HYPOTHESIS: Minimally invasive correction of pectus excavatum (PE) deformities of the anterior chest wall in children is safe and effective. DESIGN: Prospective cohort study. SETTING: Tertiary pediatric referral center. PATIENTS: Between February 1996 and July 2000, 36 patients underwent minimally invasive repair (MIR) of PE deformities, and 6 patients had traditional Ravitch repairs (RR). MAIN OUTCOME MEASURES: Morbidity, operating time, estimated blood loss, days to tolerating a regular diet, and length of hospital stay. RESULTS: Thirty-six children underwent MIR for moderate to severe chest wall deformities, with a mean operative time of 1.6 hours, a mean blood loss of 22 mL, a mean time to tolerating a regular diet of 2.9 days, no intensive care unit admissions, and a mean length of hospital stay of 5.5 days. Six children had RR procedures performed for moderate to severe deformity, with a mean operative time of 5.2 hours, a mean blood loss of 222 mL, a mean time to tolerating a regular diet of 3.3 days, 2 patients admitted to the intensive care unit, and a mean length of hospital stay of 4.5 days. Complications for both procedures consisted mainly of postoperative pneumothorax. CONCLUSIONS: The MIR technique for PE is less invasive, less morbid, and better tolerated than traditional open RR of this common pediatric chest wall deformity. Elective surgical reconstruction can be safely performed in children rather than limiting repair to only symptomatic patients with severe deformities.