Background Stringent remission criteria are crucial in rheumatoid arthritis (RA) assessment. Disease activity score in 28 joints (DAS28)-remission has not been included among American College of Rheumatology/European League Against Rheumatism definitions, because of its association with significant residual disease activity, partly due to high weighting of acute-phase reactants (APR). New, more stringent cut-points for DAS28-remission have recently been proposed that are suggested to reflect remission by clinical and simplified disease activity indices (clinical disease activity index (CDAI), simple disease activity index (SDAI)). However, their stringency in therapies directly influencing APR, like IL-6-blockers, has not been tested. We tested the new cut-points in patients with RA receiving tocilizumab. Methods We used data from randomised controlled trials of tocilizumab and evaluated patients in remission according to new DAS28-C-reactive protein (DAS-CRP) and DAS-erythrocyte sedimentation rate (DAS-ESR) cut-points (1.9 and 2.2). We assessed their disease activity state using the CDAI, SDAI and Boolean criteria and analysed their individual residual core set variables, like swollen joint counts (SJC28). Results About 50% of patients in DAS28-CRP-remission (<1.9) fell into higher disease activity states when assessed with CDAI, SDAI or Boolean criteria. Also, 15% had three or more (up to eight) SJC. Even higher disease activity was seen in patients classified as being in DAS28-ESR-remission (<2.2). Conclusions Even with new, more stringent cut-points, DAS28-remission is frequently associated with considerable residual clinical disease activity, indicating that this limitation of the DAS28 is related to score construction rather than the choice of cut-points.