Su1397 Very Low and Similar Proportions of Patients With Celiac Disease Among Patients With Normal and Elevated ALT Levels, Especially Among Non-Whites: Results of a U.S. Nationwide Sample of 16,795 Persons

Abstract

Background: Studies investigating the prevalence of celiac disease (CD) in persons with elevated aminotransferases have yielded conflicting results. It raises the question, whether clinicians should consider including celiac serology in initial evaluation of patients with abnormal liver tests. Elevated ALT in asymptomatic patients is commonly encountered in practice and a common reason for referral to gastroenterologists or hepatologists. The aim of our study is to determine the prevalence of CD in a large nationwide sample representative of the general population in the U.S. Methods: This study included 16,975 persons aged 6 years or older from the National Health and Nutrition Examination Survey (NHANES), which collects nationally representative data in the U.S. civilian population. Data from 20092012 were combined for this study. Upper limit of normal for ALT was ≥29 in men and ≥ 22 for women (Ruhl et al, Hepatol 2012). CD was defined as having positive tissue transglutaminase plus endomysial IgA Abs or a reported clinical diagnosis (reported diagnosis of CD by a doctor or a health-care professional and being on gluten-free diet). Primary study outcome was proportions of patients with CD in individuals with normal ALT and elevated ALT. Sample weights were used to determine weighted prevalence of CD. Results: The median age of the entire cohort was 36 years (range 6-80, IQR 16-57). Close to half (49.6%) were male with similar distribution between those with normal and elevated ALT. Overall, 38% were Non-Hispanic White, 22.9% were Non-Hispanic Blacks, 27.7% were Hispanic, and 11.4% had other races. Elevated ALT was seen in 29.1% of the cohort. Among the 3,725 individuals with elevated ALT, 41.5% were Non-Hispanic Whites, 16.9% were Non-Hispanic Blacks, 31.5% were Hispanic, and 10.1% were other races. CD was found in 56 individuals out of 16,975 (0.61%, 95%CI 0.43%-0.79%): 20 with elevated ALT (0.80%, 95% CI 0.42%-1.2%) and 29 with normal ALT (0.61%, 95% CI 0.30%-0.91%) (Table 1). CD prevalence was highest in non-Hispanic White with both normal and elevated ALT at 0.87% (95% CI, 0.40%-1.3%) and 1.12% (95% CI, 0.58%-1.65%), respectively. Overall, CD prevalence was much lower in other races: 0.09% (95% CI 0%-0.2%) for NonHispanic Blacks, 0.12% (95% CI 0%-0.24%) in Hispanic, and 0.05% (95% CI, 0%-0.1%) in other races, and these were also similar among individuals with normal and elevated ALT (Table 1). Conclusion: Interestingly, the prevalence of CD in individuals with elevated ALT is similar to that seen in individuals with normal liver enzymes in U.S. and was approximately 1% or less in both populations. CD prevalence is particularly low in non-White individuals. Based on our results, we do not advocate routine testing with celiac serology to determine if CD is the cause of abnormal liver enzymes, especially in asymptomatic non-White patients.