Past, present, and future of mortality risk scores in the contemporary cardiac intensive care unit

Clinical complexity of an Italian cardiovascular intensive care unit: the role of mortality and severity risk scores

AIMS

The identification of patients at greater mortality risk of death at admission into an intensive cardiovascular care unit (ICCU) has relevant consequences for clinical decision-making. We described patient characteristics at admission into an ICCU by predicted mortality risk assessed with noncardiac intensive care unit (ICU) and evaluated their performance in predicting patient outcomes.

METHODS

A total of 202 consecutive patients (130 men, 75 ± 12 years) were admitted into our tertiary-care ICCU in a 20-week period. We evaluated, on the first 24 h data, in-hospital mortality risk according to Acute Physiology and Chronic Health Evaluation II (APACHE II) and Simplified Acute Physiology Score 3 (SAPS 3); Sepsis related Organ Failure Assessment (SOFA) Score and the Mayo Cardiac intensive care unit Admission Risk Score (M-CARS) were also calculated.

RESULTS

Predicted mortality was significantly lower than observed (5% during ICCU and 7% at discharge) for APACHE II and SAPS 3 (17% for both scores). Mortality risk was associated with older age, more frequent comorbidities, severe clinical presentation and complications. The APACHE II, SAPS 3, SOFA and M-CARS had good discriminative ability in distinguishing deaths and survivors with poor calibration of risk scores predicting mortality.

CONCLUSION

In a recent contemporary cohort of patients admitted into the ICCU for a variety of acute and critical cardiovascular conditions, scoring systems used in general ICU had good discrimination for patients' clinical severity and mortality. Available scores preserve powerful discrimination but the overestimation of mortality suggests the importance of specific tailored scores to improve risk assessment of patients admitted into ICCUs.

Applicability of the Zwolle score for selection of very high-risk ST-elevation myocardial infarction patients treated with primary angioplasty

The Zwolle risk score was designed to stratify in-hospital mortality risk of ST-elevation myocardial infarction (STEMI) patients treated with primary percutaneous coronary intervention (pPCI) and for decision-making in the unit where patients are admitted. We assessed the accuracy of Zwolle risk score for in-hospital mortality estimation compared with the GRACE score in all patients (n = 4446) admitted for STEMI in 3 university hospitals. Only one fourth of the patients were classified as high-risk by the Zwolle risk score vs 60% by the GRACE score. In-hospital mortality was 10.6%. A statistically significant increase in in-hospital mortality, adjusted by age, gender, and revascularization, was observed with both scores. The assessment of the optimal cut-off points verified the accuracy of Zwolle score ≥4 as optimal threshold for high-risk categorization. In contrast, GRACE score ≥140 had very low specificity as well as percentage of patients correctly classified; GRACE score ≥175 was fairly better. The reclassification index of the Zwolle score after applying the GRACE score was 35.5%. Selection of high-risk STEMI patients treated with pPCI based on the Zwolle risk score has higher specificity than the GRACE score and might be useful in clinical practice.

Heart rate complexity helps mortality prediction in the intensive care unit: A pilot study using artificial intelligence

BACKGROUND

In intensive care units (ICUs), accurate mortality prediction is crucial for effective patient management and resource allocation. The Simplified Acute Physiology Score II (SAPS-2), though commonly used, relies heavily on comprehensive clinical data and blood samples. This study sought to develop an artificial intelligence (AI) model utilizing key hemodynamic parameters to predict ICU mortality within the first 24 h and assess its performance relative to SAPS-2.

METHODS

We conducted an analysis of select hemodynamic parameters and the structure of heart rate curves to identify potential predictors of ICU mortality. A machine-learning model was subsequently trained and validated on distinct patient cohorts. The AI algorithm's performance was then compared to the SAPS-2, focusing on classification accuracy, calibration, and generalizability.

MEASUREMENTS AND MAIN RESULTS

The study included 1298 ICU admissions from March 27th, 2015, to March 27th, 2017. An additional cohort from 2022 to 2023 comprised 590 patients, resulting in a total dataset of 1888 patients. The observed mortality rate stood at 24.0%. Key determinants of mortality were the Glasgow Coma Scale score, heart rate complexity, patient age, duration of diastolic blood pressure below 50 mmHg, heart rate variability, and specific mean and systolic blood pressure thresholds. The AI model, informed by these determinants, exhibited a performance profile in predicting mortality that was comparable, if not superior, to the SAPS-2.

CONCLUSIONS

The AI model, which integrates heart rate and blood pressure curve analyses with basic clinical parameters, provides a methodological approach to predict in-hospital mortality in ICU patients. This model offers an alternative to existing tools that depend on extensive clinical data and laboratory inputs. Its potential integration into ICU monitoring systems may facilitate more streamlined mortality prediction processes.

Outcomes of Octogenarians and Nonagenarians in a Contemporary Cardiac Care Unit - Insights From 2,242 Patients Admitted Between 2019 and 2021

Background: The number of octo- and nonagenarians admitted to cardiac care units (CCUs) has been increasing in the context of an aging society; however, clinical details and outcomes for these patients are scarce. Methods and Results: Data from 2,242 consecutive patients admitted to the CCU between 2019 and 2021 (age <80 years, 1,390 [62%]; octogenarians, 655 [29%]; nonagenarians, 197 [8.7%]) were reviewed using the in-hospital database for the Tokyo CCU Network. The primary cause of admission was acute coronary syndrome in younger patients and octogenarians (58% and 49%, respectively) and acute heart failure (AHF) in nonagenarians (42%). The proportions of females, underweight, hypertension, atrial fibrillation, myocardial infarction, stroke, previous heart failure, anemia, and malnutrition were higher among octo- and nonagenarians than among younger patients. In-hospital and 1-year mortality rates were greater in octo- and nonagenarians (younger vs. octogenarian vs. nonagenarian, 2.0% vs. 3.8% vs. 5.6% and 4.1% vs. 11.9% vs. 19.0%, respectively). Multivariate analysis revealed that 1-year mortality was associated with octo-/nonagenarian status (odds ratio [OR] 2.24 and 2.64), AHF (OR 2.88), body mass index (OR per 1-kg/m2 0.91), and albumin concentration (OR per 1-g/dL 0.27). Conclusions: Approximately 40% of patients admitted to the CCU were octo- or nonagenarians, and being an octo- or nonagenarian, having AHF, a lower body mass index, and lower albumin concentrations were associated with 1-year mortality after CCU admission.

Spontaneous coronary artery dissection in Spain: a study using the minimum data set of the Spanish National Health System

INTRODUCTION AND OBJECTIVES

Spontaneous coronary artery dissection (SCAD) is a rare cause of acute myocardial infarction (AMI). We sought to compare the results on in-hospital mortality and 30-day readmission rates among patients with AMI-SCAD vs AMI due to other causes (AMI-non-SCAD).

METHODS

Risk-standardized in-hospital mortality (rIMR) and risk-standardized 30-day readmission ratios (rRAR) were calculated using the minimum dataset of the Spanish National Health System (2016-2019).

RESULTS

A total of 806 episodes of AMI-SCAD were compared with 119 425 episodes of AMI-non-SCAD. Patients with AMI-SCAD were younger and more frequently female than those with AMI-non-SCAD. Crude in-hospital mortality was lower (3% vs 7.6%; P<.001) and rIMR higher (7.6±1.7% vs 7.4±1.7%; P=.019) in AMI-SCAD. However, after propensity score adjustment (806 pairs), the mortality rate was similar in the 2 groups (AdjOR, 1.15; 95%CI, 0.61-2,2; P=.653). Crude 30-day readmission rates were also similar in the 2 groups (4.6% vs 5%, P=.67) whereas rRAR were lower (4.7±1% vs 4.8%±1%; P=.015) in patients with AMI-SCAD. Again, after propensity score adjustment (715 pairs) readmission rates were similar in the 2 groups (AdjOR, 1.14; 95%CI, 0.67-1.98; P=.603).

CONCLUSIONS

In-hospital mortality and readmission rates are similar in patients with AMI-SCAD and AMI-non-SCAD when adjusted for the differences in baseline characteristics. These findings underscore the need to optimize the management, treatment, and clinical follow-up of patients with SCAD.

Machine Learning Approaches for Phenotyping in Cardiogenic Shock and Critical Illness: Part 2 of 2

Progress in improving cardiogenic shock (CS) outcomes may have been limited by failure to embrace the heterogeneity of pathophysiologic processes driving the underlying syndrome. To better understand the variability inherent to CS populations, recent algorithms for describing underlying CS disease subphenotypes have been described and validated. These strategies hope to identify specific patient subgroups with more favorable responses to standard therapies, as well as those who require novel treatment approaches. This paper is part 2 of a 2-part state-of-the-art review. In this second article, we present machine learning-based statistical approaches to identifying subphenotypes and discuss their strengths and limitations, as well as evidence from other critical illness syndromes and emerging applications in CS. We then discuss how staging and stratification may be considered in CS clinical trials and finally consider future directions for this emerging area of research.

Refining the stratification and prognosis of cardiogenic shock patients to improve their outcome

Risk stratification in patients with cardiogenic shock must incorporate the etiology, clinical phenotype, severity, complications, response to therapy, and non-modifiable risk factors for mortality. Tailoring the degree of hemodynamic support to the shock severity is a logical approach, but this must be guided by an in-depth understanding of the patient's underlying hemodynamics, physiology, and candidacy for advanced therapies.

Predictive ability of the sequential organ failure assessment score for in-hospital mortality in patients with cardiac critical illnesses: a nationwide observational study

AIMS

Several studies have reported a high predictive ability of the Sequential Organ Failure Assessment (SOFA) score for in-hospital mortality specifically for patients with cardiac critical illnesses, however, differences according to the admission classification (surgical or non-surgical) are unknown. The present study aimed to evaluate the predictive ability of the SOFA score in surgical and non-surgical patients with cardiac critical illnesses.

METHODS AND RESULTS

Using the Japanese nationwide Diagnosis Procedure Combination database, we identified patients with cardiac critical illnesses, defined as patients admitted to the intensive care unit (ICU) and treated by cardiologists or cardiovascular surgeons as their physicians in charge from April 2018 to March 2020. The discriminatory ability of the SOFA score for in-hospital mortality was assessed by calculating the area under the receiver operating characteristic curve (AUROC). Among 52 819 eligible patients with available data on their SOFA scores, 33 526 (64%) were postoperative cardiac surgeries. The median SOFA score on ICU admission was 5.0 (interquartile range, 2.0-8.0) and overall in-hospital mortality 6.8%. The AUROC of the SOFA score was 0.75 [95% confidence interval (CI), 0.75-0.76]. In the subgroup analyses, the AUROCs were 0.76 (95% CI, 0.74-0.77) in the surgical patients, 0.83 (95% CI, 0.83-0.84) in the non-surgical patients, and 0.88 (95% CI, 0.87-0.89) in the non-surgical acute coronary syndrome patients.

CONCLUSIONS

The predictive ability of the SOFA score on the day of ICU admission for in-hospital mortality was confirmed to be acceptable in the patients with cardiac critical illnesses and varied according to the admission classification and primary diagnoses.

Braden Skin Score Subdomains Predict Mortality Among Cardiac Intensive Care Patients

BACKGROUND

The Braden Skin Score (BSS) is a bedside nursing assessment that may be a measure of frailty and predicts mortality among patients in the cardiac intensive care unit (CICU). We examined the association between each of the 6 individual BSS subscores with hospital mortality in patients in the CICU. We hypothesized that BSS subscores reflecting patient frailty would have a stronger association with outcomes.

METHODS

Retrospective cohort study of unique adult patients admitted to the Mayo Clinic CICU from 2007 to 2018 with BSS documented on admission. Primary outcome was all-cause hospital mortality. Odds ratios (ORs) were determined using multivariable logistic regression.

RESULTS

The 11,954 included patients had a mean age of 67.4 ± 15.2 years (37.8% women). Each individual BSS subscore was lower among patients who died in the hospital (all P < .001). The total BSS was inversely associated with in-hospital mortality across admission diagnoses and among patients with coma or mechanical ventilation; each individual subscore was inversely associated with in-hospital mortality. On multivariable regression, all subscores were inversely associated with hospital mortality after full adjustment. Shear had the strongest association (adjusted OR 0.59), followed by nutrition (adjusted OR 0.67), skin moisture (adjusted OR 0.76), mobility (adjusted OR 0.76), sensory perception (adjusted OR 0.82), and activity level (adjusted OR 0.85).

CONCLUSION

BSS can serve as a rapid noninvasive screening tool for identifying poor outcomes in patients in the CICU. BSS subdomains that are more strongly associated with mortality appear to reflect physical frailty. Insofar as the BSS and its subscores measure frailty, a low BSS may identify frail patients.

Mortality Prediction in Cardiac Intensive Care Unit Patients: A Systematic Review of Existing and Artificial Intelligence Augmented Approaches

The medical complexity and high acuity of patients in the cardiac intensive care unit make for a unique patient population with high morbidity and mortality. While there are many tools for predictions of mortality in other settings, there is a lack of robust mortality prediction tools for cardiac intensive care unit patients. The ongoing advances in artificial intelligence and machine learning also pose a potential asset to the advancement of mortality prediction. Artificial intelligence algorithms have been developed for application of electrocardiogram interpretation with promising accuracy and clinical application. Additionally, artificial intelligence algorithms applied to electrocardiogram interpretation have been developed to predict various variables such as structural heart disease, left ventricular systolic dysfunction, and atrial fibrillation. These variables can be used and applied to new mortality prediction models that are dynamic with the changes in the patient's clinical course and may lead to more accurate and reliable mortality prediction. The application of artificial intelligence to mortality prediction will fill the gaps left by current mortality prediction tools.

Early risk stratification in patients with cardiogenic shock irrespective of the underlying cause - The Cardiogenic Shock Score (CSS)

AIMS

Early risk stratification is essential to guide treatment in cardiogenic shock (CS). Existing CS risk scores were derived in selected cohorts, without accounting for the heterogeneity of CS. The aim of this study was to develop a universal risk score (CSS) for all CS patients, irrespective of underlying cause.

METHODS AND RESULTS

Within a registry of 1,308 CS unselected patients admitted to a tertiary-care hospital between 2009 and 2019, a Cox regression model was fitted to derive the CSS, with 30-day mortality as main outcome. CSS's predictive ability was compared to the IABP-Shock-II score, the CardShock score and SCAI classification by C-indices and validated in an external cohort of 934 CS patients. Based on the Cox regression, 9 predictors were included in the CSS: age, sex, acute myocardial infarction (AMI-CS), systolic blood pressure, heart rate, pH, lactate, glucose and cardiac arrest. CSS had the highest C-index in the overall cohort (0.740 vs. 0.677/0.683 for IABP-Shock-II score/CardShock score), in patients with AMI-CS (0.738 vs. 0.675/0.689 for IABP-Shock-II score/CardShock score) and in patients with non-AMI-CS (0.734 vs. 0.677/0.669 for IABP-Shock-II score/CardShock score). In the external validation cohort, the CSS had a C-index of 0.73, which was higher than all other tested scores.

CONCLUSION

The CSS provides improved information on the risk of death in unselected patients with CS compared to existing scores, irrespective of its cause. Because it is based on point-of-care variables which can be obtained even in critical situations, the CSS has the potential to guide treatment decisions in CS. This article is protected by copyright. All rights reserved.