Energy restriction and the risk of spontaneous mammary tumors in mice: a meta-analysis.

Our meta-analysis was aimed at providing a systematic review of the literature regarding the effect of energy restriction on spontaneous mammary tumors in mice and at providing a more precise pooled (summary) estimate of the risk of mammary tumors. A sensitivity analysis was conducted to obtain insight in potential heterogeneity between the animal studies. A literature search was conducted with the following terms to identify relevant articles: animal studies, mammary tumors, fat restricted, dietary carbohydrates, energy restriction and calorie restriction. A criteria list for the assessment of quality items (i.e., study characteristics) in animal experiments was developed that was intended to quantitatively assess potential factors that underlie heterogeneous results of different animal experiments. Incidence figures were used to calculate the risk difference. The pooled risk difference was calculated by random effects meta regression analysis. Fourteen animal experiments were included in this meta-analysis. Publication bias could not be identified. The pooled risk difference for the 14 studies was -0.55 with a narrow 95% confidence interval (-0.69; -0.41), implying that the energy-restricted animal groups developed 55% less mammary tumors than the control groups. No heterogeneity could be detected between the studies based on study characteristics that included the age of mice at the start of intervention, duration of intervention, allocation of the mice, use of ad libitum control group, fertility of the mice and the type of energy-providing nutrient (fat, carbohydrate or protein). This meta-analysis confirms that energy restriction in itself consistently protects against the development of mammary tumor in mice, irrespective of the type of restricted nutrient or other study characteristics. Copyright 2003 Wiley-Liss, Inc.