Cardiovascular disease (CVD) is the leading cause of global mortality and a major contributor to disability, and myocardial infarction (MI) is the main cause of CVD [1]. Although CVD mortality has declined in high-income areas, it is still very elevated in developing countries [2]. Thus, early identification of people at high risk of MI is crucial.

Disorders of lipoprotein metabolism, called dyslipidemia, are one of the most important risk factors for MI [3]. Elevated levels of total cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), and a high ratio of LDL-C to high-density lipoprotein cholesterol (HDL-C) increase the risk of MI [4], [5], [6]. The atherogenic index of plasma (AIP) is calculated according to the formula log (TG/HDL) and reflects levels of TG and HDL-C cholesterol, and AIP is considered a potential biomarker of MI [7], [8], [9], [10]. One cross-sectional study reported that AIP is a potential biomarker in the early diagnosis of MI [8]. A hospital-based observational study suggested that AIP could predict the size of lipoprotein particles, showing a positive correlation with the risk of MI [9]. Additionally, AIP has been observed to have a J-shaped relationship with new-onset MI in hypertensive patients with obstructive sleep apnea [11]. AIP is a potential biomarker for CVD in schizophrenia [12] and is positively correlated with arterial stiffness in patients with essential hypertension [13]. However, previous studies have either focused on specific individuals or included a small study sample, thus lacking evidence based on a large sample in the wider population.

Most previous studies have investigated the effect of AIP on clinical outcomes in MI using AIP at one time point [10], [14], [15], [16], [17], [18], [19]. This approach has limited power to predict outcomes because changes in many biological and environmental factors during long-term follow-up could have an impact on lipid levels, and a single measurement of AIP may lead to incorrect classification of risk assessment for MI. Previous studies have explored the long-term effects of exposure and their association with various outcomes, such as changes in glomerular filtration rate or visit-to-visit variability in glycosylated hemoglobin, which were linked to the risk of CVD [20], [21]. Then AIP measured at multiple time points can better characterize the longitudinal pattern of AIP, thereby providing a more robust assessment of the associations with outcomes than a single AIP measurement. Therefore, the purpose of our study was to explore whether elevated AIP at baseline and in the long-term, including updated mean AIP and the number of visits with high AIP levels, were associated with a higher risk of MI in a large community-based prospective cohort study.