Background The diagnostic performance of endoscopic retrograde cholangiopancreatography brush cytology for malignant strictures is modest. A novel larger more abrasive brush may have improved diagnostic performance. We compared the utility of the new biliary brush with a conventional brush.

Methods The new brush was used in 51 consecutive patients (group 1) referred with a biliary stricture and matched to 102 patients who underwent sampling with a conventional brush (group 2). Demographic data, stricture characteristics, sensitivity, specificity, negative predictive values and positive predictive values were analysed and compared with final diagnosis. Analysis was performed using strict criteria (definite for cancer) and relaxed criteria (suspicious for cancer). All patients had a minimum follow-up of 12 months.

Results There was no statistically significant difference in the age and sex distribution between the two groups. Malignancy was diagnosed in 74% in both groups. There was no significant difference in sensitivity between the groups using either strict criteria (sensitivity group 1 47.4% vs group 2 52%, p=0.69) or relaxed criteria (sensitivity group 1 71.1% vs group 2 71.2%, p=1.0).

Conclusions Our data suggest that the novel brush design does not confer improved diagnostic performance in malignant biliary strictures when compared in a robust manner. This highlights the difficulties of intraductal brush sampling and the need to develop newer diagnostic techniques.