The aim of this study was to assess the early results of a three-year experience with the minimally invasive correction of pectum excavatum, which requires no cartilage incision or excision, and no sternal osteotomy. Since 2001 we have performed 35 minimally invasive pectus excavatum procedures at our hospital. A convex steel bar is inserted under the sternum through small bilateral incisions, and removed after 2 years when permanent remolding had occured, the bar is removed. Complications were pneumothorax in 5 patients (only 1 required a thoracostomy tube, the other 4 resolved spontaneously), pneumonia in 3 patients, and bar displacement in 1 patient. The mean follow-up was 3 months to 3 years. Initial excellent results were maintained in 28 patients (normal postoperative chest), good results in 5patients (mild residual pectus) and poor in 2 patients (severe recurrence requiring further treatment). Poor results occurred because the steel bar was too soft in 1 patient, and the sternum too soft in 1 patient with Marfan's syndrome. Our early results with the minimally invesive technique without cartilage incision and resection or sternal osteotomy showed that the procedure is effective with excellent preliminary results.