CRP May Be a Useful Indicator for Spinal Inflammatory Lesions in axSpA

C-reactive protein (CRP) is associated with magnetic resonance imaging (MRI)-detected inflammation in the spine, but not in the sacroiliac joints (SIJ), among patients with axial spondyloarthritis (axSpA), according to results of a systematic review and meta-analysis published in Clinical Rheumatology.

Though axSpA is characterized by structural spinal damage and inflammation, a standardized tool to detect inflammation has not been agreed upon clinically. While guidelines recommend the use of CRP and MRI, the relationship between the two is understudied. The researchers aimed to assess correlations between CRP and MRI-detected inflammation among patients with axSpA.

A systematic review and meta-analysis were performed using the Medline, Embase, and Cochrane Library. The inflammation detected by MRI was assessed using a disease activity score (DAS). The relationship between CRP and MRI-based DAS was analyzed using random-effect models.

A total of 18 studies comprised of 1392 patients were included in the analysis. Eleven studies analyzed the relationship between clinical features of CRP and changes in MRI-based DAS, 3 analyzed the predictive effects of CRP on MRI-based DAS changes via baseline scores, and 10 analyzed the associations between changes in both CRP and MRI-based DAS.

C-reactive protein was significantly associated with spinal MRI-based DAS, with a moderate correlation between CRP change and spinal MRI-based DAS change (r = 0.226; 95% CI, 0.149-0.291; P <.001). However, CRP at baseline was negatively associated with improved spinal MRI-based DAS (r = -0.327; 95% CI, -0.397 to -0.264; P <.001).

A significant association between CRP and SIJ MRI-based DAS was not observed.

Subgroup analysis on the correlation between baseline CRP and MRI-based DAS was conducted based on the follow-up period and divided into 2 categories: less than 52 weeks (short-period) or 52 weeks or more (long-period). The results showed a significant association in both the short- (r = -0.319; 95% CI, -0.414 to -0.217; P <.001) and long-period (r = -0.336; 95% CI, -0.430 to -0.235; P <.001) subgroups.

Among several study limitations, different scoring methods were used to quantify inflammation in axSpA. Additionally, study types were not added to the inclusion criteria because a limited number of studies concerning CRP and MRI in axSpA were available.

The study authors concluded, “The available evidence is in favor of CRP as an indicator and predictive parameter for spinal inflammatory lesions in axSpA. Nevertheless, SIJ MRI seems to be indispensable in disease monitoring.”