Helpfulness of the liver disease scores in cardiac surgery for cirrhotic patients

Liver cirrhosis is a well-known risk factor of mortality after cardiac surgery, but not considered in the widely used EuroSCOREII (ESII). The objective was to analyse the performance of the ESII, the Child-Pugh-Turcotte (CPT) and the Model of End-stage Liver Disease (MELD) scores to predict hospital mortality in cardiac surgery for cirrhotic patients and to analyse the survival according to the preoperative cirrhosis status.

Preoperative and cirrhosis characteristics and postoperative outcomes were compared according to hospital mortality. The performance of the 3 scores was analysed by the area under the receiver-operating characteristics (AUC-ROC) by DeLong's method. The survival of the patients who were discharged was analysed by Kaplan-Meier curves according to the preoperative cirrhosis status.

Seventy-four patients were included. Observed hospital mortality was 12%, the predictive mortality by ESII was 3.9% ± 5.2%, and AUC-ROC was 0.67 [0.44-0.90]. Only the MELD score was discriminant (AUC-ROC 0.75 [0.57-0.93]). The observed hospital mortality increased by threefold over the ESII (12% versus 3.9%, p < 0.001), except the patients with MELD < 10 for whom hospital mortality was similar as ESII (3% versus 2.6%, p = 0.89). Long-term survival was higher for the MELD < 10 patients.

The ESII did not predict hospital mortality after a cardiac surgery in cirrhotic patients and the MELD score should be considered for decision of cardiac intervention in cirrhotic patients.