EP35 SHOULD MODEL FOR END STAGE LIVER DISEASES (MELD) AND MELD-BASED SCORES BE ROUTINELY APPLIED TO PREOPERATIVE ASSESMENT OF PATIENT UNDERGOING VALVULAR CARDIAC SURGERY?

Abstract

Abstract Background: MELD and MELD-based score have been introduced to assess the risk of complications in patients with chronic liver disease. We designed this study in order to evaluate the potential usefulness of routine application of MELD-based scores in patients undergoing cardiac surgery valve procedures in predicting postoperative major and minor liver-related complications. Materials: 92 patients, without history of liver diseases and with normal liver function tests, undergoing valvular heart surgery were enrolled. Patients were subdivided in four groups: combined mitro-aortic surgery (Group A); Isolated AVR (Group B); Isolated MVR (Group C) and isolated MVP (Group D). Preoperatively, 4 MELB-based scores were calculated (MELD XI, modified IM, modMESO and modMELD-Na). Intra and inter-groups distributions of MELD-based scores, and the accuracy of 4 scores in predicting postoperative liver-related complications were analysed. Results: Preoperative MELD-based scores were slightly increased in Group C. Postoperative raise (>50%) of AST,ALT,GGT and LDH was showed in 71%,44%, 28% and 54 respectively with significant differences in group A and C. GGT raise did influenced the total hospital stay and all 4 scores showed a direct correlation between preoperative value and % of postoperative raise, especially MELD XI and modified IM (Fig 1a/b). Furthermore a preoperative high MELD-based score (especially MELD XI and modified IM) accurately predicted higher postoperative GGT increase (Fig 2a/b). Conclusions: Our study seems to confirm the usefulness of application of MELD-based scores in preoperative assessment of patients undergoing cardiac surgery. The clinical impact of postoperative transient liver function impairment needs to be further evaluate.