Abstract: Objective：To retrospectively evaluate the long-term safety and efficacy of bosentan, an oral dual endothelin receptor antagonist applied to patients with pulmonary arterial hypertension associated with congenital heart disease (CHD-PAH). Methods：The patients with CHD-PAH were enrolled. Patients were received oral bosentan and underwent a routine follow-up for analyzing the clinical outcomes and side effects. Bosentan dosage started from 62.5 mg (twice daily) and increased to 125 mg (twice daily) after 4 weeks if there was no obvious adverse reaction. Before taking the medicine, six months, one year, two years, three years, four years and five years after taking the medicine, 6 min walking test (6MWD), Doppler echocardiography, heart function grade, NT-pro BNP, and liver function were evaluated respectively. Results：A total of 33 patients with CHD-PAH were enrolled. Among the 33 CHD-PAH patients, 5 patients were treated with operations successfully, 1 patient died. Since 6-months after taking bosentan, the pulmonary artery systolic pressure and 6MWD of the patients were obviously increased, and NT-proBNP was obviously decreased. By the time of 2-year follow-up, there were significant statistical differences compared with those before taking the drug (［88.4±29.4］ mmHg ［1 mmHg=0.133 kPa］ vs ［113.7±26.7］ mmHg, P<0.05; ［364.1±92.6］ m vs ［303.0±88.1］ m, P<0.05; ［2.3±0.3］ pg/mL vs ［2.8±0.5］ pg/mL, P<0.05). NYHA classification revealed a marked improvement since 1-year follow-up (P<0.05). However, when it came to the analysis of follow-up information after 3 years, the improvement of SPAP, 6MWD, NT-proBNP, and NYHA began to slow down significantly. SPAP and 6MWD still showed a significant decrease compared to baseline (P=0.018 and P=0.02) at 5-year follow-up. Whereas there was no obvious statistical difference in NYHA and NT-proBNP compared with those before taking the medicine. ALT did not change significantly during follow-up. Conclusions：Oral bosentan can effectively decrease pulmonary hypertension and improve exercise tolerance and cardiac functional classification in CHD-PAH patients. However, its long-term efficacy still needs to be confirmed by further clinical trials.