Background/Purpose: The prevalence and characteristics of patients with nonalcoholic fatty liver disease (NAFLD) have not been well characterized in the RA population. The purpose of this analysis was to report the incidence rates and time to event by key risk factors among patients with RA.

Methods: Longitudinal data (2001- 2014) on a cohort of adult RA patients without NAFLD at the index baseline visit from the Corrona Registry were used in this retrospective analysis. Fibrosis-4 index¹ was used to determine the presence of NAFLD. Kaplan-Meier survival analysis was conducted to determine rate and time to event (NAFLD). Breslow (Generalized Wilcoxon) test was used to determine if there were significant differences in the survival distribution for the categorical variables.

Results: This incident NAFLD cohort consisted of 17,481 RA patients. As shown in table 1, higher NAFLD incidence rates were seen among the elderly (53.2), women (70.9), and Caucasians (85) per 1,000 patient years. At baseline, those in Clinical Disease Activity Index remission had the lowest incidence rate whereas the highest rates were seen among those in low disease activity (15.4 vs. 29.3 cases per 1,000 person-years). Those with metabolic syndrome (vs. without) and MTX users (vs. nonusers) had higher incidence rates of 115.3 vs. 45.4 and 69.1 vs. 25.9 cases per 1,000 person-years, respectively. Nonusers (vs. users) of steroids and biologics also had higher incidence rates (62.7 vs.30.2 and 55.5 vs. 39.4 cases per 1,000 person-years respectively) as was for nonsmokers and nonusers of alcohol. There was significant difference in the time to event (NAFLD) for the elderly (4.5 years vs. younger 11.6 years), men (6.4 years vs. 7.3 years for women), and African Americans (6.7 years vs. 8.1 years for Caucasians). Similar differences were also seen for patients with (vs. without) metabolic syndrome, diabetes, dyslipidemia, cardiovascular disease, and hypertension, as well as users of methotrexate, steroids, non users of biologics, smokers, and nonusers of alcohol.

Conclusion: In this cohort of RA patients, incidence of NAFLD varied by demographic factors (higher for Caucasians and women) as well as clinical factors with the highest among those with metabolic syndrome followed by nonusers of alcohol. However, among RA medications, only current use of methotrexate was associated with higher NAFLD rates. Onset of NAFLD appears to be sooner for those with the presence of chronic risk factors such as metabolic syndrome, diabetes, dyslipidemia, CVD, and hypertension. Further understanding for the appropriate management of RA patients at risk for NAFLD is warranted.