Background Quantitating the degree of inflammation has become essential to tailor the treatment strategy of various rheumatic diseases. Traditionally used measures to assess disease activity and treatment response in spondyloarthritis (SpA) are ESR, CRP and Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and Ankylosing Spondylitis Disease Activity Score (ASDAS). Serum amyloid A (SAA) is an acute-phase reactant predominantly synthesized in the liver by hepatocytes in response to proinflammatory cytokines. Some studies have shown that SAA can be a valuable indicator of disease activity, damage and functional impairment, however it has not been extensively used in clinical practice.
Objectives To investigate if SAA levels have better correlation with conventional assessments used to monitor anti–tumor necrosis factorα (anti-TNF) treatment in SpA than CRP and ESR.
Methods Prospective study, including SpA patients under anti-TNFα treatment at a Rheumatology Department of a Portuguese Universitary Hospital. The following parameters were collected and registered in two different evaluations 6 months apart from each other: SAA, CRP and ESR levels, BASDAI, ASDAS-PCR, ASDAS-VS, BASFI, BASMI, swollen and tender joints counts (SJC and TJC), MASES, SPARCC, and patient global assessment (PtGA), physician global assessment (PGA), total back pain (TBP), nocturnal back pain (NBP) measured in a visual analogue scale (VAS). The variation of each parameter was calculated as the difference between the levels registered at each evaluation (baseline and 6 months) and presented as Δ(parameter). We compared the correlation between ΔSAA, ΔCRP and ΔESR levels with ΔBASDAI, ΔASDAS-PCR, ΔASDAS-VS, ΔBASFI, ΔBASMI, ΔSJC, ΔTJC, ΔMASES, ΔSPARCC, ΔPtGA-VAS, ΔPGA-VAS, ΔTBP-VAS, ΔNBP-VAS. The statistical analysis was performed using SPSS 21.0 software, and p<0.05 was taken to indicate statistical significance. Correlation was calculated using the Spearman rank correlation (r).
Results 89 patients were included, 58.4% (n=52) were male. On baseline the median age was 44.0 years (range 21.0–74.6) and median disease duration was 18.8 years (2- 51.6). ΔSAA was moderately correlated with ΔCRP (r=0.65, p<0.001) and had lower correlation with ΔESR (r=0.28, p=0.009). ΔSAA correlated with ΔNBP-VAS (r=0.260, p=0.016), but ESR and CRP did not correlate with this parameter. We also found statistically significant correlation between ΔSAA and ΔASDAS-VS (r=0.257, p=0.017), ΔASDAS-CRP (r=0.387, p<0.001), ΔBASFI (r=0.301, p=0.005). ΔCRP also showed significant correlation with ΔBASFI, but it was weaker than that observed with ΔSAA (r=0.230, p=0.033). There was no statistically significant correlation between ΔSAA levels and ΔMASES, although ΔCRP had a weak correlation (r=0.217, p=0.041). There was no significant correlation between neither ΔSAA, ΔCRP or ΔESR and the following parameters: ΔTBP-VAS, ΔPtGA-VAS, ΔPGA-VAS ΔBASDAI, ΔBASMI, ΔSPARCC, ΔTJC or ΔSJC.
Conclusions This study suggests that SAA can be an useful tool in monitoring treatment with anti-TNFα and that could be introduced in clinical practice. However more studies, with larger sample sizes, should be undertaken to better assess this subject.
Disclosure of Interest None declared