The purpose of this study was to investigate in a cross-sectional design body composition, muscle fiber characteristics, cardiovascular risk factors and liver enzymes in long-term androgenic-anabolic steroids (AAS) using bodybuilders three months after drug withdrawal (AAS group; n = 16) and in non-users (CO group; n = 12). Training and dietary data were collected in all subjects. Anthropometry included weight, height, 8 skinfolds and 11 circumferences. Percentage fat (%FAT), fat mass (FM) and lean body mass (LBM) were calculated. In a muscle biopsy from the vastus lateralis muscle water content, fiber type distribution and diameters of fiber type I and type II were determined. Age, height, training characteristics, nutrition, skinfolds, %FAT and FM did not differ between the groups. The AAS group had greater BW and LBM, and larger circumferences of thorax, waist, upper arm and thigh than the CO group. Muscle biopsy data were comparable, except for muscle fiber diameter of type I which was larger in the AAS group. No differences in serum values of total cholesterol, HDL-cholesterol and triglycerides, nor in systolic and diastolic blood pressure were observed. In both groups serum alkaline phosphatase and gamma GT were within the normal range. This study suggests that in long term AAS using body-builders, after a three months AAS free period, BW is greater than in non drug users. This is reflected in larger LBM, circumferences and diameter of muscle fiber type I. In addition, no differences in fat mass, blood pressure, lipoprotein profiles and liver enzymes exist between AAS users three months after interrupted drug use and their non drug using counterparts.