes), CRC family history (no and yes), and total energy intake (kcal/day, continuous). †, P for trend was significant at Bonferroni corrected P value =0.008. BMI, body mass index; CI, confidence interval; CRA, colorectal adenoma; CRC, colorectal cancer; IQR, interquartile range; MET, metabolic equivalent of task; No., number; OR, odds ratio; Q, quartile.
Total fruit and vegetable consumption was unrelated to low-risk CRA (Table 2). However, the OR (95% CI) of high-risk CRA by increasing quartiles of total fruit and vegetable consumption was 1.00, 1.32 (0.81, 2.13), 0.76 (0.44, 1.31), and 0.58 (0.31, 1.06) (P for trend =0.02) after adjusting for age, education, smoking status, alcohol drinking, physical activity, BMI, metabolic syndrome, CRC family history, and total energy intake. The association remained even after additional adjustments for total meat consumption (Table S3).
Subgroup analysis by sex
The OR (95% CI) of high-risk CRA by increasing quartiles of total fruit and vegetable consumption was 1.00, 0.95 (0.52, 1.71), 1.00 (0.56, 1.78), and 0.52 (0.27, 1.01) (P for trend =0.05) among men and 1.00, 1.11 (0.34, 3.64), 1.48 (0.46, 4.75) and 0.73 (0.20, 2.28) (P for trend =0.63) among women, with no evidence of interaction by sex (P for interaction =0.20) (Table 3). The association remained even with additional adjustments for total meat consumption (Table S3). However, the odds of high-risk CRA by increasing quartile of fermented and salted vegetable consumption was 1.00, 0.63 (0.35, 1.12), 0.37 (0.19, 0.69), and 0.50 (0.28, 0.91) (P for trend =0.01) among men, but 1.00, 1.04 (0.26, 4.14), 1.13 (0.25, 5.14), and 3.10 (0.89, 10.75) (P for trend =0.02) among women (Table S4) with statistically significant evidence of interaction (P for interaction =0.004).
Table 3. ORs and 95% CIs according to total fruit and vegetable consumption associated with CRA by sex, smoking, and alcohol drinking.
| Subgroups of participants | Total fruit and vegetable consumption quartiles (g per 1,000 kcal/d) | P for interaction |
||||
|---|---|---|---|---|---|---|
| Q1 | Q2 | Q3 | Q4 | P for trend | ||
| Men | – | |||||
| Low-risk adenoma | ||||||
| No. of cases | 64 | 71 | 71 | 74 | ||
| OR (95% CI) | 1.00 | 1.22 (0.81, 1.85) | 1.18 (0.78, 1.79) | 1.20 (0.78, 1.82) | 0.54 | |
| High-risk adenoma | ||||||
| No. of cases | 33 | 27 | 30 | 18 | ||
| OR (95% CI) | 1.00 | 0.95 (0.52, 1.71) | 1.00 (0.56, 1.78) | 0.52 (0.27, 1.01) | 0.05 | |
| Women | 0.20 | |||||
| Low-risk adenoma | ||||||
| No. of cases | 33 | 26 | 36 | 26 | ||
| OR (95% CI) | 1.00 | 0.73 (0.40, 1.33) | 1.18 (0.68, 2.06) | 0.67 (0.37, 1.23) | 0.35 | |
| High-risk adenoma | ||||||
| No. of cases | 6 | 8 | 8 | 5 | ||
| OR (95% CI) | 1.00 | 1.11 (0.34, 3.64) | 1.48 (0.46, 4.75) | 0.73 (0.20, 2.28) | 0.63 | |
| Ever smoker | 0.95 | |||||
| Low-risk adenoma | ||||||
| No. of cases | 66 | 59 | 54 | 36 | ||
| OR (95% CI) | 1.00 | 0.95 (0.62, 1.46) | 1.09 (0.70, 1.71) | 0.97 (0.59, 1.61) | 0.99 | |
| High-risk adenoma | ||||||
| No. of cases | 30 | 34 | 15 | 10 | ||
| OR (95% CI) | 1.00 | 1.24 (0.70, 2.20) | 0.61 (0.31, 1.23) | 0.53 (0.24, 1.18) | 0.04 | |
| Never smoker | – | |||||
| Low-risk adenoma | ||||||
| No. of cases | 33 | 45 | 55 | 53 | ||
| OR (95% CI) | 1.00 | 1.73 (1.01, 2.99) | 1.52 (0.90, 2.56) | 1.22 (0.71, 2.07) | 0.88 | |
| High-risk adenoma | ||||||
| No. of cases | 13 | 11 | 13 | 9 | ||
| OR (95% CI) | 1.00 | 1.13 (0.46, 2.77) | 0.83 (0.34, 1.99) | 0.54 (0.21, 1.43) | 0.13 | |
| Alcohol current drinkers | 0.54 | |||||
| Low-risk adenoma | ||||||
| No. of cases | 72 | 79 | 78 | 52 | ||
| OR (95% CI) | 1.00 | 1.26 (0.86, 1.85) | 1.27 (0.86, 1.88) | 1.04 (0.67, 1.61) | 0.99 | |
| High-risk adenoma | ||||||
| No. of cases | 36 | 34 | 19 | 12 | ||
| OR (95% CI) | 1.00 | 1.27 (0.74, 2.18) | 0.66 (0.35, 1.23) | 0.54 (0.26, 1.13) | 0.03 | |
| Alcohol non-drinkers | – | |||||
| Low-risk adenoma | ||||||
| No. of cases | 27 | 25 | 31 | 37 | ||
| OR (95% CI) | 1.00 | 0.96 (0.50, 1.87) | 0.96 (0.51, 1.81) | 0.85 (0.46, 1.57) | 0.56 | |
| High-risk adenoma | ||||||
| No. of cases | 7 | 11 | 9 | 7 | ||
| OR (95% CI) | 1.00 | 1.70 (0.56, 5.10) | 1.24 (0.40, 3.90) | 0.76 (0.23, 2.53) | 0.37 | |
CI, confidence interval; CRA, colorectal adenoma; No., number; OR, odds ratio; Q, quartile.
Subgroup analysis by smoking status
The OR (95% CI) of high-risk CRA by increasing quartiles of total fruit and vegetable consumption was 1.00, 1.13 (0.46, 2.77), 0.83 (0.34, 1.99), and 0.54 (0.21, 1.43) (P for trend = 0.13) among never smokers, but 1.00, 1.24 (0.70, 2.20), 0.61 (0.31, 1.23), and 0.53 (0.24, 1.18) (P for trend =0.04) among ever smokers (Table 3). The association remained even with additional adjustment for total meat consumption, with no evidence of interaction (P for interaction =0.95) (Table S3). Furthermore, the odds of high-risk CRA by increasing quartile of fermented and salted vegetable consumption were 1.00, 1.13 (0.47, 2.72), 0.73 (0.26, 2.00), and 1.41 (0.59, 3.35) (P for trend =0.49) for never smokers, but 1.00, 0.58 (0.31, 1.10), 0.31 (0.16, 0.62), and 0.45 (0.23, 0.86) (P for trend =0.01) for ever smokers (Table S4), with a statistically significant evidence of interaction (P for interaction =0.03).
Subgroup analysis by current alcohol drinking status
The OR (95% CI) of high-risk CRA by increasing quartiles of total fruit and vegetable consumption was 1.00, 1.70 (0.56, 5.10), 1.24 (0.40, 3.90), and 0.76 (0.23, 2.53) (P for trend =0.37) among non-drinkers, but 1.00, 1.27 (0.74, 2.18), 0.66 (0.35, 1.23), and 0.54 (0.26, 1.13) (P for trend =0.03) among current alcohol drinkers adjusting for similar covariates (Table 3). The association remained even with additional adjustment for total meat consumption, with no evidence of interaction (P for interaction =0.54) (Table S3). Furthermore, the odds of high-risk CRA by increasing quartile of fermented and salted vegetable consumption were 1.00, 0.92 (0.29, 2.92), 0.93 (0.27, 3.13), and 1.96 (0.67, 5.78) (P for trend =0.15) non-drinkers, but 1.00, 0.75 (0.42, 1.34), 0.40 (0.21, 0.78), and 0.53 (0.29, 0.97) (P for trend =0.02) among current alcohol drinkers, with a statistically significant evidence of interaction (P for interaction =0.04) (Table S4).
Discussion
In this study, we found that higher fruit and vegetable consumption was modestly and suggestively associated with lower odds of high-risk CRA, with similar trends in some cases, although the association was statistically insignificant. Furthermore, we observed contradictory associations for the consumption of fermented and salted vegetables, with significant evidence of an interaction. Higher consumption of fermented and salted vegetables was associated with higher odds of high-risk CRA among women but not among men. Similarly, higher fermented and salted vegetable consumption was associated with lower odds of high-risk CRA among smokers and current alcohol drinkers but not among those who have never smoked or non-alcohol drinkers, respectively. Most of these subgroup analyses were primarily exploratory and did not reach the statistically significant thresholds for the Bonferroni-corrected inflated type I error.
In tandem with our modest findings, higher fruit and vegetable consumption has been linked with lower odds of CRA among participants in the PLCO Cancer Screening Trial (9). Similarly, a meta-analysis of 22 longitudinal studies involving 11,696 CRA events revealed that higher consumption of fruits and vegetables was associated with CRA events (24). The general observation with most of these studies (24-27) suggests that higher consumption of fruit, more than vegetables, drives the protective inverse association with lower odds of CRA. According to the same meta-analysis (24), a 100-gram increment in fruit consumption was associated with a statistically significant 6% reduction in CRA risk. The same pooled analysis suggested that a 100-unit gram increment in vegetable consumption was associated with a statistically insignificant 2% reduction in CRA risk.
Fruits and vegetables are rich sources of cancer-inhibiting phytochemicals and antioxidants, which are particularly viable, especially at the early stages of tumorigenesis. For example, sulforaphane, primarily found in vegetables, has been reported to inhibit the proliferation of colon cell cancer (28,29), but a large randomized trial among 1,905 subjects (with excised histologically confirmed CRA within 6 months suggested no difference in the rate of CRA recurrence) on low fat and higher fibre diet with a 3.5 serving/1,000 kcal of fruit and vegetable intakes after 4 years (30). Similarly, 1mg folic acid supplementation daily does not reduce CRA risk over a 3-year treatment period (31), contrary to the antineoplastic potentials reported in in vitro studies (31). Second, variations in intestinal microbial communities might account for the CRA tumourigenesis. A pyrosequencing study of the V1–V3 regions of the 16S rRNA gene using faecal microbiota of subjects on a higher vegetable diet suggests substantially lesser yields of short-chain fatty acids [such as acetic acid, butyric acid, and propionic acid known to prevent CRC tumorigenesis (32,33)] among those with advanced CRA compared to healthy subjects (34).
In this study, higher fruit and vegetable consumption was modestly associated with lower odds of high-risk CRA, although the association was statistically insignificant, and the odds were lower among men than among women. However, our findings suggested that higher consumption of fermented and salted vegetables was associated with higher odds of high-risk CRA among women but not among men. There are reasons for the contradictory association of fermented and salted vegetables and CRA by sex. First, the consumption of fermented and salted vegetables was higher among women than men in this study (Table S2). Higher sodium and salted food consumption have been linked with discreetly higher odds of CRC in prospective cohorts from a Japanese population (35) and the China Kadoorie Biobank (36). Also, it is worth noting that the average intake of salted vegetables is higher among men (186.8 g/day) than women (146.3 g/day) aged 20 years above in the Korea National Health and Nutrition Examination Survey datasets (2001, 2005, and 2007–2018), where salted vegetables account for almost half (16% out of 34.9% attributable to dietary factors) of CRC events in 2018 (11). Second, sex-related disparities have been reported in CRC tumorigenesis (37,38), especially with the documented historical under-representation of women in most cancer studies, thereby necessitating extrapolation from disproportionately high male-dominated reports (38). Third, in vitro studies of female mouse models revealed estrogen-related insulin-mediated apoptosis in the mucosal linings of the colon (39), which may account for sex-associated differences in this association.
Additionally, we observed that higher consumption of fermented and salted vegetables was linked with lower odds of CRA among smokers and current alcohol drinkers, but linked with higher CRA odds among never smokers and non-alcohol drinkers. On the one hand, higher consumption of fermented and salted vegetables with the absence of smoking and alcohol drinking was linked with suggestive higher odds of CRA. In tandem with this finding, salted vegetables have been linked with digestive cancers independent of smoking status (25). On the other hand, higher consumption of fermented and salted vegetables was associated with lower CRA odds, with evidence of interaction by smoking and alcohol drinking. The reason for these differences is not apparent. Still, the case for how lifestyle factors, such as smoking, influence causal interpretations in health outcomes has been established (39), necessitating a cautious interpretation of these findings. However, the following explanations hold sway. First, we observed that smokers and current alcohol drinkers generally presented lower consumption of fermented and salted vegetables in this sample (Table S2). Second, it is worth noting that the composite make-up of cigarette smoke potentially exerts epigenetic imprints that modulate immune function to suppress anti-tumour immunity, thereby hampering adaptive innate response to a cascade of activities leading to CRC events (40). Alluding to this observation and contrary to ours is the lack of a statistically significant inverse association between vegetable consumption and CRC by smoking status (41). In all, the suggestive inverse association of higher consumption of fermented and salted vegetables with CRA odds among smokers and alcohol drinkers in this current study should be interpreted with caution, considering the potential presence of residual or unknown confounding factors.
Furthermore, it remains inconclusive to determine which of the fruit/vegetable subtypes drives the inverse association with high-risk CRA, given the statistically insignificant associations in this current study, which warrants explanation. First, the consumption of these fruit/vegetable subtypes is generally low, especially among men in this current study. Second, we hypothesize that the consumption of fermented and salted vegetables attenuates these associations, given the suggestive positive association with high-risk CRA observed in the subgroup analyses, especially among never smokers, non-alcohol drinkers, and women. In tandem with this hypothesis, a recent systematic review and meta-analysis of seven observational studies has suggested an indirect positive association between high salt intake from an “unhealthy” dietary pattern and a high risk of CRA (42). Similarly, higher dietary sodium from salt intake can disrupt gut microbial homeostasis and alter the abundance of microbiota in the gastrointestinal tract, triggering inflammation (43).
In light of the suggestive modest inverse association of fruit and vegetable consumption with high-risk CRA, these findings, in addition to earlier reports from other populations, substantiate the necessity for early and timely dietary interventions to improve CRA prognosis and avert the risk of CRC potentially. Similarly, promoting regular consumption of fruits and vegetables is likely a viable dietary intervention for the primordial prevention of CRC. However, our findings must be interpreted in the context of the study’s limitations. The quartile distribution of intake in g/day per 1,000 kcal was presented for comparison of fruit and vegetable consumption, adjusted for calorie intake, to discern the distribution of consumption. We also presented the quartiles of absolute intake (g/day) in Table 2. The threshold of consumption of fruit and vegetables to directly inhibit CRA manifestation remains unclear. A causal inference could not be inferred due to the study’s cross-sectional design. Measurement errors are inherent in this current study, as with most dietary assessment studies. Residual or unknown confounding due to inherently unknown factors is likely, and participants were primarily recruited from a single hospital, among others. Additionally, the subgroup analyses (especially among women, never smokers, and non-alcohol drinkers) were primarily exploratory and did not reach the statistically significant thresholds for the Bonferroni-corrected inflated type I error. This current study did not investigate whether these associations vary according to a family history of CRC, as the distribution of participants with this information was relatively small and did not significantly impact the analysis of fruit and vegetable consumption. It is imperative to consider this phenomenon to guide interventions, including screening and clinical advice for the prevention and management of CRA. However, these limitations might be minimized as a result of the unarguably large sample size, the completion of the FFQ before colonoscopy screening, the gastroenterologist’s diagnosis of CRA based on clinical evaluation and colonoscopy, and multivariable adjustment for potential covariates, among other factors. Further studies are necessary to substantiate these associations in other populations of Asian ancestry.
Overall, this study advanced understanding of the significance of fruit and vegetable consumption in the context of CRC manifestation, while also providing viable information to support the efforts of researchers, clinicians, and dietitians in providing nutrition counselling, advisories, and interventions for the primordial prevention of CRC. However, longitudinal studies with timely repeated measurements of dietary factors, including fruit and vegetable consumption, would be promising in determining whether longitudinal increases in fruit/vegetable intake reduce the recurrence of CRA or progression to CRC.
Conclusions
Although higher fruit and vegetable consumption is modestly associated with lower odds of high-risk CRA, caution should be exercised when consuming fermented and salted vegetables, given their unclear association with CRA. Furthermore, longitudinal studies and randomized trials are necessary to clarify these associations.
Supplementary
The article’s supplementary files as
Acknowledgments
We thank the staff and volunteers of this study for their commitment and dedication.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of the Seoul National University Hospital (No. H-2302-016-1401), and individual consent was obtained during the administration of the food frequency questionnaires and the requirement for consent for retrospective extraction of data from medical records was waived.
Footnotes
Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-456/rc
Funding: None.
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-456/coif). A.P.O. was supported by the Brain Pool Program through the National Research Foundation of Korea, which is funded by the Ministry of Science and ICT (No. 2020H1D3A1A04081265). The other authors have no conflicts of interest to declare.
Data Sharing Statement
Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-456/dss
References
- 1.Keum N, Giovannucci E. Global burden of colorectal cancer: emerging trends, risk factors and prevention strategies. Nat Rev Gastroenterol Hepatol 2019;16:713-32. 10.1038/s41575-019-0189-8 [DOI] [PubMed] [Google Scholar]
- 2.Lee V, Ng SB, Salto-Tellez M. New techniques. In: Gray W, Kocjan G, editors. Diagnostic Cytopathology. 3rd ed. Edinburgh: Churchill Livingstone; 2010:891-902. [Google Scholar]
- 3.Siegel RL, Wagle NS, Cercek A, et al. Colorectal cancer statistics, 2023. CA Cancer J Clin 2023;73:233-54. 10.3322/caac.21772 [DOI] [PubMed] [Google Scholar]
- 4.Conteduca V, Sansonno D, Russi S, et al. Precancerous colorectal lesions (Review). Int J Oncol 2013;43:973-84. 10.3892/ijo.2013.2041 [DOI] [PubMed] [Google Scholar]
- 5.Corley DA, Jensen CD, Marks AR, et al. Adenoma detection rate and risk of colorectal cancer and death. N Engl J Med 2014;370:1298-306. 10.1056/NEJMoa1309086 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Global, regional, and national burden of colorectal cancer and its risk factors, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Gastroenterol Hepatol 2022;7:627-47. 10.1016/S2468-1253(22)00044-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Zhou E, Rifkin S. Colorectal Cancer and Diet: Risk Versus Prevention, Is Diet an Intervention? Gastroenterol Clin North Am 2021;50:101-11. 10.1016/j.gtc.2020.10.012 [DOI] [PubMed] [Google Scholar]
- 8.Kunzmann AT, Coleman HG, Huang WY, et al. Fruit and vegetable intakes and risk of colorectal cancer and incident and recurrent adenomas in the PLCO cancer screening trial. Int J Cancer 2016;138:1851-61. 10.1002/ijc.29922 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Millen AE, Subar AF, Graubard BI, et al. Fruit and vegetable intake and prevalence of colorectal adenoma in a cancer screening trial. Am J Clin Nutr 2007;86:1754-64. 10.1093/ajcn/86.5.1754 [DOI] [PubMed] [Google Scholar]
- 10.Randi G, Edefonti V, Ferraroni M, et al. Dietary patterns and the risk of colorectal cancer and adenomas. Nutr Rev 2010;68:389-408. 10.1111/j.1753-4887.2010.00299.x [DOI] [PubMed] [Google Scholar]
- 11.Cho HJ, Woo HD, Park S, et al. Gastric and colorectal cancer incidence attributable to dietary factors in Korea. J Gastrointest Oncol 2024;15:963-73. 10.21037/jgo-24-10 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Chung GE, Youn J, Kim YS, et al. Dietary patterns are associated with the prevalence of nonalcoholic fatty liver disease in Korean adults. Nutrition 2019;62:32-8. 10.1016/j.nut.2018.11.021 [DOI] [PubMed] [Google Scholar]
- 13.Hassan C, Quintero E, Dumonceau JM, et al. Post-polypectomy colonoscopy surveillance: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy 2013;45:842-51. 10.1055/s-0033-1344548 [DOI] [PubMed] [Google Scholar]
- 14.Kim S, Song S, Kim YS, et al. The association between predicted inflammatory status and colorectal adenoma. Sci Rep 2020;10:2433. 10.1038/s41598-020-59271-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ministry of Health and Wellfare, The Korean Nutrition Society. 2020 Dietary Reference Intakes for Koreans. Available online: https://kns.or.kr/kdris
- 16.Ainsworth BE, Haskell WL, Herrmann SD, et al. 2011 Compendium of Physical Activities: a second update of codes and MET values. Med Sci Sports Exerc 2011;43:1575-81. 10.1249/MSS.0b013e31821ece12 [DOI] [PubMed] [Google Scholar]
- 17.Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet 2004;363:157-63. 10.1016/S0140-6736(03)15268-3 [DOI] [PubMed] [Google Scholar]
- 18.Lee SY, Park HS, Kim DJ, et al. Appropriate waist circumference cutoff points for central obesity in Korean adults. Diabetes Res Clin Pract 2007;75:72-80. 10.1016/j.diabres.2006.04.013 [DOI] [PubMed] [Google Scholar]
- 19.Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18:499-502. [PubMed] [Google Scholar]
- 20.Alberti KGMM, Eckel RH, Grundy SM, et al. Harmonizing the Metabolic Syndrome. Circulation 2009;120:1640-5. 10.1161/CIRCULATIONAHA.109.192644 [DOI] [PubMed] [Google Scholar]
- 21.Chobanian AV, Bakris GL, Black HR, et al. Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension 2003;42:1206-52. 10.1161/01.HYP.0000107251.49515.c2 [DOI] [PubMed] [Google Scholar]
- 22.Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med 1998;15:539-53. [DOI] [PubMed] [Google Scholar]
- 23.Kim YS, Okekunle AP, Yang SY, et al. Fish and meat intake in relation to colorectal adenoma in asymptomatic Korean adults. Front Nutr 2024;11:1432647. 10.3389/fnut.2024.1432647 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ben Q, Zhong J, Liu J, et al. Association Between Consumption of Fruits and Vegetables and Risk of Colorectal Adenoma: A PRISMA-Compliant Meta-Analysis of Observational Studies. Medicine (Baltimore) 2015;94:e1599. 10.1097/MD.0000000000001599 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wu S, Wu Y, Hu X, et al. Fruit but not vegetable consumption is beneficial for low prevalence of colorectal polyps in a high-risk population: findings from a Chinese Lanxi Pre-colorectal Cancer Cohort study. Eur J Nutr 2024;63:1759-69. 10.1007/s00394-024-03377-z [DOI] [PubMed] [Google Scholar]
- 26.Tantamango YM, Knutsen SF, Beeson WL, et al. Foods and food groups associated with the incidence of colorectal polyps: the Adventist Health Study. Nutr Cancer 2011;63:565-72. 10.1080/01635581.2011.551988 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wu H, Dai Q, Shrubsole MJ, et al. Fruit and vegetable intakes are associated with lower risk of colorectal adenomas. J Nutr 2009;139:340-4. 10.3945/jn.108.098889 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Nishikawa T, Tsuno NH, Okaji Y, et al. Inhibition of autophagy potentiates sulforaphane-induced apoptosis in human colon cancer cells. Ann Surg Oncol 2010;17:592-602. 10.1245/s10434-009-0696-x [DOI] [PubMed] [Google Scholar]
- 29.Frydoonfar HR, McGrath DR, Spigelman AD. Sulforaphane inhibits growth of a colon cancer cell line. Colorectal Dis 2004;6:28-31. 10.1111/j.1463-1318.2004.00488.x [DOI] [PubMed] [Google Scholar]
- 30.Schatzkin A, Lanza E, Corle D, et al. Lack of effect of a low-fat, high-fiber diet on the recurrence of colorectal adenomas. Polyp Prevention Trial Study Group. N Engl J Med 2000;342:1149-55. 10.1056/NEJM200004203421601 [DOI] [PubMed] [Google Scholar]
- 31.Kim YI. Role of folate in colon cancer development and progression. J Nutr 2003;133:3731S-9S. 10.1093/jn/133.11.3731S [DOI] [PubMed] [Google Scholar]
- 32.D'Argenio G, Cosenza V, Delle Cave M, et al. Butyrate enemas in experimental colitis and protection against large bowel cancer in a rat model. Gastroenterology 1996;110:1727-34. 10.1053/gast.1996.v110.pm8964397 [DOI] [PubMed] [Google Scholar]
- 33.Lu R, Wang X, Sun DF, et al. Folic acid and sodium butyrate prevent tumorigenesis in a mouse model of colorectal cancer. Epigenetics 2008;3:330-5. 10.4161/epi.3.6.7125 [DOI] [PubMed] [Google Scholar]
- 34.Chen HM, Yu YN, Wang JL, et al. Decreased dietary fiber intake and structural alteration of gut microbiota in patients with advanced colorectal adenoma. Am J Clin Nutr 2013;97:1044-52. 10.3945/ajcn.112.046607 [DOI] [PubMed] [Google Scholar]
- 35.Takachi R, Inoue M, Shimazu T, et al. Consumption of sodium and salted foods in relation to cancer and cardiovascular disease: the Japan Public Health Center-based Prospective Study. Am J Clin Nutr 2010;91:456-64. 10.3945/ajcn.2009.28587 [DOI] [PubMed] [Google Scholar]
- 36.Yu W, Ke Y, Lv J, et al. Preserved vegetable consumption and gastrointestinal tract cancers: A prospective study. J Glob Health 2024;14:04191. 10.7189/jogh.14.04191 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Kim SE, Paik HY, Yoon H, et al. Sex- and gender-specific disparities in colorectal cancer risk. World J Gastroenterol 2015;21:5167-75. 10.3748/wjg.v21.i17.5167 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Baraibar I, Ros J, Saoudi N, et al. Sex and gender perspectives in colorectal cancer. ESMO Open 2023;8:101204. 10.1016/j.esmoop.2023.101204 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Poisson T, Dallongeville J, Evans A, et al. Fruit and vegetable intake and smoking cessation. Eur J Clin Nutr 2012;66:1247-53. 10.1038/ejcn.2012.70 [DOI] [PubMed] [Google Scholar]
- 40.Fujiyoshi K, Chen Y, Haruki K, et al. Smoking Status at Diagnosis and Colorectal Cancer Prognosis According to Tumor Lymphocytic Reaction. JNCI Cancer Spectr 2020;4:pkaa040. 10.1093/jncics/pkaa040 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Hori M, Sawada N, Kito K, et al. Vegetable and fruit intake and colorectal cancer risk by smoking status in adults: The Japan Public Health Center-based Prospective Study. Eur J Clin Nutr 2023;77:255-63. 10.1038/s41430-022-01214-2 [DOI] [PubMed] [Google Scholar]
- 42.Godos J, Bella F, Torrisi A, et al. Dietary patterns and risk of colorectal adenoma: a systematic review and meta-analysis of observational studies. J Hum Nutr Diet 2016;29:757-67. 10.1111/jhn.12395 [DOI] [PubMed] [Google Scholar]
- 43.Smiljanec K, Lennon SL. Sodium, hypertension, and the gut: does the gut microbiota go salty? Am J Physiol Heart Circ Physiol 2019;317:H1173-82. 10.1152/ajpheart.00312.2019 [DOI] [PMC free article] [PubMed] [Google Scholar]