A prospective study was designed to evaluate the possible changes in abdominal wall strength following free transverse rectus abdominis musculocutaneous (TRAM) flap surgery for breast reconstruction. Twenty-two patients were examined 1 day before surgery, and at 3, 6, and 12 months postoperatively. Trunk muscle strength was measured by the same physiotherapist using an isokinetic dynamometer (Lido Multi Joint II, Loredan Biomedical Inc., Davies, CA). The peak torque and average torque for both flexion and extension at 60 degrees per second angular velocity were recorded from the curves obtained. There was a significant reduction in trunk flexion strength at 3 months postoperatively (peak torque mean, 92% of the preoperative value; p = 0.04), but this was corrected by 6 months (mean, 96%), and improved to 98% by 12 months. The patient's ability to do curled trunk sit-ups was evaluated by the same physiotherapist and graded on a scale from 1 to 6. In 9 of 19 patients the operation had no effect on sit-up performance during follow-up. In 10 of 19 patients there was a reduction of one or two grades at 3 months that did not improve by 12 months. Magnetic resonance imaging of the abdominal wall was performed on 9 patients. The mean area of the upper third of both rectus muscles was measured on the axial images. At 3 months postoperatively the mean area of the upper third of the donor muscle was significantly larger than the contralateral (p = 0.03). There was no difference in size at 6 months, and by 12 months the donor side was smaller. This prospective study shows that harvesting of a free TRAM flap can cause a subclinical reduction in abdominal strength, although this was not noticed by the patients themselves.