Background: Several studies have found that standard risk scores inaccurately reflect risk in TAVI cohorts. The assessment of mortality risk upon post-interventional ICU admission is important to optimizing clinical management. This study sought to determine outcomes and factors affecting mortality in patients admitted to the intensive care unit (ICU) after transcatheter aortic valve implantation (TAVI), and to analyze and compare the predictive values of SAPS II and EuroSCORE.
Methods and findings: 214 consecutive patients treated with transfemoral TAVI (2006-2012) admitted to the ICU in an academic tertiary-care university hospital, were included in this retrospective data analysis. The overall 30-day mortality rate was 7%. Non-survivors at 30-days and survivors showed differences in the rates of catecholamine therapy upon ICU admission (93 vs. 29%; p<0.001), stroke (20 vs. 1%;p<0.001), sepsis (27 vs. 2%;p<0.001), kidney injury (83 vs. 56%; log-rank p<0.001), catecholamine therapy (88 vs. 61%;log-rank p<0.001) and vascular complications (60 vs. 17%; p<0.001). Mean SAPS II score and predicted mortality were higher in non-survivors (38.1±7.0 vs. 29.9±6.2;p<0.001 and 23.1±11.7 vs. 10.5±8.2;p<0.001, retrospectively), whereas the logistic EuroSCORE could not discriminate between the groups (p = 0.555). Among the biochemical parameters, the maximum values of creatinine, procalcitonin, and troponin I during the first 48 h after ICU admission were significantly higher in non-survivors. Multivariate analysis of baseline characteristics and complications associated with two-year mortality showed no significant results.
Conclusions: The SAPS II is a good tool for estimating ICU mortality immediately after performing the TAVI procedure and provides valuable information for other known predictors of mortality.