A retrospective analysis comparing the use of ProCore® with standard fine needle aspiration in endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA).

Endobronchial ultrasound has become first line in the investigation of mediastinal lesions suspicious for malignancy in keeping with National Institute for Health and Care Excellence (NICE) guidelines; however, needle size and type required to maximise diagnostic sensitivity remains unclear. Previous meta-analyses have compared the use of ProCore with standard fine needle aspiration in the assessment of pancreatic masses with differences noted only in the number of passes required. We aim to assess whether a ProCore needle improves diagnostic sensitivity in EBUS-TBNA. Complete follow-up data regarding all 235 patients undergoing EBUS-TBNA in a district general hospital has been collected since the service's inception in 2012. Results were collated and retrospectively analysed allowing for calculation of test sensitivity and specificity. Comparison was then made between procedures where standard fine needle aspiration was performed and those using a ProCore needle. Overall sensitivity of EBUS-TBNA was shown to be 85% with a specificity of 100% in keeping with quoted figures from other centres. Standard fine needle aspiration produced a sensitivity of 77% (85/110) versus ProCore sensitivity of 92% (115/125) with a p value of 0.0016. Thirty percent (33/110) of patients undergoing standard fine needle aspiration required an appropriate crossover technique such as mediastinoscopy or CT-guided FNA in order to either obtain or confirm the diagnosis compared with 15% (19/125) of the ProCore group with a p value of 0.0064. Our retrospective analysis shows a statistically significant difference in the diagnostic sensitivity of sampling mediastinal lymphadenopathy using a ProCore needle compared with standard fine needle aspiration. It also shows that a significantly fewer number of patients required further procedures in order to obtain or confirm the diagnosis. This could potentially be confounded by the retrospective nature of the study design; however, due to the statistical significance demonstrated, further study is required.