AIMS: Diabetes or insulin resistance, overweight, arterial hypertension, and dyslipidaemia are recognized risk factors for cardiovascular (CV) disease. However, their predictive value and hierarchy in elderly subjects remain uncertain. METHODS: We investigated the impact of cardiometabolic risk factors on mortality in a prospective cohort study of 331 elderly high-risk subjects (mean age+/-SD: 85+/-7 years). RESULTS: Two-year total mortality was predicted by age, diabetes, low BMI, low diastolic blood pressure (DBP), low total and HDL cholesterol, and previous CV events. The effect of diabetes was explained by previous CV events. In non-diabetic subjects, mortality was predicted by high insulin sensitivity, determined by HOMA-IR and QUICKI indices. In multivariate analyses, the strongest mortality predictors were low BMI, low HDL cholesterol and previous myocardial infarction. Albumin, a marker of malnutrition, was associated with blood pressure, total and HDL cholesterol, and HOMA-IR. The inflammation marker CRP was associated with low total and HDL cholesterol, and high HOMA-IR. CONCLUSION: In very old patients, low BMI, low DBP, low total and HDL cholesterol, and high insulin sensitivity predict total mortality, indicating a "reverse metabolic syndrome" that is probably attributable to malnutrition and/or chronic disorders. These inverse associations limit the relevance of conventional risk factors. Previous CV events and HDL cholesterol remain strong predictors of mortality. Future studies should determine if and when the prevention and treatment of malnutrition in the elderly should be incorporated into conventional CV prevention.