Surgeons who perform transverse rectus abdominis musculocutaneous (TRAM) flaps have differing opinions about how many drains are required in the breast and abdomen to prevent seroma. The authors therefore decided to review their experience to determine whether the number of drains influenced the incidence of seroma. All patients who underwent breast reconstruction using TRAM or deep inferior epigastric perforator flaps at The University of Texas M. D. Anderson Cancer Center from January 1, 1995 to June 20, 2000 and whose charts could be retrieved were included in the study. The number of drains used was correlated with the presence or absence of seroma and wound infection in both the abdomen and the breast. Significance was analyzed using the Chi-squared and Fisher's exact tests. There were 608 patients and 768 reconstructive procedures in this series (160 reconstructions were bilateral). Of patients who had only one drain in the abdomen, seroma developed in 9 patients (7.1%), whereas of those having two drains in the abdomen, seroma developed in only 10 patients (2.1%) (p = 0.006). Also, of patients who had only one drain in the breast, seroma developed in the breast in 47 patients (9.1%), and in those with two drains, seroma developed in only 11 patients (4.3%) (p = 0.02). There were no significant differences in the infection rate in either the breast or the abdomen, although the trends favored a lower infection risk when two drains were used. The authors found that using two drains in both the abdomen and the breast can reduce the risk for seroma without increasing the risk for infection. This study supports the use of two drains in both the breast (one each beneath the TRAM flap and in the axilla) and abdomen (beneath the abdominoplasty flap) for patients undergoing breast reconstruction using the TRAM flap.