OTU-03 Pancreaticobiliary endoscopic ultrasound in england 2007–2017: changing practice, benefits and harms. (June 2019)

• Record Type: Journal Article
• Authors: King, Dominic; Dosanjh, Amandeep; Patel, Prashant; Leeds, John; Nayar, Manu; Trudgill, Nigel; Oppong, Kofi
• Is Part Of: Gut. Volume 68(2019)Supplement 2
• Page Start: A1
• Page End: A2
• Publication Date: 2019-06
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gutjnl-2019-BSGAbstracts.3
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Introduction: Population level data on complications and outcomes of pancreaticobiliary (PB) endoscopic ultrasound (EUS) are limited. A perforation rate of 0.03% and attributable morbidity and mortality from pancreatic EUS fine needle aspiration (FNA) of 2.4% and 0.02% respectively are reported. We have examined PB EUS & FNA use in England, how it relates to pancreatic cancer (PC) therapy and associated mortality and adverse events. Methods: Adults undergoing PB EUS from 2007–17 were identified in Hospital Episode Statistics. A PC diagnosis within 6 months of EUS was required for PC cohort inclusion. EUS and FNA numbers per year, associated 7-day adverse events and 30-day mortality were examined. A logistic regression model examined the impact of variables on mortality and surgical resection. Results: 79, 490 PB EUS in 69, 120 subjects were identified. The number per year increased from 2, 915 (29% FNA) to 12, 764 (35% FNA) over the study period. 8, 859 subjects were diagnosed with PC. Bleeding was coded in 0.4% of PB EUS. Perforation was coded in 0.03% and in 0.05% of the PC cohort. 1.6% of PB EUS subjects and 2.8% in the PC cohort died within 30 days of their final EUS. The following factors were associated with increased mortality: increasing age (odds ratio 1.04(95%CI 1.03–1.04), p<0.001); males (1.38(1.22–1.56), p<0.001); increasing co-morbidity (1.49 (1.28–1.75), p<0.001); FNA (2.20 (1.93–2.51), p<0.001); pancreatic cancer (1.37 (1.17–1.60), p<0.001); and low provider total PB EUS volume [baseline >782 EUS] 8–111 (3.99 (2.95–5.38), p<0.001), 112–782 (1.33 (1.13–1.57), p=0.001). Less deprivation was associated with reduced mortality: quintile 5 (0.76 (0.62–0.92), p=0.006). 32.9% of PC subjects had a surgical resection, 43.1% received chemotherapy alone and 33.1% had no active therapy. Increasing age (0.97(95%CI 0.96–0.97), p<0.001), increasing co-morbidity score (0.77(0.60–0.98), p=0.034) and multiple EUS procedures prior to PC diagnosis (0.80(0.66–0.98), p=0.033) were all associated with a lower resection rate. The two least deprived quintiles were associated with an increased resection rate (quintile 4 1.33(1.11–1.59), p=0.002), quintile 5 1.33 (1.11–1.59), p=0.002). Conclusions: The number of PB-EUS undertaken in England has increased six-fold over the last decade, with an increased proportion of FNA. Deprivation and low provider volume were associated with 30 day mortality. A third of subjects undergoing PB EUS for PC did not receive surgery or chemotherapy. The results of this study support a reappraisal of the provision of and indications for PB EUS.