Venous thromboembolism (VTE) is known as a common complication in surgical and non-surgical patients. We hypothesized that according to the underlying risk factors and the acute illness, the prevalence ofVTE in non-surgical patients admitted to hospital is widely underestimated. For three months each patient admitted to the department of internal medicine with an acute illness, but without known deep venous thrombosis (DVT) was investigated by ultrasound compression sonography. Patients' history, risk factors and extent of immobilisation were documented. In patients with newly detected DVT D-dimer and fibrinogen were measured as well as computer tomography scans performed. Follow-up investigations of the DVT population were performed at four weeks and three months. Six hundred seventeen patients (49.3% men) were included. In 16 patients (men = 7) a previously unknown thrombosis (2.6%) was detected, mainly in patients with acute cardio-pulmonary disease (56%) and the elderly (mean age 75.6 years). Eight patients had femoro-popliteal (50.0%), four a femoral (25.0%), and four a popliteal vein thrombosis (25.0%). Five had pulmonary embolism (31.3%). In patients with DVT D-dimer was 875 +/- 1,228 mg/l, fibrinogen 568 +/- 215 mg/dl and C-reactive-protein 58.54 +/- 73.65 mg/dl. One patient died from sepsis during hospitalisation, one died from sudden cardiac death at home. None of the other 14 surviving patients relapsed. The study shows a 2.6% risk for DVT in outpatients with acute illness admitted to the department of internal medicine. These data demonstrate the high risk of DVT is in non-surgical patients. Early prophylaxis has to be considered in internal medicine patients especially in the elderly.