Several studies have reported an association between apolipoprotein E polymorphisms and osteoporosis, specially the genotype APO-E4. In order to analyze the APO-E polymorphisms and to identify their association with clinical and social variables, a descriptive study was undertaken that included 32 women with osteoporosis, from different regions of Colombia. The polymorphisms were detected by PCR and RFLP methods. In osteopenia and osteoporosis combined with osteopenia were observed the genotype epsilon3/epsilon3 in the 84% (n=27), and 16% (epsilon3/epsilon4=12.5%, n=4; epsilon4/epsilon4=3.1%, n=1) for the genotypes bearing the epsilon4 allele. The same tendency was observed by age of the menopause, epsilon3/epsilon3 in the 83% (n=25), and the genotypes bearing the epsilon4 allele in the 17% (n=5) (epsilon3/epsilon4=13.3%, n=4; epsilon4/epsilon4=3.3%, n=1). No association of APO-E4 was detected with socioeconomic stratum, fracture, illness, surgeries, and milk consumption. No significant differences were observed in the bone mineral density (BMD) of the lumbar column between the genotypes with or without the epsilon4 allele epsilon4+/- (epsilon3/epsilon4 0.96+/-0.14 g/cm2); epsilon4+/+ (epsilon4/epsilon4 0.87+/-0.0 g/cm2); epsilon4-/- (epsilon3/epsilon3 0.86+/-0.16 g/cm2); p=0.49, and femoral bone mineral density epsilon4+/- (epsilon3/epsilon4 0.84+/-0.03 g/cm2); epsilon4+/+ (epsilon4/epsilon4 0.84+/-0.0 g/cm2); epsilon4-/- (epsilon3/epsilon3 0.74+/-0.01 g/cm2); p=0.014. However, when exploring the differences of BMD in the femoral neck, a significant difference was observed (t=4.17, p=0.05). These results confirm epsilon4 allele frequencies similar to those reported for caucasian and Japanese, subjects. Larger studies are necessary to elucidate the effect of APO-E in bone marrow and the dose-effect relation.