Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
While enthesitis, dactylitis and nail involvement are recognised as important outcomes of psoriatic arthritis (PsA) in the core set of domains in PsA,1 2 it is still unclear how these outcomes should best be measured.1 2 We systematically reviewed the instruments and the cut-offs used to report state or improvement, for enthesitis, dactylitis and nail involvement in recent randomised controlled trials (RCTs) in PsA.
A systematic literature review of RCTs on any pharmacological intervention in patients with PsA was conducted to inform the European League Against Rheumatism (EULAR) recommendations for the management of PsA, by searching Medline, Embase and Cochrane datasets for the period 2010–2015.3 4 Only published papers and only results of the placebo-controlled phases were analysed. The presence and type of all outcome measures reflecting enthesitis, dactylitis and nail involvement were collected. Cut-offs used for each measure (either as state or change, absolute or relative) were also collected. The proportion of trials in which each of the cut-offs for each measure was reported was calculated.
Of 2278 articles screened, 14 trials met the inclusion criteria: 4 (29%) reported on non-biologic drugs (included targeted synthetic disease-modifying antirheumatic drugs, one trial), 5 (36%) on tumour necrosis factor inhibitors, 4 (29%) on other biologic modes of action and there was one strategy trial. The trials included a total of 4744 patients. Four of the trials (29%) did not report any outcome on any of the three domains of interest (table 1). Enthesitis and dactylitis outcomes were reported in the remaining 10 trials, while nail involvement was only reported in three trials (21%). These three outcomes have been measured in several different ways, none of which having been used in more than three trials (21%), and the majority of them was actually employed in only one (7%) or two (14%) trials. Different instruments have been used, different cut-offs and different statistics reported (eg, mean and median improvement or resolution of the outcome, eg, enthesitis score of zero) (table 1). It was often the case that the same outcome measure was used (eg, the Maastricht Ankylosing Spondylitis Enthesitis Score (MASES)), but then reported in such different ways (eg, percentage of change, percentage ≥1, percentage of improvement in one tendon/ligament, etc) that the potential uniformity in the measures used got diluted (table 1). There was also heterogeneity in the timing of report of the outcome measures across trials.
In summary, there is substantial lack of uniformity in the measurement of enthesitis, dactylitis and nail involvement in recent clinical trials of PsA. A similar lack of uniformity had previous been described for patient-reported outcomes in PsA,5 6 and in what concerns enthesitis, dactylitis and nail involvement measurement is the heterogeneity even larger. This relates to both the instruments used and the evaluation and interpretation of the results. An important aspect that requires attention are the ways in which the data are reported, namely the cut-offs chosen or the different statistics reported, which make the heterogeneity larger, even when one single outcome measure (see the example of MASES) is being used. Assessment of dactylitis and enthesitis needs further development taking both their resolution and appearance into account. Another methodological aspect deserving attention is the fact that these outcomes are actually only investigated in patients with active involvement at baseline, which violates the principle of intention-to-treat analysis. Consensus is necessary and more elegant solutions should be considered. An update of the PsA Core Set of domains has just been published, however, without any indication of the instruments and cut-offs to be used.2 Harmonisation of measures to be used in trials and possibly also clinical practice is desirable to allow for optimal assessment and better comparability of the efficacy of interventions.
Contributors All authors contributed to this letter and approved its last version.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.