• Ankylosing Spondylitis
• Low Back Pain
• Spondyloarthritis

We read with interest the article by Haroon et al1 indicating that approximately 40% of patients with acute anterior uveitis (AAU) have undiagnosed spondyloarthritis. Unfortunately, delay to diagnosis remains a huge challenge in axial spondyloarthritis (axSpA) despite the publication of the Assessment of SpondyloArthritis international Society (ASAS) classification criteria, which formally recognises the role of MRI in diagnosis.2 We recently reported that the mean delay in diagnosis among 1193 patients attending two large secondary care centres in the UK was 8.5 years.3 Despite a 51% increase in new diagnoses since 2009, the delay in diagnosis has not improved and the presence of AAU was associated with an even longer delay. Given that the majority of axSpA patients present with chronic back pain (CBP),4 ,5 and that AAU is the commonest extra-articular manifestation affecting approximately 26% of patients,6 recognition of the co-existence of CBP and AAU should prompt primary care physicians to consider a diagnosis of axSpA and could potentially be used as a screening tool to aid diagnosis.

We undertook an audit of a UK primary care practice in order to determine whether primary care physicians considered a diagnosis of axSpA in patients with co-existing CBP and AAU. The following information was sought: patient demographics; age at first documentation of CBP or AAU; whether specific questions relating to the ASAS inflammatory back pain classification criteria were asked and the answers documented;7 were the presence of extra-articular features (specifically inflammatory bowel disease and psoriasis) recorded; were relevant investigations (erythrocyte sedimentation rate (ESR) or C reactive protein (CRP), human leucocyte antigen (HLA)-B27) undertaken and whether patients had been referred to secondary care specialists (table 1).

The practice population comprised 9339 adults aged >18 years. The clinical records of 34 patients (19 women) satisfying the read-codes for CBP and AAU were evaluated in detail. Eleven had been referred to secondary care (eight rheumatology, one gastroenterology, two ophthalmology), four of whom were diagnosed with axSpA (three ankylosing spondylitis, one non-radiographic axSpA). The median age at time of first documentation of CBP was 48.0 years (IQR 29.1), 14 presented <45 years. The median age at time of first documentation of AAU was 46.3 years (IQR 26.4). The onset of AAU preceded CBP in 18 cases. HLA-B27 was checked in 11 patients (four positive), and 8 of 11 had been referred to secondary care. ESR or CRP was recorded in five patients, all of whom had also been referred.

These results suggest that the appropriate questions are not being asked nor are appropriate investigations being undertaken in patients with co-existing CBP and AAU. This implies that there is an educational need which will have to be addressed in order to shorten delay in diagnosis. We believe that there is likely to be a hidden burden of undiagnosed axSpA in primary care.