Background/Purpose: A diagnosis of (early) axial spondyloarthritis (axSpA) is based on pattern recognition, which can be challenging and may change over time. The aim of this study was to investigate consistency of diagnosis over two years in patients with chronic back pain (less than two years symptoms) suspected of axSpA.

Methods: In the SPACE cohort, patients over 16 years of age referred to the rheumatology outpatient clinic with chronic back pain (≥3 months and < 2 years) starting before the age of 45, suspected of axSpA were included. Follow-up was performed only in patients with at least two SpA features or one SpA feature with a positive likelihood ratio ≥6.4 (Rudwaleit, van der Heijde, Khan, Braun, & Sieper, 2004). 

According to protocol, all SpA features as well as MRI and radiographs of the sacroiliac joints were performed at baseline and two years. Physicians were asked to report whether the diagnosis was axSpA or no axSpA at both timepoints, for which they had information available on all SpA features and locally read imaging.

Only patients with complete data on diagnosis and imaging at baseline and 2 years were included.

Patients were labelled with a consistent axSpA diagnosis if they had a diagnosis of axSpA at baseline and at two-year follow-up. Those patients whose diagnosis switched from axSpA to no axSpA; or from no axSpA to axSpA were labelled inconsistent axSpA diagnosis.

Results: Over two years the diagnostic consistency rate was 84%. The rate of change from axSpA to no axSpA was 9% and from no axSpA to axSpA 7% (table 1).

The patients who only had an axSpA diagnosis at baseline were more often female and less often HLA-B27 positive compared to the other two groups (table 2).

Furthermore, both groups with an inconsistent diagnosis had fewer SpA features and a lower level of confidence of the diagnosis (LoC) compared to the group with a consistent diagnosis of axSpA, especially at baseline.

At two-year follow-up the LoC in the group with an axSpA diagnosis at 2 years only was much lower than in the other two groups. In the group that only had an axSpA diagnosis at baseline, the LoC regarding the diagnosis increased most compared to baseline: physicians were more certain of the diagnosis no axSpA at two-year follow-up than they were of the diagnosis axSpA at baseline.  

The number of patients with sacroiliitis on radiographs and MRI was much higher in the group with a consistent diagnosis of axSpA. Although the percentage of patients with sacroiliitis on MRI increased in the group with a diagnosis of axSpA at two-year follow-up only, this was still much lower (21%) as compared to the patients with a consistent diagnosis (81%). This was in line with a low LoC in this group.

Conclusion: In a cohort of patients with chronic back pain suspected of axSpA the diagnostic consistency rate was high. Interestingly, in the group that only had a diagnosis axSpA at baseline, rheumatologists were more certain about the absence of axSpA at two years than the presence of axSpA at baseline.

Table 1 Consistency of diagnosis over 2 years

Table 2 Characteristics at baseline and 2-year follow up of the group with a consistent diagnosis over 2 years and the groups whose diagnosis changed between baseline and 2-year follow up

« Back to ACR Convergence 2020

ACR Meeting Abstracts - https://acrabstracts.org/abstract/two-year-diagnostic-consistency-in-patients-with-chronic-back-pain-suspected-of-axial-spondyloarthritis-in-protocolised-follow-up-data-from-the-spondyloarthritis-caught-early-cohort/