Risk Factors for Graft Steatosis after Liver Transplantation using Controlled Attenuation Parameter Measurements

Abstract

Background: Liver steatosis is a cause of graft dysfunction after liver transplantation. The current study aims to determine the risk factors associated with the development of graft steatosis in a large cohort of liver transplant recipients.

Methods: Consecutive adult patients transplanted from 2003 to 2014 underwent liver stiffness and controlled attenuation parameter (CAP) measurements using transient elastography. Liver steatosis was defined as minimal (<5%), mild (5-33%), moderate (34-66%), and severe (≥67%) if the CAP score was <248, 248-267, 268-279, and ≥280 dB/m respectively. Longitudinal history including diabetes, hyperlipidemia, hypertension, and immunosuppressive regimen were recorded.

Results: A total of 549 liver transplant recipients underwent valid transient elastography, of which 359 (72%) were male. Using the predefined CAP cut-offs, 345 (63%), 48 (9%), 27 (5%), and 129 (23%) had minimal, mild, moderate, and severe steatosis respectively. There was a significant correlation between the CAP score and age at transplant (r=0.104), age at CAP measurement (r=0.116), and body mass index (BMI) at the time of CAP measurement (r=0.567)(all p<0.05). A higher CAP score was observed for male patients (229 vs 212 dB/m, p=0.011), on-treatment diabetes (232 vs 219 dB/m, p=0.029), on-treatment hypertension (237 vs 200 dB/m, p<0.001), and on-treatment hyperlipidemia (239 vs 217 dB/m, p<0.001). No difference in CAP score was observed for those requiring mTOR-inhibitors or mycophenolate, although a lower CAP was observed for prednisone (212 vs 227, p=0.022). A significant correlation was observed between CAP scores and age at the time of liver transplant (r=0.104, p=0.015), age at the time of CAP score measurement (r=0.116, p=0.007), and the BMI (r=0.567, p<0.001). No significant correlation was observed between CAP score and time from transplant to CAP measurement. After multivariate analysis, only hypertension (OR 0.54) and BMI (OR 1.41) remains significant factors associated with moderate-severe graft steatosis. There was no correlation between the CAP score and liver stiffness measurement (p=0.89).

Discussion: The liver stiffness and CAP score is a non-invasive method of diagnosing hepatic fibrosis and steatosis respectively, with validated cut-offs correlating to different severity. After multivariate analysis on significant univariate variables, only hypertension and BMI was significantly associated with moderate-severe steatosis. However, there was no correlation between liver stiffness and CAP score, suggesting that other causes other than steatosis were responsible for graft fibrosis.

Conclusion: Post-transplant graft steatosis was common, with 28% developing moderate-severe steatosis. Increase in BMI and the presence of hypertension was significantly associated with the development of moderate-severe graft steatosis. However, there was no correlation between liver stiffness and CAP scores to suggest increase in fibrosis.