Early clinical outcomes of robot-assisted surgery for anterior mediastinal mass: its superiority over a conventional sternotomy approach evaluated by propensity score matching.

OBJECTIVES

We performed this study to assess early clinical outcomes of robot-assisted surgery for anterior mediastinal mass by comparing results of the robot group with those of the sternotomy group after propensity score matching.

METHODS

Between 2008 and 2012, 145 patients underwent resection of anterior mediastinal mass. Robot-assisted surgery was performed in 37 patients, and conventional surgery by sternotomy in 108 patients. Propensity score matching was done between two groups with variables of age, sex, size of the mass, myasthenia gravis, resection of other organ and pathological diagnosis. Thirty-four patients from the robot group and 34 from the open group were matched, fitting the model. The clinical outcomes of matched groups were compared.

RESULTS

In the robot group, mediastinal cyst consisted of 47.1% (16 of 34), thymoma 32.4% (11 of 34), thymic carcinoma 8.8% (3 of 34), thymic hyperplasia 8.8% (3 of 34) and liposarcoma 2.9% (1 of 34). The mean duration of follow-up was 1.11 ± 0.21 and 1.85 ± 0.19 years for the robot and open groups, respectively. There were no mortality or recurrence in both groups during the follow-up. There were no significant differences in operation time, postoperative white blood cell and C-reactive protein increase, maximum visual analogue scale score for pain as well as postoperative intensive care unit care between the two groups. The robot group revealed a lesser number of drains (1.09 ± 0.1 vs 1.41 ± 0.1) and 24-h tube drainage (189.4 ± 20.5 vs 397.6 ± 52.6 ml), lower haemoglobin loss (0.54 ± 0.4 vs 1.35 ± 0.1 g/dl) and haematocrit decrease (1.92 ± 0.5 vs 3.85 ± 0.4%), shorter chest tube days (1.53 ± 0.2 vs 3.06 ± 0.2) and length of hospital stay (2.65 ± 0.2 vs 5.53 ± 0.8) after operation, which were all statistically significant. Although statistically insignificant, there were no postoperative complications in the robot group, but there were 5 (14.7%) in the open group (P = 0.063).

CONCLUSIONS

In carefully selected patients with relatively smaller sized masses, robot-assisted surgery resulted in excellent early clinical outcomes with lesser tube drainage, lower blood loss, shorter tube days and length of hospital stay without any postoperative complications, compared with the matched open group. Further investigation for long-term clinical outcomes and oncological outcomes is required for a robotic approach. Particularly, long-term follow-up for the local recurrence rate according to the pathological diagnoses is required.