Seul Ki Han1, Moon Young Kim1,2,3, Seong Hee Kang1,2, Ki Tae Suk4, Soon Koo Baik5,6,7. 1. Department of Internal Medicine, Division of Gastroenterology and Hepatology, Wonju Severance Christian Hospital, Yonsei University Wonju College of Medicine, 20, Ilsanro, Wonju, 26426, Republic of Korea. 2. Regeneration Medicine Research Center, Yonsei University Wonju College of Medicine, Wonju, Republic of Korea. 3. Cell Therapy and Tissue Engineering Center, Yonsei University Wonju College of Medicine, Wonju, Republic of Korea. 4. Department of Internal Medicine, Hanllym University College of Medicine, Chuncheon, South Korea. 5. Department of Internal Medicine, Division of Gastroenterology and Hepatology, Wonju Severance Christian Hospital, Yonsei University Wonju College of Medicine, 20, Ilsanro, Wonju, 26426, Republic of Korea. Baiksk@yonsei.ac.kr. 6. Regeneration Medicine Research Center, Yonsei University Wonju College of Medicine, Wonju, Republic of Korea. Baiksk@yonsei.ac.kr. 7. Cell Therapy and Tissue Engineering Center, Yonsei University Wonju College of Medicine, Wonju, Republic of Korea. Baiksk@yonsei.ac.kr.
Abstract
BACKGROUND AND AIMS: Long-term prospective data on hepatopulmonary syndrome (HPS) from a large number of patients, especially in Asian patients, are lacking. We evaluated the long-term prognosis of HPS and the development of acute-on-chronic liver failure (ACLF), and related factors. METHODS: A total of 142 patients with cirrhosis who underwent saline-agitated contrast echocardiography for the diagnosis of HPS were enrolled and observed prospectively from 2014 to 2019. RESULTS: A total of 59 patients (41%) were diagnosed with HPS (24 grade 1, 23 grade 2, 12 grade 3). Thirty-eight and 37 patients died in the HPS and non-HPS groups, respectively (p < 0.01). The 5-year survival rate was 47% in the HPS group and 62% in the non-HPS group. In the Cox proportional hazards model, HPS and Model for End-stage Liver Disease (MELD) score ≥ 18, and Child-Turcotte-Pugh (CTP) class B/C were significant risk factors for mortality after adjusting for other risk factors (HPS hazard ratio [HR] = 1.9, p = 0.01; MELD score ≥ 18 HR = 2.3, p < 0.01; CTP class B/C HR = 2.9, p < 0.01). Compared to that in non-HPS group, the HPS group had a significantly higher incidence of ACLF during follow-up (p < 0.01) and more frequently presented with lung involvement of ACLF (p = 0.03). CONCLUSIONS: In the long-term follow-up cohort, patients with HPS showed poorer prognosis than that of patients without HPS. HPS was a risk factor for ACLF development independent of hepatic dysfunction, and lung involvement was significantly common than without ACLF.
BACKGROUND AND AIMS: Long-term prospective data on hepatopulmonary syndrome (HPS) from a large number of patients, especially in Asian patients, are lacking. We evaluated the long-term prognosis of HPS and the development of acute-on-chronic liver failure (ACLF), and related factors. METHODS: A total of 142 patients with cirrhosis who underwent saline-agitated contrast echocardiography for the diagnosis of HPS were enrolled and observed prospectively from 2014 to 2019. RESULTS: A total of 59 patients (41%) were diagnosed with HPS (24 grade 1, 23 grade 2, 12 grade 3). Thirty-eight and 37 patients died in the HPS and non-HPS groups, respectively (p < 0.01). The 5-year survival rate was 47% in the HPS group and 62% in the non-HPS group. In the Cox proportional hazards model, HPS and Model for End-stage Liver Disease (MELD) score ≥ 18, and Child-Turcotte-Pugh (CTP) class B/C were significant risk factors for mortality after adjusting for other risk factors (HPS hazard ratio [HR] = 1.9, p = 0.01; MELD score ≥ 18 HR = 2.3, p < 0.01; CTP class B/C HR = 2.9, p < 0.01). Compared to that in non-HPS group, the HPS group had a significantly higher incidence of ACLF during follow-up (p < 0.01) and more frequently presented with lung involvement of ACLF (p = 0.03). CONCLUSIONS: In the long-term follow-up cohort, patients with HPS showed poorer prognosis than that of patients without HPS. HPS was a risk factor for ACLF development independent of hepatic dysfunction, and lung involvement was significantly common than without ACLF.