Long-term Prognosis of Patients With Metabolic (dysfunction)-Associated Fatty Liver Disease by Non-invasive Methods

Marie Decraecker; Dan Dutartre; Jean-Baptiste Hiriart; Marie Irles-Depé; Faiza Chermak; Juliette Foucher; Victor de Lédinghen


Aliment Pharmacol Ther. 2022;55(5):580-592. 

In This Article

Abstract and Introduction


Background: Non-invasive assessment of fibrosis is predictive of the prognosis of non-alcoholic and alcoholic fatty liver disease but this has not been demonstrated in metabolic (dysfunction)-associated fatty liver disease (MAFLD).

Aims: We assessed the prognosis of non-invasive methods in patients with MAFLD.

Methods: All consecutive patients with MAFLD, with liver stiffness measurements, FIB-4 (Fibrosis-4), and LIVERFASt were included in this cohort study. The primary endpoint was analysed by the Kaplan-Meier method and secondary endpoints were estimated by Gray test or logistic regression. Factors independently associated with overall mortality and morbidity were identified by a multivariate Cox model. The prognostic performance of non-invasive methods for prediction of mortality was evaluated by Harrell's C-index and for morbidity by area under the receiver operating characteristics curve (AUC).

Results: A total of 1239 patients with MAFLD were analysed (median age 56 years, male 56.5%, median body mass index 31 kg/m2 and obesity 59%). The median follow-up was 62 months [42–91 months] and 73 (5.8%) subjects died. Baseline results of non-invasive methods were correlated with overall and liver-related mortalities (P < 0.001), and with all-cause and liver-related outcomes (P < 0.001). A predictive model (composed of clinical parameters and liver stiffness measurement, FIB-4 or LIVERFASt) was an excellent predictor of overall and liver-related mortalities (C-index 0.8–0.9), and a good predictor of overall and liver-related morbidities (AUC 0.72–0.74).

Conclusion: Baseline liver stiffness measurement, FIB-4 and LIVERFASt can predict global and liver-related mortality and morbidity in patients with MAFLD and could be prognosis endpoints in clinical trials.


In recent years, there has been a large increase in the prevalence of obesity and metabolic syndrome in the general population, making non-alcoholic fatty liver disease (NAFLD) the main cause of cirrhosis and hepatocellular carcinoma over the next 5 years.[1–5] Low-to-moderate alcohol consumption is also common in the general population, and therefore probably frequent in the NAFLD population; up to 1 in 10 patients is diagnosed with NAFLD despite the existence of undetected excessive alcohol consumption.[6–8] However, concomitant alcohol consumption typically excludes a diagnosis of NAFLD in practice, but NAFLD and alcoholic fatty liver disease (AFLD) share pathogenetic mechanisms, raising the question of whether these two entities need to be distinguished.[9–12]

In this context, an international panel of experts recently proposed a new nomenclature for fatty liver diseases, representing the hepatic manifestation of a multisystemic disorder. The term metabolic (dysfunction)-associated fatty liver disease (MAFLD) is based on inclusion criteria highlighting systemic metabolic dysfunction, irrespective of alcohol consumption or other concomitant liver diseases.[13–21]

MAFLD covers a broad spectrum of liver damage, ranging from benign steatosis to steatohepatitis, fibrosis and, ultimately, cirrhosis with its critical complications. In 75% of cases, cirrhosis is diagnosed at a decompensated stage and a large number of deaths related to cirrhosis could be avoided if hepatopathy were to be diagnosed at an early, asymptomatic stage. Fibrosis is validated as an independent prognostic factor for death and liver-related outcomes.[22–24] Assessment of the severity of chronic liver diseases by non-invasive methods has been increasing, allowing for risk stratification based on prognosis; liver biopsy is now reserved for complicated cases.[25–30] Liver stiffness measurement by transient elastography (Fibroscan),[31,32] FIB-4[28,29,33] and LIVERFASt (an artificial intelligence algorithm, Fibronostics)[34,35] are accurate methods for non-invasive diagnosis of liver fibrosis. Moreover, liver stiffness measurement and FIB-4 are predictive of overall and liver-related survival and complications.

However, their utility for non-invasive assessment of fibrosis and prognosis has not been validated in clinical practice. We assessed the ability of non-invasive liver stiffness measurement (Fibroscan), FIB-4 and LIVERFASt (Fibronostics) to identify prognostic factors for MAFLD.