Minimally invasive cardiac surgery (MICS) for treating valvular disease was introduced in our division in July 1997, and we have treated a total of 236 cases by July 2002. Among the various types of surgical treatment, there were 21 cases (M-group) of double valve replacement (DVR) to treat combined valvular disease. There had been 8 cases (F-group) of DVR by means of conventional full sternotomy during the period from January 1990 to June 1997, before the introduction of MICS. A comparison of the results of these surgical treatments yielded the following results. There were no differences in operation time and blood loss during the operations between the 2 groups, whereas the aortic cross clamp time and cardiopulmonary bypass time were significantly longer in the M-group than the F-group (M-group: 189±6 and 228±7min; F-group: 132±18 and 183±16min, respectively). There were significantly more cases of concomitant maze operation in the M-group than in the F-group. There were no differences in the durations of postoperative intubation or ICU stay. The days required from operation to starting walking were significantly shorter in the M-group compared to in the F-group (M-group: 2.4±0.2 days; F-group: 3.3±0.2 days), while there were no differences in the postoperative hospitalization periods. There were no major postoperative complications, and 1 case each there was 1 death in each group during the hospitalization period. Although the aortic cross clamp time and cardiopulmonary bypass time were longer in the M-group than in the F-group, the postoperative course and surgical outcome were good. So MICS for DVR was considered acceptable. In addition, MICS was thought to provide high patient satisfaction with regard to cosmesis or thoracic fixation, although early discharge from the hospital, which was possible in cases of single valve MICS, was not obtained.