Colorectal cancer (CRC) is result of a complex interaction of variables, including external factors, such as environmental and dietary agents, or internal ones, such as somatic or hereditary disorders [1]. Some environmental or lifestyle factors, including lack of exercise and smoking, seem to act directly as carcinogens or increase somatic mutation frequency, predisposing or determining the appearance of tumour [2].

There is a clear association between dietary pattern and CRC, but a specific carcinogen has not yet been identified to fully justify this relationship [3]. High red meat intake and low consumption of vegetables, fruits and whole grains have been associated with increased risk [4-11], and low red meat intake and high consumption of vegetables, fruits and whole grains [11-13] or high consumption of chicken and fish [4], with decreased risk of CRC, respectively, in some, but not in all [14, 15] relevant cohort studies conducted in distinct ethnic populations.

Cancer is the second cause of death in the Southeast region of Brazil [16] and CRC has a high incidence and mortality in the country [17]. Conversely, the ethnic origin of the Brazilian population is highly heterogeneous, consisting of indigenous Amerindian populations and immigrants from Europe, Africa and Asia [18, 19].

Therefore, it seemed interesting to evaluate the influence of dietary patterns on the risk of sporadic colorectal adenocarcinoma (SCA) in an ethnically heterogeneous population.

SCA is the third most frequent malignant neoplasm in western developed countries, but it has a lower incidence in undeveloped ones [23]. In Brazil, the estimated risk considered is 13 cases/100.000 for men and 15 cases/100.000 for women. It ranks the third most frequent internal cancer in men in the Southeast region of the country, whereas in the North it is the nineth [24].

The differences observed in the incidence among countries and regions suggest that environmental and diet aspects play important roles in carcinogenesis. To our knowledge, there are no studies in Brazil analysing dietary factors in SCA patients.

In this study, SCA patients had an expressive increase of meat intake, in their different types (beef, pork and chicken). This finding is similar to previous reports, which associated dietary aspects, such as high intake of red meat and saturated fat with low ingestion of fruits, vegetables, and grains containing soluble fiber and folic acid, and the increase in occurrence of SCA [9-11]. Within this context it is interesting to emphasize that studies in several countries [4-11, 25, 26] found a relation between this pattern of high consumption of meat and increased occurrence of SCA. Norat et al. performed a study of dietary patterns in ten European countries and found an increase in relative risk of SCA related to high consumption of meat and an inverse association with consumption of fish [6]. The association was also observed by Rodrigo & Riestra and Sanz et al. in Spain [7, 27], and by Turner et al. in the United Kingdom [5].

In Japan, Lee et al. related the ‘westernization’ of dietary habits in that country to high intakes of meat and low fiber, and the increased incidence of SCA [8]. Conversely, Kimura et al. found no relationship between the consumption of meat and the occurrence of SCA in the Japanese population [14]. These authors observed that the high consumption of fish may be related to a lower risk of the disease. An analysis of 14 prospective studies in the U.S. and Europe, by Koushik et al., did not confirm this association with high intake of meat [15].

Comparing the intake patterns of vegetables, whole grains and fruits in patient and control groups, there was no difference, emphasizing that they were lower than those recommended in the Brazilian food pyramid (Fig. 1) criteria [22] in both groups. Voorrips et al. and Michels et al. observed similar results without a relationship between reduced risk of SCA and vegetable and fruit consumption [28, 29]. Similar results were found by Rodrigo & Riestra in Spain, and by Schneider et al. in South Africa [7, 30]. Other studies [31, 32] showed that high intake of vegetables and fruits do not decrease the SCA risk; however, very low intake may be associated with SCA.

Key et al. evidenced in a review that the consumption of at least 400 g/day of vegetables and fruits was associated with a decreased risk of SCA in the American population [25]. Similar results were observed by Turner et al. and Yeh et al. [5, 12]. A review of over 700 works by Ryan-Harshman & Aldoori [13] and studies developed by Turner et al., Pou et al., and Yeh et al. [5, 11, 12] also showed that the consumption of fruits and vegetables works as a protection factor.

The findings of this study show that the intake of red meat, chicken and pork in patients with colorectal cancer is, respectively, 1.3, 1.5 and 2.6-times higher than in the control group. Considering the total meat consumption, the frequency of servings/day in patients with SCA and in the control group is, respectively, 2.54 and 1.74. Since the amount of protein per serving considered was 120 g, the average intake of animal protein in patients and controls was 304.8 g vs. 208.8 g, respectively. There was no difference related to fish, probably because the consumption of this food in the study population was relatively low.

Some limitations of the study were not considering other risk factors, such as obesity, and not obtaining information about body mass index.

In this study, the dietary patterns of SCA patients were similar to the population from the North hemisphere and Australia [4-7, 9, 25, 26]. This can be explained by the great influence of the North American and European countries on Brazilian food habits. The characterization of dietary habits and its correlation with SCA incidence in a Brazilian population may be important in the development of education and prevention programs.

Meat consumption is greater in patients with SCA in the Brazilian population. Considering the study population – characterized by ethnic heterogeneity –, the environmental factor related to food habits may be associated with higher incidence of this disease in Brazil.