SPC365

Precision in Prognostication: A Comprehensive Comparison of Risk Scores in the Setting of Cardiogenic Shock

Session: Best Posters
Speaker: Samuel Azevedo

Congress: CPC 2024
Topic: E. Coronary Artery Disease, Acute Coronary Syndromes, Acute Cardiac Care
Theme: 14. Acute Cardiac Care
Subtheme: 14.4 Acute Cardiac Care – Cardiogenic Shock
Session Type: Cartazes
FP Number: ---
Authors: Samuel Azevedo; Rita Barbosa Sousa; João Presume; Débora da Silva Correia; Rita Almeida Carvalho; Joana Certo Pereira; Miguel Domingues; Ana Rita Bello; Catarina Brízido; Christopher Strong; Jorge Ferreira; António Tralhão

<p style="text-align:justify"><span style="font-size:medium"><span style="font-family:Calibri,sans-serif"><span style="color:#000000"><strong><span style="color:black">Introduction:</span></strong><span style="color:black"> Early and precise risk stratification of cardiogenic shock (CS) patients is essential for tailoring appropriate treatment, guiding decisions on the initiation of more aggressive therapies like mechanical circulatory support, or on discontinuing futile interventions. Additionally, current CS risk scores are often developed in specific cohorts, neglecting the diversity within CS cases.</span></span></span></span></p>

<p style="text-align:justify"><span style="font-size:medium"><span style="font-family:Calibri,sans-serif"><span style="color:#000000"><strong><span style="color:black">Objectives</span></strong><span style="color:black">: This study aimed to assess the applicability of existing CS-specific and general intensive care unit (ICU) risk scores, evaluating their performance in the overall CS population and across subgroups based on CS etiology.</span></span></span></span></p>

<p style="text-align:justify"><span style="font-size:medium"><span style="font-family:Calibri,sans-serif"><span style="color:#000000"><strong><span style="color:black">Methods</span></strong><span style="color:black">: We conducted a single-center retrospective study, enrolling consecutive CS patients admitted to the cardiac ICU (CICU) between January 2017 and October 2023. Three CS-specific scores and two general ICU scores were calculated based on admission data. ROC curve analysis was used to compare these scores to the SCAI stage at admission and against each other, regarding their discriminative power to predict 30-day all-cause mortality.</span></span></span></span></p>

<p style="text-align:justify"><span style="font-size:medium"><span style="font-family:Calibri,sans-serif"><span style="color:#000000"><strong><span style="color:black">Results</span></strong><span style="color:black">: The analysis included 281 patients (mean age 67&plusmn;16 years, 65% male, 30% with SCAI &ge;D at admission, median hospitalization duration of 11 [3; 23] days). Half of the cohort was admitted for acute myocardial infarction CS (AMI-CS). The 30-day mortality rate was 44%. The calculated scores at admission were as follows: CardShock 4.4&plusmn;1.8; IABP-SHOCK II 3.3&plusmn;1.9; Cardiogenic Shock Score (CSS) 5.9&plusmn;2.5; SOFA 7.6&plusmn;2.8; SAPS-II 50&plusmn;17.</span></span></span></span></p>

<p style="text-align:justify"><span style="font-size:medium"><span style="font-family:Calibri,sans-serif"><span style="color:#000000"><span style="color:black">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; In the overall cohort, ROC curve analysis revealed that SCAI staging at admission exhibited poor performance in predicting 30-day mortality compared to the scores (Figure 1A). Although CardShock exhibited the highest predictive value for 30-day mortality, all scores showed only moderate predictive ability (C-statistic between 0.550-0.731), and no significant differences were observed between them.</span></span></span></span></p>

<p style="text-align:justify"><span style="font-size:medium"><span style="font-family:Calibri,sans-serif"><span style="color:#000000"><span style="color:black">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; After stratifying the cohort into two subgroups based on etiology (AMI-CS or non-AMI-CS), SCAI at admission consistently showed poor predictive performance in both cohorts. Furthermore, within the AMI-CS subgroup, no notable differences were identified among the remaining scores. However, in the non-AMI-CS subgroup, ICU risk scores displayed a superior predictive value for 30-day mortality compared to CS-specific scores. Specifically, the SOFA score significantly outperformed CardShock in discriminative ability (z= 2.070; p=0.0384).</span></span></span></span></p>

<p style="text-align:justify"><span style="font-size:medium"><span style="font-family:Calibri,sans-serif"><span style="color:#000000"><strong><span style="color:black">Conclusion</span></strong><span style="color:black">: ICU and CS risk scores appear to provide additional value over SCAI staging at admission for predicting 30-day mortality. Overall, risk scores demonstrated moderate prognostic prediction, with no clearly superior score. In the non-AMI-CS population, CS-specific scores seem to underperform compared to general ICU scores. Further research is needed to enhance the discriminative abilities of existing models or develop new ones. </span></span></span></span></p>