The purpose of this study was to investigate the common belief that a microvascular transfer of a non-innervated free muscle flap loses muscle bulk over time. Sixteen patients (latissimus dorsi = 8, rectus abdominis = 7, and gracilis muscle = 1) were evaluated an average of 41 months after free flap transfer. Latissimus dorsi and lower extremity flaps displayed significantly more swelling than the other flaps. Flap bulk was measured by ultrasound. The mean thickness of upper extremity flaps was 10.3 +/- 1.8 mm (control muscles 11.8 +/- 2.8), lower-extremity 14.5 +/- 3.7 mm (control muscles 10.9 +/- 0.7), latissimus dorsi 14.3 +/- 2.2 mm (control muscles 10.3 +/- 0.8, P = 0.018), and rectus abdominis 11.2 +/- 1.2 mm (control muscles 12.4 +/- 1.9). Color Doppler ultrasonography was used to detect the pedicles of the free flaps and also to measure the peak velocity of blood flow intramuscularly and in the pedicles. In the upper extremities (n = 5) the pedicles could be found in only 20% of cases whereas in the lower extremities (n = 11) 91% of pedicles were located. (P = 0.013). Peak flow within the free flaps was significantly higher in the lower extremity (50% of the peak flow of the common femoral artery) than in the upper extremity (5% of the peak flow of the common femoral artery, P = 0.013). This study demonstrated that non-innervated free muscle flaps in the extremities maintain the original muscle thickness, although lower extremity and latissimus dorsi flaps have a trend to be thicker. Most pedicles of free muscle flaps in the upper extremities could not be located by ultrasound. However, flaps in the lower extremities most often have patent pedicles and also more vigorous intramuscular blood flow.