Aim: To establish the diagnostic accuracy and adequacy of fine needle aspiration cytology (FNAC) within a regional cancer network, and to determine what service improvements may be required to allow successful implementation of an FNAC-based, ‘one-stop’ head and neck clinic, as proposed by the current National Institute for Clinical Excellence guidelines.

Materials and methods: The Sussex cancer network serves a population of 1 200 000 and contains five hospitals within three acute trusts. In 2004, an audit was undertaken retrospectively to examine the diagnostic adequacy and accuracy of head and neck FNAC across the network. Comparisons were then made with the results of subsequent relevant surgery. For the purposes of the audit, FNAC was subdivided into three main groups: salivary gland, thyroid gland and neck node. As part of the data analysis, we also noted the clinical source of the FNAC and whether it was performed blind or under image guidance.

Results: In 2004, 712 FNAC procedures were undertaken in 647 patients, 276 of whom underwent subsequent surgery. Fine needle aspiration cytology was non-diagnostic in 52 per cent of patients in the neck node group, in 50 per cent in the salivary gland group and in 30 per cent in the thyroid group. With these non-diagnostic results removed, statistical analysis was performed on data from those patients who had undergone both FNAC and subsequent surgery. This gave a sensitivity of 89 per cent and a specificity of 57 per cent in the neck node group, a sensitivity of 64 per cent and specificity of 100 per cent in the salivary gland group, and a sensitivity of 62 per cent and specificity of 86 per cent in the thyroid group. Diagnostic problems with FNAC were noted, particularly in the differentiation of reactive nodal hyperplasia from lymphoma and in diagnosing follicular thyroid lesions. Ultrasound guidance was used in 50 per cent of the thyroid FNAC procedures but in only a minority of patients in the neck node and salivary gland groups.

Conclusion: This audit demonstrated widespread diagnostic difficulties associated with head and neck FNAC in a large patient sample. It is likely that these problems will be mirrored in other cancer networks. In order for one-stop head and neck clinics to succeed, the non-diagnostic rate of FNAC in particular must be minimised. There are strategies to enable this, depending on local resources, including increased access to cytologists or cytology technicians, diagnostic ultrasound, image guidance for FNAC and the use of ultrasound-guided core biopsy.