Article Text

Download PDFPDF

PWE-090 West midlands multi-centre trainee-LED audit in the assessment, management and prophylaxis of spontaneous bacterial peritonitis
  1. Keith Siau,
  2. B Hicken,
  3. A McCulloch,
  4. M Harborne,
  5. T Gupta,
  6. B Polewiczowska,
  7. T Troth,
  8. M McFarlane,
  9. E Mozdiak,
  10. A Thakor,
  11. Z Rehman,
  12. L O’Flynn,
  13. Monika Widlak
  1. West Midlands Research in Gastroenterology), UK


Introduction Spontaneous bacterial peritonitis (SBP) is a common but potentially fatal complication in patients with cirrhosis and ascites. In the first audit performed by West Midlands Research in Gastroenterology (WMRIG) trainees, we aimed to assess practice of assessment, management and primary and secondary prophylaxis of SBP according to national standards, in addition to the feasibility of regional project delivery.

Methods This trainee-led, retrospective, multi-centre study identified patients admitted with cirrhosis and ascites between Sep-Dec 2016. Outcomes of SBP and mortality were retrospectively followed-up for up to 1 year (median 8 months). Practice was audited against EASL, BSG and NICE standards. Heterogeneity between sites was assessed with chi2 and time-to-event analyses undertaken using Kaplan-Meier plots.

Results Trainees across 8 West Midlands hospitals identified 227 patients (mean age 58, SD 13; 65% male) with 282 admissions. Cirrhosis was attributed to alcohol (79%), NAFLD (10%), autoimmune (4%) and viral (3%), and was graded Child-Pugh B in 48% and C in 49%. 18% were elective admissions and 7% had a previous history of SBP. Ascitic aspirates were performed in 83% (range: 60%–92%, p=0.019), in <24 hours in 64% (range: 49%–85%, p=NS), and cultures sent in 55% (range: 11%–86%, p<0.001). 16.8% of aspirates met criteria for SBP: antibiotics were commenced in 92% (p<0.001), Day 1 albumin in 64% (p=NS), Day 3 albumin in 40% (p=NS), and secondary prophylaxis in 44% (p=NS). Repeat aspirate to ensure SBP resolution was performed in 33% (range: 0%–52%, p=NS). In patients without SBP, ascitic protein was measured in 46% (not available in 3 Trusts). 32 (67%) met criteria for primary prophylaxis (protein≤15 g/L); which was commenced in 4 (13%, range 0%–100%, p=NS). Mortality occurred in 51%. SBP was associated with lower median survival (79 days vs. non-SBP: 190 days, p=0.045) [figure 1]. Emergency admission (HR 8.4, p=0.039), older age (HR 1.03 per increase, p=0.017) and ascitic protein ≤15 g/L (HR 2.27, p=0.042) were multivariate predictors of reduced survival. SBP occurred after discharge in 8 patients (4%) after a median interval of 32 days.

Conclusions Our pilot study has been successful in highlighting deficiencies and variations in the assessment, management and prophylaxis of SBP. These Results will inform and prioritise future regional quality improvement strategies to improve outcomes in patients with advanced chronic liver disease.

Abstract PWE-090 Figure 1

Effect of SBP on survival

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.