Article Text
PDF
Abstract
Introduction Patients with suspected pancreatico-biliary (PB) cancers frequently undergo ERCPs to obtain brush cytology for confirmatory pathological diagnosis. The outcome of this often leads the management of the patient and avoids unnecessary invasive investigations.
Aims – To determine the diagnostic yield and accuracy of the brush cytology obtained at ERCP by performing a retrospective audit looking at all patients admitted to Guy’s and St. Thomas’ Hospital for ERCP during 2008–2013.
– To evaluate the impact of cytology results on patient management following ERCP, specifically the number of further investigations the patient requires.
Method Data was collected between January 2008 and December 2013. This involved analysing EndoSoft – the in-house software for endoscopic data entry, PathNet – the pathology database and Electronic Patient Records (EPR). The frequency of tests performed following cytology results (biopsies and imaging only) was retrospectively analysed to determine the impact of positive cytology on the diagnosis and the need for further investigations.
Results 162 patients in total had brush cytology performed at ERCP. 58 patients had positive cytology with a 72.6% diagnostic yield. With intention to treat analysis, sensitivity is 54.7%, specificity is 100.0% and negative predictive value is 53.9% with positive predictive value of 100%. In patients with a positive brush cytology result fewer investigations were required in comparison with patients with a negative cytology result (see chart)
Number of tests following positive or negative cytology to reach a diagnosis
Conclusion Our results compare favourably with previous studies in the field. Brush cytology has been ignored in recent times due to perceived poor results and efficacy. In fact it serves as a valuable tool in the diagnosis of PB malignancies and saves on further investigations suggesting that we should strive to improve sensitivity and specificity of cytology brushings through new brush designs or improvements in pathology processing.
Disclosure of interest None Declared.