Farin Kamangar, MD, PhD; Neal D. Freedman, PhD, MPH
(sent by Dr Vũ Quốc Duy)
THE ANNALS OF INTERNAL MEDICINE:In their current Annals report, Yu and colleagues (1) show that drinking high-temperature tea, when combined with tobacco or alcohol use, is associated with an increased risk for esophageal cancer. Their study was well-designed. Its findings, which are based on long-term follow-up (median, 9.2 years) in more than 450 000 participants, are an important addition to the current literature (2).
The hypothesis that drinking very hot beverages may cause esophageal cancer is not new. In the 1930s, based on clinical observations, New York physician W.L. Watson wrote, “Thermal irritation is probably the most constant factor predisposing to the cancer of the esophagus. The drinking of copious amount of excessively hot tea is a history frequently obtained from Russian-born patients coming to Memorial Hospital suffering from cancer of the esophagus” (3). Nevertheless, the association between hot beverages and esophageal cancer is not firmly established. An expert panel assembled by the International Agency for Research on Cancer (IARC) in 2016 classified drinking very hot beverages as a probable (class 2A) rather than a definite (class 1) carcinogen (2).
This uncertainty, despite more than 60 published epidemiologic studies, stems mainly from limitations in the literature. Most studies had a retrospective case–control design, with the accompanying bias and recall issues. In addition, some studies did not assess consumption frequency and temperature; others considered esophageal squamous cell cancer and esophageal adenocarcinoma together, despite clear etiologic differences (4). Even previous prospective studies suffered from bias. For example, a study from China (5) was marred by implausible responses, having been conducted amid a strong campaign against drinking very hot beverages, with substantial pressure on respondents to give the “right” answer.
The Yu group’s study overcame many of these challenges. It was a population-based prospective study with a very large number of esophageal cancer cases (1731). Detailed data were available on tea-drinking frequency, type of tea consumed, and temperature preference. Data also were available on important potential confounders, including smoking and alcohol use, 2 important risk factors for esophageal squamous cell cancer (4), which is the predominant form of esophageal cancer in China. Nevertheless, the study also has several limitations. Tea temperature was assessed only once and probably with error. Repeated assessments, available in a small subset of the cohort, showed low to moderate reliability over time (a Spearman correlation coefficient of just 0.35), perhaps resulting in an underestimation of the association. Another important limitation was that tea temperature was assessed by self-report, and the study lacked an objective measurement of tea temperature.
Despite inconsistencies in the literature, a clear pattern is emerging. The most recent meta-analysis showed an approximately 2-fold increased risk for esophageal squamous cell cancer in association with consumption of hot food and drinks (6). These results are consistent with those from an epidemiologic study that assessed tea temperatures by using an instrument validated against measured temperatures (7). Experimental studies have shown that hot water, at temperatures above 65 to 70 °C (149 to 158 °F), may increase the incidence of nitrosamine-induced esophageal tumors in animal models (8). The totality of this evidence resulted in the IARC classification of very hot drinks as “probably carcinogenic” to humans (2). The results of the study by Yu and colleagues, which show an increased risk for esophageal cancer in persons who drink very hot beverages daily, contribute to this emerging picture.
In the Yu group’s study, the associations with very hot tea were observed in men but not women and only in the presence of concomitant smoking or regular alcohol drinking. However, it is still too early to rule out associations in women. Besides biologic implausibility, too few cases of esophageal cancer (only 37) occurred in women who drank tea daily, and even fewer in those who consumed it very hot. Likewise, only 38 cases occurred in participants who reported drinking hot or burning hot tea but did not smoke or consume 15 g or more of alcohol per day. Future studies are needed, particularly those examining associations in women and in persons who do not smoke cigarettes or drink alcohol.
The study by Yu and colleagues will probably remain the largest study of hot drinks and esophageal cancer for some time. Nevertheless, an important role still exists for follow-up studies. For example, cohort studies that have measured tea-drinking temperature (7) will provide complementary findings when published. Further animal and mechanistic studies also are needed.
Tea is among the most popular beverages worldwide. Tea drinkers may be concerned about the implications of this study on their risk for cancer. Of importance, however, is that the accumulated literature suggests little risk from hot drinks at temperatures below 65 °C (149 °F). Most people drink beverages at temperatures lower than that threshold; for example, in the United States, coffee typically is consumed at around 60 °C (140 °F) (9).
For these and other reasons, these findings should be interpreted cautiously. Despite this study’s rigorous design and careful analysis, its results are observational and may still reflect confounding by other factors and chance. In addition, health decisions generally should not be made on the basis of a single end point, especially one that is uncommon in much of the world. Tea and other hot beverages, such as coffee and maté, contain many compounds that may have beneficial effects on health outcomes. In fact, coffee drinking has been inversely associated with overall mortality in many studies (10).
In conclusion, the Yu group’s study clearly adds to the literature. Perhaps those of us who drink hot beverages often should be prudent and wait for the liquid to cool a bit first. However, the results of this study should not cause people to abandon their favorite beverage. Most people drink their tea and coffee at a temperature that seems unlikely to cause cancer.