Background Conventional risk score for predicting short and long-term mortality following an ST-segment elevation myocardial infarction (STEMI) is often not population specific. Objective Apply machine learning for the prediction and identification of factors associated with short and long-term mortality in Asian STEMI patients and compare with a conventional risk score. Methods The National Cardiovascular Disease Database for Malaysia registry, of a multi-ethnic, heterogeneous Asian population was used for in-hospital (6299 patients), 30-days (3130 patients), and 1-year (2939 patients) model development. 50 variables were considered. Mortality prediction was analysed using feature selection methods with machine learning algorithms and compared to Thrombolysis in Myocardial Infarction (TIMI) score. Invasive management of varying degrees was selected as important variables that improved mortality prediction. Results Model performance using a complete and reduced variable produced an area under the receiver operating characteristic curve (AUC) from 0.73 to 0.90. The best machine learning model for in-hospital, 30 days, and 1-year outperformed TIMI risk score (AUC = 0.88, 95% CI: 0.846–0.910; vs AUC = 0.81, 95% CI:0.772–0.845, AUC = 0.90, 95% CI: 0.870–0.935; vs AUC = 0.80, 95% CI: 0.746–0.838, AUC = 0.84, 95% CI: 0.798–0.872; vs AUC = 0.76, 95% CI: 0.715–0.802, p < 0.0001 for all). TIMI score underestimates patients’ risk of mortality. 90% of non-survival patients are classified as high risk (>50%) by machine learning algorithm compared to 10–30% non-survival patients by TIMI. Common predictors identified for short- and long-term mortality were age, heart rate, Killip class, fasting blood glucose, prior primary PCI or pharmaco-invasive therapy and diuretics. The final algorithm was converted into an online tool with a database for continuous data archiving for algorithm validation. Conclusions In a multi-ethnic population, patients with STEMI were better classified using the machine learning method compared to TIMI scoring. Machine learning allows for the identification of distinct factors in individual Asian populations for better mortality prediction. Ongoing continuous testing and validation will allow for better risk stratification and potentially alter management and outcomes in the future.
Background: Cardiovascular care in Malaysia adopts a ‘spoke-and-hub’ model, leaving the majority of acute heart failure (AHF) care to non-cardiac centres. Granular data on AHF care are essential and yet remain scarce. Objectives: This study aimed to illustrate the baseline characteristics, management and outcome of AHF patients. Methods: A retrospective, cross-sectional study was conducted on 1307 AHF patients admitted between 1 January 2012 and 31 December 2016. Results: The younger and Malay-predominant population reflects the distribution of ischaemic heart disease in Malaysia, highlighting the need to tackle metabolic risks factors. The majority are precipitated by ischaemia (61.8%). Common co-morbidities include hypertension (70.9%), coronary artery disease (57.8%) and diabetes (62.2%). The majority were of New York Heart Association Class II (31.9%) and Class III (25.6%). A total of 14.5% required inotropes and vasopressors, 12.9% required intravenous nitrates and 8.6% required dialysis. A further 4.9% of patients required intubation and mechanical ventilation, and 25.9% required non-invasive ventilation. Readmission and mortality were extremely high in our population. Short inpatient stays, restricting optimisation of medication, and gaps in the provision of coronary intervention and stress testing are possible contributing factors. When compared to global and regional registries, disparities were noted specifically surrounding mortality rate and optimum use of guideline-directed medical therapy. Conclusion: Although smaller and single centred, our study provides a unique insight into a pure Malaysian-only cohort from a hospital with no cardiology services in-house, which is more reflective of the majority of hospitals in Malaysia, unlike previous studies and registries.
BackgroundFamilial hypercholesterolaemia (FH) is a genetic disorder with a high risk of developing premature coronary artery disease that should be diagnosed as early as possible. Several clinical diagnostic criteria for FH are available, with the Dutch Lipid Clinic Criteria (DLCC) being widely used. Information regarding diagnostic performances of the other criteria against the DLCC is scarce. We aimed to examine the diagnostic performance of the Simon-Broom (SB) Register criteria, the US Make Early Diagnosis to Prevent Early Deaths (US MEDPED) and the Japanese FH Management Criteria (JFHMC) compared to the DLCC.MethodsSeven hundered fifty five individuals from specialist clinics and community health screenings with LDL-c level ≥ 4.0 mmol/L were selected and diagnosed as FH using the DLCC, the SB Register criteria, the US MEDPED and the JFHMC. The sensitivity, specificity, efficiency, positive and negative predictive values of individuals screened with the SB register criteria, US MEDPED and JFHMC were assessed against the DLCC.ResultsWe found the SB register criteria identified more individuals with FH compared to the US MEDPED and the JFHMC (212 vs. 105 vs. 195; p < 0.001) when assessed against the DLCC. The SB Register criteria, the US MEDPED and the JFHMC had low sensitivity (51.1% vs. 25.3% vs. 47.0% respectively). The SB Register criteria showed better diagnostic performance than the other criteria with 98.8% specificity, 28.6% efficiency value, 98.1% and 62.3% for positive and negative predictive values respectively.ConclusionThe SB Register criteria appears to be more useful in identifying positive cases leading to genetic testing compared to the JFHMC and US MEDPED in this Asian population. However, further research looking into a suitable diagnosis criterion with high likelihood of positive genetic findings is required in the Asian population including in Malaysia.
ObjectiveHigh-sensitivity troponin (hs-Tn) assays need to be applied appropriately to improve diagnosis and patient outcomes in acute coronary syndromes (ACS).MethodsExperts from Asia Pacific convened in 2015 to provide data-driven consensus-based, region-specific recommendations and develop an algorithm for the appropriate incorporation of this assay into the ACS assessment and treatment pathway.ResultsNine recommendations were developed by the expert panel: (1) troponin is the preferred cardiac biomarker for diagnostic assessment of ACS and is indicated for patients with symptoms of possible ACS; (2) hs-Tn assays are recommended; (3) serial testing is required for all patients; (4) testing should be performed at presentation and 3 hours later; (5) gender-specific cut-off values should be used for hs-Tn I assays; (6) hs-Tn I level >10 times the upper limit of normal should be considered to ‘rule in’ a diagnosis of ACS; (7) dynamic change >50% in hs-Tn I level from presentation to 3-hour retest identifies patients at high risk for ACS; (8) where only point-of-care testing is available, patients with elevated readings should be considered at high risk, while patients with low/undetectable readings should be retested after 6 hours or sent for laboratory testing and (9) regular education on the appropriate use of troponin tests is essential.ConclusionsWe propose an algorithm that will potentially reduce delays in discharge by the accurate ‘rule out’ of non-ACS patients within 3 hours. Appropriate research should be undertaken to ensure the efficacy and safety of the algorithm in clinical practice, with the long-term goal of improvement of care of patients with ACS in Asia Pacific.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.