P‐12 A RETROSPECTIVE AUDIT OF FINE NEEDLE ASPIRATION CYTOLOGY OF PANCREAS

Abstract

Background: Fine needle aspiration cytology (FNAC) of pancreas is a widely accepted method of diagnosis of pancreatic mass lesions. We have performed a retrospective analysis of all radiological (CT/ultrasound) and endoscopic ultrasound guided procedures at our institution.Aim: (1) To review the results of FNAC of pancreas from January 2000 to April 2006. (2) To calculate the inadequate rate. (3) To account for discrepancies between the cytological and histological diagnoses. (4) To identify any false positive cases if present.Method: The results of all pancreatic FNAC reported at our institute from January 2000 to April 2006 were identified from the laboratory system. All results were classified as follows: Inadequate, inconclusive, benign, suspicious and malignant. The results were further categorised depending on whether they were CT/ultrasound guided or EUS guided. The histological diagnosis where available was used as the gold standard and where discrepancies were present the cytological preparations were reviewed.Results: Seventy‐three patients underwent pancreatic FNAC during the study period. Table 1 illustrates our results. Number Inadequate Inconclusive Benign Suspicious Malignant CT/US 53 8 (11%) 1(1%) 17(23%) 6(8%) 21(29%) Endoscopic ultrasound 20 1 (1%) 2(3%) 14(19%) 1(1%) 2(3%) Total 73 9 (12%) 3(4%) 31(42%) 7(10%) 23(32%) Three false negative cases where a later histological diagnosis of malignancy was present contained no diagnostic cellular material. To date no false positive diagnosis has occurred and all patients with malignant diagnosis have died except those diagnosed in the last 6 months.Conclusion: Pancreatic FNAC is a useful procedure in the majority of cases at our institute. Our inadequate/inconclusive rate is low. In fact our analysis shows that the inadequate rate for EUS guided FNAC pancreas is only 5% despite the fact that on site assessment of adequacy is not yet carried out. This is in a small group of patients however. No false positive diagnosis was identified. Discrepancies between cytological and histological diagnosis have been due to unrepresentative samples. We intend to proceed with on site assessment of adequacy in future cases. We will continue to review all cases where pancreatic FNA is carried out at our weekly pancreatic/hepatobiliary clinicopathological multidisciplinary team meeting and within our department at peer review meetings.