O10 UKELD score outperforms other risk scores in assessing post-surgical 90-day mortality in patients with cirrhosis. (20th September 2022)

• Record Type: Journal Article
• Authors: Abbas, Nadir; Faulkes, Rosemary; Hammond, Carl; Bordom, Daniel Clement-Osei; Roberts, Keith; Pinkney, Thomas; Evison, Miss Felicity; Clarke, George; Armstrong, Matthew J; Trivedi, Palak; Rajoriya, Neil
• Is Part Of: Gut. Volume 71(2022)Supplement 3
• Page Start: A7
• Page End: A7
• Publication Date: 2022-09-20
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gutjnl-2022-BASL.10
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Background: Patients with cirrhosis increasingly require major surgery carrying high morbidity and mortality. In Europe, surgical risk prediction remains a challenge, despite the availability of multiple prognostic systems (mainly US origin) in clinical practise to predict post-operative risk. In addition, there is lack of surgical outcome data in liver cirrhosis patients from the United Kingdom (UK). Aim: The aims of the study were (i) to investigate outcomes in patients with cirrhosis undergoing major surgery, (ii) to validate and compare current prognostic scoring models including the recent VOCAL-PENN score in a European centre, and (iii) to assess the role of United Kingdom model for End-stage liver disease (UKELD) as a novel scoring model. Methods: Retrospective cohort study was conducted of patients with cirrhosis undergoing major non-transplant surgery (1st January 2011 – 31st December 2016). We compared Child-Turcotte-Pugh (CTP), Model for End-stage liver disease (MELD), MELD-Na, UKELD, Mayo risk score (MRS) and VOCAL-PENN models in predicting mortality at 30-days, 90-days and 1-year post-surgery. Results: 277 patients (60% males, median age 61[IQR:14–47]) underwent major surgery (emergency n=94[34%], elective n=183[66%)) with the overall 30-, 90-day and 1-year mortality rates being 5.4%, 9.4% and 25.6%, respectively. Orthopaedic surgery (n=51, 18%) had the longest length of hospital stay (median:13.5 days) and highest 1-year mortality (40.1%). Emergency surgery and severity of liver disease (as defined by hypoalbuminemia, hyperbilirubinemia, presence of ascites; p<0.05 for all comparison) were significant predictors of mortality at all time points. Respiratory (OR: 2.58[95%CI:1.12–5.94]; p<0.05), cardiovascular comorbidities (OR: 3.64[95%CI:1.37–9.62]; p<0.01) and hyponatremia (OR: 0.91[95%CI: 0.84–0.99]; p<0.05) were significant factors in predicting mortality at 90-days. Older age and higher ASA (American society of Anesthesiologists) physical status class were significant predictors of mortality at 1-year (OR:1.03[95%CI:1.00–1.05); p<0.05 and 1.99[95%CI:1.11–3.56); p<0.05, respectively). The UKELD score was the best predictor of mortality at 30 and 90-days, and VOCAL-PENN score at 1-year in cirrhosis patients undergoing major surgery ( figure 1 ). Conclusion: We present the largest UK study assessing surgical outcomes in liver cirrhosis patients. We validate all existing prognostic scores for the first time in a UK cohort. The UKELD score outperforms established surgical risk scores in predicting 30- and 90-day postoperative mortality in patients with cirrhosis. Key factors including type of surgery (emergency, orthopaedic), age and ASA physical status should be considered when evaluating surgical risk in patients with cirrhosis.