Background/Purpose: The prevalence of asymptomatic carotid plaque (a-CP) is high in patients with inflammatory joint disease (IJD). Patients with a-CP should receive intensive lipid lowering (LL) treatment in cardiovascular prevention. Our aim was to evaluate presence of a-CP, if CV risk calculators predict presence of a-CP in patients with IJD and if optimizing cut off points in various risk calculators will improve this prediction.

Methods: We performed CV risk stratification in patients with IJD (n= 345), [rheumatoid arthritis (RA)(n=210), ankylosing spondylitis (AS)(n=87) and psoriatic arthritis (PsA)(n=49)] by using SCORE, Framingham and Reynolds CV risk algorithms with recommended cut off points at 5%, 10% and 10% respectively. Ultrasound of the carotid arteries was performed. Cross-tabulations, Chi² and ROC curves were used to calculate sensitivity/specificity, odds- and likelihood ratio (LR) for identifying a-CP. The ROC closest point (0.1) and 80% sensitivity was used for optimizing the identification of a-CP.  

Results: A-CP was present with similar frequency in RA, AS and PsA (48.3%, 41.4% and 41.7% (p=0.46). Two hundred and eleven patients had SCORE<5% indicating no need for LL prevention (Table 1). However, 72 (34.1%) of these patients had a-CP and should receive intensive LL treatment. In patients with SCORE ≥5% (n=124), indicating a need for moderate LL prevention, 81 (65.3%) had a-CP and should therefore be categorized to intensive LL treatment. The sensitivity for identifying a-CP was 0.53, the specificity: 0.76 and LR+: 2.24. Optimizing the cut off value by using closest point (0, 1) (SCORE: 4%: specificity: 0.65, specificity: 0.72, LR+:2.33) or 80% specificity (SCORE: 2.65: sensitivity: 0.80 specificity: 0.56 LR+:1.83) did not improve the ability of SCORE to identify a-CP. According to the new guidelines for prevention of CV disease (2011)¹, the sensitivity, specificity and LR+ for identifying a-CP was 0.47, 0.77 and 2.05 respectively. The associations between a-CP and cut off values for other risk calculators are shown in Table 2.

Conclusion: Carotid ultrasound will assist in the correct classification of patients with IJD into intensive LL in about 1/3 of patients with SCORE<5% and 2/3 of patients with SCORE≥5%. The CV risk calculators are poor predictors of a-CP in patients with IJD.

Table 1.  European heart SCORE according to presence of symptomatic carotid plaque

Table 2. Framingham and Reynolds cardiovascular risk scores

Ref

1.      Catapano AL et al. ESC/EAS Guidelines for the management of dyslipidaemias. Atherosclerosis 2011;217S:1-44