RESEARCH Meat Consumption, Animal Products, and the Risk of Bladder Cancer: A Case-Control Study in Uruguayan Men

In the time period 1996-2004, all incident cases of bladder cancer were included in a case-control study in order to study the role of meat consumption and product animals in the etiology of urothelial cancer. The study included 225 cases and 1,510 hospitalized controls with non-neoplastic conditions, not related to smoking and alcohol drinking. Relative risks, approximated by the odds ratios, were calculated in order to clarify the effect of meat consumption in the etiology of urothelial cancer. Total meat consumption (OR 1.47, 95% CI 1.02-2.11), total processed meat (OR 1.57, 95% CI 1.08-2.27), frankfurters (hot dogs) (OR 2.03, 95% CI 1.28-3.21), ham (OR 1.79, 95% CI 1.21-2.67) and salted meat (OR 2.73, 95% CI 1.78-4.18) were positively associated with risk of bladder cancer. Animal products, like cheese, whole milk, and total eggs were also associated with bladder cancer risk (OR for eggs 4.05, 95% CI 2.68-6.12). In conclusion, total meat, processed meat, and eggs could play an important role in the etiology of bladder cancer in Uruguay. in a


Introduction
Urinary bladder carcinoma (mainly transitional carcinoma) is a frequent malignancy among the male Uruguayan population (age-standardized incidence rate 23.4 per 100,000 men) (Barrios et al., 2010). According to the recent monograph by the World Cancer Research Fund/American Institute for Cancer Research (World Cancer Research Fund, 2007) there is limited evidence suggesting that milk and calcium protect against bladder cancer and that arsenic in drinking water is a cause of this malignancy. Moreover, according to this monograph (World Cancer Research Fund, 2007) meat intake is not a convincing cause of bladder cancer. Studies conducted in Iran (Rohani-Rasaf et al., 2013) and in the Pacific Islands (Moore et al., 2009) showed lower incidence rates of bladder cancer.
Conversely, a multisite case-control conducted in Uruguay (De Stefani et al., 2012) reported an increased risk of bladder cancer associated with processed meat intake. Also, the study by Tavani et al. (2000) showed an increased risk of bladder cancer associated with red meat consumption. Therefore, the relationship between red and processed meat consumption and bladder cancer deserves to be studied further.
For this reason we decided to conduct a case-control study in Uruguay, in order to explore the relationship

Selection of cases
In the time period 1996-2005, all newly diagnosed and microscopically validated cases of transitional-cell carcinoma of the urinary bladder occurring in men were considered eligible for this study. A total of 230 male cases were approached for a possible interview and five patients refused, leaving 225 cases for inclusion in the study (response rate 97.8%). All the cases were drawn from the four major public health hospitals in Montevideo, Uruguay.

Interviews and questionnaire
All the participants (cases and controls) were administered a structured questionnaire by two trained social workers shortly after admittance to the hospitals. All the interviews were performed face-to-face and proxy interviews were not allowed. The questionnaire included the following sections: sociodemographics (last name, first name, age, residence, education, monthly income, identification number), self-reported height and weight five years before the date of interview, family history of bladder cancer among first-degree relatives, a complete occupational history based on the last four jobs and their duration, a complete history of smoking (age at start, age at quit, number of cigarettes smoked per day, type of tobacco, type of cigarette, inhalation practices), a complete alcohol drinking history (age at start, age at quit, number of glasses per day or week, type of alcoholic beverage), a complete history of non-alcoholic beverages (mate, coffee, coffee with milk, tea, tea with milk, soft drinks), and a food frequency questionnaire (FFQ) on 64 food items. The FFQ recorded the consumption of foods in servings/years and the corresponding portion size per food. Therefore, we managed the consumption of foods in grams per day. The nutrients were calculated following a Table of chemical consumption of the foods. This FFQ allowed the estimation of total energy intake and was considered representative of the Uruguayan diet. Although it was not validated, it was tested for reproducibility with good results (Ronco et al., 2006).

Foods and nutrients included into the study
The following items were included in the study: total meat (beef, lamb, poultry, fish, liver, processed meat), red meat (beef, lamb), poultry, fish, total processed meat (bacon, sausage, mortadella, salami, saucisson, hot dog, ham, salted meat), liver, total vegetables and fruits, boiled eggs, fried eggs, total eggs, cheese, butter, whole milk, dietary cholesterol, sodium, nitrite, nitrate, nitroso compounds. All the food and nutrients were energy adjusted using the residuals method (Willett, 1998).

Statistical analysis
Relative risks of bladder cancer, approximated by the odds ratios, were estimated by unconditional multiple logistic regression (Rothman et al., 2008). We fitted the following model: age (continuous), residence (ordinal), education (categorical), tobacco smoking in pack years (categorical), alcohol drinking (categorical), mate consumption (categorical), total energy intake (continuous), total vegetables and fruits (continuous). This basic model further included several terms of risk factors for the study (meat intake, dairy foods, total eggs, and selected nutrients). All the estimates were two-sided and P value was considered as significant as alpha of 95%. All the calculations were estimated using the statistical software Stata, release 13.1 (Stata Corp, 2013).

Results
Distribution of cases and controls by sociodemographics and selected risk factors are shown in Table 1. Age, residence, education, and monthly income were rather similar among both groups of participants. Smoking and mate consumption showed significantly higher percentages among cases compared with controls.
The ORs of red meat for cooking methods are shown in Table 3. Barbecued red meat was positively associated with risk of bladder carcinoma (OR 1.96, 95% CI 1.34-2.87). Boiled red meat was also positively associated with bladder cancer (OR 1.52, 95% CI 1.06-2.19). Fried red meat was not associated with bladder cancer (OR 1.04, 95% 0.71-1.53).
The effect of animal products is shown in Table 4. Cheese consumption was protective (OR 0.60, 95% CI 0.41-0.86, P value for trend = 0.006) and whole milk was positively associated with bladder cancer (OR 1.45, 95% 1.01-2.08, P value for trend = 0.04). Eggs consumption (boiled eggs, fried eggs, and total eggs) displayed a high The effect of selected nutrients in bladder cancer etiology is shown in Table 5 Dietary cholesterol was strongly associated with risk of bladder cancer (OR 4.84, 95% 3.11-7.52, P value for linear trend <0.0001). Also higher risk was observed among sodium and NDMA (OR for the higher tertile vs the lower tertile of NDMA 2.14, 95% CI 1.43-3.18, P value trend = 0.0002). Nitrite intake was not associated with bladder cancer risk.

Discussion
According to the results of the present study, positive associations between meat consumption and bladder cancer risk were observed. Total meat, total processed meat, frankfurter (hot dog), ham, and salted meat showed a moderate to high risk increase of bladder cancer.
Several studies have reported results concerning meat intake and bladder cancer risk (La Vecchia and Negri, 1996;Wilkens et al., 1996;Tavani et al., 2000;Castelao et al., 2004;Wakai et al., 2004;Garcia-Closas et al., 2007;Hu et al., 2008;De Stefani et al., 2012). Associations from case-control studies of bladder cancer and meat intake have been somewhat inconsistent. A marginally significant 30% increased risk was reported for red meat and a significant 60% increased risk was reported for processed meat in a Canadian case-control study (Hu et al., 2008). Tavani et al. (2000) observed a significant 60% increased risk among the highest consumers of red meat in a study conducted in Italy. In contrast, Wakai et al. (2004) observed non-significant decreased risks for beef,     Alvaro Luis Ronco et al pork, ham and sausage, and Castelao et al. (2004) reported non-significant inverse associations for the highest intake levels of processed meat and preformed nitrosamine. In a recently published case-control study conducted in Spain, a non-significant odds ratio of 0.8 was reported for red meat intake and a non-significant odds ratio of 1.2 was reported for processed meat intake (Garcia-Closas et al., 2007). Finally, Wilkens et al. (1996) showed an increased risk of bladder cancer associated with bacon consumption. Two recent prospective studies reported conflictive results concerning processed meat consumption and bladder cancer risk (Michaud et al., 2006;Jakszyn et al., 2011). Whereas the EPIC prospective study reported null results (Jakszyn et al., 2011), the study by Michaud et al. showed an increased risk of bladder cancer associated with bacon intake (Michaud et al., 2006). A third prospective study on meat intake and risk of bladder cancer displayed again null results on the effect of meat consumption (Larsson et al., 2009).
Thus, both case-control and prospective studies showed conflicting results concerning meat intake and bladder cancer risk. A multisite case-control study conducted by Aune et al. (2009a) in Uruguay displayed a significant positive association between red meat and beef with bladder cancer. In a similar study based on a different population, red meat and beef were again associated with an increased risk of bladder cancer (Aune et al., 2009b). Both multisite studies showed a null association between processed meat and bladder cancer risk. Finally, another multisite study focused on salted meat and cancer showed a strong positive association between salted meat intake and risk of bladder cancer (De Stefani et al., 2009). Salted meat, also known as charque, is a food which is considered part of the group of processed meats and is characterized by a rich content in N-Nitroso compounds. It is thus noteworthy that the present study showed a strong positive association of charque (salted meat) with bladder cancer, replicating the findings of the previous study.
Eggs consumption and dietary cholesterol displayed elevated ORs for bladder cancer. The effect of both variables was, in fact, strongly and positively associated with bladder cancer and total eggs showed the highest risk of bladder cancer among foods studied in the final model. Previous studies (Balbi et al., 2001;Aune et al., 2009c;De Stefani et al., 2011;Hu et al., 2012;Fang et al., 2012) reported an increase in risk of bladder cancer for dietary cholesterol and total eggs consumption. Also, dairy foods were related with risk of bladder cancer. In our study, cheese intake was protective, whereas whole milk displayed a moderately positively association with bladder cancer risk. Our findings replicated the findings of Brennan et al. (2000), showing an inverse association of cheese with non-smokers in lung cancer.
As other case-control studies, the present study has strengths and limitations. The high response rates, both for cases and controls, is a major strength of the study. Selection bias and recall bias are the major limitations. Selection bias is unlikely since cases and controls were drawn from the same base population. On the other hand, recall bias is very difficult to manage, since this bias could result in misclassification, differential or non-differential, which could lead to erroneous estimates.
In conclusion, the present study showed that total meat, processed meat, hot dog, ham, salted meat, boiled eggs, fried eggs, total eggs, and whole milk were positively associated with increased risk of bladder cancer. These associations suggest that meat intake and eggs could be important risk factors in urothelial carcinoma of the urinary bladder.