Risk model of cardiovascular surgery in 845 Marfan patients using the Japan adult cardiovascular surgery database

The aim of this study was to evaluate the short-term operative results of patients with Marfan syndrome who underwent thoracic or abdominal aortic surgery in a 4-year period in Japan. Data were collected from the Japan Cardiovascular Surgery Database (JCVSD). We retrospectively analyzed the data of 845 patients with Marfan syndrome who underwent cardiovascular surgery between January 2008 and January 2011. Logistic regression was used to generate risk models. The early mortality rate was 4.4% (37/845). Odds ratios (OR), 95% confidence intervals (CI), and P values for structures and processes in the mortality prediction model were as follows: renal insufficiency (OR, 11.37; CI, 3.72-34.66; P < 0.001); respiratory disorder (OR, 11.12; CI, 3.20-38.67; P < 0.001); aortic dissection (OR, 13.02; CI, 2.80-60.60; P = 0.001); pseudoaneurysm (OR, 11.23; CI, 1.38-91.66; P = 0.024); thoracoabdominal aneurysm (OR, 2.67; CI, 1.22-5.84; P = 0.014); and aortic rupure (OR, 4.23; CI, 1.26-14.23; P = 0.002). The mortality prediction model had a C-index of 0.82 and a Hosmer-Lemeshow P value of 0.56. In conclusion, this study demonstrated that renal insufficiency and respiratory disorder had great impact on the operative mortality of Marfan patients undergoing cardiovascular surgery. Because patients with aortic dissection or aortic rupture showed high operative mortality, close follow-up to avoid emergency operation is mandatory to improve the operative results. Achieving good results from surgery of the thoraco-abdominal aorta was quite challenging, also in Marfan patients.