Physical frailty, genetic predisposition, and the risks of severe non-alcoholic fatty liver disease and cirrhosis: a cohort study

Abstract

Background: Frailty, defined as a phenotype of decreased physiological reserves and diminished ability to respond to stressors, has been linked to the development of chronic diseases. Epidemiological evidence connecting frailty to non-alcoholic fatty liver disease (NAFLD) and cirrhosis risks remain sparse. We aimed to assess the longitudinal associations of frailty with the risks of severe NAFLD and cirrhosis in middle-aged to older adults and further explore the modification role of genetic risk on these associations.

Methods: This study included a total of 398 386 participants from the UK Biobank. Incident cases of severe NAFLD and cirrhosis were ascertained through linked hospital records and death registries. Frailty status was assessed by a modified version of the frailty phenotype, encompassing five key components: weight loss, tiredness, physical activity, gait speed, and grip strength. Participants were classified as pre-frailty if they met one or two of these criteria, and as frailty if they met three or more. Genetic predisposition to NAFLD and cirrhosis was estimated by genetic risk score (GRS) and further categorized into high, intermediate, and low genetic risk levels according to tertiles of GRSs. Cox proportional hazards regression model was employed to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) for their associations.

Results: The mean (standard deviation) age of the study population was 56.6 (8.03) years. 214 408 (53.8%) of the participants was female; 14 924 (3.75%) of participants met the criteria for frailty, 170 498 (42.8%) for pre-frailty, and 212 964 (53.5%) for non-frailty. Over a median follow-up of 12.0 years, we documented 4439 incident severe NAFLD and 3323 incident cirrhosis cases, respectively. Compared with non-frailty, both pre-frailty (HR: 1.50; 95% CI: 1.40-1.60) and frailty (HR: 1.98; 95% CI: 1.77-2.21) were associated with increased risk of NAFLD. Similar associations were observed for cirrhosis, the corresponding HRs (95% CIs) for non-frailty, pre-frailty, and frailty were 1.00 (reference), 1.29 (1.20, 1.38), and 1.90 (1.66, 2.18). Such associations were consistent across all genetic risk levels, with no observed interactions between frailty and GRSs (all P for interactions ≥0.10). Compared with participants with frailty and a low level of genetic risk, the greatest risk increasement in developing severe NAFLD (HR: 3.36; 95% CI: 2.83-3.99) and cirrhosis (HR: 2.81; 95% CI: 2.29-3.44) was both observed in those with frailty and a high level of genetic risk.

Conclusions: Our findings indicate that frailty is a significant predictor of severe NAFLD and cirrhosis, irrespective of genetic predisposition.

Keywords: Cirrhosis; Fatty liver; Frailty; Genetic predisposition; NAFLD.

Figure 1

Associations between frailty status and the risks of severe NAFLD (A) and cirrhosis (B) according to genetic risk. Multivariable cox proportional regression was adjusted for age, sex, BMI, education level, Townsend deprivation index, smoking status, drinking status, a healthy diet score, hypertension, diabetes, hyperlipidaemia, CVD, cancer, NAFLD GRS (for NAFLD) or cirrhosis GRS (for cirrhosis), first 10 principal components of ancestry, and genotype measurement batch. P for interaction was calculated by involving the cross‐product term of frailty phenotype and NAFLD‐GRS or cirrhosis‐GRS in the Cox models. BMI, body mass index; CI, confidence interval; CVD, cardiovascular disease; GRS, genetic risk score; HR, hazard ratio; NAFLD, nonalcoholic fatty liver disease.