The best imaging target for remission is demonstration that no progression of structural damage has occurred. Radiographs are able only to show inhibition of structural damage over a 6 to 12 month period, and they do not demonstrate erosions as early after onset of clinical manifestations as do Magnetic Resonance Imaging [MRI] or Power Doppler Ultrasonography [PDUS]. Likely reflecting erosions and JSN already destined to happen, future radiographic progression is best determined by damage at baseline. Radiographs remain the “gold standard” by which inhibition of structural damage is demonstrated – from a regulatory point of view, this definition, also clinically meaningful, is “no progression”: either change ≤0 in vander Heijde modified Total Sharp Scores [vdHS] or ≤0.5 to account for variability between ≥2 readers. Clinically meaningful changes in vdHS scores have been variously described to be ≥3, ≥5 points or >smallest detectable change [SDC] over 12 or 24 months. However such cut- offs have little meaning unless viewed in the context of baseline damage [and thus the individual patient], and more frequently than not SDC values well exceed observed mean changes over time.

PDUS visualizes synovitis and erosions; MRI synovitis, bone edema [BE or osteitis] and erosions – both better visualize structural damage early in disease as well as its progression. Prospectively we know that both imaging methodologies demonstrate changes ascribed to inflammation that ultimately result in damage observed by x ray. Most would surmise that synovitis by PDUS and/or BE by MRI at 3 or 6 months will be reflected by progression in radiographic vdHS scores at 12 months. Although not fully prospectively validated, MRI hopefully will soon become the more sensitive standard for assessing progression of structural damage in randomized controlled trials [RCTs]. PDUS lends itself more easily to daily clinical practice but has been successfully utilized in RCTs, based on standardization of techniques and acquisition of imaging.

The opposite process is less clear. Although healing of erosions has been demonstrated longitudinally by x ray, resolution of erosions evident on MRI appears to be less reversible. Although PDUS can demonstrate synovitis and erosions, it is not able to show BE, and gray scale US has evidenced abnormalities even in healthy volunteers.

To return to the target of “remission” – do we need a separate definition based on imaging? or to add an imaging component to our remission criteria? In a majority of RCTs, a large percentage of patients considered in remission by stringent definitions such as SDAI, CDAI and Boolean do not demonstrate progression in vdHS scores and longitudinal observational studies [LOS] have shown similarly that few have persistent synovitis/erosions by PDUS or BE/erosions by MRI. This is particularly true if there are no swollen joints or elevated CRP. The question remains, are there still patients with “sub-clinical” synovitis/inflammation that can be detected by these more sensitive imaging techniques who will not do well over time or tolerate tapering of therapy?

Here again definitions of clinically meaningful progression, or the lack thereof, based on PDUS and MRI are still to be determined. Thus the importance of the planned RCTs examining attainment of remission based on clinical definitions with/without PDUS: ARCTIC and TURA, or with/without MRI: IMAGINE. Until these results are known it is best that we rely upon strict clinical definitions of remission, particularly in early disease, knowing that, if maintained over ≥12 months these will result in little progression of structural damage.

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.6234