Prevalence of Nonalcoholic Fatty Liver Disease (NAFLD) in Patients With Type 1 Diabetes Mellitus: A Systematic Review and Meta-Analysis

CONTEXT: Nonalcoholic fatty liver disease (NAFLD) prevalence is high, especially in patients with obesity and type 2 diabetes, and is expected to rise steeply in the coming decades. OBJECTIVE: We estimated NAFLD prevalence in patients with type 1 diabetes and explored associated characteristics and outcomes. DATA SOURCES: We reviewed PubMed and Embase for studies on NAFLD and type 1 diabetes to March 2020. We screened references of included articles. STUDY SELECTION: Two authors independently screened titles/abstracts. One author screened full text articles. NAFLD was defined as described in the individual studies: steatosis and/or fibrosis. Studies not reporting alternative causes of hepatic steatosis or defining NAFLD only as elevated liver enzymes, were excluded. Initially, 919 articles met the selection criteria. DATA EXTRACTION: One researcher performed data extraction and risk of bias assessment using standardized tables. DATA SYNTHESIS: We assessed pooled prevalence rates by meta-analysis using a random-effects model, subsequently exploring heterogeneity by subgroup-, meta-regression-, and sensitivity analysis. Twenty studies between 2009 and 2019 were included (n = 3901). Pooled NAFLD prevalence was 19.3% (95% CI, 12.3%-27.5%), increasing to 22.0% (95% CI, 13.9%-31.2%) in adults only. Pooled prevalence of ultrasound studies was high (27.1%, 95% CI, 18.7%-36.3%) compared to studies using magnetic resonance imaging (8.6%, 95% CI, 2.1%-18.6%), liver biopsy (19.3%, 95% CI, 10.0%-30.7%), or transient elastography (2.3%, 95% CI, 0.6%-4.8%). CONCLUSION: NAFLD prevalence in patients with type 1 diabetes is considerable and is highly dependent on the specific diagnostic modality and NAFLD definition used. These data are helpful in directing actions to standardize NAFLD diagnosis, which will help defining contributing mechanisms and outcomes.