A considerable interest in autologous tissue breast reconstructions has developed recently, especially since Food and Drug Administration (FDA) experts have raised the polemic on silicone implants. Although such enthusiasm for the transverse rectus abdominis musculocutaneous (TRAM) flap is justified in what concerns the final cosmetic result of the reconstructed breast, the risk of abdominal sequelae should be explained to the patient. Abdominal scarring, parietal weakness, strength loss, and back pain have been recorded in a series of pedicled TRAM flap reconstructions performed at the Institut Gustave-Roussy before 1991. Cosmetic results of the abdomen were tested by three independent judges on the photographs taken of 132 patients. The global appearance of the abdomen was rated as "natural" in 70 percent of the patients. The umbilical scar was unacceptable in 25 percent and very good or not visible in 21 percent. The lower abdominal scar was rated as "not acceptable" in 35 percent. Abdominal strength was tested among the 38 patients who underwent both preoperative and postoperative muscular testing, of whom 18 had single-pedicled and 20 had bipedicled TRAM flap reconstructions. In the first group, the physiotherapist observed an impairment of function of the upper portion of the rectus abdominis and of the oblique muscle in almost 50 percent. In the second group, the same muscles were more severely impaired (in 60 percent), as well as the function of the lower portion of the rectus (in 15 percent). The percentage of patients complaining of "light back pain" in the 6-month postoperative period was 55 percent in the single-pedicle group and 30 percent in the double-pedicle group. However, 20 percent of the bipedicled patients complained of "severe" back pain, while there was none in the single-pedicled group. In conclusion, abdominal sequelae after TRAM flap breast reconstruction should not be underestimated and constitute one of the drawbacks of the technique when it is compared with reconstruction with a prosthesis.