Background
This risk analysis aimed to explore all modifiable factors associated with prolonged mechanical ventilation (lasting > 24 h) after liver transplantation, based on prospectively collected data from a clinical trial.
Methods
We evaluated 306 candidates. Ninety-three patients were excluded for low risk for transfusion (preoperative haemoglobin > 130 g.l−1), and 31 patients were excluded for anticoagulation therapy, bleeding disorders, familial polyneuropathy, or emergency status. Risk factors were initially identified with a log-binomial regression model. Relative risk was then calculated and adjusted for age, sex, and disease severity (Model for End-Stage Liver Disease [MELD] score).
Results
Early tracheal extubation was performed in 149 patients (84.7%), and 27 patients (15.3%) required prolonged mechanical ventilation. Reoperations were required for 6.04% of the early extubated patients and 44% of patients who underwent prolonged ventilation (p = 0.001). A MELD score > 23 was the main risk factor for prolonged ventilation. Once modifiable risk factors were adjusted for MELD score, sex, and age, three factors were significantly associated with prolonged ventilation: tranexamic acid (p = 0.007) and red blood cell (p = 0.001) infusion and the occurrence of postreperfusion syndrome (p = 0.004). The median (IQR) ICU stay was 3 (2–4) days in the early extubation group vs. 5 (3–10) days in the prolonged ventilation group (p = 0.001). The median hospital stay was also significantly shorter after early extubation, at 14 (10–24) days, vs. 25 (14–55) days in the prolonged ventilation group (p = 0.001). Eight patients in the early-extubation group (5.52%) were readmitted to the ICU, nearly all for reoperations, with no between-group differences in ICU readmissions (prolonged ventilation group, 3.7%). Conclusion.
We conclude that bleeding and postreperfusion syndrome are the main modifiable factors associated with prolonged mechanical ventilation and length of ICU stay, suggesting that trials should explore vasopressor support strategies and other interventions prior to graft reperfusion that might prevent potential fibrinolysis.
Trial Registration.
European Clinical Trials Database (EudraCT 2018–002510-13,) and on ClinicalTrials.gov (NCT01539057).