Objective: This study evaluated and compared the adherence to the CONSORT for quality of reports on the randomized controlled trials (RCT) abstracts by four major Korean Science Citation Index (SCI) journals and The New England Journal of Medicine (NEJM). Methods: A descriptive analysis of published RCT abstracts in Korean SCI journals and NEJM from 2007/01 to 2011/06 was conducted by two reviewers, independently extracting data from a PubMed search. A modification of CONSORT for abstract was used including 16 checklist items. Reporting of checklist items for individual group was conducted to compare adherence patterns between two groups. Results: We identified the potential 57 RCT abstracts from Korean SCI and 50 from NEJM meeting our inclusion criteria; among them, three abstracts from Korean SCI and one from NEJM were excluded. Among total 16 checklist items based on CONSORT statement, Korean SCI journals and NEJM were statistically equivalent in 4 items; Korean journals were better in three items and NEJM were in nine. The methodological quality domains were inadequately reported in both journals: allocation concealment about 1.9% and 4.0%, and blinding 44.4% and 40.8%. In general, The CONSORT adherence of NEJM was better than that of Korean SCI in the method and result domain (p < 0.0001). Conclusions: The quality of NEJM reporting of RCT abstracts was better than that of Korean SCI, in general. This study on adherence of RCT reports from Korean SCI journals and NEJM abstracts to the CONSORT statement reveals that there is a need for improvement, especially Korean SCI. Further investigation on the quality of RCT reports and ways to improve reporting quality is required.