Objectives Anti-keratin antibody (AKA) is a serum antibody for patients with rheumatoid arthritis (RA), and it has a high specificity. Diagnostic role of AKA in RA was evaluated in this study. Methods PubMed, EMBASE, and Web of Science were searched to acquire eligible studies. Articles published before 15 March 2018 were considered to be included. Quality Assessment of Diagnostic Accuracy Studies 2 was used to evaluate the risk of bias and application concern of the included articles. Pooled analysis of diagnostic indicators of AKA for RA was conducted by using a random effects model. Subgroup analysis was employed to explore the potential influencing factors. RevMan 5.3, Stata 11.0, and Meta-DiSc 1.4 software were used in this study. Results A total of 15 studies (2350 positive and 2067 negative participants) were included. The pooled sensitivity was 0.46 (95% CI 0.44-0.48), pooled specificity was 0.94 (95% CI 0.93-0.95), and pooled diagnostic odds ratio was 15.86 (95% CI 9.48-26.52). In addition, the area under the curve was 0.7194. Conclusions The current evidence indicated that AKA has high diagnostic specificity in RA and may be useful for RA diagnostic application in clinic.