Article Text
Abstract
Introduction We have previously shown that non-targeted gastric biopsies seldom contributed to patient management. We developed a local protocol for indications for gastric biopsy, which was distributed to endoscopy theatres trust-wide, and provided education via presentation of results, followed by re-audit.
Methods We retrospectively analysed all gastric biopsies taken within a 3 month period in 2015, across 2 sites within Newcastle, using our histopathology and endoscopy databases. This included patient demographics, endoscopy findings, grade of endoscopist, biopsy result, and whether urease-based Helicobacter test (UBHT) was performed and the outcome. We accessed patients electronic medical records to determine whether the result of the biopsy altered diagnosis or management. We compared this to the results from our identical audit undertaken in 2013. A targeted biopsy was defined as the presence at OGD of a polyp, ulcer or any other lesion. Non-targeted biopsy was any other appearance, including gastritis. The cost of a biopsy included manpower and histopathology processing costs. We looked separately at the cost of UBHT testing plus forceps use.
Results The table below compares outcomes for the 2 cycles of audit. χ2testing showed a significant reduction in the proportion of non targeted biopsies from 2013 to 2015 (p = 0.001).
Of non-targeted biopsies in 2015, 0.07% (n = 1) showed lymphoma in a patient under surveillance. In 2013, 0.8% (n = 2) revealed adenocarcinoma; both biopsies from the same patient under surveillance for gastric carcinoma. No other serious diagnosis was made. Of non-targeted biopsies in 2013, 94% (n = 223) had no management alteration based on histology compared with 90% (n = 135) in 2015. Aside from patients under cancer surveillance, histology results leading to management alteration were based on presence of Helicobacter. The proportion of non targeted biopsies taken by nurse endoscopists reduced from 55% to 43%, and by SpRs, 44% to 28%. The development of a protocol appears to have led to a 36% reduction in non-targeted biopsies. This results in an annual saving of £36,432 (assuming single biopsy set cost of £103.51).
Conclusion The majority of non-targeted gastric biopsies taken for histology do not contribute to the management of patients who are not under cancer surveillance. Limiting these biopsies can save significant resources. Education techniques with use of protocol can safely reduce the numbers of non-targeted biopsies. Reinforcement of this message would be expected to further reduce the rate of non-targeted biopsies.
Disclosure of Interest None Declared