Vascular malformations can usually be diagnosed on clinical grounds. They have a well-defined appearance on magnetic resonance imaging, which can effectively determine their tissue and flow characteristics. However, the role of cross-sectional imaging in the management of vascular malformations is not well defined. Most reviews suggest that magnetic resonance imaging should be reserved for cases in which the extent of the lesion cannot be estimated on physical examination. However, to date no group has compared the accuracy of physical examination alone to that of magnetic resonance imaging in determining this extent. A review was performed of all the patients evaluated for vascular malformations at the New York University Trunk and Extremity Vascular Anomalies Conference between July of 1994 and August of 1999. Patients who underwent magnetic resonance evaluation at other institutions and whose images were not available for review were excluded. All study patients either underwent magnetic resonance imaging examination at New York University Medical Center or had outside films reviewed at the center. The physical examination findings were compared with the magnetic resonance findings and the surgeon and radiologist made a joint decision about whether there was a correlation between the magnetic resonance and physical examination findings. Fifty-eight patients met the study criteria, 44 (76 percent) of whom were found to have more extensive disease on magnetic resonance examination than appreciated on physical examination. Of the 51 patients with low-flow vascular malformations (venous vascular malformations, lymphatic malformations, and capillary malformations), 39 (76 percent) had more extensive disease on magnetic resonance examination than on physical examination. Of the seven patients with high-flow arteriovenous malformations, five had more extensive disease on magnetic resonance. In all of the 44 patients whose magnetic resonance imaging findings did not correlate with those of the physical examination, therapeutic decision making was affected. Contrary to the conventional wisdom of published reviews, physical examination findings significantly underestimated the extent of vascular malformations in the majority of cases. Magnetic resonance imaging should be performed in all patients with vascular malformations of the trunk and extremities before therapy is planned. In an age when physicians are asked to justify their decisions, especially where the use of expensive diagnostic modalities is concerned, the situations in which these tests are indispensable must be clearly defined or else patients will be denied access to them.