As early as 1942, it was reported that the ingestion of caffeine in men exposed to a cool ambient temperature reduced the drop in mean skin temperature (_sk) and thus ensured a warmer _sk (Scheurer and Hugo, 1942). Similarly, LeBlanc (1987) found that caffeine ingestion before retiring for the night in a cool environment significantly increased oxygen consumption and provided a warmer _sk, but without any change in T_re. Other studies that have analyzed the effect of caffeine in the cold have confirmed that it has little influence on T_re. Contrary to LeBlanc's observations, they have also reported that it tends to exaggerate the drop in _sk in cold air (Graham et al., 1991; McNaughton et al., 1990) or that it offers no beneficial effect in cold water, in spite of an important increase in M (Doubt and Hsieh, 1991).

Following up on his animal work (described earlier), Wang and colleagues showed that the drop in T_re in cold-exposed individuals can be greatly reduced with the prior administration of theophylline. This was demonstrated in acute cold air studies, with the subjects either at rest or performing intermittent exercise (Wang et al., 1986, 1987, 1989). It was suggested that, as in animals, the effectiveness of theophylline in enhancing human cold tolerance resided in an enhancement of energy substrate mobilization, a factor that was thought to be limiting for cold-induced thermogenesis and consequently cold tolerance (Wang, 1978, 1981; Wang and Anholt, 1982; Wang et al., 1986). Although this theory seems well supported by the animal studies, the corresponding metabolic data in humans are not as convincing, particularly since the marked improvements in the subjects' T_re, described above, were not accompanied by any significant increase in M (Wang et al., 1986, 1987, 1989). To clarify the mechanisms by which cold tolerance can be enhanced, Vallerand and coworkers decided to reinvestigate the concept linking energy substrate mobilization, thermoregulatory thermogenesis, and cold tolerance.

Based on the theory above, which stated that energy substrate mobilization regulates thermoregulatory thermogenesis, a commercially available recreation and sports bar (Cold Buster™) has been recently developed. It is purported to delay markedly the onset of hypothermia in humans as a result of its energy-containing substrates and theobromine content. After proper subject familiarization (i.e., prior exposure to cold and instruments) to ensure a good reproducibility between cold tests (< 5 percent variability) (Vallerand and Jacobs, 1989) and while fasting seminude (jogging shorts only), subjects were