Background and aims: Hepatitis E virus-related acute liver failure (HEV-ALF) rapidly worsens and has a high mortality. However, no simple and specific parameters for predicting short-term mortality are available.
Methods: A derivation cohort including 97 patients with HEV-ALF and another validation cohort were enrolled. Laboratory and clinical parameters were recorded. Platelet count, model for end-stage liver disease (MELD), and King's College criteria (KCC) were separately used for predicting mortality, and the levels of cytokines associated with systemic inflammation, platelet production, and platelet activation were measured.
Results: Platelet counts were significantly lower in patients with HEV-ALF, and nonsurvivors had lower platelet counts than survivors (p < 0.001). Platelet count was an independent risk factor for predicting 28- and 90-day mortality in patients with HEV-ALF. The AUROC of the baseline platelet count (cutoff, 131 × 109/L) for 28- and 90-day mortality was 0.786 and 0.764, respectively, which was superior to KCC score (p < 0.05) and comparable to MELD score. Furthermore, the platelet counts at 3 and 7 days after ALF diagnosis had similar predictive power for 28- and 90-day mortality. The value of platelet count was also confirmed in the validation cohort. Moreover, platelet-associated cytokines, including thrombopoietin, platelet factor 4, and P-selectin, were increased in patients with HEV-ALF.
Conclusions: Decreased platelet count is a simple and reliable indicator for predicting 28- and 90-day mortality in patients with HEV-ALF. Overactivation of platelets is an important risk for platelet counts decrease, and treatment aiming at platelet count recovery may be considered.
Keywords: Acute hepatitis; Acute liver failure; Hepatitis E virus; King’s College criteria; Model for end-stage liver disease; Mortality; Outcome; Platelet count; Prognosis; Systemic inflammation.
© 2022. Asian Pacific Association for the Study of the Liver.