LAUSR

an abnormal result lead to a change in treatment. In no patient was TNFi discontinued, changed, or the dosage postponed. No patients developed infections that could be linked to abnormal blood tests. The risk of cytopaenias due to TNFi in AS is unknown, but it is rare and the causal relationship to the drug unclear. A 2-year study of 402 Chinese AS patients reported no cases of cytopaenia (1). Reports in RA more clearly link rare cases of cytopaenias to TNFi (2, 3), but most patients had pre-existing haematological abnormalities and patients with RA may be at higher risk of immune-mediated cytopaenias than patients with AS. In this audit, significant neutropaenia (< 1.4 × 10/L) was not seen in any of 425 tests. Levels outside the local normal range were seen in 8%, but no patient developed an infection and no treatment was stopped. Only 2% of tests indicated a platelet count below the normal range, consistent with the background rate of such abnormalities, as is the 5% rate of elevated ALT. Other studies have failed to find a clear link between TNFi in AS and hepatotoxicity (4). Most centres recommend monitoring AS patients on TNFi at 3-monthly intervals but this seems to be an arbitrary decision invoking guidelines for traditional DMARDs. Most patients in our audit had ALT and renal function checked routinely, although there is no suggestion that TNFis adversely affect these parameters. Cytopaenias may occur as a result of TNFi therapy but such idiosyncratic effects seem exceedingly rare and probably no more common than may occur with other drugs for which routine testing is not recommended. The cost of blood tests in UK National Health Service (NHS) hospitals has recently (5) been reported as: FBC GBP 2.65, renal function GBP 2.12; liver function GBP 2.78. These values are significant when large numbers of patients are considered. Additional costs include the time taken to chase up results and follow up any abnormalities discovered (which may be unrelated to the drug). Rather than routine testing of all patients at 3-monthly intervals, we believe this audit supports a policy of testing when clinically indicated, perhaps when the patient is seen in the rheumatology department (rather than requiring GP involvement) or prompted by a change in the patient’s status. The British Society for Rheumatology Biologics Register for AS (BSRBR-AS) may offer further information on the incidence of TNFi-related morbidity in AS.