Preliminary validation of the Simplified Psoriatic Arthritis Radiographic Score (SPARS)

Skeletal Radiol. 2019 Jul;48(7):1033-1041. doi: 10.1007/s00256-018-3124-0. Epub 2018 Dec 7.

Abstract

Objective: To develop and validate in a preliminary way a novel radiographic scoring system for psoriatic arthritis (PsA), called Simplified Psoriatic Arthritis Radiographic Score (SPARS).

Materials and methods: Radiographs of hands and feet were obtained from consecutive PsA patients and assessed by two readers. For each joint (30 joints in the hands, ten joints in the feet), the combination of the erosions, joint narrowing space (JNS) and bony proliferation (BP) has been assessed, giving a value of 1 for erosions presence, 1 for JNS presence, and 1 for BP presence (SPARS score range, 0-120). Reliability was assessed by calculating the intraclass correlation coefficient (ICC) and smallest detectable difference (SDD) of the readings. To determine the convergent validity, SPARS was compared to the modified Sharp/van der Heijde Score (mSvdHS) and to the Ratingen scoring system (PARS).

Results: One-hundred and five hands and feet radiographs have been assessed. The inter- and intra-rater reliability were excellent (inter-rater reliability 0.934, and intra-rater reliability for both readers 0.845 and 0.876). The SDD for the average SPARS scores of the two readers was 8.0. SPARS correlated strongly with mSvdHS (r = 0.926; p < 0.0001), and PARS (r = 0.904; p < 0.0001). The mean time to score each of the mSvdHS, PARS, and SPARS was 14.4, 10.1, and 4.5 min, respectively.

Conclusions: The SPARS properties are close to those of the mSvdHS and PARS and is quicker to calculate.

Keywords: Diagnostic imaging; Psoriatic arthritis; Radiography.

Publication types

  • Validation Study

MeSH terms

  • Arthritis, Psoriatic / classification
  • Arthritis, Psoriatic / diagnostic imaging*
  • Cross-Sectional Studies
  • Female
  • Foot Joints / diagnostic imaging*
  • Hand Joints / diagnostic imaging*
  • Humans
  • Male
  • Middle Aged
  • Reproducibility of Results