Background

In the United States, an estimated 70,530 new cases of bladder cancer will be diagnosed and approximately 14,680 deaths were attributed to bladder cancer in 2010 [1]. Bladder cancer is the most expensive cancer to survey and treat because of the need for frequent interval cystourethroscopy, urine cytology and radiological evaluations [2]. Therefore, more and more attention has been given to chemoprevention. Cancer chemoprevention is defined as the use of natural, synthetic, or biologic chemical agents to reverse, suppress, or prevent carcinogenic progression to invasive cancer. Bladder cancer has a protracted course of progression and may be ideal for chemoprevention strategies [3].

Tea, derived from the plant Camellia sinensis, is one of the most common beverages consumed worldwide, especially in China. Multiple lines of evidence support a protective effect of tea on various cancers [4]. The emperor of China, ShenNung, is credited with first describing the therapeutic effects of tea in 2737 BC [5]. Studies conducted on cell-culture systems and animal models show that tea or the active ingredient in tea, polyphenols, could afford protection against a variety of cancer types [4]. However, the results based on epidemiological studies on the association of tea consumption with bladder cancer were inconsistent. A meta-analysis conducted in 2001 suggested the consumption of tea seems not to be related to an increased risk of urinary tract cancer [6].

The purpose of the present study was to update and quantitatively assess the association between tea consumption and the risk of bladder cancer by summarizing the results of published cohort and case-control studies. We also sought to address the unresolved issue of whether this relationship differs across the tea type.

Results

In total, we identified 35 papers examining the risk of bladder cancer with tea consumption published between 1966 and December 2011, and these were reviewed by 2 authors. We found no relevant non-English language papers in this field. Six studies were excluded because of insufficient information to compute its relative risk (RR) and 95% confidence interval (CI) [7–9], or a summary odds ratio (OR) adjusted for at least age, sex and smoking [10–12]. Six studies were excluded because they were found to be subsets of other studies or have overlapping data and were excluded [13–18]. Thus, 23 studies [19–41] were included in the meta-analysis on the association of tea consumption with bladder cancer risk. There were six cohorts [26, 29, 31, 33, 38, 40] and seventeen case–control studies; nine of these were population-based [19–21, 23, 27, 30, 34, 37, 41] and eight were hospital-based case-control studies [22, 24, 25, 28, 32, 35, 36, 39]. Of the twenty-three studies, seventeen were conducted in Western countries [17, 20–27, 29, 30, 32–34, 36, 37, 40, 41], and six were conducted in Asia [19, 28, 31, 35, 38, 39]. Only eight articles reported association between consumption of specific tea types (green or black tea) and the risk of bladder cancer [19, 23, 25, 26, 31, 35, 38, 39]. Information on tea consumption was obtained by interview, self-administered questionnaire or both techniques.

Table 1 presents the basic characteristics of each study included in our meta-analysis. Of the seventeen case-control studies, most studies found no significant association between tea consumption and bladder cancer, whereas four studies reported significantly increased risks [20, 28, 36, 41], and three found inverse associations [19, 30, 35]. Of the six cohort studies, five reported no significant association between tea consumption and bladder cancer [26, 29, 31, 38, 40], whereas one found significantly decreased risk [33].

Table 1 Study characteristics of published cohort and case-control studies on tea consumption and bladder cancer
Full size table

Figure 1 plots the pooled risk estimates for overall tea consumption by study design. When all these studies were analyzed together, no association was observed for tea consumption with bladder cancer (OR 0.94, 95% CI 0.85, 1.04) and the summary ORs were similar across study design and source of the controls in case-control studies. No statistically significant heterogeneity was observed when all the studies were analyzed together. However, we noted some heterogeneity in the population-based case-control studies (I2 58.2%, P = 0.014). After excluding one study by Slattery et al. [41], which reported the highest point estimates, the P-value for heterogeneity in the subgroup was no longer statistically significant (I2 15.2%, P 0.311), and the summary OR was not significantly changed (OR 0.94, 95% CI 0.83, 1.05). There was no indication of publication bias from the Begg funnel plot (Figure 2).

Figure 1
figure 1

A forest plot showing risk estimates from case-control and cohort studies estimating the association between tea consumption and risk for bladder cancer.

Full size image
Figure 2
figure 2

Funnel plot of tea consumption and bladder cancer risk.

Full size image

The results of subgroup analyzes by sex (men and women), geographical region (US/Canada, Europe, Uruguay and Asia) and tea type (green tea and black tea) are shown in Table 2. The OR estimates from subgroup analysis varied little, showing tea consumption was not associated with the likelihood of bladder cancer when separately analyzed by sex, geographical regions or tea type. In the subgroup analysis by tea type (Figure 3), we noted that green tea or black tea consumption was not associated with bladder cancer risk (OR 0.97, 95% CI 0.73, 1.21; OR 0.79, 95% CI 0.59, 0.99). There was weak heterogeneity among studies for black tea. We performed a sensitivity analysis, which removed one study at a time. This analysis confirmed the stability of the results for black tea. No indication of publication bias was observed from either with the Egger or Begg test in any subgroup.

Table 2 Summary of pooled odds ratios (ORs)for bladder cancer by sex, geographical region, and tea type
Full size table
Figure 3
figure 3

Forest plots showing the risk estimates from each study and the pooled risk estimates for green tea and black tea.

Full size image

Discussion

There has been considerable interest in the possible impact of tea consumption on bladder cancer risk due to the modifiable nature of tea consumption. In this meta-analysis of epidemiological studies of the association between tea and risk of bladder cancer including six cohorts and seventeen case-control studies, we found that tea consumption was not associated with reduced risk of bladder cancer.

A previous meta-analysis by Zeegers et al. [6] found no association between tea consumption and bladder cancer risk. The authors reported that the consumption of tea seems not to be related to an increased risk of urinary tract cancer. Our finding is consistent with their results. However, our study included 11 studies published after 2000 that were not included in the previous meta-analysis. We pooled the OR estimates by sex (men and women), tea type (green tea and black tea), geographical region (US/Canada, Europe, Uruguay and Asia), and study design (cohort or case-control studies).

Studies conducted on cell-culture systems and animal models as well as human epidemiological studies show that tea could afford protection against a variety of cancer types. Several laboratory studies have tried to investigate the link between tea and bladder cancer. Most of tea research on bladder cancer to date has focused on the effect and mechanism of green tea. It is generally agreed that many of the chemoprevention effects of green tea are mediated by polyphenols. The major catechins in green tea are epigallocatechin-3-gallate (EGCG), epicatechin-3-gallate, epigallocatechin, and epicatechin. EGCG accounts for 50% to 80% of catechin in green tea. Our previous study indicates that EGCG induces apoptosis in human bladder cancer T24 cells [42]. Kemberling et al. found that green tea (15% of which are polyphenols) have shown efficacy against rat bladder cancer induced by N-(4-hydroxybutyl)-N-bytyl-nitrosame (OH-BBN) [5]. The non-significant findings regarding the effects of tea consumption on bladder cancer in our meta-analyses contradict the results of previous experimental studies on this topic using in vitro bladder cancer cell lines and in vivo animal models. The difference between the results from experimental studies and our meta-analyses is likely to be due to the lower quantities of human tea consumption compared to the doses used in experimental studies and the fact that bioavailability is an important factor for consideration.

As a meta-analysis of previously published observational studies, our study has several limitations that need to be taken into account when considering its contributions. First, our meta-analysis only included published articles. Unpublished studies were not searched for our analysis. Second, we did not include studies with insufficient information to estimate an adjusted OR, which could bring publication bias even though no significant evidence of publication bias was observed in the Egger or the Begg tests. Third, our meta-analysis is likely affected by some misclassification of tea consumption. Tea exposure is mostly assessed regarding the number of cups of tea consumed daily or weekly. However, cup size may vary considerably. Fourth, only eight studies performed analyses and reported the RR separately for black tea and green tea. All other studies referred simply to tea. Finally, studies included in our meta-analysis were mainly conducted in Europe, US, Canada, Uruguay, and Asia; therefore, we are not able to generalize our findings for all populations. Also, most studies just included bladder cancer without specifying the type. Transitional cell carcinoma is the most common type of cancer in these regions and we found no relevant papers on squamous cell carcinoma. So the results are mainly based on transitional cell carcinoma but not squamous cell carcinoma.

Conclusions

In conclusion, in this pooled analysis of six cohort and seventeen case-control studies, we did not find that tea consumption was associated with decreased risk of bladder cancer. Given the small number of cohort studies included in this meta-analysis, further research from large epidemiological studies is needed in this area.

Methods

Literature research

We searched and reviewed the MEDLINE database using PubMed, Web of Science and the Cochrane Library, using selected common key words related to tea consumption and bladder cancer risk in case-control and cohort studies. We also scanned bibliographies of relevant articles in order to identify additional studies. As the key words for the literature search, we selected tea for the exposure factors, and bladder cancer for the outcome factors. The articles evaluating the relationship of urinary tract cancer and tea consumption were also retrieved, because the overwhelming majority of tumors occurred in the bladder, and the renal pelvis and ureter are covered by the same urothelium. The term bladder cancer was used as a synonym for these neoplasms.

Each identified publication was reviewed and included in the analysis if all the following criteria were met: first, they had to be case-control or cohort studies; second, papers reported in English between 1966 and December 2011; last, the result of each study was expressed as relative risk (RR) or odds ratio (OR) together with its corresponding 95% confidence interval (95% CI) adjusted for age, sex and smoking at the least, or sufficient information allowing us to compute them.

Data extraction

Data from all articles were retrieved independently by JQ and QM while the methods and results sections were removed and coded to blind the assessors to this information. The following data were collected: the first author’s name, the year of publication, country of origin, the study design (cohort or case-control), number of participants (cases and cohort size, or cases and controls), anatomical site of the neoplasm, adjusted effects estimates, exposure assessment and adjusted covariates. Considering that bladder cancer is a rare disease, the RR was assumed to be approximately the same as the OR, and the OR was used as the study outcome. Adjusted ORs were extracted directly from the original reports. If studies reported sex-stratified age- and smoking-adjusted ORs, we calculated the overall age-, smoking- and sex-adjusted OR by combining these estimates using the method of Mantel and Haenszel [43].

Statistical analysis

We pooled data using the DerSimonian and Laird random effects models [44], which considers both within-study and between-study variation. Subgroup analyses were performed according to sex (male or female), study design (cohort or case-control studies), the study location (US/Canada, Europe, Uruguay or Asia) and tea types (green tea, black tea). We quantified the extent of heterogeneity using the Q-test [44] and the I2 score [45], and P < 0.05 was considered statistically significant. Publication bias was assessed using the tests of Egger [46] and Begg [47]. All statistical analyzes were performed using Stata Statistical Software, version 10.0.