Objective, We have shown that the presence of 5 or more effusions and high
medication use at first clinic visit predicted clinical progression in pati
ents with psoriatic arthritis (PsA), while a low erythrocyte sedimentation
rate (ESR) was "protective." These clinical indicators will change over the
course of a patient's disease. We investigated whether there is additional
prognostic information available through monitoring these indicators at ea
ch clinic visit.
Methods. A total of 365 patients with at least 2 followup visits at the PsA
Clinic who did not have 10 or more damaged joints at first visit were incl
uded. Clinical assessments including the number of actively inflamed and da
maged joints were carried out every 6-12 mo according to a standard protoco
l. The analysis used a generalized linear model that relates the number of
damaged joints that developed between consecutive clinic visits to the info
rmation available at the first of the 2 visits, and in which the informatio
n was added to a baseline model including the first visit variables previou
sly shown to be important.
Results. Single factor analysis suggested that the addition of functional c
lass, number of actively inflamed joints, and Lansbury index provide progno
stic information for subsequent damage. The final multivariate model includ
es time varying information on the number of actively inflamed joints, func
tional class, and current damage, as well as first visit information on pri
or medication as predictive of progression of damage, plus male sex and a l
ow ESR at first visit as "protective,"
Conclusion. Time varying predictors for damage are important and should be
monitored longitudinally in patients with PsA.