When a patient experiences chest pain and is rushed to the hospital, an electrocardiogram (ECG) is typically the first diagnostic tool used to confirm or rule out a heart attack. In an ECG, healthcare providers can often identify a specific pattern indicative of a severe type of heart attack known as STEMI, caused by a complete coronary artery blockage requiring immediate intervention. However, nearly two-thirds of heart attacks, even those caused by severe blockages, fail to present this distinct ECG pattern. A new tool is now available to detect subtle ECG indicators that clinicians might find challenging to identify, making it easier for them to classify chest pain patients.

Researchers at University of Pittsburgh (Pittsburgh, PA, USA) have developed a machine learning model that utilizes ECG data for faster, more accurate heart attack diagnosis and classification compared to existing methodologies. The model was developed using ECGs from 4,026 chest pain patients from three Pittsburgh hospitals, and later externally validated with 3,287 patients from a separate hospital system. The researchers compared their model to three established cardiac event assessment methods: experienced clinician ECG interpretation, commercial ECG algorithms, and the HEART score, a scoring system that considers patient history at presentation, ECG interpretation, age, risk factors like smoking, diabetes, and high cholesterol, and troponin protein levels in the blood. The model beat all three methods, accurately reclassifying one in every three chest pain patients as low, intermediate, or high risk.

According to the researchers, the algorithm can aid EMS personnel and emergency department providers in identifying heart attack patients and those experiencing reduced blood flow to the heart, compared to traditional ECG analysis. Such insights can inform EMS medical decisions, like initiating field treatments, notifying hospitals about incoming high-risk patients, or determining which low-risk patients need not be transported to a specialized cardiac facility. This could greatly enhance pre-hospital triage. In the next research phase, the team is focusing on optimizing the model's deployment and developing a cloud-based system that interfaces with hospital command centers receiving EMS ECG readings. The model will analyze the ECG and provide a patient risk assessment, thereby guiding medical decisions in real time.

“When a patient comes into the hospital with chest pain, the first question we ask is whether the patient is having a heart attack or not. It seems like that should be straightforward, but when it’s not clear from the ECG, it can take up to 24 hours to complete additional tests,” said lead author Salah Al-Zaiti, Ph.D., R.N., associate professor in the Pitt School of Nursing and of emergency medicine and cardiology in the School of Medicine. “Our model helps address this major challenge by improving risk assessment so that patients can get appropriate care without delay.” 

http://www.gzjiayumed.com/en/index.asp