Intramuscular venous malformations are often mistaken for tumors because of a similar presentation and improper nomenclature. This is a review of 176 patients with venous malformations localized to skeletal muscle compiled from the Vascular Anomalies Center at Children's Hospital from 1980 through 1999. The female-to-male ratio was 2:1. Two-thirds of skeletal muscle venous malformations were noted at birth; the remainder manifested in childhood and adolescence. Venous malformations occurred in every muscle group, most often in the head and neck and extremities. Pain and swelling were the usual presenting complaints. Skeletal problems, such as fracture, deformation, or growth abnormalities, were rare. Hormonal exacerbation and intralesional bleeding were infrequent. Magnetic resonance imaging showed the lesions to be isointense to surrounding muscle on T1-weighted sequences and hyperintense on T2-weighted images. Characteristic tubular or serpentine components were oriented along the muscular long axis. Thrombi were hyperintense on T1-weighted and hypointense on T2-weighted sequences; phleboliths were seen as signal voids on all sequences. Gross examination of resected specimens revealed multicolored tissue with dilated vascular channels, frequently containing phleboliths. Light microscopy showed aggregates of primarily medium-sized, thin-walled vascular channels with flat endothelium and variable smooth muscle, most closely resembling dysplastic veins. Three lesions had a different histologic appearance consisting predominantly of small vessels with capillary structure and proliferative activity admixed with large feeding and draining vessels, similar to a lesion called intramuscular capillary hemangioma in the literature. The endothelium in these three lesions was negative for glucose transporter-1 by immunostaining. Eight percent of the patients, who had minor or no symptoms, were not treated. Twenty-four percent of the patients were managed conservatively (with aspirin and compressive garments); for 17 of these patients (10 percent of 176), noninvasive therapy was not successful, and they proceeded to sclerotherapy, excision, or both. A total of 31 percent of the patients had sclerotherapy, 20 percent had excision, and 27 percent had combined sclerotherapy and excision. Sclerotherapy was used for diffuse lesions, except for those with multiple intralesional thromboses, neurologic impairment, or compressive signs and symptoms. Resection was preferred for venous malformations well localized to a single muscle or muscle group, particularly if the muscles are expendable. Therapeutic outcomes were recorded in the charts or obtained by telephone interview in 122 of the patients (69 percent). Of these, compression garment and aspirin, resection, sclerotherapy, or combined excision and sclerotherapy improved symptoms in 121 patients (92 percent); no change was noted in 10 patients (8 percent). Only one patient was worse (self-reported) after intervention.