Although nonalcoholic fatty liver disease (NAFLD) and lipoprotein(a) [Lp(a)] are associated with cardiovascular diseases, existing data on Lp(a) in NAFLD are conflicting. The aim of this systematic review and meta-analysis was to summarize and compare data on circulating Lp(a) between NAFLD patients and non-NAFLD controls. A systematic literature search was performed in PubMed, Scopus, and Cochrane Library. This meta-analysis included 18 studies containing data on 74 691 individuals (20 220 patients with NAFLD and 54 471 controls). Circulating Lp(a) was similar between patients with NAFLD and controls (standardized mean difference [SMD] 0.09; 95% confidence interval [95% CI] -0.21, 0.38). The heterogeneity among studies was high (I2=100%); no publication bias was detected (Egger's test P=0.941). However, in subgroup analysis, Lp(a) was lower in NAFLD patients than controls, when Lp(a) was measured with nephelometry (SMD -0.26; 95% CI -0.46, -0.06), but not turbidimetry; this analysis also resulted in mild reduction of heterogeneity within the subgroup of nephelometry (I2=87%). The sensitivity analyses, based on the exclusion of studies with Newcastle-Ottawa Scale score ≤6 (n=5), studies in which liver biopsy was used for NAFLD diagnosis (n=4) or studies that adopted the criteria of metabolic dysfunction-associated fatty liver disease (n=2), and meta-regression analysis did not explain the high heterogeneity among studies. Overall, circulating Lp(a) was similar between NAFLD patients and non-NAFLD controls; however, patients with NAFLD had lower circulating Lp(a) compared with controls, when Lp(a) was measured with nephelometry. These results should be cautiously interpreted, because of the high heterogeneity among studies.