Association of Vitamin D Deficiency with ST-Elevation Myocardial Infarction: Evidence from Statistical Modeling and Machine Learning Classification

Heart attacks are not a single event but a spectrum of emergencies, ranging from partial blockages of the heart's arteries to complete, sudden stoppages. When a coronary artery is fully blocked, the heart muscle begins to die within minutes, a condition doctors call an ST-elevation myocardial infarction, or STEMI. This is distinct from other types of heart attacks where the blockage is less severe or the heart is still receiving some blood flow. Understanding why some people suffer the most severe form of heart attack while others experience a milder version is a central challenge in modern medicine. One factor that has recently come under intense scrutiny is vitamin D, a nutrient most people know only as a helper for strong bones. In reality, this vitamin acts as a powerful regulator throughout the body, influencing how blood vessels function, how the immune system responds to injury, and how the heart muscle handles stress. When the body lacks this nutrient, the delicate balance that keeps arteries clear and blood flowing smoothly can be disrupted, potentially setting the stage for more dangerous cardiac events.

Researchers in Peshawar, Pakistan, set out to investigate whether a lack of vitamin D could help explain why some patients arrive at the hospital with the most severe type of heart attack while others have a less critical condition. They focused their study on two groups of patients: fifty individuals who had suffered a STEMI and fifty who had experienced a less severe type of acute coronary syndrome, known as NSTEMI or unstable angina. These patients were treated at the Hayatabad Medical Complex between late 2024 and early 2025. The team did not rely on guesswork; they measured specific details about each patient, including their age, blood pressure, heart rate, and levels of various chemicals in their blood, such as cholesterol, sugar, and the vitamin D itself. To make sense of this complex web of data, the scientists used two different approaches. First, they applied traditional statistical methods to find simple links between the patients' conditions and their blood test results. Then, they employed advanced computer programs, known as machine learning, which are designed to find hidden patterns in large sets of information that human eyes might miss.

The investigation revealed a clear and significant difference between the two groups of heart attack patients. While factors like age, heart rate, and the presence of diabetes or high blood pressure were similar in both groups, two specific measurements stood out. The patients who suffered the severe STEMI heart attacks had significantly lower levels of oxygen in their blood and much lower levels of high-density lipoprotein, often called "good" cholesterol, compared to those with the milder heart conditions. More importantly, the study found that vitamin D deficiency was far more common among the patients with the severe STEMI heart attacks. When the researchers used a computer model to screen for the most important predictors of which type of heart attack a patient would have, vitamin D levels emerged as a top factor, alongside the levels of good cholesterol and oxygen saturation. This suggests that a lack of vitamin D is not just a background issue but a key player in determining the severity of the cardiac event.

To see if these findings could be used to predict heart attack types in the future, the team trained several machine learning models to act as classifiers. These digital tools were fed the patient data and asked to guess whether a new patient would have a severe STEMI or a milder condition. The results showed that these computer models could distinguish between the two types of heart attacks with moderate success. One of the most effective models, known as a Random Forest, demonstrated a strong ability to differentiate between the two groups, achieving an area under the curve (AUC) of 0.792. This metric indicates that the model has a high capacity to rank a randomly selected STEMI case higher than a randomly selected NSTEMI/USAP case, significantly exceeding random chance. Another model, using a different mathematical approach, achieved an accuracy of about 67 percent. While these numbers indicate that the models are not yet perfect and cannot replace a doctor's diagnosis, they demonstrate that combining clinical data with computer analysis can reveal useful patterns. The computer also confirmed what the human researchers found: vitamin D levels, along with blood pressure and the presence of diabetes, were among the most influential factors in making these predictions.

The study does not claim that low vitamin D causes heart attacks on its own, nor does it suggest that taking supplements will instantly cure the condition. Instead, the findings point to a strong association, where patients with the most severe heart attacks are significantly more likely to be deficient in this vital nutrient. The researchers noted that their study had limitations, including a relatively small number of participants and the fact that it was conducted in a single hospital, which means the results might look different in other populations. They also could not track the patients over a long period to see how their conditions changed. However, the consistency of the data—showing that vitamin D deficiency, low good cholesterol, and low oxygen levels cluster together in the most severe cases—provides a compelling reason for further investigation. By identifying these specific biological markers, the study offers a clearer picture of the complex factors that drive the severity of heart disease, suggesting that nutritional status may be a critical piece of the puzzle in understanding why some heart attacks are far more dangerous than others.