White matter hyperintensity (WMH) refers to white matter lesions with a signal intensity brighter than the surrounding white matter on the fluid attenuation inversion recovery (FLAIR) sequence of brain magnetic resonance imaging (MRI) [1]. WMH usually occurs with microvascular ischemic changes in the cerebrovascular vessels [2]. Although WMH is commonly found in older adults, the clinical manifestations of WMH vary from asymptomatic to cognitive impairment, dementia, or cerebrovascular accident (CVA), including ischemic stroke and transient ischemic attack [3]. Therefore, early detection and management of cerebrovascular diseases should be emphasized. Although brain MRI is the most precise and accurate non-invasive method for investigating cerebral microvascular diseases, including WMH, it is time-consuming and expensive. Hence, it is not routinely performed in clinical practice [4]. Much effort has been made to develop cost-effective and simple surrogate markers for the prediction of cerebral microvascular diseases [5].

Chronic inflammation and insulin resistance worsen the lipid profile, which contributes to atherosclerotic complications [6]. Recently, the atherogenic index of plasma (AIP), the base 10 logarithm of the ratio of triglycerides to high-density lipoprotein (HDL) cholesterol, has been developed [7]. It has been recognized as a useful predictive marker for plasma dyslipidemia and its related diseases, such as CVA, coronary heart disease, hypertension (HTN), diabetes mellitus (DM), metabolic syndrome, chronic kidney disease, and hyperuricemia [8,9]. Despite its correlation with atherosclerotic systemic diseases, there is limited evidence supporting the relationship between AIP and cerebral microvascular changes, especially in the Korean population. Therefore, we aimed to verify the association between AIP and WMH in a sample of Korean adults without CVA.

We found that AIP was independently and positively associated with WMH in an apparently healthy population. This relationship persisted even after adjusting for confounding factors. Since small and dense LDL cholesterol is a major factor in the development of atherosclerosis, and is technically less convenient to measure, AIP is a potentially cost-effective alternative surrogate marker for the assessment of WMH.

AIP is independently associated with the occurrence of ischemic stroke [8,15]. Wang et al. [8] demonstrated a significant and positive association between AIP and ischemic stroke. The prevalence of ischemic stroke showed a substantially increasing trend with increasing AIP quartiles. Woo et al. [15] reported that elevated triglyceride levels and reduced the HDL level ratio (THR), and may be associated with intracranial atherosclerosis and a higher risk of early recurrent stroke events. However, their results did not demonstrate a significant association between THR and WMH. Most previous studies were dedicated to confirming the relationship between AIP and ischemic stroke [8,15]. Only a few studies investigated the relationship between AIP and WMH. Compared with previous studies, our study was based on the results of a relatively healthy population without CVA from screening examinations. Considering that WMH is a risk factor for incident ischemic stroke, our findings that AIP was related to WMH even in individuals without CVA support the results of previous studies [16,17].

While we could not explain the exact mechanism by which AIP is related to WMH, possible explanations include insulin resistance and chronic inflammation [16]. Higher serum triglyceride levels reduce insulin sensitivity and islet beta-cell function via superoxide dismutase [18]. Moreover, hypertriglyceridemia and low HDL cholesterolemia are usually accompanied by insulin resistance because insulin affects very-low-density lipoprotein (VLDL) and HDL cholesterol metabolism. Insulin resistance plays an important role in VLDL metabolism and increased hepatic triglyceride synthesis [19]. Increased VLDL triglyceride synthesis is linked to increased hepatic apolipoprotein (Apo) B-100 production [20]. This metabolism results in hypertriglyceridemia, which is reflected by increase in VLDL, Apo B-100, and decrease in HDL concentrations [21]. Park et al. [22] reported that triglycerides and LDL cholesterol correlated with the homeostasis model assessment of insulin resistance and beta-cell function, and thus, is a useful marker for predicting insulin resistance. Therefore, insulin resistance may affect the association between AIP and WMH.

Chronic inflammation is a major pathophysiological factor affecting metabolic syndrome and insulin resistance. Elevated triglyceride levels contribute to triglyceride-rich lipoproteins. Apo C-III, as a potent modular of plasma triglyceride levels, affects the pathway that activates necrosis factor-beta and increases inflammatory components such as vascular cell adhesion molecules (VCAM1), leading to the development of fatty streaks and advanced atherosclerosis [23]. Apo C-III is a glycoprotein synthesized by the liver and intestines. This glycoprotein affects the increase in triglyceride levels and stimulation of VLDL synthesis. VCAM1 is a membrane-bound adhesion receptor that plays a critical role in leukocyte accumulation [24]. The link between chronic inflammation and insulin resistance could be a possible explanation for the relationship between AIP and WHM.

The relationship between triglyceride levels and blood pressure should also be considered. Triglycerides are regarded as an independent risk factor for the early development of endothelial dysfunction [25]. As endothelial dysfunction and HTN have a bidirectional relationship, an increased serum triglyceride level may contribute to the elevation of blood pressure. Moreover, the relationship between triglycerides and blood pressure exists among participants with lower levels of circulating CD34-positive cells which have a beneficial role in endothelial repair [26].

HDL cholesterol exhibits anti-atherosclerotic effects by transporting cholesterol from peripheral vessels to the liver [27]. HDL cholesterol is also antioxidative. This is mediated by inhibiting the oxidation of LDL cholesterol, which results in the reduced cellular uptake of oxidized LDL cholesterol by the monocyte macrophage system [28]. In addition, HDL provides a pathway for detoxification or excretion of lipid peroxidation products, derived from oxidized LDL cholesterol, by accepting hydroperoxides from oxidized membranes [29]. In contrast, hypertriglyceridemia and low HDL cholesterolemia contribute to the development of WMH.

Our study had several limitations. First, we could not verify the causal relationship between AIP and WMH because of the retrospective cross-sectional design of this study. Further cohort studies are needed to elucidate the causal relationship between AIP and the incidence of WMH. Second, specific information regarding the medications used by patients with HTN, DM, or dyslipidemia was not evaluated. However, we adjusted for the major risk factors of atherosclerosis to minimize information bias. Third, the results could not be generalized because we analyzed only Korean data. As the Korean population has a higher prevalence of ischemic stroke than Westerner populations [30], the positive relationship between AIP and WMH may be weaker in other ethnicities. Finally, the relatively small sample size of this study may limit its generalizability. Future large-scale, multi-ethnic studies to confirm the association between AIP and WMH in the general population are warranted.

In conclusion, our analyses showed that AIP was independently and positively associated with WMH in an apparently healthy population. AIP could be a cost-effective method for screening and identifying individuals with WMH who require early intervention.