Six symptom clusters were identified in patients investigated for ACS, with the anginal cluster most strongly associated with AMI, but no symptom cluster provided sufficient discriminatory power to rule in or rule out AMI, and while some individual symptom differences were observed for Aboriginal and Torres Strait Islander patients, overall cluster patterns were comparable across groups.
Key Findings
Results
Six symptom clusters were identified in patients investigated for ACS: pleuritic, numbness, anginal, palpitations, epigastric, and infectious.
Factor analysis was used to identify the symptom clusters from data collected in the LEGEND trial.
The analysis included 2185 patients, 22.7% of whom identified as Aboriginal and Torres Strait Islander.
Data on presenting symptoms and outcomes were collected from patient medical records.
The six clusters represent distinct groupings of co-occurring symptoms in the ACS investigation population.
Results
The anginal symptom cluster was most strongly associated with acute myocardial infarction.
Risk ratio for the anginal cluster and AMI: RR = 1.46; 95% CI: 1.31–1.62.
This was the strongest positive association among all six identified symptom clusters.
Regression analyses were used to estimate associations between each symptom cluster and AMI.
Results
The pleuritic symptom cluster was associated with a reduced risk of acute myocardial infarction.
Risk ratio for the pleuritic cluster and AMI: RR = 0.47; 95% CI: 0.35–0.62.
This was the only cluster associated with a meaningfully reduced AMI risk.
The confidence interval does not cross 1.0, indicating a statistically significant inverse association.
Results
The numbness and infectious symptom clusters were each associated with an increased risk of AMI.
Numbness cluster: RR = 1.29; 95% CI: 1.11–1.49.
Infectious cluster: RR = 1.21; 95% CI: 1.05–1.41.
Both confidence intervals exclude 1.0, indicating statistically significant positive associations.
These clusters represent less 'classic' presentations of AMI, suggesting atypical symptom patterns can still predict AMI.
Results
No symptom cluster provided sufficient discriminatory power to rule in or rule out AMI.
Despite statistically significant associations, none of the six clusters had adequate performance for clinical decision-making on their own.
The authors note that 'symptom presentation in AMI is heterogeneous, and reliance on classic symptoms alone may miss at-risk patients.'
This finding applies broadly across the full study population.
Results
Some individual symptoms differed in their associations with AMI for Aboriginal and Torres Strait Islander patients compared to non-Indigenous Australians, but cluster-level predictive performance was similar across groups.
Aboriginal and Torres Strait Islander patients comprised 22.7% of the 2185-patient sample (approximately 496 patients).
Regression analyses examined whether individual symptoms predicted AMI separately for Aboriginal and Torres Strait Islander and non-Indigenous patients.
Specific individual symptoms showed different associations by group, though the paper does not detail which symptoms in the abstract.
Overall, 'cluster patterns were comparable' across Indigenous and non-Indigenous groups.
Methods
The study was an analysis of the LEGEND trial, which included a substantial proportion of Aboriginal and Torres Strait Islander participants.
Total sample size: 2185 patients.
22.7% of participants identified as Aboriginal and Torres Strait Islander people.
Data were collected from patient medical records.
The study used both factor analysis for cluster identification and regression analyses for predictive associations.
What This Means
This research suggests that when patients come to the emergency department with possible heart attack symptoms, their symptoms tend to group into six patterns or 'clusters': pleuritic (sharp, breathing-related chest pain), numbness, anginal (classic chest pressure or tightness), palpitations, epigastric (stomach-area discomfort), and infectious (flu-like symptoms). The cluster most strongly linked to a confirmed heart attack was the anginal cluster, while the pleuritic cluster was actually associated with a lower chance of heart attack. Importantly, clusters involving numbness and infectious-type symptoms were also linked to higher heart attack risk, suggesting that non-classic presentations should not be dismissed.
The study also looked specifically at whether Aboriginal and Torres Strait Islander Australians — who made up nearly a quarter of the 2,185 patients studied — experienced different symptom patterns. Some differences were found at the level of individual symptoms, but when symptoms were considered as clusters, the patterns and their ability to predict heart attacks were broadly similar between Aboriginal and Torres Strait Islander and non-Indigenous patients.
A key practical implication of this research is that no single symptom cluster was good enough on its own to confidently confirm or rule out a heart attack. This suggests that healthcare providers should not rely solely on 'classic' heart attack symptoms (like chest pressure radiating to the arm) when assessing patients, as doing so may cause them to miss people who are actually having a heart attack but presenting with less typical symptoms. This is particularly relevant for ensuring equitable care for Aboriginal and Torres Strait Islander patients.