3.1 Risk factors for mortality
Ten patients died in the hospital, and 2 patients died after discharge from the hospital during follow-up (8 and 10 months after the operation, respectively). The early mortality rate was 13.9% (10/72), and the overall mortality rate was 16.7% (12/72) at the median follow-up time of 48 months (9.25-73.75), with 10 patients lost to follow-up. Two patients in our series died within 24 hours postoperatively due to coronary malperfusion, 1 patient died iatrogenically due to the inotropic transfusion line being accidentally disconnected during patient transfer from the operating room to the ICU, and 1 patient died from intractable heart failure due to the misdiagnosis of left ventricular outflow tract obstruction caused by accessory mitral valve tissue. Additionally, 1 patient died on postoperative day (POD) #5 because of uncontrollable arrhythmia (junctional ectopic tachycardia). The cause of death in the remaining 5 patients was related to nosocomial infection, with most of these patients requiring preoperative ventilation. The details of early mortality are described in Table 3.
There were two cases of late mortality. In the first case, the patient died suddenly at 8 months postoperatively, despite a recent echocardiogram showing normal biventricular function before death. The second case of late death occurred at 10 months after the operation; due to severe sepsis from pneumonia but had normal cardiac function at the latest follow-up. The estimated 1-year, 5-year, and 9-year freedom from death for the 72 patients were 83.2% (SD=4.4), 83.2% (SD=4.4) and 83.2% (SD=4.4), (95% CI, 0.7237 to 0.9012), respectively (Figure 2).
Competing risk analysis revealed that: intraoperative ventricular septal defect enlargement [hazard ratio (HR) 7.23, 95% confidence interval (CI) 3.1294-16.7167; P < 0.001], secondary aortic cross clamping (HR 28.38, 95% CI 4.8427-166.3484; P < 0.001), post-operative pneumonia (HR 5.64, 95% CI 1.2724-24.9917;P = 0.023), and post-operative sepsis (HR 5.28, 95% CI 1.3512-20.6553; P = 0.017) were factors associated to overall mortality. By Cox multivariable analysis, the Ao arch hypoplasia (HR 2.96, 95% CI 0.9170-9.5288; P = 0.07) are approaching significant factors associated to overall mortality.