Background. Autoantibodies to citrullinated peptides have been shown to be valuable in the diagnosis of rheumatoid arthritis (RA). The expanding repertoire of antibodies to citrullinated peptide antigens (ACPA) has been a topic of great interest in recent reviews and research studies, as has the ability of these autoantibodies to predict disease outcome. Objectives. The aim of this review was to provide an update on the relevance of ACPA as prognostic markers in RA. The ability to identify patients predisposed to an aggressive outcome at the time of initial diagnosis greatly facilitates the selection of appropriate and cost-effective treatment. Methods. A systematic review of the literature was carried out. Studies from 1967 up to June 2014 with data on prognostic value of ACPA were included. Quality assessment was done by using the modified Hayden list for prognostic studies. Meta-analysis was performed using BioStat software. Results. The results of 25 studies were selected for the final review. A total of 6421 patients with RA were included, mainly in inception cohorts, with follow-up duration ranging from one year to ten years. All studies carried prognostic data on all available isotypes of anticyclic citrullinated protein (CCP), while four had data on antimutated citrullinated vimentin (MCV). There was a single relevant study each on anticitrullinated enolase peptide 1 (CEP1) and antichimaeric fibrin/filaggrin citrullinated peptide 1 (CFFCP1). All studies showed ACPA to be strong predictors of joint erosions in RA. Other factors, particularly baseline erosions, showed an additive effect. Anti-MCV appeared to be a marker of a more aggressive form of disease. Ten studies had data on which a meta-analysis could be performed. This gave an overall odds ratio of 4.85 for ACPA (anti-CCP/MCV) positivity being predictive for the development of joint erosions. Two studies with data on anti-CEP1 and anti-CFFCP1 also showed this positive predictive role of ACPA for joint erosions. Conclusions. ACPA are strong predictors of severity in RA. Their use should be part of routine rheumatology practice.