Postoperative deep vein thrombosis is usually asymptomatic so that pulmonary embolism is often the first clinical manifestation of venous thromboembolism. The diagnostic accuracy of impedance plethysmography, 125I-fibrinogen leg scanning and real-time B-mode ultrasonography has been extensively evaluated in patients with asymptomatic deep vein thrombosis. Impedance plethysmography has been evaluated in a number of studies and its sensitivity has been invariably found to be approximately 20% or less. These results seem to be due to the high prevalence in asymptomatic patients of distal, small and nonocclusive thrombi, unable to produce a critical obstruction of the venous flow. The accuracy of 125I-fibrinogen leg scanning has been assessed in a number of studies and found to be significantly different in the initial and more recent studies. This difference has been found to be due to the quality of the study design: the better the methodology, the lower the diagnostic accuracy of 125I-fibrinogen leg scanning. The association of impedance plethysmography and 125I-fibrinogen leg scanning do not result in an improvement of the results obtained by the single diagnostic method. Study methodology strongly influenced the results obtained with real-time B-mode ultrasonography with a reported sensitivity of approximately 50% in the studies performed adopting an appropriate methodology. In conclusion, noninvasive diagnostic methods are inaccurate in the diagnosis of asymptomatic deep vein thrombosis. Thus, venography remains the only accurate diagnostic method for the diagnosis of asymptomatic deep vein thrombosis.