Background and aims: Prognostic factors for poor evolution are critical in the setting of limited access to liver transplantation for patients with cirrhosis. We aimed to investigate the impact of hypoxaemia on the outcome in cirrhosis and the evolution of arterial oxygen tension during long-term follow-up in these patients.
Methods: Consecutive cirrhotic patients were prospectively enroled and followed-up in our tertiary referral center. Clinical features, biological tests, arterial blood gases, NT-proBNP levels, pulse oximetry measurements, 12-lead ECG, and transthoracic contrast echocardiography were documented on enrolment. The main outcomes were death and decompensation due to liver disease.
Results: 87 cirrhotic patients were included in the analysis and followed-up for a mean of 16 months. At enrolment, 27 (31%) patients were hypoxaemic, 19 had hepatopulmonary syndrome (HPS), but only 6 of those who were sampled at follow-up had persistent hypoxaemia. During the study period, 22 patients died of liver-related complications. Nine of them (41%) were hypoxaemic on enrolment but none had severe hypoxaemia. Hypoxaemia present at enrollment was not a risk factor for death (p=0.29) or decompensation of liver disease (p=0.7). A higher MELD score at baseline or increase during follow-up was a risk factor for death (p=0.02) and correlated with the presence of hypoxaemia. Normalization of the arterial oxygen levels was accompanied by a significant decrease in NT-proBNP (83 pg/ml vs 0 pg/mL, p=0.023).
Conclusion: Mild and moderate hypoxaemia was frequent in our patients but was not associated with adverse outcome in cirrhosis. Repeated arterial blood gas sampling is advisable, especially in patients diagnosed with hepatopulmonary syndrome.